202 files changed, 7441 insertions, 3196 deletions

diff --git a/MAINTAINERS b/MAINTAINERS
index d1d9ae6173b8..1b71f6ceae0a 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -5910,7 +5910,7 @@ M:	Ingo Molnar <mingo@redhat.com>
 M:      Peter Zijlstra <peterz@infradead.org>
 T:      git git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git sched/core
 S:      Maintained
-F:      kernel/sched*
+F:      kernel/sched/
 F:      include/linux/sched.h
 
 SCORE ARCHITECTURE
diff --git a/Makefile b/Makefile
index bf02707a5e9b..aa8e315f26ef 100644
--- a/Makefile
+++ b/Makefile
@@ -1,7 +1,7 @@
 VERSION = 3
 PATCHLEVEL = 5
 SUBLEVEL = 0
-EXTRAVERSION = -rc6
+EXTRAVERSION = -rc7
 NAME = Saber-toothed Squirrel
 
 # *DOCUMENTATION*
diff --git a/arch/arm/mach-exynos/pm_domains.c b/arch/arm/mach-exynos/pm_domains.c
index e9fafcf163de..373c3c00d24c 100644
--- a/arch/arm/mach-exynos/pm_domains.c
+++ b/arch/arm/mach-exynos/pm_domains.c
@@ -119,7 +119,9 @@ static __init void exynos_pm_add_dev_to_genpd(struct platform_device *pdev,
                                                 struct exynos_pm_domain *pd)
 {
         if (pdev->dev.bus) {
-                if (pm_genpd_add_device(&pd->pd, &pdev->dev))
+                if (!pm_genpd_add_device(&pd->pd, &pdev->dev))
+                        pm_genpd_dev_need_restore(&pdev->dev, true);
+                else
                         pr_info("%s: error in adding %s device to %s power"
                                 "domain\n", __func__, dev_name(&pdev->dev),
                                 pd->name);
@@ -151,9 +153,12 @@ static __init int exynos4_pm_init_power_domain(void)
         if (of_have_populated_dt())
                 return exynos_pm_dt_parse_domains();
 
-        for (idx = 0; idx < ARRAY_SIZE(exynos4_pm_domains); idx++)
-                pm_genpd_init(&exynos4_pm_domains[idx]->pd, NULL,
-                                exynos4_pm_domains[idx]->is_off);
+        for (idx = 0; idx < ARRAY_SIZE(exynos4_pm_domains); idx++) {
+                struct exynos_pm_domain *pd = exynos4_pm_domains[idx];
+                int on = __raw_readl(pd->base + 0x4) & S5P_INT_LOCAL_PWR_EN;
+
+                pm_genpd_init(&pd->pd, NULL, !on);
+        }
 
 #ifdef CONFIG_S5P_DEV_FIMD0
         exynos_pm_add_dev_to_genpd(&s5p_device_fimd0, &exynos4_pd_lcd0);
diff --git a/arch/arm/mach-s3c24xx/clock-s3c2440.c b/arch/arm/mach-s3c24xx/clock-s3c2440.c
index 414364eb426c..cb2883d553b5 100644
--- a/arch/arm/mach-s3c24xx/clock-s3c2440.c
+++ b/arch/arm/mach-s3c24xx/clock-s3c2440.c
@@ -106,7 +106,7 @@ static struct clk s3c2440_clk_cam_upll = {
 static struct clk s3c2440_clk_ac97 = {
         .name           = "ac97",
         .enable         = s3c2410_clkcon_enable,
-        .ctrlbit        = S3C2440_CLKCON_CAMERA,
+        .ctrlbit        = S3C2440_CLKCON_AC97,
 };
 
 static unsigned long  s3c2440_fclk_n_getrate(struct clk *clk)
diff --git a/arch/arm/plat-samsung/adc.c b/arch/arm/plat-samsung/adc.c
index 33ecd0c9f0c3..b1e05ccff3ac 100644
--- a/arch/arm/plat-samsung/adc.c
+++ b/arch/arm/plat-samsung/adc.c
@@ -157,11 +157,13 @@ int s3c_adc_start(struct s3c_adc_client *client,
                 return -EINVAL;
         }
 
-        if (client->is_ts && adc->ts_pend)
-                return -EAGAIN;
-
         spin_lock_irqsave(&adc->lock, flags);
 
+        if (client->is_ts && adc->ts_pend) {
+                spin_unlock_irqrestore(&adc->lock, flags);
+                return -EAGAIN;
+        }
+
         client->channel = channel;
         client->nr_samples = nr_samples;
 
diff --git a/arch/arm/plat-samsung/devs.c b/arch/arm/plat-samsung/devs.c
index 1d214cb9d770..6303974c2ee0 100644
--- a/arch/arm/plat-samsung/devs.c
+++ b/arch/arm/plat-samsung/devs.c
@@ -126,7 +126,8 @@ struct platform_device s3c_device_adc = {
 #ifdef CONFIG_CPU_S3C2440
 static struct resource s3c_camif_resource[] = {
         [0] = DEFINE_RES_MEM(S3C2440_PA_CAMIF, S3C2440_SZ_CAMIF),
-        [1] = DEFINE_RES_IRQ(IRQ_CAM),
+        [1] = DEFINE_RES_IRQ(IRQ_S3C2440_CAM_C),
+        [2] = DEFINE_RES_IRQ(IRQ_S3C2440_CAM_P),
 };
 
 struct platform_device s3c_device_camif = {
diff --git a/arch/arm/plat-samsung/s5p-clock.c b/arch/arm/plat-samsung/s5p-clock.c
index 031a61899bef..48a159911037 100644
--- a/arch/arm/plat-samsung/s5p-clock.c
+++ b/arch/arm/plat-samsung/s5p-clock.c
@@ -37,6 +37,7 @@ struct clk clk_ext_xtal_mux = {
 struct clk clk_xusbxti = {
         .name           = "xusbxti",
         .id             = -1,
+        .rate           = 24000000,
 };
 
 struct clk s5p_clk_27m = {
diff --git a/arch/um/drivers/mconsole_kern.c b/arch/um/drivers/mconsole_kern.c
index 88e466b159dc..43b39d61b538 100644
--- a/arch/um/drivers/mconsole_kern.c
+++ b/arch/um/drivers/mconsole_kern.c
@@ -705,7 +705,6 @@ static void stack_proc(void *arg)
         struct task_struct *from = current, *to = arg;
 
         to->thread.saved_task = from;
-        rcu_switch_from(from);
         switch_to(from, to, from);
 }
 
diff --git a/arch/x86/kernel/vsyscall_64.c b/arch/x86/kernel/vsyscall_64.c
index 7515cf0e1805..5db36caf4289 100644
--- a/arch/x86/kernel/vsyscall_64.c
+++ b/arch/x86/kernel/vsyscall_64.c
@@ -139,6 +139,19 @@ static int addr_to_vsyscall_nr(unsigned long addr)
         return nr;
 }
 
+#ifdef CONFIG_SECCOMP
+static int vsyscall_seccomp(struct task_struct *tsk, int syscall_nr)
+{
+        if (!seccomp_mode(&tsk->seccomp))
+                return 0;
+        task_pt_regs(tsk)->orig_ax = syscall_nr;
+        task_pt_regs(tsk)->ax = syscall_nr;
+        return __secure_computing(syscall_nr);
+}
+#else
+#define vsyscall_seccomp(_tsk, _nr) 0
+#endif
+
 static bool write_ok_or_segv(unsigned long ptr, size_t size)
 {
         /*
@@ -174,6 +187,7 @@ bool emulate_vsyscall(struct pt_regs *regs, unsigned long address)
         int vsyscall_nr;
         int prev_sig_on_uaccess_error;
         long ret;
+        int skip;
 
         /*
          * No point in checking CS -- the only way to get here is a user mode
@@ -205,9 +219,6 @@ bool emulate_vsyscall(struct pt_regs *regs, unsigned long address)
         }
 
         tsk = current;
-        if (seccomp_mode(&tsk->seccomp))
-                do_exit(SIGKILL);
-
         /*
          * With a real vsyscall, page faults cause SIGSEGV.  We want to
          * preserve that behavior to make writing exploits harder.
@@ -222,8 +233,13 @@ bool emulate_vsyscall(struct pt_regs *regs, unsigned long address)
          * address 0".
          */
         ret = -EFAULT;
+        skip = 0;
         switch (vsyscall_nr) {
         case 0:
+                skip = vsyscall_seccomp(tsk, __NR_gettimeofday);
+                if (skip)
+                        break;
+
                 if (!write_ok_or_segv(regs->di, sizeof(struct timeval)) ||
                     !write_ok_or_segv(regs->si, sizeof(struct timezone)))
                         break;
@@ -234,6 +250,10 @@ bool emulate_vsyscall(struct pt_regs *regs, unsigned long address)
                 break;
 
         case 1:
+                skip = vsyscall_seccomp(tsk, __NR_time);
+                if (skip)
+                        break;
+
                 if (!write_ok_or_segv(regs->di, sizeof(time_t)))
                         break;
 
@@ -241,6 +261,10 @@ bool emulate_vsyscall(struct pt_regs *regs, unsigned long address)
                 break;
 
         case 2:
+                skip = vsyscall_seccomp(tsk, __NR_getcpu);
+                if (skip)
+                        break;
+
                 if (!write_ok_or_segv(regs->di, sizeof(unsigned)) ||
                     !write_ok_or_segv(regs->si, sizeof(unsigned)))
                         break;
@@ -253,6 +277,12 @@ bool emulate_vsyscall(struct pt_regs *regs, unsigned long address)
 
         current_thread_info()->sig_on_uaccess_error = prev_sig_on_uaccess_error;
 
+        if (skip) {
+                if ((long)regs->ax <= 0L) /* seccomp errno emulation */
+                        goto do_ret;
+                goto done; /* seccomp trace/trap */
+        }
+
         if (ret == -EFAULT) {
                 /* Bad news -- userspace fed a bad pointer to a vsyscall. */
                 warn_bad_vsyscall(KERN_INFO, regs,
@@ -271,10 +301,11 @@ bool emulate_vsyscall(struct pt_regs *regs, unsigned long address)
 
         regs->ax = ret;
 
+do_ret:
         /* Emulate a ret instruction. */
         regs->ip = caller;
         regs->sp += 8;
-
+done:
         return true;
 
 sigsegv:
diff --git a/drivers/acpi/acpica/nspredef.c b/drivers/acpi/acpica/nspredef.c
index 23ce09686418..fe6626035495 100644
--- a/drivers/acpi/acpica/nspredef.c
+++ b/drivers/acpi/acpica/nspredef.c
@@ -638,7 +638,7 @@ acpi_ns_check_package(struct acpi_predefined_data *data,
                         /* Create the new outer package and populate it */
 
                         status =
-                            acpi_ns_wrap_with_package(data, *elements,
+                            acpi_ns_wrap_with_package(data, return_object,
                                                       return_object_ptr);
                         if (ACPI_FAILURE(status)) {
                                 return (status);
diff --git a/drivers/acpi/processor_core.c b/drivers/acpi/processor_core.c
index c850de4c9a14..eff722278ff5 100644
--- a/drivers/acpi/processor_core.c
+++ b/drivers/acpi/processor_core.c
@@ -189,10 +189,12 @@ int acpi_get_cpuid(acpi_handle handle, int type, u32 acpi_id)
                  *     Processor (CPU3, 0x03, 0x00000410, 0x06) {}
                  * }
                  *
-                 * Ignores apic_id and always return 0 for CPU0's handle.
+                 * Ignores apic_id and always returns 0 for the processor
+                 * handle with acpi id 0 if nr_cpu_ids is 1.
+                 * This should be the case if SMP tables are not found.
                  * Return -1 for other CPU's handle.
                  */
-                if (acpi_id == 0)
+                if (nr_cpu_ids <= 1 && acpi_id == 0)
                         return acpi_id;
                 else
                         return apic_id;
diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index bbca966f8f66..3bba65510d23 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -1597,14 +1597,12 @@ static int loop_add(struct loop_device **l, int i)
         struct gendisk *disk;
         int err;
 
+        err = -ENOMEM;
         lo = kzalloc(sizeof(*lo), GFP_KERNEL);
-        if (!lo) {
-                err = -ENOMEM;
+        if (!lo)
                 goto out;
-        }
 
-        err = idr_pre_get(&loop_index_idr, GFP_KERNEL);
-        if (err < 0)
+        if (!idr_pre_get(&loop_index_idr, GFP_KERNEL))
                 goto out_free_dev;
 
         if (i >= 0) {
diff --git a/drivers/char/agp/intel-agp.c b/drivers/char/agp/intel-agp.c
index 0a4185279417..b130df0a1958 100644
--- a/drivers/char/agp/intel-agp.c
+++ b/drivers/char/agp/intel-agp.c
@@ -12,6 +12,7 @@
 #include <asm/smp.h>
 #include "agp.h"
 #include "intel-agp.h"
+#include <drm/intel-gtt.h>
 
 int intel_agp_enabled;
 EXPORT_SYMBOL(intel_agp_enabled);
@@ -747,7 +748,7 @@ static int __devinit agp_intel_probe(struct pci_dev *pdev,
 
         bridge->capndx = cap_ptr;
 
-        if (intel_gmch_probe(pdev, bridge))
+        if (intel_gmch_probe(pdev, NULL, bridge))
                 goto found_gmch;
 
         for (i = 0; intel_agp_chipsets[i].name != NULL; i++) {
@@ -824,7 +825,7 @@ static void __devexit agp_intel_remove(struct pci_dev *pdev)
 
         agp_remove_bridge(bridge);
 
-        intel_gmch_remove(pdev);
+        intel_gmch_remove();
 
         agp_put_bridge(bridge);
 }
@@ -902,17 +903,6 @@ static struct pci_device_id agp_intel_pci_table[] = {
         ID(PCI_DEVICE_ID_INTEL_IRONLAKE_M_HB),
         ID(PCI_DEVICE_ID_INTEL_IRONLAKE_MA_HB),
         ID(PCI_DEVICE_ID_INTEL_IRONLAKE_MC2_HB),
-        ID(PCI_DEVICE_ID_INTEL_SANDYBRIDGE_HB),
-        ID(PCI_DEVICE_ID_INTEL_SANDYBRIDGE_M_HB),
-        ID(PCI_DEVICE_ID_INTEL_SANDYBRIDGE_S_HB),
-        ID(PCI_DEVICE_ID_INTEL_IVYBRIDGE_HB),
-        ID(PCI_DEVICE_ID_INTEL_IVYBRIDGE_M_HB),
-        ID(PCI_DEVICE_ID_INTEL_IVYBRIDGE_S_HB),
-        ID(PCI_DEVICE_ID_INTEL_VALLEYVIEW_HB),
-        ID(PCI_DEVICE_ID_INTEL_HASWELL_HB),
-        ID(PCI_DEVICE_ID_INTEL_HASWELL_M_HB),
-        ID(PCI_DEVICE_ID_INTEL_HASWELL_S_HB),
-        ID(PCI_DEVICE_ID_INTEL_HASWELL_E_HB),
         { }
 };
 
diff --git a/drivers/char/agp/intel-agp.h b/drivers/char/agp/intel-agp.h
index 8e2d9140f300..57226424690c 100644
--- a/drivers/char/agp/intel-agp.h
+++ b/drivers/char/agp/intel-agp.h
@@ -251,7 +251,4 @@
 #define PCI_DEVICE_ID_INTEL_HASWELL_SDV         0x0c16 /* SDV */
 #define PCI_DEVICE_ID_INTEL_HASWELL_E_HB                        0x0c04
 
-int intel_gmch_probe(struct pci_dev *pdev,
-                               struct agp_bridge_data *bridge);
-void intel_gmch_remove(struct pci_dev *pdev);
 #endif
diff --git a/drivers/char/agp/intel-gtt.c b/drivers/char/agp/intel-gtt.c
index 1237e7575c3f..9ed92ef5829b 100644
--- a/drivers/char/agp/intel-gtt.c
+++ b/drivers/char/agp/intel-gtt.c
@@ -66,7 +66,6 @@ static struct _intel_private {
         struct pci_dev *bridge_dev;
         u8 __iomem *registers;
         phys_addr_t gtt_bus_addr;
-        phys_addr_t gma_bus_addr;
         u32 PGETBL_save;
         u32 __iomem *gtt;               /* I915G */
         bool clear_fake_agp; /* on first access via agp, fill with scratch */
@@ -76,6 +75,7 @@ static struct _intel_private {
         struct resource ifp_resource;
         int resource_valid;
         struct page *scratch_page;
+        int refcount;
 } intel_private;
 
 #define INTEL_GTT_GEN   intel_private.driver->gen
@@ -648,6 +648,7 @@ static void intel_gtt_cleanup(void)
 
 static int intel_gtt_init(void)
 {
+        u32 gma_addr;
         u32 gtt_map_size;
         int ret;
 
@@ -694,6 +695,15 @@ static int intel_gtt_init(void)
                 return ret;
         }
 
+        if (INTEL_GTT_GEN <= 2)
+                pci_read_config_dword(intel_private.pcidev, I810_GMADDR,
+                                      &gma_addr);
+        else
+                pci_read_config_dword(intel_private.pcidev, I915_GMADDR,
+                                      &gma_addr);
+
+        intel_private.base.gma_bus_addr = (gma_addr & PCI_BASE_ADDRESS_MEM_MASK);
+
         return 0;
 }
 
@@ -767,20 +777,10 @@ static void i830_write_entry(dma_addr_t addr, unsigned int entry,
         writel(addr | pte_flags, intel_private.gtt + entry);
 }
 
-static bool intel_enable_gtt(void)
+bool intel_enable_gtt(void)
 {
-        u32 gma_addr;
         u8 __iomem *reg;
 
-        if (INTEL_GTT_GEN <= 2)
-                pci_read_config_dword(intel_private.pcidev, I810_GMADDR,
-                                      &gma_addr);
-        else
-                pci_read_config_dword(intel_private.pcidev, I915_GMADDR,
-                                      &gma_addr);
-
-        intel_private.gma_bus_addr = (gma_addr & PCI_BASE_ADDRESS_MEM_MASK);
-
         if (INTEL_GTT_GEN >= 6)
             return true;
 
@@ -823,6 +823,7 @@ static bool intel_enable_gtt(void)
 
         return true;
 }
+EXPORT_SYMBOL(intel_enable_gtt);
 
 static int i830_setup(void)
 {
@@ -860,7 +861,7 @@ static int intel_fake_agp_configure(void)
             return -EIO;
 
         intel_private.clear_fake_agp = true;
-        agp_bridge->gart_bus_addr = intel_private.gma_bus_addr;
+        agp_bridge->gart_bus_addr = intel_private.base.gma_bus_addr;
 
         return 0;
 }
@@ -1182,9 +1183,17 @@ static void gen6_write_entry(dma_addr_t addr, unsigned int entry,
 static void valleyview_write_entry(dma_addr_t addr, unsigned int entry,
                                    unsigned int flags)
 {
+        unsigned int type_mask = flags & ~AGP_USER_CACHED_MEMORY_GFDT;
+        unsigned int gfdt = flags & AGP_USER_CACHED_MEMORY_GFDT;
         u32 pte_flags;
 
-        pte_flags = GEN6_PTE_UNCACHED | I810_PTE_VALID;
+        if (type_mask == AGP_USER_MEMORY)
+                pte_flags = GEN6_PTE_UNCACHED | I810_PTE_VALID;
+        else {
+                pte_flags = GEN6_PTE_LLC | I810_PTE_VALID;
+                if (gfdt)
+                        pte_flags |= GEN6_PTE_GFDT;
+        }
 
         /* gen6 has bit11-4 for physical addr bit39-32 */
         addr |= (addr >> 28) & 0xff0;
@@ -1244,6 +1253,7 @@ static int i9xx_setup(void)
                 switch (INTEL_GTT_GEN) {
                 case 5:
                 case 6:
+                case 7:
                         gtt_offset = MB(2);
                         break;
                 case 4:
@@ -1379,7 +1389,6 @@ static const struct intel_gtt_driver valleyview_gtt_driver = {
         .write_entry = valleyview_write_entry,
         .dma_mask_size = 40,
         .check_flags = gen6_check_flags,
-        .chipset_flush = i9xx_chipset_flush,
 };
 
 /* Table to describe Intel GMCH and AGP/PCIE GART drivers.  At least one of
@@ -1523,14 +1532,32 @@ static int find_gmch(u16 device)
         return 1;
 }
 
-int intel_gmch_probe(struct pci_dev *pdev,
-                                      struct agp_bridge_data *bridge)
+int intel_gmch_probe(struct pci_dev *bridge_pdev, struct pci_dev *gpu_pdev,
+                     struct agp_bridge_data *bridge)
 {
         int i, mask;
-        intel_private.driver = NULL;
+
+        /*
+         * Can be called from the fake agp driver but also directly from
+         * drm/i915.ko. Hence we need to check whether everything is set up
+         * already.
+         */
+        if (intel_private.driver) {
+                intel_private.refcount++;
+                return 1;
+        }
 
         for (i = 0; intel_gtt_chipsets[i].name != NULL; i++) {
-                if (find_gmch(intel_gtt_chipsets[i].gmch_chip_id)) {
+                if (gpu_pdev) {
+                        if (gpu_pdev->device ==
+                            intel_gtt_chipsets[i].gmch_chip_id) {
+                                intel_private.pcidev = pci_dev_get(gpu_pdev);
+                                intel_private.driver =
+                                        intel_gtt_chipsets[i].gtt_driver;
+
+                                break;
+                        }
+                } else if (find_gmch(intel_gtt_chipsets[i].gmch_chip_id)) {
                         intel_private.driver =
                                 intel_gtt_chipsets[i].gtt_driver;
                         break;
@@ -1540,13 +1567,17 @@ int intel_gmch_probe(struct pci_dev *pdev,
         if (!intel_private.driver)
                 return 0;
 
-        bridge->driver = &intel_fake_agp_driver;
-        bridge->dev_private_data = &intel_private;
-        bridge->dev = pdev;
+        intel_private.refcount++;
+
+        if (bridge) {
+                bridge->driver = &intel_fake_agp_driver;
+                bridge->dev_private_data = &intel_private;
+                bridge->dev = bridge_pdev;
+        }
 
-        intel_private.bridge_dev = pci_dev_get(pdev);
+        intel_private.bridge_dev = pci_dev_get(bridge_pdev);
 
-        dev_info(&pdev->dev, "Intel %s Chipset\n", intel_gtt_chipsets[i].name);
+        dev_info(&bridge_pdev->dev, "Intel %s Chipset\n", intel_gtt_chipsets[i].name);
 
         mask = intel_private.driver->dma_mask_size;
         if (pci_set_dma_mask(intel_private.pcidev, DMA_BIT_MASK(mask)))
@@ -1556,11 +1587,11 @@ int intel_gmch_probe(struct pci_dev *pdev,
                 pci_set_consistent_dma_mask(intel_private.pcidev,
                                             DMA_BIT_MASK(mask));
 
-        /*if (bridge->driver == &intel_810_driver)
-                return 1;*/
+        if (intel_gtt_init() != 0) {
+                intel_gmch_remove();
 
-        if (intel_gtt_init() != 0)
                 return 0;
+        }
 
         return 1;
 }
@@ -1579,12 +1610,16 @@ void intel_gtt_chipset_flush(void)
 }
 EXPORT_SYMBOL(intel_gtt_chipset_flush);
 
-void intel_gmch_remove(struct pci_dev *pdev)
+void intel_gmch_remove(void)
 {
+        if (--intel_private.refcount)
+                return;
+
         if (intel_private.pcidev)
                 pci_dev_put(intel_private.pcidev);
         if (intel_private.bridge_dev)
                 pci_dev_put(intel_private.bridge_dev);
+        intel_private.driver = NULL;
 }
 EXPORT_SYMBOL(intel_gmch_remove);
 
diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c
index 65f9d231af14..7282c081fb53 100644
--- a/drivers/gpu/drm/ast/ast_mode.c
+++ b/drivers/gpu/drm/ast/ast_mode.c
@@ -460,8 +460,8 @@ static void ast_crtc_dpms(struct drm_crtc *crtc, int mode)
 }
 
 static bool ast_crtc_mode_fixup(struct drm_crtc *crtc,
-                                  struct drm_display_mode *mode,
-                                  struct drm_display_mode *adjusted_mode)
+                                const struct drm_display_mode *mode,
+                                struct drm_display_mode *adjusted_mode)
 {
         return true;
 }
@@ -680,7 +680,7 @@ static void ast_encoder_dpms(struct drm_encoder *encoder, int mode)
 }
 
 static bool ast_mode_fixup(struct drm_encoder *encoder,
-                           struct drm_display_mode *mode,
+                           const struct drm_display_mode *mode,
                            struct drm_display_mode *adjusted_mode)
 {
         return true;
diff --git a/drivers/gpu/drm/cirrus/cirrus_mode.c b/drivers/gpu/drm/cirrus/cirrus_mode.c
index 100f6308c509..a44d31aa4e3c 100644
--- a/drivers/gpu/drm/cirrus/cirrus_mode.c
+++ b/drivers/gpu/drm/cirrus/cirrus_mode.c
@@ -97,7 +97,7 @@ static void cirrus_crtc_dpms(struct drm_crtc *crtc, int mode)
  * to just pass that straight through, so this does nothing
  */
 static bool cirrus_crtc_mode_fixup(struct drm_crtc *crtc,
-                                   struct drm_display_mode *mode,
+                                   const struct drm_display_mode *mode,
                                    struct drm_display_mode *adjusted_mode)
 {
         return true;
@@ -429,8 +429,8 @@ void cirrus_crtc_fb_gamma_get(struct drm_crtc *crtc, u16 *red, u16 *green,
 
 
 static bool cirrus_encoder_mode_fixup(struct drm_encoder *encoder,
-                                  struct drm_display_mode *mode,
-                                  struct drm_display_mode *adjusted_mode)
+                                      const struct drm_display_mode *mode,
+                                      struct drm_display_mode *adjusted_mode)
 {
         return true;
 }
diff --git a/drivers/gpu/drm/drm_bufs.c b/drivers/gpu/drm/drm_bufs.c
index 348b367debeb..b356c719f2f1 100644
--- a/drivers/gpu/drm/drm_bufs.c
+++ b/drivers/gpu/drm/drm_bufs.c
@@ -641,8 +641,6 @@ int drm_addbufs_agp(struct drm_device * dev, struct drm_buf_desc * request)
 
         if (order < DRM_MIN_ORDER || order > DRM_MAX_ORDER)
                 return -EINVAL;
-        if (dev->queue_count)
-                return -EBUSY;  /* Not while in use */
 
         /* Make sure buffers are located in AGP memory that we own */
         valid = 0;
@@ -704,7 +702,6 @@ int drm_addbufs_agp(struct drm_device * dev, struct drm_buf_desc * request)
                 buf->next = NULL;
                 buf->waiting = 0;
                 buf->pending = 0;
-                init_waitqueue_head(&buf->dma_wait);
                 buf->file_priv = NULL;
 
                 buf->dev_priv_size = dev->driver->dev_priv_size;
@@ -796,13 +793,11 @@ int drm_addbufs_pci(struct drm_device * dev, struct drm_buf_desc * request)
         order = drm_order(request->size);
         size = 1 << order;
 
-        DRM_DEBUG("count=%d, size=%d (%d), order=%d, queue_count=%d\n",
-                  request->count, request->size, size, order, dev->queue_count);
+        DRM_DEBUG("count=%d, size=%d (%d), order=%d\n",
+                  request->count, request->size, size, order);
 
         if (order < DRM_MIN_ORDER || order > DRM_MAX_ORDER)
                 return -EINVAL;
-        if (dev->queue_count)
-                return -EBUSY;  /* Not while in use */
 
         alignment = (request->flags & _DRM_PAGE_ALIGN)
             ? PAGE_ALIGN(size) : size;
@@ -904,7 +899,6 @@ int drm_addbufs_pci(struct drm_device * dev, struct drm_buf_desc * request)
                         buf->next = NULL;
                         buf->waiting = 0;
                         buf->pending = 0;
-                        init_waitqueue_head(&buf->dma_wait);
                         buf->file_priv = NULL;
 
                         buf->dev_priv_size = dev->driver->dev_priv_size;
@@ -1019,8 +1013,6 @@ static int drm_addbufs_sg(struct drm_device * dev, struct drm_buf_desc * request
 
         if (order < DRM_MIN_ORDER || order > DRM_MAX_ORDER)
                 return -EINVAL;
-        if (dev->queue_count)
-                return -EBUSY;  /* Not while in use */
 
         spin_lock(&dev->count_lock);
         if (dev->buf_use) {
@@ -1071,7 +1063,6 @@ static int drm_addbufs_sg(struct drm_device * dev, struct drm_buf_desc * request
                 buf->next = NULL;
                 buf->waiting = 0;
                 buf->pending = 0;
-                init_waitqueue_head(&buf->dma_wait);
                 buf->file_priv = NULL;
 
                 buf->dev_priv_size = dev->driver->dev_priv_size;
@@ -1177,8 +1168,6 @@ static int drm_addbufs_fb(struct drm_device * dev, struct drm_buf_desc * request
 
         if (order < DRM_MIN_ORDER || order > DRM_MAX_ORDER)
                 return -EINVAL;
-        if (dev->queue_count)
-                return -EBUSY;  /* Not while in use */
 
         spin_lock(&dev->count_lock);
         if (dev->buf_use) {
@@ -1228,7 +1217,6 @@ static int drm_addbufs_fb(struct drm_device * dev, struct drm_buf_desc * request
                 buf->next = NULL;
                 buf->waiting = 0;
                 buf->pending = 0;
-                init_waitqueue_head(&buf->dma_wait);
                 buf->file_priv = NULL;
 
                 buf->dev_priv_size = dev->driver->dev_priv_size;
diff --git a/drivers/gpu/drm/drm_debugfs.c b/drivers/gpu/drm/drm_debugfs.c
index 1c7a1c0d3edd..70b13fc19396 100644
--- a/drivers/gpu/drm/drm_debugfs.c
+++ b/drivers/gpu/drm/drm_debugfs.c
@@ -46,7 +46,6 @@ static struct drm_info_list drm_debugfs_list[] = {
         {"name", drm_name_info, 0},
         {"vm", drm_vm_info, 0},
         {"clients", drm_clients_info, 0},
-        {"queues", drm_queues_info, 0},
         {"bufs", drm_bufs_info, 0},
         {"gem_names", drm_gem_name_info, DRIVER_GEM},
 #if DRM_DEBUG_CODE
diff --git a/drivers/gpu/drm/drm_dma.c b/drivers/gpu/drm/drm_dma.c
index cfb4e333ec0f..08f5e5309b22 100644
--- a/drivers/gpu/drm/drm_dma.c
+++ b/drivers/gpu/drm/drm_dma.c
@@ -120,11 +120,6 @@ void drm_free_buffer(struct drm_device *dev, struct drm_buf * buf)
         buf->pending = 0;
         buf->file_priv = NULL;
         buf->used = 0;
-
-        if (drm_core_check_feature(dev, DRIVER_DMA_QUEUE)
-            && waitqueue_active(&buf->dma_wait)) {
-                wake_up_interruptible(&buf->dma_wait);
-        }
 }
 
 /**
diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c
index 8a9d0792e4ec..9238de4009fa 100644
--- a/drivers/gpu/drm/drm_drv.c
+++ b/drivers/gpu/drm/drm_drv.c
@@ -182,7 +182,6 @@ static struct drm_ioctl_desc drm_ioctls[] = {
 int drm_lastclose(struct drm_device * dev)
 {
         struct drm_vma_entry *vma, *vma_temp;
-        int i;
 
         DRM_DEBUG("\n");
 
@@ -228,16 +227,6 @@ int drm_lastclose(struct drm_device * dev)
                 kfree(vma);
         }
 
-        if (drm_core_check_feature(dev, DRIVER_DMA_QUEUE) && dev->queuelist) {
-                for (i = 0; i < dev->queue_count; i++) {
-                        kfree(dev->queuelist[i]);
-                        dev->queuelist[i] = NULL;
-                }
-                kfree(dev->queuelist);
-                dev->queuelist = NULL;
-        }
-        dev->queue_count = 0;
-
         if (drm_core_check_feature(dev, DRIVER_HAVE_DMA) &&
             !drm_core_check_feature(dev, DRIVER_MODESET))
                 drm_dma_takedown(dev);
@@ -486,7 +475,7 @@ long drm_ioctl(struct file *filp,
                 kfree(kdata);
         atomic_dec(&dev->ioctl_count);
         if (retcode)
-                DRM_DEBUG("ret = %x\n", retcode);
+                DRM_DEBUG("ret = %d\n", retcode);
         return retcode;
 }
 
diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c
index 5683b7fdd746..f546d1e8af82 100644
--- a/drivers/gpu/drm/drm_fb_helper.c
+++ b/drivers/gpu/drm/drm_fb_helper.c
@@ -228,7 +228,7 @@ bool drm_fb_helper_restore_fbdev_mode(struct drm_fb_helper *fb_helper)
         int i, ret;
         for (i = 0; i < fb_helper->crtc_count; i++) {
                 struct drm_mode_set *mode_set = &fb_helper->crtc_info[i].mode_set;
-                ret = drm_crtc_helper_set_config(mode_set);
+                ret = mode_set->crtc->funcs->set_config(mode_set);
                 if (ret)
                         error = true;
         }
@@ -1353,7 +1353,7 @@ int drm_fb_helper_hotplug_event(struct drm_fb_helper *fb_helper)
         struct drm_device *dev = fb_helper->dev;
         int count = 0;
         u32 max_width, max_height, bpp_sel;
-        bool bound = false, crtcs_bound = false;
+        int bound = 0, crtcs_bound = 0;
         struct drm_crtc *crtc;
 
         if (!fb_helper->fb)
@@ -1362,12 +1362,12 @@ int drm_fb_helper_hotplug_event(struct drm_fb_helper *fb_helper)
         mutex_lock(&dev->mode_config.mutex);
         list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
                 if (crtc->fb)
-                        crtcs_bound = true;
+                        crtcs_bound++;
                 if (crtc->fb == fb_helper->fb)
-                        bound = true;
+                        bound++;
         }
 
-        if (!bound && crtcs_bound) {
+        if (bound < crtcs_bound) {
                 fb_helper->delayed_hotplug = true;
                 mutex_unlock(&dev->mode_config.mutex);
                 return 0;
diff --git a/drivers/gpu/drm/drm_fops.c b/drivers/gpu/drm/drm_fops.c
index 123de28f94ef..5062eec673f1 100644
--- a/drivers/gpu/drm/drm_fops.c
+++ b/drivers/gpu/drm/drm_fops.c
@@ -75,10 +75,6 @@ static int drm_setup(struct drm_device * dev)
 
         dev->sigdata.lock = NULL;
 
-        dev->queue_count = 0;
-        dev->queue_reserved = 0;
-        dev->queue_slots = 0;
-        dev->queuelist = NULL;
         dev->context_flag = 0;
         dev->interrupt_flag = 0;
         dev->dma_flag = 0;
@@ -144,12 +140,12 @@ int drm_open(struct inode *inode, struct file *filp)
         }
         if (!retcode) {
                 mutex_lock(&dev->struct_mutex);
-                if (minor->type == DRM_MINOR_LEGACY) {
-                        if (dev->dev_mapping == NULL)
-                                dev->dev_mapping = inode->i_mapping;
-                        else if (dev->dev_mapping != inode->i_mapping)
-                                retcode = -ENODEV;
-                }
+                if (dev->dev_mapping == NULL)
+                        dev->dev_mapping = &inode->i_data;
+                /* ihold ensures nobody can remove inode with our i_data */
+                ihold(container_of(dev->dev_mapping, struct inode, i_data));
+                inode->i_mapping = dev->dev_mapping;
+                filp->f_mapping = dev->dev_mapping;
                 mutex_unlock(&dev->struct_mutex);
         }
 
@@ -370,72 +366,16 @@ int drm_fasync(int fd, struct file *filp, int on)
 }
 EXPORT_SYMBOL(drm_fasync);
 
-/*
- * Reclaim locked buffers; note that this may be a bad idea if the current
- * context doesn't have the hw lock...
- */
-static void drm_reclaim_locked_buffers(struct drm_device *dev, struct file *f)
-{
-        struct drm_file *file_priv = f->private_data;
-
-        if (drm_i_have_hw_lock(dev, file_priv)) {
-                dev->driver->reclaim_buffers_locked(dev, file_priv);
-        } else {
-                unsigned long _end = jiffies + 3 * DRM_HZ;
-                int locked = 0;
-
-                drm_idlelock_take(&file_priv->master->lock);
-
-                /*
-                 * Wait for a while.
-                 */
-                do {
-                        spin_lock_bh(&file_priv->master->lock.spinlock);
-                        locked = file_priv->master->lock.idle_has_lock;
-                        spin_unlock_bh(&file_priv->master->lock.spinlock);
-                        if (locked)
-                                break;
-                        schedule();
-                } while (!time_after_eq(jiffies, _end));
-
-                if (!locked) {
-                        DRM_ERROR("reclaim_buffers_locked() deadlock. Please rework this\n"
-                                  "\tdriver to use reclaim_buffers_idlelocked() instead.\n"
-                                  "\tI will go on reclaiming the buffers anyway.\n");
-                }
-
-                dev->driver->reclaim_buffers_locked(dev, file_priv);
-                drm_idlelock_release(&file_priv->master->lock);
-        }
-}
-
 static void drm_master_release(struct drm_device *dev, struct file *filp)
 {
         struct drm_file *file_priv = filp->private_data;
 
-        if (dev->driver->reclaim_buffers_locked &&
-            file_priv->master->lock.hw_lock)
-                drm_reclaim_locked_buffers(dev, filp);
-
-        if (dev->driver->reclaim_buffers_idlelocked &&
-            file_priv->master->lock.hw_lock) {
-                drm_idlelock_take(&file_priv->master->lock);
-                dev->driver->reclaim_buffers_idlelocked(dev, file_priv);
-                drm_idlelock_release(&file_priv->master->lock);
-        }
-
-
         if (drm_i_have_hw_lock(dev, file_priv)) {
                 DRM_DEBUG("File %p released, freeing lock for context %d\n",
                           filp, _DRM_LOCKING_CONTEXT(file_priv->master->lock.hw_lock->lock));
                 drm_lock_free(&file_priv->master->lock,
                               _DRM_LOCKING_CONTEXT(file_priv->master->lock.hw_lock->lock));
         }
-
-        if (drm_core_check_feature(dev, DRIVER_HAVE_DMA) &&
-            !dev->driver->reclaim_buffers_locked) {
-                dev->driver->reclaim_buffers(dev, file_priv);
-        }
 }
 
 static void drm_events_release(struct drm_file *file_priv)
@@ -505,6 +445,9 @@ int drm_release(struct inode *inode, struct file *filp)
         if (file_priv->minor->master)
                 drm_master_release(dev, filp);
 
+        if (drm_core_check_feature(dev, DRIVER_HAVE_DMA))
+                drm_core_reclaim_buffers(dev, file_priv);
+
         drm_events_release(file_priv);
 
         if (dev->driver->driver_features & DRIVER_MODESET)
@@ -566,6 +509,9 @@ int drm_release(struct inode *inode, struct file *filp)
                 }
         }
 
+        BUG_ON(dev->dev_mapping == NULL);
+        iput(container_of(dev->dev_mapping, struct inode, i_data));
+
         /* drop the reference held my the file priv */
         drm_master_put(&file_priv->master);
         file_priv->is_master = 0;
diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c
index d58e69da1fb5..fbe0842038b5 100644
--- a/drivers/gpu/drm/drm_gem.c
+++ b/drivers/gpu/drm/drm_gem.c
@@ -354,7 +354,7 @@ drm_gem_create_mmap_offset(struct drm_gem_object *obj)
 
         /* Get a DRM GEM mmap offset allocated... */
         list->file_offset_node = drm_mm_search_free(&mm->offset_manager,
-                        obj->size / PAGE_SIZE, 0, 0);
+                        obj->size / PAGE_SIZE, 0, false);
 
         if (!list->file_offset_node) {
                 DRM_ERROR("failed to allocate offset for bo %d\n", obj->name);
diff --git a/drivers/gpu/drm/drm_info.c b/drivers/gpu/drm/drm_info.c
index ab1162da70f8..8928edbb94c7 100644
--- a/drivers/gpu/drm/drm_info.c
+++ b/drivers/gpu/drm/drm_info.c
@@ -110,42 +110,6 @@ int drm_vm_info(struct seq_file *m, void *data)
 }
 
 /**
- * Called when "/proc/dri/.../queues" is read.
- */
-int drm_queues_info(struct seq_file *m, void *data)
-{
-        struct drm_info_node *node = (struct drm_info_node *) m->private;
-        struct drm_device *dev = node->minor->dev;
-        int i;
-        struct drm_queue *q;
-
-        mutex_lock(&dev->struct_mutex);
-        seq_printf(m, "  ctx/flags   use   fin"
-                   "   blk/rw/rwf  wait    flushed         queued"
-                   "      locks\n\n");
-        for (i = 0; i < dev->queue_count; i++) {
-                q = dev->queuelist[i];
-                atomic_inc(&q->use_count);
-                seq_printf(m,   "%5d/0x%03x %5d %5d"
-                           " %5d/%c%c/%c%c%c %5Zd\n",
-                           i,
-                           q->flags,
-                           atomic_read(&q->use_count),
-                           atomic_read(&q->finalization),
-                           atomic_read(&q->block_count),
-                           atomic_read(&q->block_read) ? 'r' : '-',
-                           atomic_read(&q->block_write) ? 'w' : '-',
-                           waitqueue_active(&q->read_queue) ? 'r' : '-',
-                           waitqueue_active(&q->write_queue) ? 'w' : '-',
-                           waitqueue_active(&q->flush_queue) ? 'f' : '-',
-                           DRM_BUFCOUNT(&q->waitlist));
-                atomic_dec(&q->use_count);
-        }
-        mutex_unlock(&dev->struct_mutex);
-        return 0;
-}
-
-/**
  * Called when "/proc/dri/.../bufs" is read.
  */
 int drm_bufs_info(struct seq_file *m, void *data)
@@ -235,7 +199,7 @@ int drm_clients_info(struct seq_file *m, void *data)
 }
 
 
-int drm_gem_one_name_info(int id, void *ptr, void *data)
+static int drm_gem_one_name_info(int id, void *ptr, void *data)
 {
         struct drm_gem_object *obj = ptr;
         struct seq_file *m = data;
diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index c798eeae0a03..03f16f352fe2 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -974,7 +974,6 @@ EXPORT_SYMBOL(drm_vblank_off);
  * drm_vblank_pre_modeset - account for vblanks across mode sets
  * @dev: DRM device
  * @crtc: CRTC in question
- * @post: post or pre mode set?
  *
  * Account for vblank events across mode setting events, which will likely
  * reset the hardware frame counter.
@@ -1037,6 +1036,10 @@ int drm_modeset_ctl(struct drm_device *dev, void *data,
         if (!dev->num_crtcs)
                 return 0;
 
+        /* KMS drivers handle this internally */
+        if (drm_core_check_feature(dev, DRIVER_MODESET))
+                return 0;
+
         crtc = modeset->crtc;
         if (crtc >= dev->num_crtcs)
                 return -EINVAL;
diff --git a/drivers/gpu/drm/drm_lock.c b/drivers/gpu/drm/drm_lock.c
index 521152041691..32039553e172 100644
--- a/drivers/gpu/drm/drm_lock.c
+++ b/drivers/gpu/drm/drm_lock.c
@@ -70,10 +70,6 @@ int drm_lock(struct drm_device *dev, void *data, struct drm_file *file_priv)
                   lock->context, task_pid_nr(current),
                   master->lock.hw_lock->lock, lock->flags);
 
-        if (drm_core_check_feature(dev, DRIVER_DMA_QUEUE))
-                if (lock->context < 0)
-                        return -EINVAL;
-
         add_wait_queue(&master->lock.lock_queue, &entry);
         spin_lock_bh(&master->lock.spinlock);
         master->lock.user_waiters++;
diff --git a/drivers/gpu/drm/drm_mm.c b/drivers/gpu/drm/drm_mm.c
index 961fb54f4266..9bb82f7f0061 100644
--- a/drivers/gpu/drm/drm_mm.c
+++ b/drivers/gpu/drm/drm_mm.c
@@ -118,45 +118,53 @@ static inline unsigned long drm_mm_hole_node_end(struct drm_mm_node *hole_node)
 
 static void drm_mm_insert_helper(struct drm_mm_node *hole_node,
                                  struct drm_mm_node *node,
-                                 unsigned long size, unsigned alignment)
+                                 unsigned long size, unsigned alignment,
+                                 unsigned long color)
 {
         struct drm_mm *mm = hole_node->mm;
-        unsigned long tmp = 0, wasted = 0;
         unsigned long hole_start = drm_mm_hole_node_start(hole_node);
         unsigned long hole_end = drm_mm_hole_node_end(hole_node);
+        unsigned long adj_start = hole_start;
+        unsigned long adj_end = hole_end;
 
         BUG_ON(!hole_node->hole_follows || node->allocated);
 
-        if (alignment)
-                tmp = hole_start % alignment;
+        if (mm->color_adjust)
+                mm->color_adjust(hole_node, color, &adj_start, &adj_end);
 
-        if (!tmp) {
+        if (alignment) {
+                unsigned tmp = adj_start % alignment;
+                if (tmp)
+                        adj_start += alignment - tmp;
+        }
+
+        if (adj_start == hole_start) {
                 hole_node->hole_follows = 0;
-                list_del_init(&hole_node->hole_stack);
-        } else
-                wasted = alignment - tmp;
+                list_del(&hole_node->hole_stack);
+        }
 
-        node->start = hole_start + wasted;
+        node->start = adj_start;
         node->size = size;
         node->mm = mm;
+        node->color = color;
         node->allocated = 1;
 
         INIT_LIST_HEAD(&node->hole_stack);
         list_add(&node->node_list, &hole_node->node_list);
 
-        BUG_ON(node->start + node->size > hole_end);
+        BUG_ON(node->start + node->size > adj_end);
 
+        node->hole_follows = 0;
         if (node->start + node->size < hole_end) {
                 list_add(&node->hole_stack, &mm->hole_stack);
                 node->hole_follows = 1;
-        } else {
-                node->hole_follows = 0;
         }
 }
 
 struct drm_mm_node *drm_mm_get_block_generic(struct drm_mm_node *hole_node,
                                              unsigned long size,
                                              unsigned alignment,
+                                             unsigned long color,
                                              int atomic)
 {
         struct drm_mm_node *node;
@@ -165,7 +173,7 @@ struct drm_mm_node *drm_mm_get_block_generic(struct drm_mm_node *hole_node,
         if (unlikely(node == NULL))
                 return NULL;
 
-        drm_mm_insert_helper(hole_node, node, size, alignment);
+        drm_mm_insert_helper(hole_node, node, size, alignment, color);
 
         return node;
 }
@@ -181,11 +189,11 @@ int drm_mm_insert_node(struct drm_mm *mm, struct drm_mm_node *node,
 {
         struct drm_mm_node *hole_node;
 
-        hole_node = drm_mm_search_free(mm, size, alignment, 0);
+        hole_node = drm_mm_search_free(mm, size, alignment, false);
         if (!hole_node)
                 return -ENOSPC;
 
-        drm_mm_insert_helper(hole_node, node, size, alignment);
+        drm_mm_insert_helper(hole_node, node, size, alignment, 0);
 
         return 0;
 }
@@ -194,50 +202,57 @@ EXPORT_SYMBOL(drm_mm_insert_node);
 static void drm_mm_insert_helper_range(struct drm_mm_node *hole_node,
                                        struct drm_mm_node *node,
                                        unsigned long size, unsigned alignment,
+                                       unsigned long color,
                                        unsigned long start, unsigned long end)
 {
         struct drm_mm *mm = hole_node->mm;
-        unsigned long tmp = 0, wasted = 0;
         unsigned long hole_start = drm_mm_hole_node_start(hole_node);
         unsigned long hole_end = drm_mm_hole_node_end(hole_node);
+        unsigned long adj_start = hole_start;
+        unsigned long adj_end = hole_end;
 
         BUG_ON(!hole_node->hole_follows || node->allocated);
 
-        if (hole_start < start)
-                wasted += start - hole_start;
-        if (alignment)
-                tmp = (hole_start + wasted) % alignment;
+        if (mm->color_adjust)
+                mm->color_adjust(hole_node, color, &adj_start, &adj_end);
 
-        if (tmp)
-                wasted += alignment - tmp;
+        if (adj_start < start)
+                adj_start = start;
+
+        if (alignment) {
+                unsigned tmp = adj_start % alignment;
+                if (tmp)
+                        adj_start += alignment - tmp;
+        }
 
-        if (!wasted) {
+        if (adj_start == hole_start) {
                 hole_node->hole_follows = 0;
-                list_del_init(&hole_node->hole_stack);
+                list_del(&hole_node->hole_stack);
         }
 
-        node->start = hole_start + wasted;
+        node->start = adj_start;
         node->size = size;
         node->mm = mm;
+        node->color = color;
         node->allocated = 1;
 
         INIT_LIST_HEAD(&node->hole_stack);
         list_add(&node->node_list, &hole_node->node_list);
 
-        BUG_ON(node->start + node->size > hole_end);
+        BUG_ON(node->start + node->size > adj_end);
         BUG_ON(node->start + node->size > end);
 
+        node->hole_follows = 0;
         if (node->start + node->size < hole_end) {
                 list_add(&node->hole_stack, &mm->hole_stack);
                 node->hole_follows = 1;
-        } else {
-                node->hole_follows = 0;
         }
 }
 
 struct drm_mm_node *drm_mm_get_block_range_generic(struct drm_mm_node *hole_node,
                                                 unsigned long size,
                                                 unsigned alignment,
+                                                unsigned long color,
                                                 unsigned long start,
                                                 unsigned long end,
                                                 int atomic)
@@ -248,7 +263,7 @@ struct drm_mm_node *drm_mm_get_block_range_generic(struct drm_mm_node *hole_node
         if (unlikely(node == NULL))
                 return NULL;
 
-        drm_mm_insert_helper_range(hole_node, node, size, alignment,
+        drm_mm_insert_helper_range(hole_node, node, size, alignment, color,
                                    start, end);
 
         return node;
@@ -267,11 +282,11 @@ int drm_mm_insert_node_in_range(struct drm_mm *mm, struct drm_mm_node *node,
         struct drm_mm_node *hole_node;
 
         hole_node = drm_mm_search_free_in_range(mm, size, alignment,
-                                                start, end, 0);
+                                                start, end, false);
         if (!hole_node)
                 return -ENOSPC;
 
-        drm_mm_insert_helper_range(hole_node, node, size, alignment,
+        drm_mm_insert_helper_range(hole_node, node, size, alignment, 0,
                                    start, end);
 
         return 0;
@@ -336,27 +351,23 @@ EXPORT_SYMBOL(drm_mm_put_block);
 static int check_free_hole(unsigned long start, unsigned long end,
                            unsigned long size, unsigned alignment)
 {
-        unsigned wasted = 0;
-
         if (end - start < size)
                 return 0;
 
         if (alignment) {
                 unsigned tmp = start % alignment;
                 if (tmp)
-                        wasted = alignment - tmp;
-        }
-
-        if (end >= start + size + wasted) {
-                return 1;
+                        start += alignment - tmp;
         }
 
-        return 0;
+        return end >= start + size;
 }
 
-struct drm_mm_node *drm_mm_search_free(const struct drm_mm *mm,
-                                       unsigned long size,
-                                       unsigned alignment, int best_match)
+struct drm_mm_node *drm_mm_search_free_generic(const struct drm_mm *mm,
+                                               unsigned long size,
+                                               unsigned alignment,
+                                               unsigned long color,
+                                               bool best_match)
 {
         struct drm_mm_node *entry;
         struct drm_mm_node *best;
@@ -368,10 +379,17 @@ struct drm_mm_node *drm_mm_search_free(const struct drm_mm *mm,
         best_size = ~0UL;
 
         list_for_each_entry(entry, &mm->hole_stack, hole_stack) {
+                unsigned long adj_start = drm_mm_hole_node_start(entry);
+                unsigned long adj_end = drm_mm_hole_node_end(entry);
+
+                if (mm->color_adjust) {
+                        mm->color_adjust(entry, color, &adj_start, &adj_end);
+                        if (adj_end <= adj_start)
+                                continue;
+                }
+
                 BUG_ON(!entry->hole_follows);
-                if (!check_free_hole(drm_mm_hole_node_start(entry),
-                                     drm_mm_hole_node_end(entry),
-                                     size, alignment))
+                if (!check_free_hole(adj_start, adj_end, size, alignment))
                         continue;
 
                 if (!best_match)
@@ -385,14 +403,15 @@ struct drm_mm_node *drm_mm_search_free(const struct drm_mm *mm,
 
         return best;
 }
-EXPORT_SYMBOL(drm_mm_search_free);
-
-struct drm_mm_node *drm_mm_search_free_in_range(const struct drm_mm *mm,
-                                                unsigned long size,
-                                                unsigned alignment,
-                                                unsigned long start,
-                                                unsigned long end,
-                                                int best_match)
+EXPORT_SYMBOL(drm_mm_search_free_generic);
+
+struct drm_mm_node *drm_mm_search_free_in_range_generic(const struct drm_mm *mm,
+                                                        unsigned long size,
+                                                        unsigned alignment,
+                                                        unsigned long color,
+                                                        unsigned long start,
+                                                        unsigned long end,
+                                                        bool best_match)
 {
         struct drm_mm_node *entry;
         struct drm_mm_node *best;
@@ -410,6 +429,13 @@ struct drm_mm_node *drm_mm_search_free_in_range(const struct drm_mm *mm,
                         end : drm_mm_hole_node_end(entry);
 
                 BUG_ON(!entry->hole_follows);
+
+                if (mm->color_adjust) {
+                        mm->color_adjust(entry, color, &adj_start, &adj_end);
+                        if (adj_end <= adj_start)
+                                continue;
+                }
+
                 if (!check_free_hole(adj_start, adj_end, size, alignment))
                         continue;
 
@@ -424,7 +450,7 @@ struct drm_mm_node *drm_mm_search_free_in_range(const struct drm_mm *mm,
 
         return best;
 }
-EXPORT_SYMBOL(drm_mm_search_free_in_range);
+EXPORT_SYMBOL(drm_mm_search_free_in_range_generic);
 
 /**
  * Moves an allocation. To be used with embedded struct drm_mm_node.
@@ -437,6 +463,7 @@ void drm_mm_replace_node(struct drm_mm_node *old, struct drm_mm_node *new)
         new->mm = old->mm;
         new->start = old->start;
         new->size = old->size;
+        new->color = old->color;
 
         old->allocated = 0;
         new->allocated = 1;
@@ -452,9 +479,12 @@ EXPORT_SYMBOL(drm_mm_replace_node);
  * Warning: As long as the scan list is non-empty, no other operations than
  * adding/removing nodes to/from the scan list are allowed.
  */
-void drm_mm_init_scan(struct drm_mm *mm, unsigned long size,
-                      unsigned alignment)
+void drm_mm_init_scan(struct drm_mm *mm,
+                      unsigned long size,
+                      unsigned alignment,
+                      unsigned long color)
 {
+        mm->scan_color = color;
         mm->scan_alignment = alignment;
         mm->scan_size = size;
         mm->scanned_blocks = 0;
@@ -474,11 +504,14 @@ EXPORT_SYMBOL(drm_mm_init_scan);
  * Warning: As long as the scan list is non-empty, no other operations than
  * adding/removing nodes to/from the scan list are allowed.
  */
-void drm_mm_init_scan_with_range(struct drm_mm *mm, unsigned long size,
+void drm_mm_init_scan_with_range(struct drm_mm *mm,
+                                 unsigned long size,
                                  unsigned alignment,
+                                 unsigned long color,
                                  unsigned long start,
                                  unsigned long end)
 {
+        mm->scan_color = color;
         mm->scan_alignment = alignment;
         mm->scan_size = size;
         mm->scanned_blocks = 0;
@@ -522,17 +555,21 @@ int drm_mm_scan_add_block(struct drm_mm_node *node)
 
         hole_start = drm_mm_hole_node_start(prev_node);
         hole_end = drm_mm_hole_node_end(prev_node);
+
+        adj_start = hole_start;
+        adj_end = hole_end;
+
+        if (mm->color_adjust)
+                mm->color_adjust(prev_node, mm->scan_color, &adj_start, &adj_end);
+
         if (mm->scan_check_range) {
-                adj_start = hole_start < mm->scan_start ?
-                        mm->scan_start : hole_start;
-                adj_end = hole_end > mm->scan_end ?
-                        mm->scan_end : hole_end;
-        } else {
-                adj_start = hole_start;
-                adj_end = hole_end;
+                if (adj_start < mm->scan_start)
+                        adj_start = mm->scan_start;
+                if (adj_end > mm->scan_end)
+                        adj_end = mm->scan_end;
         }
 
-        if (check_free_hole(adj_start , adj_end,
+        if (check_free_hole(adj_start, adj_end,
                             mm->scan_size, mm->scan_alignment)) {
                 mm->scan_hit_start = hole_start;
                 mm->scan_hit_size = hole_end;
@@ -616,6 +653,8 @@ int drm_mm_init(struct drm_mm * mm, unsigned long start, unsigned long size)
         mm->head_node.size = start - mm->head_node.start;
         list_add_tail(&mm->head_node.hole_stack, &mm->hole_stack);
 
+        mm->color_adjust = NULL;
+
         return 0;
 }
 EXPORT_SYMBOL(drm_mm_init);
diff --git a/drivers/gpu/drm/drm_pci.c b/drivers/gpu/drm/drm_pci.c
index 13f3d936472f..5320364582ce 100644
--- a/drivers/gpu/drm/drm_pci.c
+++ b/drivers/gpu/drm/drm_pci.c
@@ -465,3 +465,52 @@ void drm_pci_exit(struct drm_driver *driver, struct pci_driver *pdriver)
         DRM_INFO("Module unloaded\n");
 }
 EXPORT_SYMBOL(drm_pci_exit);
+
+int drm_pcie_get_speed_cap_mask(struct drm_device *dev, u32 *mask)
+{
+        struct pci_dev *root;
+        int pos;
+        u32 lnkcap, lnkcap2;
+
+        *mask = 0;
+        if (!dev->pdev)
+                return -EINVAL;
+
+        if (!pci_is_pcie(dev->pdev))
+                return -EINVAL;
+
+        root = dev->pdev->bus->self;
+
+        pos = pci_pcie_cap(root);
+        if (!pos)
+                return -EINVAL;
+
+        /* we've been informed via and serverworks don't make the cut */
+        if (root->vendor == PCI_VENDOR_ID_VIA ||
+            root->vendor == PCI_VENDOR_ID_SERVERWORKS)
+                return -EINVAL;
+
+        pci_read_config_dword(root, pos + PCI_EXP_LNKCAP, &lnkcap);
+        pci_read_config_dword(root, pos + PCI_EXP_LNKCAP2, &lnkcap2);
+
+        lnkcap &= PCI_EXP_LNKCAP_SLS;
+        lnkcap2 &= 0xfe;
+
+        if (lnkcap2) { /* PCIE GEN 3.0 */
+                if (lnkcap2 & PCI_EXP_LNKCAP2_SLS_2_5GB)
+                        *mask |= DRM_PCIE_SPEED_25;
+                if (lnkcap2 & PCI_EXP_LNKCAP2_SLS_5_0GB)
+                        *mask |= DRM_PCIE_SPEED_50;
+                if (lnkcap2 & PCI_EXP_LNKCAP2_SLS_8_0GB)
+                        *mask |= DRM_PCIE_SPEED_80;
+        } else {
+                if (lnkcap & 1)
+                        *mask |= DRM_PCIE_SPEED_25;
+                if (lnkcap & 2)
+                        *mask |= DRM_PCIE_SPEED_50;
+        }
+
+        DRM_INFO("probing gen 2 caps for device %x:%x = %x/%x\n", root->vendor, root->device, lnkcap, lnkcap2);
+        return 0;
+}
+EXPORT_SYMBOL(drm_pcie_get_speed_cap_mask);
diff --git a/drivers/gpu/drm/drm_proc.c b/drivers/gpu/drm/drm_proc.c
index fff87221f9e9..371c695322d9 100644
--- a/drivers/gpu/drm/drm_proc.c
+++ b/drivers/gpu/drm/drm_proc.c
@@ -53,7 +53,6 @@ static struct drm_info_list drm_proc_list[] = {
         {"name", drm_name_info, 0},
         {"vm", drm_vm_info, 0},
         {"clients", drm_clients_info, 0},
-        {"queues", drm_queues_info, 0},
         {"bufs", drm_bufs_info, 0},
         {"gem_names", drm_gem_name_info, DRIVER_GEM},
 #if DRM_DEBUG_CODE
diff --git a/drivers/gpu/drm/drm_sysfs.c b/drivers/gpu/drm/drm_sysfs.c
index 45cf1dd3eb9c..45ac8d6c92b7 100644
--- a/drivers/gpu/drm/drm_sysfs.c
+++ b/drivers/gpu/drm/drm_sysfs.c
@@ -134,6 +134,7 @@ void drm_sysfs_destroy(void)
                 return;
         class_remove_file(drm_class, &class_attr_version.attr);
         class_destroy(drm_class);
+        drm_class = NULL;
 }
 
 /**
@@ -554,6 +555,9 @@ void drm_sysfs_device_remove(struct drm_minor *minor)
 
 int drm_class_device_register(struct device *dev)
 {
+        if (!drm_class || IS_ERR(drm_class))
+                return -ENOENT;
+
         dev->class = drm_class;
         return device_register(dev);
 }
diff --git a/drivers/gpu/drm/exynos/exynos_drm_crtc.c b/drivers/gpu/drm/exynos/exynos_drm_crtc.c
index 4afb625128d7..32a34c85899b 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_crtc.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_crtc.c
@@ -237,7 +237,7 @@ static void exynos_drm_crtc_commit(struct drm_crtc *crtc)
 
 static bool
 exynos_drm_crtc_mode_fixup(struct drm_crtc *crtc,
-                            struct drm_display_mode *mode,
+                            const struct drm_display_mode *mode,
                             struct drm_display_mode *adjusted_mode)
 {
         DRM_DEBUG_KMS("%s\n", __FILE__);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.h b/drivers/gpu/drm/exynos/exynos_drm_drv.h
index c82c90c443e7..277653d5fda0 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_drv.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_drv.h
@@ -174,7 +174,7 @@ struct exynos_drm_manager_ops {
         void (*apply)(struct device *subdrv_dev);
         void (*mode_fixup)(struct device *subdrv_dev,
                                 struct drm_connector *connector,
-                                struct drm_display_mode *mode,
+                                const struct drm_display_mode *mode,
                                 struct drm_display_mode *adjusted_mode);
         void (*mode_set)(struct device *subdrv_dev, void *mode);
         void (*get_max_resol)(struct device *subdrv_dev, unsigned int *width,
diff --git a/drivers/gpu/drm/exynos/exynos_drm_encoder.c b/drivers/gpu/drm/exynos/exynos_drm_encoder.c
index 23d5ad379f86..4a13a747f5d4 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_encoder.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_encoder.c
@@ -108,7 +108,7 @@ static void exynos_drm_encoder_dpms(struct drm_encoder *encoder, int mode)
 
 static bool
 exynos_drm_encoder_mode_fixup(struct drm_encoder *encoder,
-                               struct drm_display_mode *mode,
+                               const struct drm_display_mode *mode,
                                struct drm_display_mode *adjusted_mode)
 {
         struct drm_device *dev = encoder->dev;
diff --git a/drivers/gpu/drm/exynos/exynos_drm_hdmi.c b/drivers/gpu/drm/exynos/exynos_drm_hdmi.c
index 5d9d2c2f8f3f..8ffcdf8b9e22 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_hdmi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_hdmi.c
@@ -142,7 +142,7 @@ static void drm_hdmi_disable_vblank(struct device *subdrv_dev)
 
 static void drm_hdmi_mode_fixup(struct device *subdrv_dev,
                                 struct drm_connector *connector,
-                                struct drm_display_mode *mode,
+                                const struct drm_display_mode *mode,
                                 struct drm_display_mode *adjusted_mode)
 {
         struct drm_hdmi_context *ctx = to_context(subdrv_dev);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_hdmi.h b/drivers/gpu/drm/exynos/exynos_drm_hdmi.h
index bd8126996e52..a91c42088e42 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_hdmi.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_hdmi.h
@@ -51,7 +51,7 @@ struct exynos_hdmi_ops {
 
         /* manager */
         void (*mode_fixup)(void *ctx, struct drm_connector *connector,
-                                struct drm_display_mode *mode,
+                                const struct drm_display_mode *mode,
                                 struct drm_display_mode *adjusted_mode);
         void (*mode_set)(void *ctx, void *mode);
         void (*get_max_resol)(void *ctx, unsigned int *width,
diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c
index a137e9e39a33..066bde3f19c4 100644
--- a/drivers/gpu/drm/exynos/exynos_hdmi.c
+++ b/drivers/gpu/drm/exynos/exynos_hdmi.c
@@ -1940,7 +1940,7 @@ static void hdmi_conf_apply(struct hdmi_context *hdata)
 }
 
 static void hdmi_mode_fixup(void *ctx, struct drm_connector *connector,
-                                struct drm_display_mode *mode,
+                                const struct drm_display_mode *mode,
                                 struct drm_display_mode *adjusted_mode)
 {
         struct drm_display_mode *m;
diff --git a/drivers/gpu/drm/gma500/cdv_intel_crt.c b/drivers/gpu/drm/gma500/cdv_intel_crt.c
index 187422018601..8c175345d85c 100644
--- a/drivers/gpu/drm/gma500/cdv_intel_crt.c
+++ b/drivers/gpu/drm/gma500/cdv_intel_crt.c
@@ -82,7 +82,7 @@ static int cdv_intel_crt_mode_valid(struct drm_connector *connector,
 }
 
 static bool cdv_intel_crt_mode_fixup(struct drm_encoder *encoder,
-                                 struct drm_display_mode *mode,
+                                 const struct drm_display_mode *mode,
                                  struct drm_display_mode *adjusted_mode)
 {
         return true;
diff --git a/drivers/gpu/drm/gma500/cdv_intel_display.c b/drivers/gpu/drm/gma500/cdv_intel_display.c
index c3e9a0f701df..a68509ba22a8 100644
--- a/drivers/gpu/drm/gma500/cdv_intel_display.c
+++ b/drivers/gpu/drm/gma500/cdv_intel_display.c
@@ -913,7 +913,7 @@ static void cdv_intel_crtc_commit(struct drm_crtc *crtc)
 }
 
 static bool cdv_intel_crtc_mode_fixup(struct drm_crtc *crtc,
-                                  struct drm_display_mode *mode,
+                                  const struct drm_display_mode *mode,
                                   struct drm_display_mode *adjusted_mode)
 {
         return true;
diff --git a/drivers/gpu/drm/gma500/cdv_intel_hdmi.c b/drivers/gpu/drm/gma500/cdv_intel_hdmi.c
index 88b59d4a7b7f..a86f87b9ddde 100644
--- a/drivers/gpu/drm/gma500/cdv_intel_hdmi.c
+++ b/drivers/gpu/drm/gma500/cdv_intel_hdmi.c
@@ -90,7 +90,7 @@ static void cdv_hdmi_mode_set(struct drm_encoder *encoder,
 }
 
 static bool cdv_hdmi_mode_fixup(struct drm_encoder *encoder,
-                                  struct drm_display_mode *mode,
+                                  const struct drm_display_mode *mode,
                                   struct drm_display_mode *adjusted_mode)
 {
         return true;
diff --git a/drivers/gpu/drm/gma500/cdv_intel_lvds.c b/drivers/gpu/drm/gma500/cdv_intel_lvds.c
index ff5b58eb878c..c7f9468b74ba 100644
--- a/drivers/gpu/drm/gma500/cdv_intel_lvds.c
+++ b/drivers/gpu/drm/gma500/cdv_intel_lvds.c
@@ -270,7 +270,7 @@ static int cdv_intel_lvds_mode_valid(struct drm_connector *connector,
 }
 
 static bool cdv_intel_lvds_mode_fixup(struct drm_encoder *encoder,
-                                  struct drm_display_mode *mode,
+                                  const struct drm_display_mode *mode,
                                   struct drm_display_mode *adjusted_mode)
 {
         struct drm_device *dev = encoder->dev;
diff --git a/drivers/gpu/drm/gma500/intel_bios.c b/drivers/gpu/drm/gma500/intel_bios.c
index 973d7f6d66b7..8d7caf0f363e 100644
--- a/drivers/gpu/drm/gma500/intel_bios.c
+++ b/drivers/gpu/drm/gma500/intel_bios.c
@@ -427,7 +427,7 @@ parse_device_mapping(struct drm_psb_private *dev_priv,
  *
  * Returns 0 on success, nonzero on failure.
  */
-bool psb_intel_init_bios(struct drm_device *dev)
+int psb_intel_init_bios(struct drm_device *dev)
 {
         struct drm_psb_private *dev_priv = dev->dev_private;
         struct pci_dev *pdev = dev->pdev;
diff --git a/drivers/gpu/drm/gma500/intel_bios.h b/drivers/gpu/drm/gma500/intel_bios.h
index 0a738663eb5a..2e95523b84b1 100644
--- a/drivers/gpu/drm/gma500/intel_bios.h
+++ b/drivers/gpu/drm/gma500/intel_bios.h
@@ -431,7 +431,7 @@ struct bdb_driver_features {
         u8 custom_vbt_version;
 } __attribute__((packed));
 
-extern bool psb_intel_init_bios(struct drm_device *dev);
+extern int psb_intel_init_bios(struct drm_device *dev);
 extern void psb_intel_destroy_bios(struct drm_device *dev);
 
 /*
diff --git a/drivers/gpu/drm/gma500/mdfld_dsi_dpi.c b/drivers/gpu/drm/gma500/mdfld_dsi_dpi.c
index b34ff097b979..d4813e03f5ee 100644
--- a/drivers/gpu/drm/gma500/mdfld_dsi_dpi.c
+++ b/drivers/gpu/drm/gma500/mdfld_dsi_dpi.c
@@ -684,7 +684,7 @@ void mdfld_dsi_dpi_dpms(struct drm_encoder *encoder, int mode)
 }
 
 bool mdfld_dsi_dpi_mode_fixup(struct drm_encoder *encoder,
-                                     struct drm_display_mode *mode,
+                                     const struct drm_display_mode *mode,
                                      struct drm_display_mode *adjusted_mode)
 {
         struct mdfld_dsi_encoder *dsi_encoder = mdfld_dsi_encoder(encoder);
diff --git a/drivers/gpu/drm/gma500/mdfld_dsi_dpi.h b/drivers/gpu/drm/gma500/mdfld_dsi_dpi.h
index 6f762478b959..2b40663e1696 100644
--- a/drivers/gpu/drm/gma500/mdfld_dsi_dpi.h
+++ b/drivers/gpu/drm/gma500/mdfld_dsi_dpi.h
@@ -65,7 +65,7 @@ extern struct mdfld_dsi_encoder *mdfld_dsi_dpi_init(struct drm_device *dev,
 /* MDFLD DPI helper functions */
 extern void mdfld_dsi_dpi_dpms(struct drm_encoder *encoder, int mode);
 extern bool mdfld_dsi_dpi_mode_fixup(struct drm_encoder *encoder,
-                                struct drm_display_mode *mode,
+                                const struct drm_display_mode *mode,
                                 struct drm_display_mode *adjusted_mode);
 extern void mdfld_dsi_dpi_prepare(struct drm_encoder *encoder);
 extern void mdfld_dsi_dpi_commit(struct drm_encoder *encoder);
diff --git a/drivers/gpu/drm/gma500/mdfld_intel_display.c b/drivers/gpu/drm/gma500/mdfld_intel_display.c
index 3f3cd619c79f..dec6a9aea3c6 100644
--- a/drivers/gpu/drm/gma500/mdfld_intel_display.c
+++ b/drivers/gpu/drm/gma500/mdfld_intel_display.c
@@ -117,7 +117,7 @@ static void psb_intel_crtc_commit(struct drm_crtc *crtc)
 }
 
 static bool psb_intel_crtc_mode_fixup(struct drm_crtc *crtc,
-                                  struct drm_display_mode *mode,
+                                  const struct drm_display_mode *mode,
                                   struct drm_display_mode *adjusted_mode)
 {
         return true;
diff --git a/drivers/gpu/drm/gma500/oaktrail_crtc.c b/drivers/gpu/drm/gma500/oaktrail_crtc.c
index f821c835ca90..cdafd2acc72f 100644
--- a/drivers/gpu/drm/gma500/oaktrail_crtc.c
+++ b/drivers/gpu/drm/gma500/oaktrail_crtc.c
@@ -487,7 +487,7 @@ oaktrail_crtc_mode_set_exit:
 }
 
 static bool oaktrail_crtc_mode_fixup(struct drm_crtc *crtc,
-                                  struct drm_display_mode *mode,
+                                  const struct drm_display_mode *mode,
                                   struct drm_display_mode *adjusted_mode)
 {
         return true;
diff --git a/drivers/gpu/drm/gma500/oaktrail_hdmi.c b/drivers/gpu/drm/gma500/oaktrail_hdmi.c
index c10899c953b9..2eb3dc4e9c9b 100644
--- a/drivers/gpu/drm/gma500/oaktrail_hdmi.c
+++ b/drivers/gpu/drm/gma500/oaktrail_hdmi.c
@@ -191,7 +191,7 @@ static int oaktrail_hdmi_mode_valid(struct drm_connector *connector,
 }
 
 static bool oaktrail_hdmi_mode_fixup(struct drm_encoder *encoder,
-                                 struct drm_display_mode *mode,
+                                 const struct drm_display_mode *mode,
                                  struct drm_display_mode *adjusted_mode)
 {
         return true;
diff --git a/drivers/gpu/drm/gma500/psb_drv.c b/drivers/gpu/drm/gma500/psb_drv.c
index caba6e08693c..cd1dd1b14c76 100644
--- a/drivers/gpu/drm/gma500/psb_drv.c
+++ b/drivers/gpu/drm/gma500/psb_drv.c
@@ -632,7 +632,6 @@ static struct drm_driver driver = {
         .open = psb_driver_open,
         .preclose = psb_driver_preclose,
         .postclose = psb_driver_close,
-        .reclaim_buffers = drm_core_reclaim_buffers,
 
         .gem_init_object = psb_gem_init_object,
         .gem_free_object = psb_gem_free_object,
diff --git a/drivers/gpu/drm/gma500/psb_intel_display.c b/drivers/gpu/drm/gma500/psb_intel_display.c
index 36c3c99612f6..30dc22a7156c 100644
--- a/drivers/gpu/drm/gma500/psb_intel_display.c
+++ b/drivers/gpu/drm/gma500/psb_intel_display.c
@@ -543,7 +543,7 @@ void psb_intel_encoder_destroy(struct drm_encoder *encoder)
 }
 
 static bool psb_intel_crtc_mode_fixup(struct drm_crtc *crtc,
-                                  struct drm_display_mode *mode,
+                                  const struct drm_display_mode *mode,
                                   struct drm_display_mode *adjusted_mode)
 {
         return true;
diff --git a/drivers/gpu/drm/gma500/psb_intel_drv.h b/drivers/gpu/drm/gma500/psb_intel_drv.h
index 2515f83248cb..ebe1a28f60e1 100644
--- a/drivers/gpu/drm/gma500/psb_intel_drv.h
+++ b/drivers/gpu/drm/gma500/psb_intel_drv.h
@@ -268,7 +268,7 @@ extern struct drm_framebuffer *psb_intel_framebuffer_create(struct drm_device
                                                         *mode_cmd,
                                                         void *mm_private);
 extern bool psb_intel_lvds_mode_fixup(struct drm_encoder *encoder,
-                                      struct drm_display_mode *mode,
+                                      const struct drm_display_mode *mode,
                                       struct drm_display_mode *adjusted_mode);
 extern int psb_intel_lvds_mode_valid(struct drm_connector *connector,
                                      struct drm_display_mode *mode);
diff --git a/drivers/gpu/drm/gma500/psb_intel_lvds.c b/drivers/gpu/drm/gma500/psb_intel_lvds.c
index c83f5b5d1057..37adc9edf974 100644
--- a/drivers/gpu/drm/gma500/psb_intel_lvds.c
+++ b/drivers/gpu/drm/gma500/psb_intel_lvds.c
@@ -375,7 +375,7 @@ int psb_intel_lvds_mode_valid(struct drm_connector *connector,
 }
 
 bool psb_intel_lvds_mode_fixup(struct drm_encoder *encoder,
-                                  struct drm_display_mode *mode,
+                                  const struct drm_display_mode *mode,
                                   struct drm_display_mode *adjusted_mode)
 {
         struct drm_device *dev = encoder->dev;
diff --git a/drivers/gpu/drm/gma500/psb_intel_sdvo.c b/drivers/gpu/drm/gma500/psb_intel_sdvo.c
index d39b15be7649..0466c7b985f8 100644
--- a/drivers/gpu/drm/gma500/psb_intel_sdvo.c
+++ b/drivers/gpu/drm/gma500/psb_intel_sdvo.c
@@ -901,7 +901,7 @@ static bool psb_intel_sdvo_set_tv_format(struct psb_intel_sdvo *psb_intel_sdvo)
 
 static bool
 psb_intel_sdvo_set_output_timings_from_mode(struct psb_intel_sdvo *psb_intel_sdvo,
-                                        struct drm_display_mode *mode)
+                                        const struct drm_display_mode *mode)
 {
         struct psb_intel_sdvo_dtd output_dtd;
 
@@ -918,7 +918,7 @@ psb_intel_sdvo_set_output_timings_from_mode(struct psb_intel_sdvo *psb_intel_sdv
 
 static bool
 psb_intel_sdvo_set_input_timings_for_mode(struct psb_intel_sdvo *psb_intel_sdvo,
-                                        struct drm_display_mode *mode,
+                                        const struct drm_display_mode *mode,
                                         struct drm_display_mode *adjusted_mode)
 {
         /* Reset the input timing to the screen. Assume always input 0. */
@@ -942,7 +942,7 @@ psb_intel_sdvo_set_input_timings_for_mode(struct psb_intel_sdvo *psb_intel_sdvo,
 }
 
 static bool psb_intel_sdvo_mode_fixup(struct drm_encoder *encoder,
-                                  struct drm_display_mode *mode,
+                                  const struct drm_display_mode *mode,
                                   struct drm_display_mode *adjusted_mode)
 {
         struct psb_intel_sdvo *psb_intel_sdvo = to_psb_intel_sdvo(encoder);
diff --git a/drivers/gpu/drm/i2c/ch7006_drv.c b/drivers/gpu/drm/i2c/ch7006_drv.c
index d3f2e8785010..36d952280c50 100644
--- a/drivers/gpu/drm/i2c/ch7006_drv.c
+++ b/drivers/gpu/drm/i2c/ch7006_drv.c
@@ -88,7 +88,7 @@ static void ch7006_encoder_restore(struct drm_encoder *encoder)
 }
 
 static bool ch7006_encoder_mode_fixup(struct drm_encoder *encoder,
-                                      struct drm_display_mode *mode,
+                                      const struct drm_display_mode *mode,
                                       struct drm_display_mode *adjusted_mode)
 {
         struct ch7006_priv *priv = to_ch7006_priv(encoder);
diff --git a/drivers/gpu/drm/i2c/ch7006_mode.c b/drivers/gpu/drm/i2c/ch7006_mode.c
index c860f24a5afc..9b83574141a6 100644
--- a/drivers/gpu/drm/i2c/ch7006_mode.c
+++ b/drivers/gpu/drm/i2c/ch7006_mode.c
@@ -172,7 +172,7 @@ struct ch7006_mode ch7006_modes[] = {
 };
 
 struct ch7006_mode *ch7006_lookup_mode(struct drm_encoder *encoder,
-                                       struct drm_display_mode *drm_mode)
+                                       const struct drm_display_mode *drm_mode)
 {
         struct ch7006_priv *priv = to_ch7006_priv(encoder);
         struct ch7006_mode *mode;
diff --git a/drivers/gpu/drm/i2c/ch7006_priv.h b/drivers/gpu/drm/i2c/ch7006_priv.h
index 17667b7d57e7..09599f4c0c9a 100644
--- a/drivers/gpu/drm/i2c/ch7006_priv.h
+++ b/drivers/gpu/drm/i2c/ch7006_priv.h
@@ -111,7 +111,7 @@ extern struct ch7006_tv_norm_info ch7006_tv_norms[];
 extern struct ch7006_mode ch7006_modes[];
 
 struct ch7006_mode *ch7006_lookup_mode(struct drm_encoder *encoder,
-                                       struct drm_display_mode *drm_mode);
+                                       const struct drm_display_mode *drm_mode);
 
 void ch7006_setup_levels(struct drm_encoder *encoder);
 void ch7006_setup_subcarrier(struct drm_encoder *encoder);
diff --git a/drivers/gpu/drm/i2c/sil164_drv.c b/drivers/gpu/drm/i2c/sil164_drv.c
index b7d45ab4ba69..30b8ae5e5c4a 100644
--- a/drivers/gpu/drm/i2c/sil164_drv.c
+++ b/drivers/gpu/drm/i2c/sil164_drv.c
@@ -254,7 +254,7 @@ sil164_encoder_restore(struct drm_encoder *encoder)
 
 static bool
 sil164_encoder_mode_fixup(struct drm_encoder *encoder,
-                          struct drm_display_mode *mode,
+                          const struct drm_display_mode *mode,
                           struct drm_display_mode *adjusted_mode)
 {
         return true;
diff --git a/drivers/gpu/drm/i810/i810_dma.c b/drivers/gpu/drm/i810/i810_dma.c
index fa9439159ebd..57d892eaaa6e 100644
--- a/drivers/gpu/drm/i810/i810_dma.c
+++ b/drivers/gpu/drm/i810/i810_dma.c
@@ -881,7 +881,7 @@ static int i810_flush_queue(struct drm_device *dev)
 }
 
 /* Must be called with the lock held */
-static void i810_reclaim_buffers(struct drm_device *dev,
+void i810_driver_reclaim_buffers(struct drm_device *dev,
                                  struct drm_file *file_priv)
 {
         struct drm_device_dma *dma = dev->dma;
@@ -1220,12 +1220,17 @@ void i810_driver_preclose(struct drm_device *dev, struct drm_file *file_priv)
                 if (dev_priv->page_flipping)
                         i810_do_cleanup_pageflip(dev);
         }
-}
 
-void i810_driver_reclaim_buffers_locked(struct drm_device *dev,
-                                        struct drm_file *file_priv)
-{
-        i810_reclaim_buffers(dev, file_priv);
+        if (file_priv->master && file_priv->master->lock.hw_lock) {
+                drm_idlelock_take(&file_priv->master->lock);
+                i810_driver_reclaim_buffers(dev, file_priv);
+                drm_idlelock_release(&file_priv->master->lock);
+        } else {
+                /* master disappeared, clean up stuff anyway and hope nothing
+                 * goes wrong */
+                i810_driver_reclaim_buffers(dev, file_priv);
+        }
+
 }
 
 int i810_driver_dma_quiescent(struct drm_device *dev)
diff --git a/drivers/gpu/drm/i810/i810_drv.c b/drivers/gpu/drm/i810/i810_drv.c
index ec12f7dc717a..f9924ad04d09 100644
--- a/drivers/gpu/drm/i810/i810_drv.c
+++ b/drivers/gpu/drm/i810/i810_drv.c
@@ -57,13 +57,12 @@ static const struct file_operations i810_driver_fops = {
 static struct drm_driver driver = {
         .driver_features =
             DRIVER_USE_AGP | DRIVER_REQUIRE_AGP | DRIVER_USE_MTRR |
-            DRIVER_HAVE_DMA | DRIVER_DMA_QUEUE,
+            DRIVER_HAVE_DMA,
         .dev_priv_size = sizeof(drm_i810_buf_priv_t),
         .load = i810_driver_load,
         .lastclose = i810_driver_lastclose,
         .preclose = i810_driver_preclose,
         .device_is_agp = i810_driver_device_is_agp,
-        .reclaim_buffers_locked = i810_driver_reclaim_buffers_locked,
         .dma_quiescent = i810_driver_dma_quiescent,
         .ioctls = i810_ioctls,
         .fops = &i810_driver_fops,
diff --git a/drivers/gpu/drm/i810/i810_drv.h b/drivers/gpu/drm/i810/i810_drv.h
index c9339f481795..6e0acad9e0f5 100644
--- a/drivers/gpu/drm/i810/i810_drv.h
+++ b/drivers/gpu/drm/i810/i810_drv.h
@@ -116,14 +116,12 @@ typedef struct drm_i810_private {
 
                                 /* i810_dma.c */
 extern int i810_driver_dma_quiescent(struct drm_device *dev);
-extern void i810_driver_reclaim_buffers_locked(struct drm_device *dev,
-                                               struct drm_file *file_priv);
+void i810_driver_reclaim_buffers(struct drm_device *dev,
+                                 struct drm_file *file_priv);
 extern int i810_driver_load(struct drm_device *, unsigned long flags);
 extern void i810_driver_lastclose(struct drm_device *dev);
 extern void i810_driver_preclose(struct drm_device *dev,
                                  struct drm_file *file_priv);
-extern void i810_driver_reclaim_buffers_locked(struct drm_device *dev,
-                                               struct drm_file *file_priv);
 extern int i810_driver_device_is_agp(struct drm_device *dev);
 
 extern long i810_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index 2e9268da58d8..b0bacdba6d7e 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -7,6 +7,7 @@ i915-y := i915_drv.o i915_dma.o i915_irq.o \
           i915_debugfs.o \
           i915_suspend.o \
           i915_gem.o \
+          i915_gem_context.o \
           i915_gem_debug.o \
           i915_gem_evict.o \
           i915_gem_execbuffer.o \
diff --git a/drivers/gpu/drm/i915/dvo.h b/drivers/gpu/drm/i915/dvo.h
index 8c2ad014c47f..58914691a77b 100644
--- a/drivers/gpu/drm/i915/dvo.h
+++ b/drivers/gpu/drm/i915/dvo.h
@@ -86,7 +86,7 @@ struct intel_dvo_dev_ops {
          * buses with clock limitations.
          */
         bool (*mode_fixup)(struct intel_dvo_device *dvo,
-                           struct drm_display_mode *mode,
+                           const struct drm_display_mode *mode,
                            struct drm_display_mode *adjusted_mode);
 
         /*
diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 5363e9c66c27..2909b123baf5 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -713,6 +713,7 @@ static int i915_error_state(struct seq_file *m, void *unused)
         seq_printf(m, "EIR: 0x%08x\n", error->eir);
         seq_printf(m, "IER: 0x%08x\n", error->ier);
         seq_printf(m, "PGTBL_ER: 0x%08x\n", error->pgtbl_er);
+        seq_printf(m, "CCID: 0x%08x\n", error->ccid);
 
         for (i = 0; i < dev_priv->num_fence_regs; i++)
                 seq_printf(m, "  fence[%d] = %08llx\n", i, error->fence[i]);
@@ -1765,6 +1766,64 @@ static const struct file_operations i915_max_freq_fops = {
 };
 
 static ssize_t
+i915_min_freq_read(struct file *filp, char __user *ubuf, size_t max,
+                   loff_t *ppos)
+{
+        struct drm_device *dev = filp->private_data;
+        drm_i915_private_t *dev_priv = dev->dev_private;
+        char buf[80];
+        int len;
+
+        len = snprintf(buf, sizeof(buf),
+                       "min freq: %d\n", dev_priv->min_delay * 50);
+
+        if (len > sizeof(buf))
+                len = sizeof(buf);
+
+        return simple_read_from_buffer(ubuf, max, ppos, buf, len);
+}
+
+static ssize_t
+i915_min_freq_write(struct file *filp, const char __user *ubuf, size_t cnt,
+                    loff_t *ppos)
+{
+        struct drm_device *dev = filp->private_data;
+        struct drm_i915_private *dev_priv = dev->dev_private;
+        char buf[20];
+        int val = 1;
+
+        if (cnt > 0) {
+                if (cnt > sizeof(buf) - 1)
+                        return -EINVAL;
+
+                if (copy_from_user(buf, ubuf, cnt))
+                        return -EFAULT;
+                buf[cnt] = 0;
+
+                val = simple_strtoul(buf, NULL, 0);
+        }
+
+        DRM_DEBUG_DRIVER("Manually setting min freq to %d\n", val);
+
+        /*
+         * Turbo will still be enabled, but won't go below the set value.
+         */
+        dev_priv->min_delay = val / 50;
+
+        gen6_set_rps(dev, val / 50);
+
+        return cnt;
+}
+
+static const struct file_operations i915_min_freq_fops = {
+        .owner = THIS_MODULE,
+        .open = simple_open,
+        .read = i915_min_freq_read,
+        .write = i915_min_freq_write,
+        .llseek = default_llseek,
+};
+
+static ssize_t
 i915_cache_sharing_read(struct file *filp,
                    char __user *ubuf,
                    size_t max,
@@ -1997,6 +2056,12 @@ int i915_debugfs_init(struct drm_minor *minor)
                 return ret;
 
         ret = i915_debugfs_create(minor->debugfs_root, minor,
+                                  "i915_min_freq",
+                                  &i915_min_freq_fops);
+        if (ret)
+                return ret;
+
+        ret = i915_debugfs_create(minor->debugfs_root, minor,
                                   "i915_cache_sharing",
                                   &i915_cache_sharing_fops);
         if (ret)
@@ -2028,6 +2093,8 @@ void i915_debugfs_cleanup(struct drm_minor *minor)
                                  1, minor);
         drm_debugfs_remove_files((struct drm_info_list *) &i915_max_freq_fops,
                                  1, minor);
+        drm_debugfs_remove_files((struct drm_info_list *) &i915_min_freq_fops,
+                                 1, minor);
         drm_debugfs_remove_files((struct drm_info_list *) &i915_cache_sharing_fops,
                                  1, minor);
         drm_debugfs_remove_files((struct drm_info_list *) &i915_ring_stop_fops,
diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index 36822b924eb1..f64ef4b723fd 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -1006,6 +1006,9 @@ static int i915_getparam(struct drm_device *dev, void *data,
         case I915_PARAM_HAS_ALIASING_PPGTT:
                 value = dev_priv->mm.aliasing_ppgtt ? 1 : 0;
                 break;
+        case I915_PARAM_HAS_WAIT_TIMEOUT:
+                value = 1;
+                break;
         default:
                 DRM_DEBUG_DRIVER("Unknown parameter %d\n",
                                  param->param);
@@ -1082,8 +1085,8 @@ static int i915_set_status_page(struct drm_device *dev, void *data,
 
         ring->status_page.gfx_addr = hws->addr & (0x1ffff<<12);
 
-        dev_priv->dri1.gfx_hws_cpu_addr = ioremap_wc(dev->agp->base + hws->addr,
-                                                     4096);
+        dev_priv->dri1.gfx_hws_cpu_addr =
+                ioremap_wc(dev_priv->mm.gtt_base_addr + hws->addr, 4096);
         if (dev_priv->dri1.gfx_hws_cpu_addr == NULL) {
                 i915_dma_cleanup(dev);
                 ring->status_page.gfx_addr = 0;
@@ -1411,7 +1414,7 @@ static void i915_kick_out_firmware_fb(struct drm_i915_private *dev_priv)
         if (!ap)
                 return;
 
-        ap->ranges[0].base = dev_priv->dev->agp->base;
+        ap->ranges[0].base = dev_priv->mm.gtt->gma_bus_addr;
         ap->ranges[0].size =
                 dev_priv->mm.gtt->gtt_mappable_entries << PAGE_SHIFT;
         primary =
@@ -1467,11 +1470,18 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
                 goto free_priv;
         }
 
+        ret = intel_gmch_probe(dev_priv->bridge_dev, dev->pdev, NULL);
+        if (!ret) {
+                DRM_ERROR("failed to set up gmch\n");
+                ret = -EIO;
+                goto put_bridge;
+        }
+
         dev_priv->mm.gtt = intel_gtt_get();
         if (!dev_priv->mm.gtt) {
                 DRM_ERROR("Failed to initialize GTT\n");
                 ret = -ENODEV;
-                goto put_bridge;
+                goto put_gmch;
         }
 
         i915_kick_out_firmware_fb(dev_priv);
@@ -1498,19 +1508,22 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
         if (!dev_priv->regs) {
                 DRM_ERROR("failed to map registers\n");
                 ret = -EIO;
-                goto put_bridge;
+                goto put_gmch;
         }
 
         aperture_size = dev_priv->mm.gtt->gtt_mappable_entries << PAGE_SHIFT;
+        dev_priv->mm.gtt_base_addr = dev_priv->mm.gtt->gma_bus_addr;
 
         dev_priv->mm.gtt_mapping =
-                io_mapping_create_wc(dev->agp->base, aperture_size);
+                io_mapping_create_wc(dev_priv->mm.gtt_base_addr,
+                                     aperture_size);
         if (dev_priv->mm.gtt_mapping == NULL) {
                 ret = -EIO;
                 goto out_rmmap;
         }
 
-        i915_mtrr_setup(dev_priv, dev->agp->base, aperture_size);
+        i915_mtrr_setup(dev_priv, dev_priv->mm.gtt_base_addr,
+                        aperture_size);
 
         /* The i915 workqueue is primarily used for batched retirement of
          * requests (and thus managing bo) once the task has been completed
@@ -1534,7 +1547,11 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
                 goto out_mtrrfree;
         }
 
+        /* This must be called before any calls to HAS_PCH_* */
+        intel_detect_pch(dev);
+
         intel_irq_init(dev);
+        intel_gt_init(dev);
 
         /* Try to make sure MCHBAR is enabled before poking at it */
         intel_setup_mchbar(dev);
@@ -1567,7 +1584,6 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
         if (!IS_I945G(dev) && !IS_I945GM(dev))
                 pci_enable_msi(dev->pdev);
 
-        spin_lock_init(&dev_priv->gt_lock);
         spin_lock_init(&dev_priv->irq_lock);
         spin_lock_init(&dev_priv->error_lock);
         spin_lock_init(&dev_priv->rps_lock);
@@ -1586,8 +1602,6 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
         /* Start out suspended */
         dev_priv->mm.suspended = 1;
 
-        intel_detect_pch(dev);
-
         if (drm_core_check_feature(dev, DRIVER_MODESET)) {
                 ret = i915_load_modeset_init(dev);
                 if (ret < 0) {
@@ -1622,13 +1636,16 @@ out_gem_unload:
         destroy_workqueue(dev_priv->wq);
 out_mtrrfree:
         if (dev_priv->mm.gtt_mtrr >= 0) {
-                mtrr_del(dev_priv->mm.gtt_mtrr, dev->agp->base,
-                         dev->agp->agp_info.aper_size * 1024 * 1024);
+                mtrr_del(dev_priv->mm.gtt_mtrr,
+                         dev_priv->mm.gtt_base_addr,
+                         aperture_size);
                 dev_priv->mm.gtt_mtrr = -1;
         }
         io_mapping_free(dev_priv->mm.gtt_mapping);
 out_rmmap:
         pci_iounmap(dev->pdev, dev_priv->regs);
+put_gmch:
+        intel_gmch_remove();
 put_bridge:
         pci_dev_put(dev_priv->bridge_dev);
 free_priv:
@@ -1660,8 +1677,9 @@ int i915_driver_unload(struct drm_device *dev)
 
         io_mapping_free(dev_priv->mm.gtt_mapping);
         if (dev_priv->mm.gtt_mtrr >= 0) {
-                mtrr_del(dev_priv->mm.gtt_mtrr, dev->agp->base,
-                         dev->agp->agp_info.aper_size * 1024 * 1024);
+                mtrr_del(dev_priv->mm.gtt_mtrr,
+                         dev_priv->mm.gtt_base_addr,
+                         dev_priv->mm.gtt->gtt_mappable_entries * PAGE_SIZE);
                 dev_priv->mm.gtt_mtrr = -1;
         }
 
@@ -1702,6 +1720,7 @@ int i915_driver_unload(struct drm_device *dev)
                 mutex_lock(&dev->struct_mutex);
                 i915_gem_free_all_phys_object(dev);
                 i915_gem_cleanup_ringbuffer(dev);
+                i915_gem_context_fini(dev);
                 mutex_unlock(&dev->struct_mutex);
                 i915_gem_cleanup_aliasing_ppgtt(dev);
                 i915_gem_cleanup_stolen(dev);
@@ -1741,6 +1760,8 @@ int i915_driver_open(struct drm_device *dev, struct drm_file *file)
         spin_lock_init(&file_priv->mm.lock);
         INIT_LIST_HEAD(&file_priv->mm.request_list);
 
+        idr_init(&file_priv->context_idr);
+
         return 0;
 }
 
@@ -1773,6 +1794,7 @@ void i915_driver_lastclose(struct drm_device * dev)
 
 void i915_driver_preclose(struct drm_device * dev, struct drm_file *file_priv)
 {
+        i915_gem_context_close(dev, file_priv);
         i915_gem_release(dev, file_priv);
 }
 
@@ -1826,6 +1848,9 @@ struct drm_ioctl_desc i915_ioctls[] = {
         DRM_IOCTL_DEF_DRV(I915_OVERLAY_ATTRS, intel_overlay_attrs, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED),
         DRM_IOCTL_DEF_DRV(I915_SET_SPRITE_COLORKEY, intel_sprite_set_colorkey, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED),
         DRM_IOCTL_DEF_DRV(I915_GET_SPRITE_COLORKEY, intel_sprite_get_colorkey, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED),
+        DRM_IOCTL_DEF_DRV(I915_GEM_WAIT, i915_gem_wait_ioctl, DRM_AUTH|DRM_UNLOCKED),
+        DRM_IOCTL_DEF_DRV(I915_GEM_CONTEXT_CREATE, i915_gem_context_create_ioctl, DRM_UNLOCKED),
+        DRM_IOCTL_DEF_DRV(I915_GEM_CONTEXT_DESTROY, i915_gem_context_destroy_ioctl, DRM_UNLOCKED),
 };
 
 int i915_max_ioctl = DRM_ARRAY_SIZE(i915_ioctls);
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 9fe9ebe52a7a..ed22612bc847 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -32,6 +32,7 @@
 #include "drm.h"
 #include "i915_drm.h"
 #include "i915_drv.h"
+#include "i915_trace.h"
 #include "intel_drv.h"
 
 #include <linux/console.h>
@@ -215,7 +216,6 @@ static const struct intel_device_info intel_ironlake_d_info = {
         .gen = 5,
         .need_gfx_hws = 1, .has_hotplug = 1,
         .has_bsd_ring = 1,
-        .has_pch_split = 1,
 };
 
 static const struct intel_device_info intel_ironlake_m_info = {
@@ -223,7 +223,6 @@ static const struct intel_device_info intel_ironlake_m_info = {
         .need_gfx_hws = 1, .has_hotplug = 1,
         .has_fbc = 1,
         .has_bsd_ring = 1,
-        .has_pch_split = 1,
 };
 
 static const struct intel_device_info intel_sandybridge_d_info = {
@@ -232,7 +231,6 @@ static const struct intel_device_info intel_sandybridge_d_info = {
         .has_bsd_ring = 1,
         .has_blt_ring = 1,
         .has_llc = 1,
-        .has_pch_split = 1,
         .has_force_wake = 1,
 };
 
@@ -243,7 +241,6 @@ static const struct intel_device_info intel_sandybridge_m_info = {
         .has_bsd_ring = 1,
         .has_blt_ring = 1,
         .has_llc = 1,
-        .has_pch_split = 1,
         .has_force_wake = 1,
 };
 
@@ -253,7 +250,6 @@ static const struct intel_device_info intel_ivybridge_d_info = {
         .has_bsd_ring = 1,
         .has_blt_ring = 1,
         .has_llc = 1,
-        .has_pch_split = 1,
         .has_force_wake = 1,
 };
 
@@ -264,7 +260,6 @@ static const struct intel_device_info intel_ivybridge_m_info = {
         .has_bsd_ring = 1,
         .has_blt_ring = 1,
         .has_llc = 1,
-        .has_pch_split = 1,
         .has_force_wake = 1,
 };
 
@@ -292,7 +287,6 @@ static const struct intel_device_info intel_haswell_d_info = {
         .has_bsd_ring = 1,
         .has_blt_ring = 1,
         .has_llc = 1,
-        .has_pch_split = 1,
         .has_force_wake = 1,
 };
 
@@ -302,7 +296,6 @@ static const struct intel_device_info intel_haswell_m_info = {
         .has_bsd_ring = 1,
         .has_blt_ring = 1,
         .has_llc = 1,
-        .has_pch_split = 1,
         .has_force_wake = 1,
 };
 
@@ -358,6 +351,9 @@ static const struct pci_device_id pciidlist[] = {		/* aka */
         INTEL_VGA_DEVICE(0x0406, &intel_haswell_m_info), /* GT1 mobile */
         INTEL_VGA_DEVICE(0x0416, &intel_haswell_m_info), /* GT2 mobile */
         INTEL_VGA_DEVICE(0x0c16, &intel_haswell_d_info), /* SDV */
+        INTEL_VGA_DEVICE(0x0f30, &intel_valleyview_m_info),
+        INTEL_VGA_DEVICE(0x0157, &intel_valleyview_m_info),
+        INTEL_VGA_DEVICE(0x0155, &intel_valleyview_d_info),
         {0, 0, 0}
 };
 
@@ -429,135 +425,6 @@ bool i915_semaphore_is_enabled(struct drm_device *dev)
         return 1;
 }
 
-void __gen6_gt_force_wake_get(struct drm_i915_private *dev_priv)
-{
-        int count;
-
-        count = 0;
-        while (count++ < 50 && (I915_READ_NOTRACE(FORCEWAKE_ACK) & 1))
-                udelay(10);
-
-        I915_WRITE_NOTRACE(FORCEWAKE, 1);
-        POSTING_READ(FORCEWAKE);
-
-        count = 0;
-        while (count++ < 50 && (I915_READ_NOTRACE(FORCEWAKE_ACK) & 1) == 0)
-                udelay(10);
-}
-
-void __gen6_gt_force_wake_mt_get(struct drm_i915_private *dev_priv)
-{
-        int count;
-
-        count = 0;
-        while (count++ < 50 && (I915_READ_NOTRACE(FORCEWAKE_MT_ACK) & 1))
-                udelay(10);
-
-        I915_WRITE_NOTRACE(FORCEWAKE_MT, _MASKED_BIT_ENABLE(1));
-        POSTING_READ(FORCEWAKE_MT);
-
-        count = 0;
-        while (count++ < 50 && (I915_READ_NOTRACE(FORCEWAKE_MT_ACK) & 1) == 0)
-                udelay(10);
-}
-
-/*
- * Generally this is called implicitly by the register read function. However,
- * if some sequence requires the GT to not power down then this function should
- * be called at the beginning of the sequence followed by a call to
- * gen6_gt_force_wake_put() at the end of the sequence.
- */
-void gen6_gt_force_wake_get(struct drm_i915_private *dev_priv)
-{
-        unsigned long irqflags;
-
-        spin_lock_irqsave(&dev_priv->gt_lock, irqflags);
-        if (dev_priv->forcewake_count++ == 0)
-                dev_priv->display.force_wake_get(dev_priv);
-        spin_unlock_irqrestore(&dev_priv->gt_lock, irqflags);
-}
-
-static void gen6_gt_check_fifodbg(struct drm_i915_private *dev_priv)
-{
-        u32 gtfifodbg;
-        gtfifodbg = I915_READ_NOTRACE(GTFIFODBG);
-        if (WARN(gtfifodbg & GT_FIFO_CPU_ERROR_MASK,
-             "MMIO read or write has been dropped %x\n", gtfifodbg))
-                I915_WRITE_NOTRACE(GTFIFODBG, GT_FIFO_CPU_ERROR_MASK);
-}
-
-void __gen6_gt_force_wake_put(struct drm_i915_private *dev_priv)
-{
-        I915_WRITE_NOTRACE(FORCEWAKE, 0);
-        /* The below doubles as a POSTING_READ */
-        gen6_gt_check_fifodbg(dev_priv);
-}
-
-void __gen6_gt_force_wake_mt_put(struct drm_i915_private *dev_priv)
-{
-        I915_WRITE_NOTRACE(FORCEWAKE_MT, _MASKED_BIT_DISABLE(1));
-        /* The below doubles as a POSTING_READ */
-        gen6_gt_check_fifodbg(dev_priv);
-}
-
-/*
- * see gen6_gt_force_wake_get()
- */
-void gen6_gt_force_wake_put(struct drm_i915_private *dev_priv)
-{
-        unsigned long irqflags;
-
-        spin_lock_irqsave(&dev_priv->gt_lock, irqflags);
-        if (--dev_priv->forcewake_count == 0)
-                dev_priv->display.force_wake_put(dev_priv);
-        spin_unlock_irqrestore(&dev_priv->gt_lock, irqflags);
-}
-
-int __gen6_gt_wait_for_fifo(struct drm_i915_private *dev_priv)
-{
-        int ret = 0;
-
-        if (dev_priv->gt_fifo_count < GT_FIFO_NUM_RESERVED_ENTRIES) {
-                int loop = 500;
-                u32 fifo = I915_READ_NOTRACE(GT_FIFO_FREE_ENTRIES);
-                while (fifo <= GT_FIFO_NUM_RESERVED_ENTRIES && loop--) {
-                        udelay(10);
-                        fifo = I915_READ_NOTRACE(GT_FIFO_FREE_ENTRIES);
-                }
-                if (WARN_ON(loop < 0 && fifo <= GT_FIFO_NUM_RESERVED_ENTRIES))
-                        ++ret;
-                dev_priv->gt_fifo_count = fifo;
-        }
-        dev_priv->gt_fifo_count--;
-
-        return ret;
-}
-
-void vlv_force_wake_get(struct drm_i915_private *dev_priv)
-{
-        int count;
-
-        count = 0;
-
-        /* Already awake? */
-        if ((I915_READ(0x130094) & 0xa1) == 0xa1)
-                return;
-
-        I915_WRITE_NOTRACE(FORCEWAKE_VLV, 0xffffffff);
-        POSTING_READ(FORCEWAKE_VLV);
-
-        count = 0;
-        while (count++ < 50 && (I915_READ_NOTRACE(FORCEWAKE_ACK_VLV) & 1) == 0)
-                udelay(10);
-}
-
-void vlv_force_wake_put(struct drm_i915_private *dev_priv)
-{
-        I915_WRITE_NOTRACE(FORCEWAKE_VLV, 0xffff0000);
-        /* FIXME: confirm VLV behavior with Punit folks */
-        POSTING_READ(FORCEWAKE_VLV);
-}
-
 static int i915_drm_freeze(struct drm_device *dev)
 {
         struct drm_i915_private *dev_priv = dev->dev_private;
@@ -637,7 +504,7 @@ static int i915_drm_thaw(struct drm_device *dev)
 
         /* KMS EnterVT equivalent */
         if (drm_core_check_feature(dev, DRIVER_MODESET)) {
-                if (HAS_PCH_SPLIT(dev))
+                if (HAS_PCH_IBX(dev) || HAS_PCH_CPT(dev))
                         ironlake_init_pch_refclk(dev);
 
                 mutex_lock(&dev->struct_mutex);
@@ -794,9 +661,9 @@ static int gen6_do_reset(struct drm_device *dev)
 
         /* If reset with a user forcewake, try to restore, otherwise turn it off */
         if (dev_priv->forcewake_count)
-                dev_priv->display.force_wake_get(dev_priv);
+                dev_priv->gt.force_wake_get(dev_priv);
         else
-                dev_priv->display.force_wake_put(dev_priv);
+                dev_priv->gt.force_wake_put(dev_priv);
 
         /* Restore fifo count */
         dev_priv->gt_fifo_count = I915_READ_NOTRACE(GT_FIFO_FREE_ENTRIES);
@@ -805,7 +672,7 @@ static int gen6_do_reset(struct drm_device *dev)
         return ret;
 }
 
-static int intel_gpu_reset(struct drm_device *dev)
+int intel_gpu_reset(struct drm_device *dev)
 {
         struct drm_i915_private *dev_priv = dev->dev_private;
         int ret = -ENODEV;
@@ -863,10 +730,7 @@ int i915_reset(struct drm_device *dev)
         if (!i915_try_reset)
                 return 0;
 
-        if (!mutex_trylock(&dev->struct_mutex))
-                return -EBUSY;
-
-        dev_priv->stop_rings = 0;
+        mutex_lock(&dev->struct_mutex);
 
         i915_gem_reset(dev);
 
@@ -909,12 +773,16 @@ int i915_reset(struct drm_device *dev)
                 for_each_ring(ring, dev_priv, i)
                         ring->init(ring);
 
+                i915_gem_context_init(dev);
                 i915_gem_init_ppgtt(dev);
 
-                mutex_unlock(&dev->struct_mutex);
+                /*
+                 * It would make sense to re-init all the other hw state, at
+                 * least the rps/rc6/emon init done within modeset_init_hw. For
+                 * some unknown reason, this blows up my ilk, so don't.
+                 */
 
-                if (drm_core_check_feature(dev, DRIVER_MODESET))
-                        intel_modeset_init_hw(dev);
+                mutex_unlock(&dev->struct_mutex);
 
                 drm_irq_uninstall(dev);
                 drm_irq_install(dev);
@@ -925,10 +793,12 @@ int i915_reset(struct drm_device *dev)
         return 0;
 }
 
-
 static int __devinit
 i915_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
+        struct intel_device_info *intel_info =
+                (struct intel_device_info *) ent->driver_data;
+
         /* Only bind to function 0 of the device. Early generations
          * used function 1 as a placeholder for multi-head. This causes
          * us confusion instead, especially on the systems where both
@@ -937,6 +807,18 @@ i915_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
         if (PCI_FUNC(pdev->devfn))
                 return -ENODEV;
 
+        /* We've managed to ship a kms-enabled ddx that shipped with an XvMC
+         * implementation for gen3 (and only gen3) that used legacy drm maps
+         * (gasp!) to share buffers between X and the client. Hence we need to
+         * keep around the fake agp stuff for gen3, even when kms is enabled. */
+        if (intel_info->gen != 3) {
+                driver.driver_features &=
+                        ~(DRIVER_USE_AGP | DRIVER_REQUIRE_AGP);
+        } else if (!intel_agp_enabled) {
+                DRM_ERROR("drm/i915 can't work without intel_agp module!\n");
+                return -ENODEV;
+        }
+
         return drm_get_pci_dev(pdev, ent, &driver);
 }
 
@@ -1058,7 +940,6 @@ static struct drm_driver driver = {
         .resume = i915_resume,
 
         .device_is_agp = i915_driver_device_is_agp,
-        .reclaim_buffers = drm_core_reclaim_buffers,
         .master_create = i915_master_create,
         .master_destroy = i915_master_destroy,
 #if defined(CONFIG_DEBUG_FS)
@@ -1097,11 +978,6 @@ static struct pci_driver i915_pci_driver = {
 
 static int __init i915_init(void)
 {
-        if (!intel_agp_enabled) {
-                DRM_ERROR("drm/i915 can't work without intel_agp module!\n");
-                return -ENODEV;
-        }
-
         driver.num_ioctls = i915_max_ioctl;
 
         /*
@@ -1149,6 +1025,84 @@ MODULE_LICENSE("GPL and additional rights");
          ((reg) < 0x40000) &&            \
          ((reg) != FORCEWAKE))
 
+static bool IS_DISPLAYREG(u32 reg)
+{
+        /*
+         * This should make it easier to transition modules over to the
+         * new register block scheme, since we can do it incrementally.
+         */
+        if (reg >= 0x180000)
+                return false;
+
+        if (reg >= RENDER_RING_BASE &&
+            reg < RENDER_RING_BASE + 0xff)
+                return false;
+        if (reg >= GEN6_BSD_RING_BASE &&
+            reg < GEN6_BSD_RING_BASE + 0xff)
+                return false;
+        if (reg >= BLT_RING_BASE &&
+            reg < BLT_RING_BASE + 0xff)
+                return false;
+
+        if (reg == PGTBL_ER)
+                return false;
+
+        if (reg >= IPEIR_I965 &&
+            reg < HWSTAM)
+                return false;
+
+        if (reg == MI_MODE)
+                return false;
+
+        if (reg == GFX_MODE_GEN7)
+                return false;
+
+        if (reg == RENDER_HWS_PGA_GEN7 ||
+            reg == BSD_HWS_PGA_GEN7 ||
+            reg == BLT_HWS_PGA_GEN7)
+                return false;
+
+        if (reg == GEN6_BSD_SLEEP_PSMI_CONTROL ||
+            reg == GEN6_BSD_RNCID)
+                return false;
+
+        if (reg == GEN6_BLITTER_ECOSKPD)
+                return false;
+
+        if (reg >= 0x4000c &&
+            reg <= 0x4002c)
+                return false;
+
+        if (reg >= 0x4f000 &&
+            reg <= 0x4f08f)
+                return false;
+
+        if (reg >= 0x4f100 &&
+            reg <= 0x4f11f)
+                return false;
+
+        if (reg >= VLV_MASTER_IER &&
+            reg <= GEN6_PMIER)
+                return false;
+
+        if (reg >= FENCE_REG_SANDYBRIDGE_0 &&
+            reg < (FENCE_REG_SANDYBRIDGE_0 + (16*8)))
+                return false;
+
+        if (reg >= VLV_IIR_RW &&
+            reg <= VLV_ISR)
+                return false;
+
+        if (reg == FORCEWAKE_VLV ||
+            reg == FORCEWAKE_ACK_VLV)
+                return false;
+
+        if (reg == GEN6_GDRST)
+                return false;
+
+        return true;
+}
+
 #define __i915_read(x, y) \
 u##x i915_read##x(struct drm_i915_private *dev_priv, u32 reg) { \
         u##x val = 0; \
@@ -1156,11 +1110,13 @@ u##x i915_read##x(struct drm_i915_private *dev_priv, u32 reg) { \
                 unsigned long irqflags; \
                 spin_lock_irqsave(&dev_priv->gt_lock, irqflags); \
                 if (dev_priv->forcewake_count == 0) \
-                        dev_priv->display.force_wake_get(dev_priv); \
+                        dev_priv->gt.force_wake_get(dev_priv); \
                 val = read##y(dev_priv->regs + reg); \
                 if (dev_priv->forcewake_count == 0) \
-                        dev_priv->display.force_wake_put(dev_priv); \
+                        dev_priv->gt.force_wake_put(dev_priv); \
                 spin_unlock_irqrestore(&dev_priv->gt_lock, irqflags); \
+        } else if (IS_VALLEYVIEW(dev_priv->dev) && IS_DISPLAYREG(reg)) { \
+                val = read##y(dev_priv->regs + reg + 0x180000);         \
         } else { \
                 val = read##y(dev_priv->regs + reg); \
         } \
@@ -1181,7 +1137,11 @@ void i915_write##x(struct drm_i915_private *dev_priv, u32 reg, u##x val) { \
         if (NEEDS_FORCE_WAKE((dev_priv), (reg))) { \
                 __fifo_ret = __gen6_gt_wait_for_fifo(dev_priv); \
         } \
-        write##y(val, dev_priv->regs + reg); \
+        if (IS_VALLEYVIEW(dev_priv->dev) && IS_DISPLAYREG(reg)) { \
+                write##y(val, dev_priv->regs + reg + 0x180000);         \
+        } else {                                                        \
+                write##y(val, dev_priv->regs + reg);                    \
+        }                                                               \
         if (unlikely(__fifo_ret)) { \
                 gen6_gt_check_fifodbg(dev_priv); \
         } \
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index b0b676abde0d..476c64c4844c 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -79,6 +79,10 @@ enum port {
 
 #define for_each_pipe(p) for ((p) = 0; (p) < dev_priv->num_pipe; (p)++)
 
+#define for_each_encoder_on_crtc(dev, __crtc, intel_encoder) \
+        list_for_each_entry((intel_encoder), &(dev)->mode_config.encoder_list, base.head) \
+                if ((intel_encoder)->base.crtc == (__crtc))
+
 struct intel_pch_pll {
         int refcount; /* count of number of CRTCs sharing this PLL */
         int active; /* count of number of active CRTCs (i.e. DPMS on) */
@@ -176,6 +180,7 @@ struct drm_i915_error_state {
         u32 eir;
         u32 pgtbl_er;
         u32 ier;
+        u32 ccid;
         bool waiting[I915_NUM_RINGS];
         u32 pipestat[I915_MAX_PIPES];
         u32 tail[I915_NUM_RINGS];
@@ -261,8 +266,6 @@ struct drm_i915_display_funcs {
                           struct drm_i915_gem_object *obj);
         int (*update_plane)(struct drm_crtc *crtc, struct drm_framebuffer *fb,
                             int x, int y);
-        void (*force_wake_get)(struct drm_i915_private *dev_priv);
-        void (*force_wake_put)(struct drm_i915_private *dev_priv);
         /* clock updates for mode set */
         /* cursor updates */
         /* render clock increase/decrease */
@@ -270,6 +273,11 @@ struct drm_i915_display_funcs {
         /* pll clock increase/decrease */
 };
 
+struct drm_i915_gt_funcs {
+        void (*force_wake_get)(struct drm_i915_private *dev_priv);
+        void (*force_wake_put)(struct drm_i915_private *dev_priv);
+};
+
 struct intel_device_info {
         u8 gen;
         u8 is_mobile:1;
@@ -284,7 +292,6 @@ struct intel_device_info {
         u8 is_crestline:1;
         u8 is_ivybridge:1;
         u8 is_valleyview:1;
-        u8 has_pch_split:1;
         u8 has_force_wake:1;
         u8 is_haswell:1;
         u8 has_fbc:1;
@@ -309,6 +316,17 @@ struct i915_hw_ppgtt {
         dma_addr_t scratch_page_dma_addr;
 };
 
+
+/* This must match up with the value previously used for execbuf2.rsvd1. */
+#define DEFAULT_CONTEXT_ID 0
+struct i915_hw_context {
+        int id;
+        bool is_initialized;
+        struct drm_i915_file_private *file_priv;
+        struct intel_ring_buffer *ring;
+        struct drm_i915_gem_object *obj;
+};
+
 enum no_fbc_reason {
         FBC_NO_OUTPUT, /* no outputs enabled to compress */
         FBC_STOLEN_TOO_SMALL, /* not enough space to hold compressed buffers */
@@ -321,6 +339,7 @@ enum no_fbc_reason {
 };
 
 enum intel_pch {
+        PCH_NONE = 0,   /* No PCH present */
         PCH_IBX,        /* Ibexpeak PCH */
         PCH_CPT,        /* Cougarpoint PCH */
         PCH_LPT,        /* Lynxpoint PCH */
@@ -350,6 +369,8 @@ typedef struct drm_i915_private {
         int relative_constants_mode;
 
         void __iomem *regs;
+
+        struct drm_i915_gt_funcs gt;
         /** gt_fifo_count and the subsequent register write are synchronized
          * with dev->struct_mutex. */
         unsigned gt_fifo_count;
@@ -652,11 +673,14 @@ typedef struct drm_i915_private {
                 unsigned long gtt_end;
 
                 struct io_mapping *gtt_mapping;
+                phys_addr_t gtt_base_addr;
                 int gtt_mtrr;
 
                 /** PPGTT used for aliasing the PPGTT with the GTT */
                 struct i915_hw_ppgtt *aliasing_ppgtt;
 
+                u32 *l3_remap_info;
+
                 struct shrinker inactive_shrinker;
 
                 /**
@@ -817,6 +841,10 @@ typedef struct drm_i915_private {
 
         struct drm_property *broadcast_rgb_property;
         struct drm_property *force_audio_property;
+
+        struct work_struct parity_error_work;
+        bool hw_contexts_disabled;
+        uint32_t hw_context_size;
 } drm_i915_private_t;
 
 /* Iterate over initialised rings */
@@ -1026,6 +1054,7 @@ struct drm_i915_file_private {
                 struct spinlock lock;
                 struct list_head request_list;
         } mm;
+        struct idr context_idr;
 };
 
 #define INTEL_INFO(dev) (((struct drm_i915_private *) (dev)->dev_private)->info)
@@ -1071,7 +1100,8 @@ struct drm_i915_file_private {
 #define HAS_LLC(dev)            (INTEL_INFO(dev)->has_llc)
 #define I915_NEED_GFX_HWS(dev)  (INTEL_INFO(dev)->need_gfx_hws)
 
-#define HAS_ALIASING_PPGTT(dev) (INTEL_INFO(dev)->gen >=6)
+#define HAS_HW_CONTEXTS(dev)    (INTEL_INFO(dev)->gen >= 6)
+#define HAS_ALIASING_PPGTT(dev) (INTEL_INFO(dev)->gen >=6 && !IS_VALLEYVIEW(dev))
 
 #define HAS_OVERLAY(dev)                (INTEL_INFO(dev)->has_overlay)
 #define OVERLAY_NEEDS_PHYSICAL(dev)     (INTEL_INFO(dev)->overlay_needs_physical)
@@ -1094,13 +1124,13 @@ struct drm_i915_file_private {
 #define HAS_PIPE_CXSR(dev) (INTEL_INFO(dev)->has_pipe_cxsr)
 #define I915_HAS_FBC(dev) (INTEL_INFO(dev)->has_fbc)
 
-#define HAS_PCH_SPLIT(dev) (INTEL_INFO(dev)->has_pch_split)
 #define HAS_PIPE_CONTROL(dev) (INTEL_INFO(dev)->gen >= 5)
 
 #define INTEL_PCH_TYPE(dev) (((struct drm_i915_private *)(dev)->dev_private)->pch_type)
 #define HAS_PCH_LPT(dev) (INTEL_PCH_TYPE(dev) == PCH_LPT)
 #define HAS_PCH_CPT(dev) (INTEL_PCH_TYPE(dev) == PCH_CPT)
 #define HAS_PCH_IBX(dev) (INTEL_PCH_TYPE(dev) == PCH_IBX)
+#define HAS_PCH_SPLIT(dev) (INTEL_PCH_TYPE(dev) != PCH_NONE)
 
 #define HAS_FORCE_WAKE(dev) (INTEL_INFO(dev)->has_force_wake)
 
@@ -1166,6 +1196,7 @@ extern long i915_compat_ioctl(struct file *filp, unsigned int cmd,
 extern int i915_emit_box(struct drm_device *dev,
                          struct drm_clip_rect *box,
                          int DR1, int DR4);
+extern int intel_gpu_reset(struct drm_device *dev);
 extern int i915_reset(struct drm_device *dev);
 extern unsigned long i915_chipset_val(struct drm_i915_private *dev_priv);
 extern unsigned long i915_mch_val(struct drm_i915_private *dev_priv);
@@ -1178,6 +1209,7 @@ void i915_hangcheck_elapsed(unsigned long data);
 void i915_handle_error(struct drm_device *dev, bool wedged);
 
 extern void intel_irq_init(struct drm_device *dev);
+extern void intel_gt_init(struct drm_device *dev);
 
 void i915_error_state_free(struct kref *error_ref);
 
@@ -1237,6 +1269,8 @@ int i915_gem_get_tiling(struct drm_device *dev, void *data,
                         struct drm_file *file_priv);
 int i915_gem_get_aperture_ioctl(struct drm_device *dev, void *data,
                                 struct drm_file *file_priv);
+int i915_gem_wait_ioctl(struct drm_device *dev, void *data,
+                        struct drm_file *file_priv);
 void i915_gem_load(struct drm_device *dev);
 int i915_gem_init_object(struct drm_gem_object *obj);
 int __must_check i915_gem_flush_ring(struct intel_ring_buffer *ring,
@@ -1306,6 +1340,8 @@ i915_gem_object_unpin_fence(struct drm_i915_gem_object *obj)
 
 void i915_gem_retire_requests(struct drm_device *dev);
 void i915_gem_retire_requests_ring(struct intel_ring_buffer *ring);
+int __must_check i915_gem_check_wedge(struct drm_i915_private *dev_priv,
+                                      bool interruptible);
 
 void i915_gem_reset(struct drm_device *dev);
 void i915_gem_clflush_object(struct drm_i915_gem_object *obj);
@@ -1315,6 +1351,7 @@ int __must_check i915_gem_object_set_domain(struct drm_i915_gem_object *obj,
 int __must_check i915_gem_object_finish_gpu(struct drm_i915_gem_object *obj);
 int __must_check i915_gem_init(struct drm_device *dev);
 int __must_check i915_gem_init_hw(struct drm_device *dev);
+void i915_gem_l3_remap(struct drm_device *dev);
 void i915_gem_init_swizzling(struct drm_device *dev);
 void i915_gem_init_ppgtt(struct drm_device *dev);
 void i915_gem_cleanup_ringbuffer(struct drm_device *dev);
@@ -1323,8 +1360,8 @@ int __must_check i915_gem_idle(struct drm_device *dev);
 int __must_check i915_add_request(struct intel_ring_buffer *ring,
                                   struct drm_file *file,
                                   struct drm_i915_gem_request *request);
-int __must_check i915_wait_request(struct intel_ring_buffer *ring,
-                                   uint32_t seqno);
+int __must_check i915_wait_seqno(struct intel_ring_buffer *ring,
+                                 uint32_t seqno);
 int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf);
 int __must_check
 i915_gem_object_set_to_gtt_domain(struct drm_i915_gem_object *obj,
@@ -1358,6 +1395,16 @@ struct drm_gem_object *i915_gem_prime_import(struct drm_device *dev,
 struct dma_buf *i915_gem_prime_export(struct drm_device *dev,
                                 struct drm_gem_object *gem_obj, int flags);
 
+/* i915_gem_context.c */
+void i915_gem_context_init(struct drm_device *dev);
+void i915_gem_context_fini(struct drm_device *dev);
+void i915_gem_context_close(struct drm_device *dev, struct drm_file *file);
+int i915_switch_context(struct intel_ring_buffer *ring,
+                        struct drm_file *file, int to_id);
+int i915_gem_context_create_ioctl(struct drm_device *dev, void *data,
+                                  struct drm_file *file);
+int i915_gem_context_destroy_ioctl(struct drm_device *dev, void *data,
+                                   struct drm_file *file);
 
 /* i915_gem_gtt.c */
 int __must_check i915_gem_init_aliasing_ppgtt(struct drm_device *dev);
@@ -1475,20 +1522,12 @@ extern bool intel_fbc_enabled(struct drm_device *dev);
 extern void intel_disable_fbc(struct drm_device *dev);
 extern bool ironlake_set_drps(struct drm_device *dev, u8 val);
 extern void ironlake_init_pch_refclk(struct drm_device *dev);
-extern void ironlake_enable_rc6(struct drm_device *dev);
 extern void gen6_set_rps(struct drm_device *dev, u8 val);
 extern void intel_detect_pch(struct drm_device *dev);
 extern int intel_trans_dp_port_sel(struct drm_crtc *crtc);
 extern int intel_enable_rc6(const struct drm_device *dev);
 
 extern bool i915_semaphore_is_enabled(struct drm_device *dev);
-extern void __gen6_gt_force_wake_get(struct drm_i915_private *dev_priv);
-extern void __gen6_gt_force_wake_mt_get(struct drm_i915_private *dev_priv);
-extern void __gen6_gt_force_wake_put(struct drm_i915_private *dev_priv);
-extern void __gen6_gt_force_wake_mt_put(struct drm_i915_private *dev_priv);
-
-extern void vlv_force_wake_get(struct drm_i915_private *dev_priv);
-extern void vlv_force_wake_put(struct drm_i915_private *dev_priv);
 
 /* overlay */
 #ifdef CONFIG_DEBUG_FS
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 288d7b8f49ae..0fdb3d29cbbb 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -96,9 +96,18 @@ i915_gem_wait_for_error(struct drm_device *dev)
         if (!atomic_read(&dev_priv->mm.wedged))
                 return 0;
 
-        ret = wait_for_completion_interruptible(x);
-        if (ret)
+        /*
+         * Only wait 10 seconds for the gpu reset to complete to avoid hanging
+         * userspace. If it takes that long something really bad is going on and
+         * we should simply try to bail out and fail as gracefully as possible.
+         */
+        ret = wait_for_completion_interruptible_timeout(x, 10*HZ);
+        if (ret == 0) {
+                DRM_ERROR("Timed out waiting for the gpu reset to complete\n");
+                return -EIO;
+        } else if (ret < 0) {
                 return ret;
+        }
 
         if (atomic_read(&dev_priv->mm.wedged)) {
                 /* GPU is hung, bump the completion count to account for
@@ -1122,7 +1131,7 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 
         obj->fault_mappable = true;
 
-        pfn = ((dev->agp->base + obj->gtt_offset) >> PAGE_SHIFT) +
+        pfn = ((dev_priv->mm.gtt_base_addr + obj->gtt_offset) >> PAGE_SHIFT) +
                 page_offset;
 
         /* Finally, remap it using the new GTT offset */
@@ -1132,6 +1141,11 @@ unlock:
 out:
         switch (ret) {
         case -EIO:
+                /* If this -EIO is due to a gpu hang, give the reset code a
+                 * chance to clean up the mess. Otherwise return the proper
+                 * SIGBUS. */
+                if (!atomic_read(&dev_priv->mm.wedged))
+                        return VM_FAULT_SIGBUS;
         case -EAGAIN:
                 /* Give the error handler a chance to run and move the
                  * objects off the GPU active list. Next time we service the
@@ -1568,6 +1582,21 @@ i915_add_request(struct intel_ring_buffer *ring,
         int was_empty;
         int ret;
 
+        /*
+         * Emit any outstanding flushes - execbuf can fail to emit the flush
+         * after having emitted the batchbuffer command. Hence we need to fix
+         * things up similar to emitting the lazy request. The difference here
+         * is that the flush _must_ happen before the next request, no matter
+         * what.
+         */
+        if (ring->gpu_caches_dirty) {
+                ret = i915_gem_flush_ring(ring, 0, I915_GEM_GPU_DOMAINS);
+                if (ret)
+                        return ret;
+
+                ring->gpu_caches_dirty = false;
+        }
+
         BUG_ON(request == NULL);
         seqno = i915_gem_next_request_seqno(ring);
 
@@ -1613,6 +1642,9 @@ i915_add_request(struct intel_ring_buffer *ring,
                         queue_delayed_work(dev_priv->wq,
                                            &dev_priv->mm.retire_work, HZ);
         }
+
+        WARN_ON(!list_empty(&ring->gpu_write_list));
+
         return 0;
 }
 
@@ -1827,14 +1859,11 @@ i915_gem_retire_work_handler(struct work_struct *work)
          */
         idle = true;
         for_each_ring(ring, dev_priv, i) {
-                if (!list_empty(&ring->gpu_write_list)) {
+                if (ring->gpu_caches_dirty) {
                         struct drm_i915_gem_request *request;
-                        int ret;
 
-                        ret = i915_gem_flush_ring(ring,
-                                                  0, I915_GEM_GPU_DOMAINS);
                         request = kzalloc(sizeof(*request), GFP_KERNEL);
-                        if (ret || request == NULL ||
+                        if (request == NULL ||
                             i915_add_request(ring, NULL, request))
                             kfree(request);
                 }
@@ -1848,11 +1877,10 @@ i915_gem_retire_work_handler(struct work_struct *work)
         mutex_unlock(&dev->struct_mutex);
 }
 
-static int
-i915_gem_check_wedge(struct drm_i915_private *dev_priv)
+int
+i915_gem_check_wedge(struct drm_i915_private *dev_priv,
+                     bool interruptible)
 {
-        BUG_ON(!mutex_is_locked(&dev_priv->dev->struct_mutex));
-
         if (atomic_read(&dev_priv->mm.wedged)) {
                 struct completion *x = &dev_priv->error_completion;
                 bool recovery_complete;
@@ -1863,7 +1891,16 @@ i915_gem_check_wedge(struct drm_i915_private *dev_priv)
                 recovery_complete = x->done > 0;
                 spin_unlock_irqrestore(&x->wait.lock, flags);
 
-                return recovery_complete ? -EIO : -EAGAIN;
+                /* Non-interruptible callers can't handle -EAGAIN, hence return
+                 * -EIO unconditionally for these. */
+                if (!interruptible)
+                        return -EIO;
+
+                /* Recovery complete, but still wedged means reset failure. */
+                if (recovery_complete)
+                        return -EIO;
+
+                return -EAGAIN;
         }
 
         return 0;
@@ -1899,34 +1936,84 @@ i915_gem_check_olr(struct intel_ring_buffer *ring, u32 seqno)
         return ret;
 }
 
+/**
+ * __wait_seqno - wait until execution of seqno has finished
+ * @ring: the ring expected to report seqno
+ * @seqno: duh!
+ * @interruptible: do an interruptible wait (normally yes)
+ * @timeout: in - how long to wait (NULL forever); out - how much time remaining
+ *
+ * Returns 0 if the seqno was found within the alloted time. Else returns the
+ * errno with remaining time filled in timeout argument.
+ */
 static int __wait_seqno(struct intel_ring_buffer *ring, u32 seqno,
-                        bool interruptible)
+                        bool interruptible, struct timespec *timeout)
 {
         drm_i915_private_t *dev_priv = ring->dev->dev_private;
-        int ret = 0;
+        struct timespec before, now, wait_time={1,0};
+        unsigned long timeout_jiffies;
+        long end;
+        bool wait_forever = true;
+        int ret;
 
         if (i915_seqno_passed(ring->get_seqno(ring), seqno))
                 return 0;
 
         trace_i915_gem_request_wait_begin(ring, seqno);
+
+        if (timeout != NULL) {
+                wait_time = *timeout;
+                wait_forever = false;
+        }
+
+        timeout_jiffies = timespec_to_jiffies(&wait_time);
+
         if (WARN_ON(!ring->irq_get(ring)))
                 return -ENODEV;
 
+        /* Record current time in case interrupted by signal, or wedged * */
+        getrawmonotonic(&before);
+
 #define EXIT_COND \
         (i915_seqno_passed(ring->get_seqno(ring), seqno) || \
         atomic_read(&dev_priv->mm.wedged))
+        do {
+                if (interruptible)
+                        end = wait_event_interruptible_timeout(ring->irq_queue,
+                                                               EXIT_COND,
+                                                               timeout_jiffies);
+                else
+                        end = wait_event_timeout(ring->irq_queue, EXIT_COND,
+                                                 timeout_jiffies);
 
-        if (interruptible)
-                ret = wait_event_interruptible(ring->irq_queue,
-                                               EXIT_COND);
-        else
-                wait_event(ring->irq_queue, EXIT_COND);
+                ret = i915_gem_check_wedge(dev_priv, interruptible);
+                if (ret)
+                        end = ret;
+        } while (end == 0 && wait_forever);
+
+        getrawmonotonic(&now);
 
         ring->irq_put(ring);
         trace_i915_gem_request_wait_end(ring, seqno);
 #undef EXIT_COND
 
-        return ret;
+        if (timeout) {
+                struct timespec sleep_time = timespec_sub(now, before);
+                *timeout = timespec_sub(*timeout, sleep_time);
+        }
+
+        switch (end) {
+        case -EAGAIN: /* Wedged */
+        case -ERESTARTSYS: /* Signal */
+                return (int)end;
+        case 0: /* Timeout */
+                if (timeout)
+                        set_normalized_timespec(timeout, 0, 0);
+                return -ETIME;
+        default: /* Completed */
+                WARN_ON(end < 0); /* We're not aware of other errors */
+                return 0;
+        }
 }
 
 /**
@@ -1934,15 +2021,14 @@ static int __wait_seqno(struct intel_ring_buffer *ring, u32 seqno,
  * request and object lists appropriately for that event.
  */
 int
-i915_wait_request(struct intel_ring_buffer *ring,
-                  uint32_t seqno)
+i915_wait_seqno(struct intel_ring_buffer *ring, uint32_t seqno)
 {
         drm_i915_private_t *dev_priv = ring->dev->dev_private;
         int ret = 0;
 
         BUG_ON(seqno == 0);
 
-        ret = i915_gem_check_wedge(dev_priv);
+        ret = i915_gem_check_wedge(dev_priv, dev_priv->mm.interruptible);
         if (ret)
                 return ret;
 
@@ -1950,9 +2036,7 @@ i915_wait_request(struct intel_ring_buffer *ring,
         if (ret)
                 return ret;
 
-        ret = __wait_seqno(ring, seqno, dev_priv->mm.interruptible);
-        if (atomic_read(&dev_priv->mm.wedged))
-                ret = -EAGAIN;
+        ret = __wait_seqno(ring, seqno, dev_priv->mm.interruptible, NULL);
 
         return ret;
 }
@@ -1975,7 +2059,32 @@ i915_gem_object_wait_rendering(struct drm_i915_gem_object *obj)
          * it.
          */
         if (obj->active) {
-                ret = i915_wait_request(obj->ring, obj->last_rendering_seqno);
+                ret = i915_wait_seqno(obj->ring, obj->last_rendering_seqno);
+                if (ret)
+                        return ret;
+                i915_gem_retire_requests_ring(obj->ring);
+        }
+
+        return 0;
+}
+
+/**
+ * Ensures that an object will eventually get non-busy by flushing any required
+ * write domains, emitting any outstanding lazy request and retiring and
+ * completed requests.
+ */
+static int
+i915_gem_object_flush_active(struct drm_i915_gem_object *obj)
+{
+        int ret;
+
+        if (obj->active) {
+                ret = i915_gem_object_flush_gpu_write_domain(obj);
+                if (ret)
+                        return ret;
+
+                ret = i915_gem_check_olr(obj->ring,
+                                         obj->last_rendering_seqno);
                 if (ret)
                         return ret;
                 i915_gem_retire_requests_ring(obj->ring);
@@ -1985,6 +2094,90 @@ i915_gem_object_wait_rendering(struct drm_i915_gem_object *obj)
 }
 
 /**
+ * i915_gem_wait_ioctl - implements DRM_IOCTL_I915_GEM_WAIT
+ * @DRM_IOCTL_ARGS: standard ioctl arguments
+ *
+ * Returns 0 if successful, else an error is returned with the remaining time in
+ * the timeout parameter.
+ *  -ETIME: object is still busy after timeout
+ *  -ERESTARTSYS: signal interrupted the wait
+ *  -ENONENT: object doesn't exist
+ * Also possible, but rare:
+ *  -EAGAIN: GPU wedged
+ *  -ENOMEM: damn
+ *  -ENODEV: Internal IRQ fail
+ *  -E?: The add request failed
+ *
+ * The wait ioctl with a timeout of 0 reimplements the busy ioctl. With any
+ * non-zero timeout parameter the wait ioctl will wait for the given number of
+ * nanoseconds on an object becoming unbusy. Since the wait itself does so
+ * without holding struct_mutex the object may become re-busied before this
+ * function completes. A similar but shorter * race condition exists in the busy
+ * ioctl
+ */
+int
+i915_gem_wait_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
+{
+        struct drm_i915_gem_wait *args = data;
+        struct drm_i915_gem_object *obj;
+        struct intel_ring_buffer *ring = NULL;
+        struct timespec timeout_stack, *timeout = NULL;
+        u32 seqno = 0;
+        int ret = 0;
+
+        if (args->timeout_ns >= 0) {
+                timeout_stack = ns_to_timespec(args->timeout_ns);
+                timeout = &timeout_stack;
+        }
+
+        ret = i915_mutex_lock_interruptible(dev);
+        if (ret)
+                return ret;
+
+        obj = to_intel_bo(drm_gem_object_lookup(dev, file, args->bo_handle));
+        if (&obj->base == NULL) {
+                mutex_unlock(&dev->struct_mutex);
+                return -ENOENT;
+        }
+
+        /* Need to make sure the object gets inactive eventually. */
+        ret = i915_gem_object_flush_active(obj);
+        if (ret)
+                goto out;
+
+        if (obj->active) {
+                seqno = obj->last_rendering_seqno;
+                ring = obj->ring;
+        }
+
+        if (seqno == 0)
+                 goto out;
+
+        /* Do this after OLR check to make sure we make forward progress polling
+         * on this IOCTL with a 0 timeout (like busy ioctl)
+         */
+        if (!args->timeout_ns) {
+                ret = -ETIME;
+                goto out;
+        }
+
+        drm_gem_object_unreference(&obj->base);
+        mutex_unlock(&dev->struct_mutex);
+
+        ret = __wait_seqno(ring, seqno, true, timeout);
+        if (timeout) {
+                WARN_ON(!timespec_valid(timeout));
+                args->timeout_ns = timespec_to_ns(timeout);
+        }
+        return ret;
+
+out:
+        drm_gem_object_unreference(&obj->base);
+        mutex_unlock(&dev->struct_mutex);
+        return ret;
+}
+
+/**
  * i915_gem_object_sync - sync an object to a ring.
  *
  * @obj: object which may be in use on another ring.
@@ -2160,7 +2353,7 @@ static int i915_ring_idle(struct intel_ring_buffer *ring)
                         return ret;
         }
 
-        return i915_wait_request(ring, i915_gem_next_request_seqno(ring));
+        return i915_wait_seqno(ring, i915_gem_next_request_seqno(ring));
 }
 
 int i915_gpu_idle(struct drm_device *dev)
@@ -2178,6 +2371,10 @@ int i915_gpu_idle(struct drm_device *dev)
                 /* Is the device fubar? */
                 if (WARN_ON(!list_empty(&ring->gpu_write_list)))
                         return -EBUSY;
+
+                ret = i915_switch_context(ring, NULL, DEFAULT_CONTEXT_ID);
+                if (ret)
+                        return ret;
         }
 
         return 0;
@@ -2364,7 +2561,7 @@ i915_gem_object_flush_fence(struct drm_i915_gem_object *obj)
         }
 
         if (obj->last_fenced_seqno) {
-                ret = i915_wait_request(obj->ring, obj->last_fenced_seqno);
+                ret = i915_wait_seqno(obj->ring, obj->last_fenced_seqno);
                 if (ret)
                         return ret;
 
@@ -2551,8 +2748,8 @@ i915_gem_object_bind_to_gtt(struct drm_i915_gem_object *obj,
         if (map_and_fenceable)
                 free_space =
                         drm_mm_search_free_in_range(&dev_priv->mm.gtt_space,
-                                                    size, alignment, 0,
-                                                    dev_priv->mm.gtt_mappable_end,
+                                                    size, alignment,
+                                                    0, dev_priv->mm.gtt_mappable_end,
                                                     0);
         else
                 free_space = drm_mm_search_free(&dev_priv->mm.gtt_space,
@@ -2563,7 +2760,7 @@ i915_gem_object_bind_to_gtt(struct drm_i915_gem_object *obj,
                         obj->gtt_space =
                                 drm_mm_get_block_range_generic(free_space,
                                                                size, alignment, 0,
-                                                               dev_priv->mm.gtt_mappable_end,
+                                                               0, dev_priv->mm.gtt_mappable_end,
                                                                0);
                 else
                         obj->gtt_space =
@@ -3030,7 +3227,7 @@ i915_gem_ring_throttle(struct drm_device *dev, struct drm_file *file)
         if (seqno == 0)
                 return 0;
 
-        ret = __wait_seqno(ring, seqno, true);
+        ret = __wait_seqno(ring, seqno, true, NULL);
         if (ret == 0)
                 queue_delayed_work(dev_priv->wq, &dev_priv->mm.retire_work, 0);
 
@@ -3199,30 +3396,9 @@ i915_gem_busy_ioctl(struct drm_device *dev, void *data,
          * become non-busy without any further actions, therefore emit any
          * necessary flushes here.
          */
-        args->busy = obj->active;
-        if (args->busy) {
-                /* Unconditionally flush objects, even when the gpu still uses this
-                 * object. Userspace calling this function indicates that it wants to
-                 * use this buffer rather sooner than later, so issuing the required
-                 * flush earlier is beneficial.
-                 */
-                if (obj->base.write_domain & I915_GEM_GPU_DOMAINS) {
-                        ret = i915_gem_flush_ring(obj->ring,
-                                                  0, obj->base.write_domain);
-                } else {
-                        ret = i915_gem_check_olr(obj->ring,
-                                                 obj->last_rendering_seqno);
-                }
-
-                /* Update the active list for the hardware's current position.
-                 * Otherwise this only updates on a delayed timer or when irqs
-                 * are actually unmasked, and our working set ends up being
-                 * larger than required.
-                 */
-                i915_gem_retire_requests_ring(obj->ring);
+        ret = i915_gem_object_flush_active(obj);
 
-                args->busy = obj->active;
-        }
+        args->busy = obj->active;
 
         drm_gem_object_unreference(&obj->base);
 unlock:
@@ -3435,6 +3611,38 @@ i915_gem_idle(struct drm_device *dev)
         return 0;
 }
 
+void i915_gem_l3_remap(struct drm_device *dev)
+{
+        drm_i915_private_t *dev_priv = dev->dev_private;
+        u32 misccpctl;
+        int i;
+
+        if (!IS_IVYBRIDGE(dev))
+                return;
+
+        if (!dev_priv->mm.l3_remap_info)
+                return;
+
+        misccpctl = I915_READ(GEN7_MISCCPCTL);
+        I915_WRITE(GEN7_MISCCPCTL, misccpctl & ~GEN7_DOP_CLOCK_GATE_ENABLE);
+        POSTING_READ(GEN7_MISCCPCTL);
+
+        for (i = 0; i < GEN7_L3LOG_SIZE; i += 4) {
+                u32 remap = I915_READ(GEN7_L3LOG_BASE + i);
+                if (remap && remap != dev_priv->mm.l3_remap_info[i/4])
+                        DRM_DEBUG("0x%x was already programmed to %x\n",
+                                  GEN7_L3LOG_BASE + i, remap);
+                if (remap && !dev_priv->mm.l3_remap_info[i/4])
+                        DRM_DEBUG_DRIVER("Clearing remapped register\n");
+                I915_WRITE(GEN7_L3LOG_BASE + i, dev_priv->mm.l3_remap_info[i/4]);
+        }
+
+        /* Make sure all the writes land before disabling dop clock gating */
+        POSTING_READ(GEN7_L3LOG_BASE);
+
+        I915_WRITE(GEN7_MISCCPCTL, misccpctl);
+}
+
 void i915_gem_init_swizzling(struct drm_device *dev)
 {
         drm_i915_private_t *dev_priv = dev->dev_private;
@@ -3524,6 +3732,11 @@ i915_gem_init_hw(struct drm_device *dev)
         drm_i915_private_t *dev_priv = dev->dev_private;
         int ret;
 
+        if (!intel_enable_gtt())
+                return -EIO;
+
+        i915_gem_l3_remap(dev);
+
         i915_gem_init_swizzling(dev);
 
         ret = intel_init_render_ring_buffer(dev);
@@ -3544,6 +3757,11 @@ i915_gem_init_hw(struct drm_device *dev)
 
         dev_priv->next_seqno = 1;
 
+        /*
+         * XXX: There was some w/a described somewhere suggesting loading
+         * contexts before PPGTT.
+         */
+        i915_gem_context_init(dev);
         i915_gem_init_ppgtt(dev);
 
         return 0;
diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c
new file mode 100644
index 000000000000..9ae3f2cf414e
--- /dev/null
+++ b/drivers/gpu/drm/i915/i915_gem_context.c
@@ -0,0 +1,533 @@
+/*
+ * Copyright © 2011-2012 Intel Corporation
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the next
+ * paragraph) shall be included in all copies or substantial portions of the
+ * Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ *
+ * Authors:
+ *    Ben Widawsky <ben@bwidawsk.net>
+ *
+ */
+
+/*
+ * This file implements HW context support. On gen5+ a HW context consists of an
+ * opaque GPU object which is referenced at times of context saves and restores.
+ * With RC6 enabled, the context is also referenced as the GPU enters and exists
+ * from RC6 (GPU has it's own internal power context, except on gen5). Though
+ * something like a context does exist for the media ring, the code only
+ * supports contexts for the render ring.
+ *
+ * In software, there is a distinction between contexts created by the user,
+ * and the default HW context. The default HW context is used by GPU clients
+ * that do not request setup of their own hardware context. The default
+ * context's state is never restored to help prevent programming errors. This
+ * would happen if a client ran and piggy-backed off another clients GPU state.
+ * The default context only exists to give the GPU some offset to load as the
+ * current to invoke a save of the context we actually care about. In fact, the
+ * code could likely be constructed, albeit in a more complicated fashion, to
+ * never use the default context, though that limits the driver's ability to
+ * swap out, and/or destroy other contexts.
+ *
+ * All other contexts are created as a request by the GPU client. These contexts
+ * store GPU state, and thus allow GPU clients to not re-emit state (and
+ * potentially query certain state) at any time. The kernel driver makes
+ * certain that the appropriate commands are inserted.
+ *
+ * The context life cycle is semi-complicated in that context BOs may live
+ * longer than the context itself because of the way the hardware, and object
+ * tracking works. Below is a very crude representation of the state machine
+ * describing the context life.
+ *                                         refcount     pincount     active
+ * S0: initial state                          0            0           0
+ * S1: context created                        1            0           0
+ * S2: context is currently running           2            1           X
+ * S3: GPU referenced, but not current        2            0           1
+ * S4: context is current, but destroyed      1            1           0
+ * S5: like S3, but destroyed                 1            0           1
+ *
+ * The most common (but not all) transitions:
+ * S0->S1: client creates a context
+ * S1->S2: client submits execbuf with context
+ * S2->S3: other clients submits execbuf with context
+ * S3->S1: context object was retired
+ * S3->S2: clients submits another execbuf
+ * S2->S4: context destroy called with current context
+ * S3->S5->S0: destroy path
+ * S4->S5->S0: destroy path on current context
+ *
+ * There are two confusing terms used above:
+ *  The "current context" means the context which is currently running on the
+ *  GPU. The GPU has loaded it's state already and has stored away the gtt
+ *  offset of the BO. The GPU is not actively referencing the data at this
+ *  offset, but it will on the next context switch. The only way to avoid this
+ *  is to do a GPU reset.
+ *
+ *  An "active context' is one which was previously the "current context" and is
+ *  on the active list waiting for the next context switch to occur. Until this
+ *  happens, the object must remain at the same gtt offset. It is therefore
+ *  possible to destroy a context, but it is still active.
+ *
+ */
+
+#include "drmP.h"
+#include "i915_drm.h"
+#include "i915_drv.h"
+
+/* This is a HW constraint. The value below is the largest known requirement
+ * I've seen in a spec to date, and that was a workaround for a non-shipping
+ * part. It should be safe to decrease this, but it's more future proof as is.
+ */
+#define CONTEXT_ALIGN (64<<10)
+
+static struct i915_hw_context *
+i915_gem_context_get(struct drm_i915_file_private *file_priv, u32 id);
+static int do_switch(struct drm_i915_gem_object *from_obj,
+                     struct i915_hw_context *to, u32 seqno);
+
+static int get_context_size(struct drm_device *dev)
+{
+        struct drm_i915_private *dev_priv = dev->dev_private;
+        int ret;
+        u32 reg;
+
+        switch (INTEL_INFO(dev)->gen) {
+        case 6:
+                reg = I915_READ(CXT_SIZE);
+                ret = GEN6_CXT_TOTAL_SIZE(reg) * 64;
+                break;
+        case 7:
+                reg = I915_READ(GEN7_CTX_SIZE);
+                ret = GEN7_CTX_TOTAL_SIZE(reg) * 64;
+                break;
+        default:
+                BUG();
+        }
+
+        return ret;
+}
+
+static void do_destroy(struct i915_hw_context *ctx)
+{
+        struct drm_device *dev = ctx->obj->base.dev;
+        struct drm_i915_private *dev_priv = dev->dev_private;
+
+        if (ctx->file_priv)
+                idr_remove(&ctx->file_priv->context_idr, ctx->id);
+        else
+                BUG_ON(ctx != dev_priv->ring[RCS].default_context);
+
+        drm_gem_object_unreference(&ctx->obj->base);
+        kfree(ctx);
+}
+
+static struct i915_hw_context *
+create_hw_context(struct drm_device *dev,
+                  struct drm_i915_file_private *file_priv)
+{
+        struct drm_i915_private *dev_priv = dev->dev_private;
+        struct i915_hw_context *ctx;
+        int ret, id;
+
+        ctx = kzalloc(sizeof(struct drm_i915_file_private), GFP_KERNEL);
+        if (ctx == NULL)
+                return ERR_PTR(-ENOMEM);
+
+        ctx->obj = i915_gem_alloc_object(dev, dev_priv->hw_context_size);
+        if (ctx->obj == NULL) {
+                kfree(ctx);
+                DRM_DEBUG_DRIVER("Context object allocated failed\n");
+                return ERR_PTR(-ENOMEM);
+        }
+
+        /* The ring associated with the context object is handled by the normal
+         * object tracking code. We give an initial ring value simple to pass an
+         * assertion in the context switch code.
+         */
+        ctx->ring = &dev_priv->ring[RCS];
+
+        /* Default context will never have a file_priv */
+        if (file_priv == NULL)
+                return ctx;
+
+        ctx->file_priv = file_priv;
+
+again:
+        if (idr_pre_get(&file_priv->context_idr, GFP_KERNEL) == 0) {
+                ret = -ENOMEM;
+                DRM_DEBUG_DRIVER("idr allocation failed\n");
+                goto err_out;
+        }
+
+        ret = idr_get_new_above(&file_priv->context_idr, ctx,
+                                DEFAULT_CONTEXT_ID + 1, &id);
+        if (ret == 0)
+                ctx->id = id;
+
+        if (ret == -EAGAIN)
+                goto again;
+        else if (ret)
+                goto err_out;
+
+        return ctx;
+
+err_out:
+        do_destroy(ctx);
+        return ERR_PTR(ret);
+}
+
+static inline bool is_default_context(struct i915_hw_context *ctx)
+{
+        return (ctx == ctx->ring->default_context);
+}
+
+/**
+ * The default context needs to exist per ring that uses contexts. It stores the
+ * context state of the GPU for applications that don't utilize HW contexts, as
+ * well as an idle case.
+ */
+static int create_default_context(struct drm_i915_private *dev_priv)
+{
+        struct i915_hw_context *ctx;
+        int ret;
+
+        BUG_ON(!mutex_is_locked(&dev_priv->dev->struct_mutex));
+
+        ctx = create_hw_context(dev_priv->dev, NULL);
+        if (IS_ERR(ctx))
+                return PTR_ERR(ctx);
+
+        /* We may need to do things with the shrinker which require us to
+         * immediately switch back to the default context. This can cause a
+         * problem as pinning the default context also requires GTT space which
+         * may not be available. To avoid this we always pin the
+         * default context.
+         */
+        dev_priv->ring[RCS].default_context = ctx;
+        ret = i915_gem_object_pin(ctx->obj, CONTEXT_ALIGN, false);
+        if (ret) {
+                do_destroy(ctx);
+                return ret;
+        }
+
+        ret = do_switch(NULL, ctx, 0);
+        if (ret) {
+                i915_gem_object_unpin(ctx->obj);
+                do_destroy(ctx);
+        } else {
+                DRM_DEBUG_DRIVER("Default HW context loaded\n");
+        }
+
+        return ret;
+}
+
+void i915_gem_context_init(struct drm_device *dev)
+{
+        struct drm_i915_private *dev_priv = dev->dev_private;
+        uint32_t ctx_size;
+
+        if (!HAS_HW_CONTEXTS(dev)) {
+                dev_priv->hw_contexts_disabled = true;
+                return;
+        }
+
+        /* If called from reset, or thaw... we've been here already */
+        if (dev_priv->hw_contexts_disabled ||
+            dev_priv->ring[RCS].default_context)
+                return;
+
+        ctx_size = get_context_size(dev);
+        dev_priv->hw_context_size = get_context_size(dev);
+        dev_priv->hw_context_size = round_up(dev_priv->hw_context_size, 4096);
+
+        if (ctx_size <= 0 || ctx_size > (1<<20)) {
+                dev_priv->hw_contexts_disabled = true;
+                return;
+        }
+
+        if (create_default_context(dev_priv)) {
+                dev_priv->hw_contexts_disabled = true;
+                return;
+        }
+
+        DRM_DEBUG_DRIVER("HW context support initialized\n");
+}
+
+void i915_gem_context_fini(struct drm_device *dev)
+{
+        struct drm_i915_private *dev_priv = dev->dev_private;
+
+        if (dev_priv->hw_contexts_disabled)
+                return;
+
+        /* The only known way to stop the gpu from accessing the hw context is
+         * to reset it. Do this as the very last operation to avoid confusing
+         * other code, leading to spurious errors. */
+        intel_gpu_reset(dev);
+
+        i915_gem_object_unpin(dev_priv->ring[RCS].default_context->obj);
+
+        do_destroy(dev_priv->ring[RCS].default_context);
+}
+
+static int context_idr_cleanup(int id, void *p, void *data)
+{
+        struct i915_hw_context *ctx = p;
+
+        BUG_ON(id == DEFAULT_CONTEXT_ID);
+
+        do_destroy(ctx);
+
+        return 0;
+}
+
+void i915_gem_context_close(struct drm_device *dev, struct drm_file *file)
+{
+        struct drm_i915_file_private *file_priv = file->driver_priv;
+
+        mutex_lock(&dev->struct_mutex);
+        idr_for_each(&file_priv->context_idr, context_idr_cleanup, NULL);
+        idr_destroy(&file_priv->context_idr);
+        mutex_unlock(&dev->struct_mutex);
+}
+
+static struct i915_hw_context *
+i915_gem_context_get(struct drm_i915_file_private *file_priv, u32 id)
+{
+        return (struct i915_hw_context *)idr_find(&file_priv->context_idr, id);
+}
+
+static inline int
+mi_set_context(struct intel_ring_buffer *ring,
+               struct i915_hw_context *new_context,
+               u32 hw_flags)
+{
+        int ret;
+
+        /* w/a: If Flush TLB Invalidation Mode is enabled, driver must do a TLB
+         * invalidation prior to MI_SET_CONTEXT. On GEN6 we don't set the value
+         * explicitly, so we rely on the value at ring init, stored in
+         * itlb_before_ctx_switch.
+         */
+        if (IS_GEN6(ring->dev) && ring->itlb_before_ctx_switch) {
+                ret = ring->flush(ring, 0, 0);
+                if (ret)
+                        return ret;
+        }
+
+        ret = intel_ring_begin(ring, 6);
+        if (ret)
+                return ret;
+
+        if (IS_GEN7(ring->dev))
+                intel_ring_emit(ring, MI_ARB_ON_OFF | MI_ARB_DISABLE);
+        else
+                intel_ring_emit(ring, MI_NOOP);
+
+        intel_ring_emit(ring, MI_NOOP);
+        intel_ring_emit(ring, MI_SET_CONTEXT);
+        intel_ring_emit(ring, new_context->obj->gtt_offset |
+                        MI_MM_SPACE_GTT |
+                        MI_SAVE_EXT_STATE_EN |
+                        MI_RESTORE_EXT_STATE_EN |
+                        hw_flags);
+        /* w/a: MI_SET_CONTEXT must always be followed by MI_NOOP */
+        intel_ring_emit(ring, MI_NOOP);
+
+        if (IS_GEN7(ring->dev))
+                intel_ring_emit(ring, MI_ARB_ON_OFF | MI_ARB_ENABLE);
+        else
+                intel_ring_emit(ring, MI_NOOP);
+
+        intel_ring_advance(ring);
+
+        return ret;
+}
+
+static int do_switch(struct drm_i915_gem_object *from_obj,
+                     struct i915_hw_context *to,
+                     u32 seqno)
+{
+        struct intel_ring_buffer *ring = NULL;
+        u32 hw_flags = 0;
+        int ret;
+
+        BUG_ON(to == NULL);
+        BUG_ON(from_obj != NULL && from_obj->pin_count == 0);
+
+        ret = i915_gem_object_pin(to->obj, CONTEXT_ALIGN, false);
+        if (ret)
+                return ret;
+
+        if (!to->obj->has_global_gtt_mapping)
+                i915_gem_gtt_bind_object(to->obj, to->obj->cache_level);
+
+        if (!to->is_initialized || is_default_context(to))
+                hw_flags |= MI_RESTORE_INHIBIT;
+        else if (WARN_ON_ONCE(from_obj == to->obj)) /* not yet expected */
+                hw_flags |= MI_FORCE_RESTORE;
+
+        ring = to->ring;
+        ret = mi_set_context(ring, to, hw_flags);
+        if (ret) {
+                i915_gem_object_unpin(to->obj);
+                return ret;
+        }
+
+        /* The backing object for the context is done after switching to the
+         * *next* context. Therefore we cannot retire the previous context until
+         * the next context has already started running. In fact, the below code
+         * is a bit suboptimal because the retiring can occur simply after the
+         * MI_SET_CONTEXT instead of when the next seqno has completed.
+         */
+        if (from_obj != NULL) {
+                from_obj->base.read_domains = I915_GEM_DOMAIN_INSTRUCTION;
+                i915_gem_object_move_to_active(from_obj, ring, seqno);
+                /* As long as MI_SET_CONTEXT is serializing, ie. it flushes the
+                 * whole damn pipeline, we don't need to explicitly mark the
+                 * object dirty. The only exception is that the context must be
+                 * correct in case the object gets swapped out. Ideally we'd be
+                 * able to defer doing this until we know the object would be
+                 * swapped, but there is no way to do that yet.
+                 */
+                from_obj->dirty = 1;
+                BUG_ON(from_obj->ring != to->ring);
+                i915_gem_object_unpin(from_obj);
+        }
+
+        ring->last_context_obj = to->obj;
+        to->is_initialized = true;
+
+        return 0;
+}
+
+/**
+ * i915_switch_context() - perform a GPU context switch.
+ * @ring: ring for which we'll execute the context switch
+ * @file_priv: file_priv associated with the context, may be NULL
+ * @id: context id number
+ * @seqno: sequence number by which the new context will be switched to
+ * @flags:
+ *
+ * The context life cycle is simple. The context refcount is incremented and
+ * decremented by 1 and create and destroy. If the context is in use by the GPU,
+ * it will have a refoucnt > 1. This allows us to destroy the context abstract
+ * object while letting the normal object tracking destroy the backing BO.
+ */
+int i915_switch_context(struct intel_ring_buffer *ring,
+                        struct drm_file *file,
+                        int to_id)
+{
+        struct drm_i915_private *dev_priv = ring->dev->dev_private;
+        struct drm_i915_file_private *file_priv = NULL;
+        struct i915_hw_context *to;
+        struct drm_i915_gem_object *from_obj = ring->last_context_obj;
+        int ret;
+
+        if (dev_priv->hw_contexts_disabled)
+                return 0;
+
+        if (ring != &dev_priv->ring[RCS])
+                return 0;
+
+        if (file)
+                file_priv = file->driver_priv;
+
+        if (to_id == DEFAULT_CONTEXT_ID) {
+                to = ring->default_context;
+        } else {
+                to = i915_gem_context_get(file_priv, to_id);
+                if (to == NULL)
+                        return -ENOENT;
+        }
+
+        if (from_obj == to->obj)
+                return 0;
+
+        ret = do_switch(from_obj, to, i915_gem_next_request_seqno(to->ring));
+        if (ret)
+                return ret;
+
+        /* Just to make the code a little cleaner we take the object reference
+         * after the switch was successful. It would be more intuitive to ref
+         * the 'to' object before the switch but we know the refcount must be >0
+         * if context_get() succeeded, and we hold struct mutex. So it's safe to
+         * do this here/now
+         */
+        drm_gem_object_reference(&to->obj->base);
+        if (from_obj != NULL)
+                drm_gem_object_unreference(&from_obj->base);
+        return ret;
+}
+
+int i915_gem_context_create_ioctl(struct drm_device *dev, void *data,
+                                  struct drm_file *file)
+{
+        struct drm_i915_private *dev_priv = dev->dev_private;
+        struct drm_i915_gem_context_create *args = data;
+        struct drm_i915_file_private *file_priv = file->driver_priv;
+        struct i915_hw_context *ctx;
+        int ret;
+
+        if (!(dev->driver->driver_features & DRIVER_GEM))
+                return -ENODEV;
+
+        if (dev_priv->hw_contexts_disabled)
+                return -ENODEV;
+
+        ret = i915_mutex_lock_interruptible(dev);
+        if (ret)
+                return ret;
+
+        ctx = create_hw_context(dev, file_priv);
+        mutex_unlock(&dev->struct_mutex);
+
+        args->ctx_id = ctx->id;
+        DRM_DEBUG_DRIVER("HW context %d created\n", args->ctx_id);
+
+        return PTR_RET(ctx);
+}
+
+int i915_gem_context_destroy_ioctl(struct drm_device *dev, void *data,
+                                   struct drm_file *file)
+{
+        struct drm_i915_gem_context_destroy *args = data;
+        struct drm_i915_file_private *file_priv = file->driver_priv;
+        struct i915_hw_context *ctx;
+        int ret;
+
+        if (!(dev->driver->driver_features & DRIVER_GEM))
+                return -ENODEV;
+
+        ret = i915_mutex_lock_interruptible(dev);
+        if (ret)
+                return ret;
+
+        ctx = i915_gem_context_get(file_priv, args->ctx_id);
+        if (!ctx) {
+                mutex_unlock(&dev->struct_mutex);
+                return -ENOENT;
+        }
+
+        do_destroy(ctx);
+
+        mutex_unlock(&dev->struct_mutex);
+
+        DRM_DEBUG_DRIVER("HW context %d destroyed\n", args->ctx_id);
+        return 0;
+}
diff --git a/drivers/gpu/drm/i915/i915_gem_debug.c b/drivers/gpu/drm/i915/i915_gem_debug.c
index a4f6aaabca99..bddf7bed183f 100644
--- a/drivers/gpu/drm/i915/i915_gem_debug.c
+++ b/drivers/gpu/drm/i915/i915_gem_debug.c
@@ -132,7 +132,8 @@ i915_gem_object_check_coherency(struct drm_i915_gem_object *obj, int handle)
                  __func__, obj, obj->gtt_offset, handle,
                  obj->size / 1024);
 
-        gtt_mapping = ioremap(dev->agp->base + obj->gtt_offset, obj->base.size);
+        gtt_mapping = ioremap(dev_priv->mm.gtt_base_addr + obj->gtt_offset,
+                              obj->base.size);
         if (gtt_mapping == NULL) {
                 DRM_ERROR("failed to map GTT space\n");
                 return;
diff --git a/drivers/gpu/drm/i915/i915_gem_evict.c b/drivers/gpu/drm/i915/i915_gem_evict.c
index ae7c24e12e52..eba0308f10e3 100644
--- a/drivers/gpu/drm/i915/i915_gem_evict.c
+++ b/drivers/gpu/drm/i915/i915_gem_evict.c
@@ -78,11 +78,12 @@ i915_gem_evict_something(struct drm_device *dev, int min_size,
 
         INIT_LIST_HEAD(&unwind_list);
         if (mappable)
-                drm_mm_init_scan_with_range(&dev_priv->mm.gtt_space, min_size,
-                                            alignment, 0,
-                                            dev_priv->mm.gtt_mappable_end);
+                drm_mm_init_scan_with_range(&dev_priv->mm.gtt_space,
+                                            min_size, alignment, 0,
+                                            0, dev_priv->mm.gtt_mappable_end);
         else
-                drm_mm_init_scan(&dev_priv->mm.gtt_space, min_size, alignment);
+                drm_mm_init_scan(&dev_priv->mm.gtt_space,
+                                 min_size, alignment, 0);
 
         /* First see if there is a large enough contiguous idle region... */
         list_for_each_entry(obj, &dev_priv->mm.inactive_list, mm_list) {
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index 974a9f1068a3..88e2e114189c 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -810,33 +810,16 @@ err:
         return ret;
 }
 
-static int
+static void
 i915_gem_execbuffer_flush(struct drm_device *dev,
                           uint32_t invalidate_domains,
-                          uint32_t flush_domains,
-                          uint32_t flush_rings)
+                          uint32_t flush_domains)
 {
-        drm_i915_private_t *dev_priv = dev->dev_private;
-        int i, ret;
-
         if (flush_domains & I915_GEM_DOMAIN_CPU)
                 intel_gtt_chipset_flush();
 
         if (flush_domains & I915_GEM_DOMAIN_GTT)
                 wmb();
-
-        if ((flush_domains | invalidate_domains) & I915_GEM_GPU_DOMAINS) {
-                for (i = 0; i < I915_NUM_RINGS; i++)
-                        if (flush_rings & (1 << i)) {
-                                ret = i915_gem_flush_ring(&dev_priv->ring[i],
-                                                          invalidate_domains,
-                                                          flush_domains);
-                                if (ret)
-                                        return ret;
-                        }
-        }
-
-        return 0;
 }
 
 static int
@@ -885,12 +868,9 @@ i915_gem_execbuffer_move_to_gpu(struct intel_ring_buffer *ring,
                 i915_gem_object_set_to_gpu_domain(obj, ring, &cd);
 
         if (cd.invalidate_domains | cd.flush_domains) {
-                ret = i915_gem_execbuffer_flush(ring->dev,
-                                                cd.invalidate_domains,
-                                                cd.flush_domains,
-                                                cd.flush_rings);
-                if (ret)
-                        return ret;
+                i915_gem_execbuffer_flush(ring->dev,
+                                          cd.invalidate_domains,
+                                          cd.flush_domains);
         }
 
         if (cd.flips) {
@@ -905,6 +885,11 @@ i915_gem_execbuffer_move_to_gpu(struct intel_ring_buffer *ring,
                         return ret;
         }
 
+        /* Unconditionally invalidate gpu caches. */
+        ret = i915_gem_flush_ring(ring, I915_GEM_GPU_DOMAINS, 0);
+        if (ret)
+                return ret;
+
         return 0;
 }
 
@@ -983,26 +968,13 @@ i915_gem_execbuffer_retire_commands(struct drm_device *dev,
                                     struct intel_ring_buffer *ring)
 {
         struct drm_i915_gem_request *request;
-        u32 invalidate;
 
-        /*
-         * Ensure that the commands in the batch buffer are
-         * finished before the interrupt fires.
-         *
-         * The sampler always gets flushed on i965 (sigh).
-         */
-        invalidate = I915_GEM_DOMAIN_COMMAND;
-        if (INTEL_INFO(dev)->gen >= 4)
-                invalidate |= I915_GEM_DOMAIN_SAMPLER;
-        if (ring->flush(ring, invalidate, 0)) {
-                i915_gem_next_request_seqno(ring);
-                return;
-        }
+        /* Unconditionally force add_request to emit a full flush. */
+        ring->gpu_caches_dirty = true;
 
         /* Add a breadcrumb for the completion of the batch buffer */
         request = kzalloc(sizeof(*request), GFP_KERNEL);
         if (request == NULL || i915_add_request(ring, file, request)) {
-                i915_gem_next_request_seqno(ring);
                 kfree(request);
         }
 }
@@ -1044,6 +1016,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
         struct drm_i915_gem_object *batch_obj;
         struct drm_clip_rect *cliprects = NULL;
         struct intel_ring_buffer *ring;
+        u32 ctx_id = i915_execbuffer2_get_context_id(*args);
         u32 exec_start, exec_len;
         u32 seqno;
         u32 mask;
@@ -1065,9 +1038,19 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
                 break;
         case I915_EXEC_BSD:
                 ring = &dev_priv->ring[VCS];
+                if (ctx_id != 0) {
+                        DRM_DEBUG("Ring %s doesn't support contexts\n",
+                                  ring->name);
+                        return -EPERM;
+                }
                 break;
         case I915_EXEC_BLT:
                 ring = &dev_priv->ring[BCS];
+                if (ctx_id != 0) {
+                        DRM_DEBUG("Ring %s doesn't support contexts\n",
+                                  ring->name);
+                        return -EPERM;
+                }
                 break;
         default:
                 DRM_DEBUG("execbuf with unknown ring: %d\n",
@@ -1261,6 +1244,10 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
                         goto err;
         }
 
+        ret = i915_switch_context(ring, file, ctx_id);
+        if (ret)
+                goto err;
+
         trace_i915_gem_ring_dispatch(ring, seqno);
 
         exec_start = batch_obj->gtt_offset + args->batch_start_offset;
@@ -1367,6 +1354,7 @@ i915_gem_execbuffer(struct drm_device *dev, void *data,
         exec2.num_cliprects = args->num_cliprects;
         exec2.cliprects_ptr = args->cliprects_ptr;
         exec2.flags = I915_EXEC_RENDER;
+        i915_execbuffer2_set_context_id(exec2, 0);
 
         ret = i915_gem_do_execbuffer(dev, data, file, &exec2, exec2_list);
         if (!ret) {
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index ed3224c37423..05014bc20ee8 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -375,6 +375,86 @@ static void gen6_pm_rps_work(struct work_struct *work)
         mutex_unlock(&dev_priv->dev->struct_mutex);
 }
 
+
+/**
+ * ivybridge_parity_work - Workqueue called when a parity error interrupt
+ * occurred.
+ * @work: workqueue struct
+ *
+ * Doesn't actually do anything except notify userspace. As a consequence of
+ * this event, userspace should try to remap the bad rows since statistically
+ * it is likely the same row is more likely to go bad again.
+ */
+static void ivybridge_parity_work(struct work_struct *work)
+{
+        drm_i915_private_t *dev_priv = container_of(work, drm_i915_private_t,
+                                                    parity_error_work);
+        u32 error_status, row, bank, subbank;
+        char *parity_event[5];
+        uint32_t misccpctl;
+        unsigned long flags;
+
+        /* We must turn off DOP level clock gating to access the L3 registers.
+         * In order to prevent a get/put style interface, acquire struct mutex
+         * any time we access those registers.
+         */
+        mutex_lock(&dev_priv->dev->struct_mutex);
+
+        misccpctl = I915_READ(GEN7_MISCCPCTL);
+        I915_WRITE(GEN7_MISCCPCTL, misccpctl & ~GEN7_DOP_CLOCK_GATE_ENABLE);
+        POSTING_READ(GEN7_MISCCPCTL);
+
+        error_status = I915_READ(GEN7_L3CDERRST1);
+        row = GEN7_PARITY_ERROR_ROW(error_status);
+        bank = GEN7_PARITY_ERROR_BANK(error_status);
+        subbank = GEN7_PARITY_ERROR_SUBBANK(error_status);
+
+        I915_WRITE(GEN7_L3CDERRST1, GEN7_PARITY_ERROR_VALID |
+                                    GEN7_L3CDERRST1_ENABLE);
+        POSTING_READ(GEN7_L3CDERRST1);
+
+        I915_WRITE(GEN7_MISCCPCTL, misccpctl);
+
+        spin_lock_irqsave(&dev_priv->irq_lock, flags);
+        dev_priv->gt_irq_mask &= ~GT_GEN7_L3_PARITY_ERROR_INTERRUPT;
+        I915_WRITE(GTIMR, dev_priv->gt_irq_mask);
+        spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
+
+        mutex_unlock(&dev_priv->dev->struct_mutex);
+
+        parity_event[0] = "L3_PARITY_ERROR=1";
+        parity_event[1] = kasprintf(GFP_KERNEL, "ROW=%d", row);
+        parity_event[2] = kasprintf(GFP_KERNEL, "BANK=%d", bank);
+        parity_event[3] = kasprintf(GFP_KERNEL, "SUBBANK=%d", subbank);
+        parity_event[4] = NULL;
+
+        kobject_uevent_env(&dev_priv->dev->primary->kdev.kobj,
+                           KOBJ_CHANGE, parity_event);
+
+        DRM_DEBUG("Parity error: Row = %d, Bank = %d, Sub bank = %d.\n",
+                  row, bank, subbank);
+
+        kfree(parity_event[3]);
+        kfree(parity_event[2]);
+        kfree(parity_event[1]);
+}
+
+static void ivybridge_handle_parity_error(struct drm_device *dev)
+{
+        drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
+        unsigned long flags;
+
+        if (!IS_IVYBRIDGE(dev))
+                return;
+
+        spin_lock_irqsave(&dev_priv->irq_lock, flags);
+        dev_priv->gt_irq_mask |= GT_GEN7_L3_PARITY_ERROR_INTERRUPT;
+        I915_WRITE(GTIMR, dev_priv->gt_irq_mask);
+        spin_unlock_irqrestore(&dev_priv->irq_lock, flags);
+
+        queue_work(dev_priv->wq, &dev_priv->parity_error_work);
+}
+
 static void snb_gt_irq_handler(struct drm_device *dev,
                                struct drm_i915_private *dev_priv,
                                u32 gt_iir)
@@ -394,6 +474,9 @@ static void snb_gt_irq_handler(struct drm_device *dev,
                 DRM_ERROR("GT error interrupt 0x%08x\n", gt_iir);
                 i915_handle_error(dev, false);
         }
+
+        if (gt_iir & GT_GEN7_L3_PARITY_ERROR_INTERRUPT)
+                ivybridge_handle_parity_error(dev);
 }
 
 static void gen6_queue_rps_work(struct drm_i915_private *dev_priv,
@@ -429,15 +512,10 @@ static irqreturn_t valleyview_irq_handler(DRM_IRQ_ARGS)
         unsigned long irqflags;
         int pipe;
         u32 pipe_stats[I915_MAX_PIPES];
-        u32 vblank_status;
-        int vblank = 0;
         bool blc_event;
 
         atomic_inc(&dev_priv->irq_received);
 
-        vblank_status = PIPE_START_VBLANK_INTERRUPT_STATUS |
-                PIPE_VBLANK_INTERRUPT_STATUS;
-
         while (true) {
                 iir = I915_READ(VLV_IIR);
                 gt_iir = I915_READ(GTIIR);
@@ -467,6 +545,16 @@ static irqreturn_t valleyview_irq_handler(DRM_IRQ_ARGS)
                 }
                 spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
 
+                for_each_pipe(pipe) {
+                        if (pipe_stats[pipe] & PIPE_VBLANK_INTERRUPT_STATUS)
+                                drm_handle_vblank(dev, pipe);
+
+                        if (pipe_stats[pipe] & PLANE_FLIPDONE_INT_STATUS_VLV) {
+                                intel_prepare_page_flip(dev, pipe);
+                                intel_finish_page_flip(dev, pipe);
+                        }
+                }
+
                 /* Consume port.  Then clear IIR or we'll miss events */
                 if (iir & I915_DISPLAY_PORT_INTERRUPT) {
                         u32 hotplug_status = I915_READ(PORT_HOTPLUG_STAT);
@@ -481,19 +569,6 @@ static irqreturn_t valleyview_irq_handler(DRM_IRQ_ARGS)
                         I915_READ(PORT_HOTPLUG_STAT);
                 }
 
-
-                if (iir & I915_DISPLAY_PIPE_A_VBLANK_INTERRUPT) {
-                        drm_handle_vblank(dev, 0);
-                        vblank++;
-                        intel_finish_page_flip(dev, 0);
-                }
-
-                if (iir & I915_DISPLAY_PIPE_B_VBLANK_INTERRUPT) {
-                        drm_handle_vblank(dev, 1);
-                        vblank++;
-                        intel_finish_page_flip(dev, 0);
-                }
-
                 if (pipe_stats[pipe] & PIPE_LEGACY_BLC_EVENT_STATUS)
                         blc_event = true;
 
@@ -1104,6 +1179,7 @@ static void i915_capture_error_state(struct drm_device *dev)
         kref_init(&error->ref);
         error->eir = I915_READ(EIR);
         error->pgtbl_er = I915_READ(PGTBL_ER);
+        error->ccid = I915_READ(CCID);
 
         if (HAS_PCH_SPLIT(dev))
                 error->ier = I915_READ(DEIER) | I915_READ(GTIER);
@@ -1426,23 +1502,20 @@ static int valleyview_enable_vblank(struct drm_device *dev, int pipe)
 {
         drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
         unsigned long irqflags;
-        u32 dpfl, imr;
+        u32 imr;
 
         if (!i915_pipe_enabled(dev, pipe))
                 return -EINVAL;
 
         spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-        dpfl = I915_READ(VLV_DPFLIPSTAT);
         imr = I915_READ(VLV_IMR);
-        if (pipe == 0) {
-                dpfl |= PIPEA_VBLANK_INT_EN;
+        if (pipe == 0)
                 imr &= ~I915_DISPLAY_PIPE_A_VBLANK_INTERRUPT;
-        } else {
-                dpfl |= PIPEA_VBLANK_INT_EN;
+        else
                 imr &= ~I915_DISPLAY_PIPE_B_VBLANK_INTERRUPT;
-        }
-        I915_WRITE(VLV_DPFLIPSTAT, dpfl);
         I915_WRITE(VLV_IMR, imr);
+        i915_enable_pipestat(dev_priv, pipe,
+                             PIPE_START_VBLANK_INTERRUPT_ENABLE);
         spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
 
         return 0;
@@ -1492,20 +1565,17 @@ static void valleyview_disable_vblank(struct drm_device *dev, int pipe)
 {
         drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
         unsigned long irqflags;
-        u32 dpfl, imr;
+        u32 imr;
 
         spin_lock_irqsave(&dev_priv->irq_lock, irqflags);
-        dpfl = I915_READ(VLV_DPFLIPSTAT);
+        i915_disable_pipestat(dev_priv, pipe,
+                              PIPE_START_VBLANK_INTERRUPT_ENABLE);
         imr = I915_READ(VLV_IMR);
-        if (pipe == 0) {
-                dpfl &= ~PIPEA_VBLANK_INT_EN;
+        if (pipe == 0)
                 imr |= I915_DISPLAY_PIPE_A_VBLANK_INTERRUPT;
-        } else {
-                dpfl &= ~PIPEB_VBLANK_INT_EN;
+        else
                 imr |= I915_DISPLAY_PIPE_B_VBLANK_INTERRUPT;
-        }
         I915_WRITE(VLV_IMR, imr);
-        I915_WRITE(VLV_DPFLIPSTAT, dpfl);
         spin_unlock_irqrestore(&dev_priv->irq_lock, irqflags);
 }
 
@@ -1648,7 +1718,6 @@ static void ironlake_irq_preinstall(struct drm_device *dev)
 
         atomic_set(&dev_priv->irq_received, 0);
 
-
         I915_WRITE(HWSTAM, 0xeffe);
 
         /* XXX hotplug from PCH */
@@ -1811,13 +1880,13 @@ static int ivybridge_irq_postinstall(struct drm_device *dev)
                    DE_PIPEA_VBLANK_IVB);
         POSTING_READ(DEIER);
 
-        dev_priv->gt_irq_mask = ~0;
+        dev_priv->gt_irq_mask = ~GT_GEN7_L3_PARITY_ERROR_INTERRUPT;
 
         I915_WRITE(GTIIR, I915_READ(GTIIR));
         I915_WRITE(GTIMR, dev_priv->gt_irq_mask);
 
         render_irqs = GT_USER_INTERRUPT | GEN6_BSD_USER_INTERRUPT |
-                GEN6_BLITTER_USER_INTERRUPT;
+                GEN6_BLITTER_USER_INTERRUPT | GT_GEN7_L3_PARITY_ERROR_INTERRUPT;
         I915_WRITE(GTIER, render_irqs);
         POSTING_READ(GTIER);
 
@@ -1840,16 +1909,24 @@ static int ivybridge_irq_postinstall(struct drm_device *dev)
 static int valleyview_irq_postinstall(struct drm_device *dev)
 {
         drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
-        u32 render_irqs;
         u32 enable_mask;
         u32 hotplug_en = I915_READ(PORT_HOTPLUG_EN);
+        u32 pipestat_enable = PLANE_FLIP_DONE_INT_EN_VLV;
         u16 msid;
 
         enable_mask = I915_DISPLAY_PORT_INTERRUPT;
-        enable_mask |= I915_DISPLAY_PIPE_A_VBLANK_INTERRUPT |
+        enable_mask |= I915_DISPLAY_PIPE_A_EVENT_INTERRUPT |
+                I915_DISPLAY_PIPE_A_VBLANK_INTERRUPT |
+                I915_DISPLAY_PIPE_B_EVENT_INTERRUPT |
                 I915_DISPLAY_PIPE_B_VBLANK_INTERRUPT;
 
-        dev_priv->irq_mask = ~enable_mask;
+        /*
+         *Leave vblank interrupts masked initially.  enable/disable will
+         * toggle them based on usage.
+         */
+        dev_priv->irq_mask = (~enable_mask) |
+                I915_DISPLAY_PIPE_A_VBLANK_INTERRUPT |
+                I915_DISPLAY_PIPE_B_VBLANK_INTERRUPT;
 
         dev_priv->pipestat[0] = 0;
         dev_priv->pipestat[1] = 0;
@@ -1868,26 +1945,27 @@ static int valleyview_irq_postinstall(struct drm_device *dev)
         I915_WRITE(PIPESTAT(1), 0xffff);
         POSTING_READ(VLV_IER);
 
+        i915_enable_pipestat(dev_priv, 0, pipestat_enable);
+        i915_enable_pipestat(dev_priv, 1, pipestat_enable);
+
         I915_WRITE(VLV_IIR, 0xffffffff);
         I915_WRITE(VLV_IIR, 0xffffffff);
 
-        render_irqs = GT_GEN6_BLT_FLUSHDW_NOTIFY_INTERRUPT |
-                GT_GEN6_BLT_CS_ERROR_INTERRUPT |
-                GT_GEN6_BLT_USER_INTERRUPT |
-                GT_GEN6_BSD_USER_INTERRUPT |
-                GT_GEN6_BSD_CS_ERROR_INTERRUPT |
-                GT_GEN7_L3_PARITY_ERROR_INTERRUPT |
-                GT_PIPE_NOTIFY |
-                GT_RENDER_CS_ERROR_INTERRUPT |
-                GT_SYNC_STATUS |
-                GT_USER_INTERRUPT;
-
-        dev_priv->gt_irq_mask = ~render_irqs;
+        dev_priv->gt_irq_mask = ~0;
 
         I915_WRITE(GTIIR, I915_READ(GTIIR));
         I915_WRITE(GTIIR, I915_READ(GTIIR));
-        I915_WRITE(GTIMR, 0);
-        I915_WRITE(GTIER, render_irqs);
+        I915_WRITE(GTIMR, dev_priv->gt_irq_mask);
+        I915_WRITE(GTIER, GT_GEN6_BLT_FLUSHDW_NOTIFY_INTERRUPT |
+                   GT_GEN6_BLT_CS_ERROR_INTERRUPT |
+                   GT_GEN6_BLT_USER_INTERRUPT |
+                   GT_GEN6_BSD_USER_INTERRUPT |
+                   GT_GEN6_BSD_CS_ERROR_INTERRUPT |
+                   GT_GEN7_L3_PARITY_ERROR_INTERRUPT |
+                   GT_PIPE_NOTIFY |
+                   GT_RENDER_CS_ERROR_INTERRUPT |
+                   GT_SYNC_STATUS |
+                   GT_USER_INTERRUPT);
         POSTING_READ(GTIER);
 
         /* ack & enable invalid PTE error interrupts */
@@ -2166,9 +2244,9 @@ static int i915_irq_postinstall(struct drm_device *dev)
                         hotplug_en |= HDMIC_HOTPLUG_INT_EN;
                 if (dev_priv->hotplug_supported_mask & HDMID_HOTPLUG_INT_STATUS)
                         hotplug_en |= HDMID_HOTPLUG_INT_EN;
-                if (dev_priv->hotplug_supported_mask & SDVOC_HOTPLUG_INT_STATUS)
+                if (dev_priv->hotplug_supported_mask & SDVOC_HOTPLUG_INT_STATUS_I915)
                         hotplug_en |= SDVOC_HOTPLUG_INT_EN;
-                if (dev_priv->hotplug_supported_mask & SDVOB_HOTPLUG_INT_STATUS)
+                if (dev_priv->hotplug_supported_mask & SDVOB_HOTPLUG_INT_STATUS_I915)
                         hotplug_en |= SDVOB_HOTPLUG_INT_EN;
                 if (dev_priv->hotplug_supported_mask & CRT_HOTPLUG_INT_STATUS) {
                         hotplug_en |= CRT_HOTPLUG_INT_EN;
@@ -2328,10 +2406,8 @@ static void i965_irq_preinstall(struct drm_device * dev)
 
         atomic_set(&dev_priv->irq_received, 0);
 
-        if (I915_HAS_HOTPLUG(dev)) {
-                I915_WRITE(PORT_HOTPLUG_EN, 0);
-                I915_WRITE(PORT_HOTPLUG_STAT, I915_READ(PORT_HOTPLUG_STAT));
-        }
+        I915_WRITE(PORT_HOTPLUG_EN, 0);
+        I915_WRITE(PORT_HOTPLUG_STAT, I915_READ(PORT_HOTPLUG_STAT));
 
         I915_WRITE(HWSTAM, 0xeffe);
         for_each_pipe(pipe)
@@ -2344,11 +2420,13 @@ static void i965_irq_preinstall(struct drm_device * dev)
 static int i965_irq_postinstall(struct drm_device *dev)
 {
         drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
+        u32 hotplug_en;
         u32 enable_mask;
         u32 error_mask;
 
         /* Unmask the interrupts that we always want on. */
         dev_priv->irq_mask = ~(I915_ASLE_INTERRUPT |
+                               I915_DISPLAY_PORT_INTERRUPT |
                                I915_DISPLAY_PIPE_A_EVENT_INTERRUPT |
                                I915_DISPLAY_PIPE_B_EVENT_INTERRUPT |
                                I915_DISPLAY_PLANE_A_FLIP_PENDING_INTERRUPT |
@@ -2364,13 +2442,6 @@ static int i965_irq_postinstall(struct drm_device *dev)
         dev_priv->pipestat[0] = 0;
         dev_priv->pipestat[1] = 0;
 
-        if (I915_HAS_HOTPLUG(dev)) {
-                /* Enable in IER... */
-                enable_mask |= I915_DISPLAY_PORT_INTERRUPT;
-                /* and unmask in IMR */
-                dev_priv->irq_mask &= ~I915_DISPLAY_PORT_INTERRUPT;
-        }
-
         /*
          * Enable some error detection, note the instruction error mask
          * bit is reserved, so we leave it masked.
@@ -2390,36 +2461,40 @@ static int i965_irq_postinstall(struct drm_device *dev)
         I915_WRITE(IER, enable_mask);
         POSTING_READ(IER);
 
-        if (I915_HAS_HOTPLUG(dev)) {
-                u32 hotplug_en = I915_READ(PORT_HOTPLUG_EN);
-
-                /* Note HDMI and DP share bits */
-                if (dev_priv->hotplug_supported_mask & HDMIB_HOTPLUG_INT_STATUS)
-                        hotplug_en |= HDMIB_HOTPLUG_INT_EN;
-                if (dev_priv->hotplug_supported_mask & HDMIC_HOTPLUG_INT_STATUS)
-                        hotplug_en |= HDMIC_HOTPLUG_INT_EN;
-                if (dev_priv->hotplug_supported_mask & HDMID_HOTPLUG_INT_STATUS)
-                        hotplug_en |= HDMID_HOTPLUG_INT_EN;
-                if (dev_priv->hotplug_supported_mask & SDVOC_HOTPLUG_INT_STATUS)
+        /* Note HDMI and DP share hotplug bits */
+        hotplug_en = 0;
+        if (dev_priv->hotplug_supported_mask & HDMIB_HOTPLUG_INT_STATUS)
+                hotplug_en |= HDMIB_HOTPLUG_INT_EN;
+        if (dev_priv->hotplug_supported_mask & HDMIC_HOTPLUG_INT_STATUS)
+                hotplug_en |= HDMIC_HOTPLUG_INT_EN;
+        if (dev_priv->hotplug_supported_mask & HDMID_HOTPLUG_INT_STATUS)
+                hotplug_en |= HDMID_HOTPLUG_INT_EN;
+        if (IS_G4X(dev)) {
+                if (dev_priv->hotplug_supported_mask & SDVOC_HOTPLUG_INT_STATUS_G4X)
                         hotplug_en |= SDVOC_HOTPLUG_INT_EN;
-                if (dev_priv->hotplug_supported_mask & SDVOB_HOTPLUG_INT_STATUS)
+                if (dev_priv->hotplug_supported_mask & SDVOB_HOTPLUG_INT_STATUS_G4X)
                         hotplug_en |= SDVOB_HOTPLUG_INT_EN;
-                if (dev_priv->hotplug_supported_mask & CRT_HOTPLUG_INT_STATUS) {
-                        hotplug_en |= CRT_HOTPLUG_INT_EN;
+        } else {
+                if (dev_priv->hotplug_supported_mask & SDVOC_HOTPLUG_INT_STATUS_I965)
+                        hotplug_en |= SDVOC_HOTPLUG_INT_EN;
+                if (dev_priv->hotplug_supported_mask & SDVOB_HOTPLUG_INT_STATUS_I965)
+                        hotplug_en |= SDVOB_HOTPLUG_INT_EN;
+        }
+        if (dev_priv->hotplug_supported_mask & CRT_HOTPLUG_INT_STATUS) {
+                hotplug_en |= CRT_HOTPLUG_INT_EN;
 
-                        /* Programming the CRT detection parameters tends
-                           to generate a spurious hotplug event about three
-                           seconds later.  So just do it once.
-                        */
-                        if (IS_G4X(dev))
-                                hotplug_en |= CRT_HOTPLUG_ACTIVATION_PERIOD_64;
-                        hotplug_en |= CRT_HOTPLUG_VOLTAGE_COMPARE_50;
-                }
+                /* Programming the CRT detection parameters tends
+                   to generate a spurious hotplug event about three
+                   seconds later.  So just do it once.
+                   */
+                if (IS_G4X(dev))
+                        hotplug_en |= CRT_HOTPLUG_ACTIVATION_PERIOD_64;
+                hotplug_en |= CRT_HOTPLUG_VOLTAGE_COMPARE_50;
+        }
 
-                /* Ignore TV since it's buggy */
+        /* Ignore TV since it's buggy */
 
-                I915_WRITE(PORT_HOTPLUG_EN, hotplug_en);
-        }
+        I915_WRITE(PORT_HOTPLUG_EN, hotplug_en);
 
         intel_opregion_enable_asle(dev);
 
@@ -2477,8 +2552,7 @@ static irqreturn_t i965_irq_handler(DRM_IRQ_ARGS)
                 ret = IRQ_HANDLED;
 
                 /* Consume port.  Then clear IIR or we'll miss events */
-                if ((I915_HAS_HOTPLUG(dev)) &&
-                    (iir & I915_DISPLAY_PORT_INTERRUPT)) {
+                if (iir & I915_DISPLAY_PORT_INTERRUPT) {
                         u32 hotplug_status = I915_READ(PORT_HOTPLUG_STAT);
 
                         DRM_DEBUG_DRIVER("hotplug event received, stat 0x%08x\n",
@@ -2551,10 +2625,8 @@ static void i965_irq_uninstall(struct drm_device * dev)
         if (!dev_priv)
                 return;
 
-        if (I915_HAS_HOTPLUG(dev)) {
-                I915_WRITE(PORT_HOTPLUG_EN, 0);
-                I915_WRITE(PORT_HOTPLUG_STAT, I915_READ(PORT_HOTPLUG_STAT));
-        }
+        I915_WRITE(PORT_HOTPLUG_EN, 0);
+        I915_WRITE(PORT_HOTPLUG_STAT, I915_READ(PORT_HOTPLUG_STAT));
 
         I915_WRITE(HWSTAM, 0xffffffff);
         for_each_pipe(pipe)
@@ -2575,6 +2647,7 @@ void intel_irq_init(struct drm_device *dev)
         INIT_WORK(&dev_priv->hotplug_work, i915_hotplug_work_func);
         INIT_WORK(&dev_priv->error_work, i915_error_work_func);
         INIT_WORK(&dev_priv->rps_work, gen6_pm_rps_work);
+        INIT_WORK(&dev_priv->parity_error_work, ivybridge_parity_work);
 
         dev->driver->get_vblank_counter = i915_get_vblank_counter;
         dev->max_vblank_count = 0xffffff; /* only 24 bits of frame count */
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 48d5e8e051cf..1218069c7f66 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -217,6 +217,9 @@
 #define   MI_DISPLAY_FLIP_IVB_SPRITE_B (3 << 19)
 #define   MI_DISPLAY_FLIP_IVB_PLANE_C  (4 << 19)
 #define   MI_DISPLAY_FLIP_IVB_SPRITE_C (5 << 19)
+#define MI_ARB_ON_OFF           MI_INSTR(0x08, 0)
+#define   MI_ARB_ENABLE                 (1<<0)
+#define   MI_ARB_DISABLE                (0<<0)
 
 #define MI_SET_CONTEXT          MI_INSTR(0x18, 0)
 #define   MI_MM_SPACE_GTT               (1<<8)
@@ -299,6 +302,7 @@
 #define   DISPLAY_PLANE_B           (1<<20)
 #define GFX_OP_PIPE_CONTROL(len)        ((0x3<<29)|(0x3<<27)|(0x2<<24)|(len-2))
 #define   PIPE_CONTROL_CS_STALL                         (1<<20)
+#define   PIPE_CONTROL_TLB_INVALIDATE                   (1<<18)
 #define   PIPE_CONTROL_QW_WRITE                         (1<<14)
 #define   PIPE_CONTROL_DEPTH_STALL                      (1<<13)
 #define   PIPE_CONTROL_WRITE_FLUSH                      (1<<12)
@@ -908,6 +912,7 @@
 #define   DPLL_P2_CLOCK_DIV_MASK        0x03000000 /* i915 */
 #define   DPLL_FPA01_P1_POST_DIV_MASK   0x00ff0000 /* i915 */
 #define   DPLL_FPA01_P1_POST_DIV_MASK_PINEVIEW  0x00ff8000 /* Pineview */
+#define   DPLL_LOCK_VLV                 (1<<15)
 #define   DPLL_INTEGRATED_CLOCK_VLV     (1<<13)
 
 #define SRX_INDEX               0x3c4
@@ -1453,6 +1458,10 @@
 #define DDRMPLL1                0X12c20
 #define PEG_BAND_GAP_DATA       0x14d68
 
+#define GEN6_GT_THREAD_STATUS_REG 0x13805c
+#define GEN6_GT_THREAD_STATUS_CORE_MASK 0x7
+#define GEN6_GT_THREAD_STATUS_CORE_MASK_HSW (0x7 | (0x07 << 16))
+
 #define GEN6_GT_PERF_STATUS     0x145948
 #define GEN6_RP_STATE_LIMITS    0x145994
 #define GEN6_RP_STATE_CAP       0x145998
@@ -1462,6 +1471,27 @@
  */
 #define CCID                    0x2180
 #define   CCID_EN               (1<<0)
+#define CXT_SIZE                0x21a0
+#define GEN6_CXT_POWER_SIZE(cxt_reg)    ((cxt_reg >> 24) & 0x3f)
+#define GEN6_CXT_RING_SIZE(cxt_reg)     ((cxt_reg >> 18) & 0x3f)
+#define GEN6_CXT_RENDER_SIZE(cxt_reg)   ((cxt_reg >> 12) & 0x3f)
+#define GEN6_CXT_EXTENDED_SIZE(cxt_reg) ((cxt_reg >> 6) & 0x3f)
+#define GEN6_CXT_PIPELINE_SIZE(cxt_reg) ((cxt_reg >> 0) & 0x3f)
+#define GEN6_CXT_TOTAL_SIZE(cxt_reg)    (GEN6_CXT_POWER_SIZE(cxt_reg) + \
+                                        GEN6_CXT_RING_SIZE(cxt_reg) + \
+                                        GEN6_CXT_RENDER_SIZE(cxt_reg) + \
+                                        GEN6_CXT_EXTENDED_SIZE(cxt_reg) + \
+                                        GEN6_CXT_PIPELINE_SIZE(cxt_reg))
+#define GEN7_CTX_SIZE           0x21a8
+#define GEN7_CTX_RENDER_SIZE(ctx_reg)   ((ctx_reg >> 16) & 0x3f)
+#define GEN7_CTX_EXTENDED_SIZE(ctx_reg) ((ctx_reg >> 9) & 0x7f)
+#define GEN7_CTX_GT1_SIZE(ctx_reg)      ((ctx_reg >> 6) & 0x7)
+#define GEN7_CTX_VFSTATE_SIZE(ctx_reg)  ((ctx_reg >> 0) & 0x3f)
+#define GEN7_CTX_TOTAL_SIZE(ctx_reg)    (GEN7_CTX_RENDER_SIZE(ctx_reg) + \
+                                         GEN7_CTX_EXTENDED_SIZE(ctx_reg) + \
+                                         GEN7_CTX_GT1_SIZE(ctx_reg) + \
+                                         GEN7_CTX_VFSTATE_SIZE(ctx_reg))
+
 /*
  * Overlay regs
  */
@@ -1566,20 +1596,34 @@
 #define CRT_HOTPLUG_DETECT_VOLTAGE_475MV        (1 << 2)
 
 #define PORT_HOTPLUG_STAT       0x61114
-#define   HDMIB_HOTPLUG_INT_STATUS              (1 << 29)
-#define   DPB_HOTPLUG_INT_STATUS                (1 << 29)
-#define   HDMIC_HOTPLUG_INT_STATUS              (1 << 28)
-#define   DPC_HOTPLUG_INT_STATUS                (1 << 28)
-#define   HDMID_HOTPLUG_INT_STATUS              (1 << 27)
-#define   DPD_HOTPLUG_INT_STATUS                (1 << 27)
+/* HDMI/DP bits are gen4+ */
+#define   DPB_HOTPLUG_LIVE_STATUS               (1 << 29)
+#define   DPC_HOTPLUG_LIVE_STATUS               (1 << 28)
+#define   DPD_HOTPLUG_LIVE_STATUS               (1 << 27)
+#define   DPD_HOTPLUG_INT_STATUS                (3 << 21)
+#define   DPC_HOTPLUG_INT_STATUS                (3 << 19)
+#define   DPB_HOTPLUG_INT_STATUS                (3 << 17)
+/* HDMI bits are shared with the DP bits */
+#define   HDMIB_HOTPLUG_LIVE_STATUS             (1 << 29)
+#define   HDMIC_HOTPLUG_LIVE_STATUS             (1 << 28)
+#define   HDMID_HOTPLUG_LIVE_STATUS             (1 << 27)
+#define   HDMID_HOTPLUG_INT_STATUS              (3 << 21)
+#define   HDMIC_HOTPLUG_INT_STATUS              (3 << 19)
+#define   HDMIB_HOTPLUG_INT_STATUS              (3 << 17)
+/* CRT/TV common between gen3+ */
 #define   CRT_HOTPLUG_INT_STATUS                (1 << 11)
 #define   TV_HOTPLUG_INT_STATUS                 (1 << 10)
 #define   CRT_HOTPLUG_MONITOR_MASK              (3 << 8)
 #define   CRT_HOTPLUG_MONITOR_COLOR             (3 << 8)
 #define   CRT_HOTPLUG_MONITOR_MONO              (2 << 8)
 #define   CRT_HOTPLUG_MONITOR_NONE              (0 << 8)
-#define   SDVOC_HOTPLUG_INT_STATUS              (1 << 7)
-#define   SDVOB_HOTPLUG_INT_STATUS              (1 << 6)
+/* SDVO is different across gen3/4 */
+#define   SDVOC_HOTPLUG_INT_STATUS_G4X          (1 << 3)
+#define   SDVOB_HOTPLUG_INT_STATUS_G4X          (1 << 2)
+#define   SDVOC_HOTPLUG_INT_STATUS_I965         (3 << 4)
+#define   SDVOB_HOTPLUG_INT_STATUS_I965         (3 << 2)
+#define   SDVOC_HOTPLUG_INT_STATUS_I915         (1 << 7)
+#define   SDVOB_HOTPLUG_INT_STATUS_I915         (1 << 6)
 
 /* SDVO port control */
 #define SDVOB                   0x61140
@@ -1711,8 +1755,10 @@
 #define   VIDEO_DIP_PORT_C              (2 << 29)
 #define   VIDEO_DIP_PORT_D              (3 << 29)
 #define   VIDEO_DIP_PORT_MASK           (3 << 29)
+#define   VIDEO_DIP_ENABLE_GCP          (1 << 25)
 #define   VIDEO_DIP_ENABLE_AVI          (1 << 21)
 #define   VIDEO_DIP_ENABLE_VENDOR       (2 << 21)
+#define   VIDEO_DIP_ENABLE_GAMUT        (4 << 21)
 #define   VIDEO_DIP_ENABLE_SPD          (8 << 21)
 #define   VIDEO_DIP_SELECT_AVI          (0 << 19)
 #define   VIDEO_DIP_SELECT_VENDOR       (1 << 19)
@@ -1723,7 +1769,11 @@
 #define   VIDEO_DIP_FREQ_2VSYNC         (2 << 16)
 #define   VIDEO_DIP_FREQ_MASK           (3 << 16)
 /* HSW and later: */
+#define   VIDEO_DIP_ENABLE_VSC_HSW      (1 << 20)
+#define   VIDEO_DIP_ENABLE_GCP_HSW      (1 << 16)
 #define   VIDEO_DIP_ENABLE_AVI_HSW      (1 << 12)
+#define   VIDEO_DIP_ENABLE_VS_HSW       (1 << 8)
+#define   VIDEO_DIP_ENABLE_GMP_HSW      (1 << 4)
 #define   VIDEO_DIP_ENABLE_SPD_HSW      (1 << 0)
 
 /* Panel power sequencing */
@@ -1795,18 +1845,35 @@
 #define PFIT_AUTO_RATIOS 0x61238
 
 /* Backlight control */
-#define BLC_PWM_CTL             0x61254
-#define   BACKLIGHT_MODULATION_FREQ_SHIFT               (17)
 #define BLC_PWM_CTL2            0x61250 /* 965+ only */
-#define   BLM_COMBINATION_MODE (1 << 30)
+#define   BLM_PWM_ENABLE                (1 << 31)
+#define   BLM_COMBINATION_MODE          (1 << 30) /* gen4 only */
+#define   BLM_PIPE_SELECT               (1 << 29)
+#define   BLM_PIPE_SELECT_IVB           (3 << 29)
+#define   BLM_PIPE_A                    (0 << 29)
+#define   BLM_PIPE_B                    (1 << 29)
+#define   BLM_PIPE_C                    (2 << 29) /* ivb + */
+#define   BLM_PIPE(pipe)                ((pipe) << 29)
+#define   BLM_POLARITY_I965             (1 << 28) /* gen4 only */
+#define   BLM_PHASE_IN_INTERUPT_STATUS  (1 << 26)
+#define   BLM_PHASE_IN_ENABLE           (1 << 25)
+#define   BLM_PHASE_IN_INTERUPT_ENABL   (1 << 24)
+#define   BLM_PHASE_IN_TIME_BASE_SHIFT  (16)
+#define   BLM_PHASE_IN_TIME_BASE_MASK   (0xff << 16)
+#define   BLM_PHASE_IN_COUNT_SHIFT      (8)
+#define   BLM_PHASE_IN_COUNT_MASK       (0xff << 8)
+#define   BLM_PHASE_IN_INCR_SHIFT       (0)
+#define   BLM_PHASE_IN_INCR_MASK        (0xff << 0)
+#define BLC_PWM_CTL             0x61254
 /*
  * This is the most significant 15 bits of the number of backlight cycles in a
  * complete cycle of the modulated backlight control.
  *
  * The actual value is this field multiplied by two.
  */
-#define   BACKLIGHT_MODULATION_FREQ_MASK                (0x7fff << 17)
-#define   BLM_LEGACY_MODE                               (1 << 16)
+#define   BACKLIGHT_MODULATION_FREQ_SHIFT       (17)
+#define   BACKLIGHT_MODULATION_FREQ_MASK        (0x7fff << 17)
+#define   BLM_LEGACY_MODE                       (1 << 16) /* gen2 only */
 /*
  * This is the number of cycles out of the backlight modulation cycle for which
  * the backlight is on.
@@ -1816,9 +1883,24 @@
  */
 #define   BACKLIGHT_DUTY_CYCLE_SHIFT            (0)
 #define   BACKLIGHT_DUTY_CYCLE_MASK             (0xffff)
+#define   BACKLIGHT_DUTY_CYCLE_MASK_PNV         (0xfffe)
+#define   BLM_POLARITY_PNV                      (1 << 0) /* pnv only */
 
 #define BLC_HIST_CTL            0x61260
 
+/* New registers for PCH-split platforms. Safe where new bits show up, the
+ * register layout machtes with gen4 BLC_PWM_CTL[12]. */
+#define BLC_PWM_CPU_CTL2        0x48250
+#define BLC_PWM_CPU_CTL         0x48254
+
+/* PCH CTL1 is totally different, all but the below bits are reserved. CTL2 is
+ * like the normal CTL from gen4 and earlier. Hooray for confusing naming. */
+#define BLC_PWM_PCH_CTL1        0xc8250
+#define   BLM_PCH_PWM_ENABLE                    (1 << 30)
+#define   BLM_PCH_OVERRIDE_ENABLE               (1 << 30)
+#define   BLM_PCH_POLARITY                      (1 << 29)
+#define BLC_PWM_PCH_CTL2        0xc8254
+
 /* TV port control */
 #define TV_CTL                  0x68000
 /** Enables the TV encoder */
@@ -2583,13 +2665,13 @@
 #define PIPESTAT(pipe) _PIPE(pipe, _PIPEASTAT, _PIPEBSTAT)
 
 #define VLV_DPFLIPSTAT                          0x70028
-#define   PIPEB_LINE_COMPARE_STATUS             (1<<29)
+#define   PIPEB_LINE_COMPARE_INT_EN             (1<<29)
 #define   PIPEB_HLINE_INT_EN                    (1<<28)
 #define   PIPEB_VBLANK_INT_EN                   (1<<27)
 #define   SPRITED_FLIPDONE_INT_EN               (1<<26)
 #define   SPRITEC_FLIPDONE_INT_EN               (1<<25)
 #define   PLANEB_FLIPDONE_INT_EN                (1<<24)
-#define   PIPEA_LINE_COMPARE_STATUS             (1<<21)
+#define   PIPEA_LINE_COMPARE_INT_EN             (1<<21)
 #define   PIPEA_HLINE_INT_EN                    (1<<20)
 #define   PIPEA_VBLANK_INT_EN                   (1<<19)
 #define   SPRITEB_FLIPDONE_INT_EN               (1<<18)
@@ -2897,13 +2979,14 @@
 #define DSPSIZE(plane) _PIPE(plane, _DSPASIZE, _DSPBSIZE)
 #define DSPSURF(plane) _PIPE(plane, _DSPASURF, _DSPBSURF)
 #define DSPTILEOFF(plane) _PIPE(plane, _DSPATILEOFF, _DSPBTILEOFF)
+#define DSPLINOFF(plane) DSPADDR(plane)
 
 /* Display/Sprite base address macros */
 #define DISP_BASEADDR_MASK      (0xfffff000)
 #define I915_LO_DISPBASE(val)   (val & ~DISP_BASEADDR_MASK)
 #define I915_HI_DISPBASE(val)   (val & DISP_BASEADDR_MASK)
 #define I915_MODIFY_DISPBASE(reg, gfx_addr) \
-                (I915_WRITE(reg, gfx_addr | I915_LO_DISPBASE(I915_READ(reg))))
+                (I915_WRITE((reg), (gfx_addr) | I915_LO_DISPBASE(I915_READ(reg))))
 
 /* VBIOS flags */
 #define SWF00                   0x71410
@@ -3771,6 +3854,9 @@
 #define _FDI_RXA_TUSIZE2         0xf0038
 #define _FDI_RXB_TUSIZE1         0xf1030
 #define _FDI_RXB_TUSIZE2         0xf1038
+#define  FDI_RX_TP1_TO_TP2_48   (2<<20)
+#define  FDI_RX_TP1_TO_TP2_64   (3<<20)
+#define  FDI_RX_FDI_DELAY_90    (0x90<<0)
 #define FDI_RX_MISC(pipe) _PIPE(pipe, _FDI_RXA_MISC, _FDI_RXB_MISC)
 #define FDI_RX_TUSIZE1(pipe) _PIPE(pipe, _FDI_RXA_TUSIZE1, _FDI_RXB_TUSIZE1)
 #define FDI_RX_TUSIZE2(pipe) _PIPE(pipe, _FDI_RXA_TUSIZE2, _FDI_RXB_TUSIZE2)
@@ -3824,7 +3910,6 @@
 #define  ADPA_CRT_HOTPLUG_FORCE_TRIGGER (1<<16)
 
 /* or SDVOB */
-#define VLV_HDMIB 0x61140
 #define HDMIB   0xe1140
 #define  PORT_ENABLE    (1 << 31)
 #define  TRANSCODER(pipe)       ((pipe) << 30)
@@ -3855,20 +3940,18 @@
 #define PCH_LVDS        0xe1180
 #define  LVDS_DETECTED  (1 << 1)
 
-#define BLC_PWM_CPU_CTL2        0x48250
-#define  PWM_ENABLE             (1 << 31)
-#define  PWM_PIPE_A             (0 << 29)
-#define  PWM_PIPE_B             (1 << 29)
-#define BLC_PWM_CPU_CTL         0x48254
+/* vlv has 2 sets of panel control regs. */
+#define PIPEA_PP_STATUS         0x61200
+#define PIPEA_PP_CONTROL        0x61204
+#define PIPEA_PP_ON_DELAYS      0x61208
+#define PIPEA_PP_OFF_DELAYS     0x6120c
+#define PIPEA_PP_DIVISOR        0x61210
 
-#define BLC_PWM_PCH_CTL1        0xc8250
-#define  PWM_PCH_ENABLE         (1 << 31)
-#define  PWM_POLARITY_ACTIVE_LOW        (1 << 29)
-#define  PWM_POLARITY_ACTIVE_HIGH       (0 << 29)
-#define  PWM_POLARITY_ACTIVE_LOW2       (1 << 28)
-#define  PWM_POLARITY_ACTIVE_HIGH2      (0 << 28)
-
-#define BLC_PWM_PCH_CTL2        0xc8254
+#define PIPEB_PP_STATUS         0x61300
+#define PIPEB_PP_CONTROL        0x61304
+#define PIPEB_PP_ON_DELAYS      0x61308
+#define PIPEB_PP_OFF_DELAYS     0x6130c
+#define PIPEB_PP_DIVISOR        0x61310
 
 #define PCH_PP_STATUS           0xc7200
 #define PCH_PP_CONTROL          0xc7204
@@ -3992,6 +4075,7 @@
 #define  FORCEWAKE                              0xA18C
 #define  FORCEWAKE_VLV                          0x1300b0
 #define  FORCEWAKE_ACK_VLV                      0x1300b4
+#define  FORCEWAKE_ACK_HSW                      0x130044
 #define  FORCEWAKE_ACK                          0x130090
 #define  FORCEWAKE_MT                           0xa188 /* multi-threaded */
 #define  FORCEWAKE_MT_ACK                       0x130040
@@ -4012,10 +4096,15 @@
 # define GEN6_CSUNIT_CLOCK_GATE_DISABLE                 (1 << 7)
 
 #define GEN6_UCGCTL2                            0x9404
+# define GEN7_VDSUNIT_CLOCK_GATE_DISABLE                (1 << 30)
+# define GEN7_TDLUNIT_CLOCK_GATE_DISABLE                (1 << 22)
 # define GEN6_RCZUNIT_CLOCK_GATE_DISABLE                (1 << 13)
 # define GEN6_RCPBUNIT_CLOCK_GATE_DISABLE               (1 << 12)
 # define GEN6_RCCUNIT_CLOCK_GATE_DISABLE                (1 << 11)
 
+#define GEN7_UCGCTL4                            0x940c
+#define  GEN7_L3BANK2X_CLOCK_GATE_DISABLE       (1<<25)
+
 #define GEN6_RPNSWREQ                           0xA008
 #define   GEN6_TURBO_DISABLE                    (1<<31)
 #define   GEN6_FREQUENCY(x)                     ((x)<<25)
@@ -4047,6 +4136,7 @@
 #define   GEN6_RP_UP_IDLE_MIN                   (0x1<<3)
 #define   GEN6_RP_UP_BUSY_AVG                   (0x2<<3)
 #define   GEN6_RP_UP_BUSY_CONT                  (0x4<<3)
+#define   GEN7_RP_DOWN_IDLE_AVG                 (0x2<<0)
 #define   GEN6_RP_DOWN_IDLE_CONT                (0x1<<0)
 #define GEN6_RP_UP_THRESHOLD                    0xA02C
 #define GEN6_RP_DOWN_THRESHOLD                  0xA030
@@ -4111,6 +4201,26 @@
 #define   GEN6_RC6                      3
 #define   GEN6_RC7                      4
 
+#define GEN7_MISCCPCTL                  (0x9424)
+#define   GEN7_DOP_CLOCK_GATE_ENABLE    (1<<0)
+
+/* IVYBRIDGE DPF */
+#define GEN7_L3CDERRST1                 0xB008 /* L3CD Error Status 1 */
+#define   GEN7_L3CDERRST1_ROW_MASK      (0x7ff<<14)
+#define   GEN7_PARITY_ERROR_VALID       (1<<13)
+#define   GEN7_L3CDERRST1_BANK_MASK     (3<<11)
+#define   GEN7_L3CDERRST1_SUBBANK_MASK  (7<<8)
+#define GEN7_PARITY_ERROR_ROW(reg) \
+                ((reg & GEN7_L3CDERRST1_ROW_MASK) >> 14)
+#define GEN7_PARITY_ERROR_BANK(reg) \
+                ((reg & GEN7_L3CDERRST1_BANK_MASK) >> 11)
+#define GEN7_PARITY_ERROR_SUBBANK(reg) \
+                ((reg & GEN7_L3CDERRST1_SUBBANK_MASK) >> 8)
+#define   GEN7_L3CDERRST1_ENABLE        (1<<7)
+
+#define GEN7_L3LOG_BASE                 0xB070
+#define GEN7_L3LOG_SIZE                 0x80
+
 #define G4X_AUD_VID_DID                 0x62020
 #define INTEL_AUDIO_DEVCL               0x808629FB
 #define INTEL_AUDIO_DEVBLC              0x80862801
@@ -4177,7 +4287,7 @@
                                         PIPE_DDI_FUNC_CTL_B)
 #define  PIPE_DDI_FUNC_ENABLE           (1<<31)
 /* Those bits are ignored by pipe EDP since it can only connect to DDI A */
-#define  PIPE_DDI_PORT_MASK                             (0xf<<28)
+#define  PIPE_DDI_PORT_MASK                     (7<<28)
 #define  PIPE_DDI_SELECT_PORT(x)                ((x)<<28)
 #define  PIPE_DDI_MODE_SELECT_HDMI              (0<<24)
 #define  PIPE_DDI_MODE_SELECT_DVI               (1<<24)
@@ -4335,7 +4445,7 @@
 #define PIPE_WM_LINETIME_B              0x45274
 #define PIPE_WM_LINETIME(pipe) _PIPE(pipe, \
                                         PIPE_WM_LINETIME_A, \
-                                        PIPE_WM_LINETIME_A)
+                                        PIPE_WM_LINETIME_B)
 #define   PIPE_WM_LINETIME_MASK         (0x1ff)
 #define   PIPE_WM_LINETIME_TIME(x)                      ((x))
 #define   PIPE_WM_LINETIME_IPS_LINETIME_MASK    (0x1ff<<16)
@@ -4347,4 +4457,9 @@
 #define  SFUSE_STRAP_DDIC_DETECTED      (1<<1)
 #define  SFUSE_STRAP_DDID_DETECTED      (1<<0)
 
+#define WM_DBG                          0x45280
+#define  WM_DBG_DISALLOW_MULTIPLE_LP    (1<<0)
+#define  WM_DBG_DISALLOW_MAXFIFO        (1<<1)
+#define  WM_DBG_DISALLOW_SPRITE         (1<<2)
+
 #endif /* _I915_REG_H_ */
diff --git a/drivers/gpu/drm/i915/i915_suspend.c b/drivers/gpu/drm/i915/i915_suspend.c
index a748e5cabe14..4776ccf1b3cd 100644
--- a/drivers/gpu/drm/i915/i915_suspend.c
+++ b/drivers/gpu/drm/i915/i915_suspend.c
@@ -828,10 +828,7 @@ int i915_save_state(struct drm_device *dev)
                 dev_priv->saveIMR = I915_READ(IMR);
         }
 
-        if (IS_IRONLAKE_M(dev))
-                ironlake_disable_drps(dev);
-        if (INTEL_INFO(dev)->gen >= 6)
-                gen6_disable_rps(dev);
+        intel_disable_gt_powersave(dev);
 
         /* Cache mode state */
         dev_priv->saveCACHE_MODE_0 = I915_READ(CACHE_MODE_0);
diff --git a/drivers/gpu/drm/i915/i915_sysfs.c b/drivers/gpu/drm/i915/i915_sysfs.c
index 79f83445afa0..2f5388af8df9 100644
--- a/drivers/gpu/drm/i915/i915_sysfs.c
+++ b/drivers/gpu/drm/i915/i915_sysfs.c
@@ -29,6 +29,7 @@
 #include <linux/module.h>
 #include <linux/stat.h>
 #include <linux/sysfs.h>
+#include "intel_drv.h"
 #include "i915_drv.h"
 
 static u32 calc_residency(struct drm_device *dev, const u32 reg)
@@ -92,20 +93,134 @@ static struct attribute_group rc6_attr_group = {
         .attrs =  rc6_attrs
 };
 
-void i915_setup_sysfs(struct drm_device *dev)
+static int l3_access_valid(struct drm_device *dev, loff_t offset)
+{
+        if (!IS_IVYBRIDGE(dev))
+                return -EPERM;
+
+        if (offset % 4 != 0)
+                return -EINVAL;
+
+        if (offset >= GEN7_L3LOG_SIZE)
+                return -ENXIO;
+
+        return 0;
+}
+
+static ssize_t
+i915_l3_read(struct file *filp, struct kobject *kobj,
+             struct bin_attribute *attr, char *buf,
+             loff_t offset, size_t count)
+{
+        struct device *dev = container_of(kobj, struct device, kobj);
+        struct drm_minor *dminor = container_of(dev, struct drm_minor, kdev);
+        struct drm_device *drm_dev = dminor->dev;
+        struct drm_i915_private *dev_priv = drm_dev->dev_private;
+        uint32_t misccpctl;
+        int i, ret;
+
+        ret = l3_access_valid(drm_dev, offset);
+        if (ret)
+                return ret;
+
+        ret = i915_mutex_lock_interruptible(drm_dev);
+        if (ret)
+                return ret;
+
+        misccpctl = I915_READ(GEN7_MISCCPCTL);
+        I915_WRITE(GEN7_MISCCPCTL, misccpctl & ~GEN7_DOP_CLOCK_GATE_ENABLE);
+
+        for (i = offset; count >= 4 && i < GEN7_L3LOG_SIZE; i += 4, count -= 4)
+                *((uint32_t *)(&buf[i])) = I915_READ(GEN7_L3LOG_BASE + i);
+
+        I915_WRITE(GEN7_MISCCPCTL, misccpctl);
+
+        mutex_unlock(&drm_dev->struct_mutex);
+
+        return i - offset;
+}
+
+static ssize_t
+i915_l3_write(struct file *filp, struct kobject *kobj,
+              struct bin_attribute *attr, char *buf,
+              loff_t offset, size_t count)
 {
+        struct device *dev = container_of(kobj, struct device, kobj);
+        struct drm_minor *dminor = container_of(dev, struct drm_minor, kdev);
+        struct drm_device *drm_dev = dminor->dev;
+        struct drm_i915_private *dev_priv = drm_dev->dev_private;
+        u32 *temp = NULL; /* Just here to make handling failures easy */
         int ret;
 
-        /* ILK doesn't have any residency information */
-        if (INTEL_INFO(dev)->gen < 6)
-                return;
+        ret = l3_access_valid(drm_dev, offset);
+        if (ret)
+                return ret;
 
-        ret = sysfs_merge_group(&dev->primary->kdev.kobj, &rc6_attr_group);
+        ret = i915_mutex_lock_interruptible(drm_dev);
         if (ret)
-                DRM_ERROR("sysfs setup failed\n");
+                return ret;
+
+        if (!dev_priv->mm.l3_remap_info) {
+                temp = kzalloc(GEN7_L3LOG_SIZE, GFP_KERNEL);
+                if (!temp) {
+                        mutex_unlock(&drm_dev->struct_mutex);
+                        return -ENOMEM;
+                }
+        }
+
+        ret = i915_gpu_idle(drm_dev);
+        if (ret) {
+                kfree(temp);
+                mutex_unlock(&drm_dev->struct_mutex);
+                return ret;
+        }
+
+        /* TODO: Ideally we really want a GPU reset here to make sure errors
+         * aren't propagated. Since I cannot find a stable way to reset the GPU
+         * at this point it is left as a TODO.
+        */
+        if (temp)
+                dev_priv->mm.l3_remap_info = temp;
+
+        memcpy(dev_priv->mm.l3_remap_info + (offset/4),
+               buf + (offset/4),
+               count);
+
+        i915_gem_l3_remap(drm_dev);
+
+        mutex_unlock(&drm_dev->struct_mutex);
+
+        return count;
+}
+
+static struct bin_attribute dpf_attrs = {
+        .attr = {.name = "l3_parity", .mode = (S_IRUSR | S_IWUSR)},
+        .size = GEN7_L3LOG_SIZE,
+        .read = i915_l3_read,
+        .write = i915_l3_write,
+        .mmap = NULL
+};
+
+void i915_setup_sysfs(struct drm_device *dev)
+{
+        int ret;
+
+        if (INTEL_INFO(dev)->gen >= 6) {
+                ret = sysfs_merge_group(&dev->primary->kdev.kobj,
+                                        &rc6_attr_group);
+                if (ret)
+                        DRM_ERROR("RC6 residency sysfs setup failed\n");
+        }
+
+        if (IS_IVYBRIDGE(dev)) {
+                ret = device_create_bin_file(&dev->primary->kdev, &dpf_attrs);
+                if (ret)
+                        DRM_ERROR("l3 parity sysfs setup failed\n");
+        }
 }
 
 void i915_teardown_sysfs(struct drm_device *dev)
 {
+        device_remove_bin_file(&dev->primary->kdev,  &dpf_attrs);
         sysfs_unmerge_group(&dev->primary->kdev.kobj, &rc6_attr_group);
 }
diff --git a/drivers/gpu/drm/i915/i915_trace.h b/drivers/gpu/drm/i915/i915_trace.h
index dac7bba4d9da..fe90b3a84a6d 100644
--- a/drivers/gpu/drm/i915/i915_trace.h
+++ b/drivers/gpu/drm/i915/i915_trace.h
@@ -311,9 +311,33 @@ DEFINE_EVENT(i915_gem_request, i915_gem_request_retire,
             TP_ARGS(ring, seqno)
 );
 
-DEFINE_EVENT(i915_gem_request, i915_gem_request_wait_begin,
+TRACE_EVENT(i915_gem_request_wait_begin,
             TP_PROTO(struct intel_ring_buffer *ring, u32 seqno),
-            TP_ARGS(ring, seqno)
+            TP_ARGS(ring, seqno),
+
+            TP_STRUCT__entry(
+                             __field(u32, dev)
+                             __field(u32, ring)
+                             __field(u32, seqno)
+                             __field(bool, blocking)
+                             ),
+
+            /* NB: the blocking information is racy since mutex_is_locked
+             * doesn't check that the current thread holds the lock. The only
+             * other option would be to pass the boolean information of whether
+             * or not the class was blocking down through the stack which is
+             * less desirable.
+             */
+            TP_fast_assign(
+                           __entry->dev = ring->dev->primary->index;
+                           __entry->ring = ring->id;
+                           __entry->seqno = seqno;
+                           __entry->blocking = mutex_is_locked(&ring->dev->struct_mutex);
+                           ),
+
+            TP_printk("dev=%u, ring=%u, seqno=%u, blocking=%s",
+                      __entry->dev, __entry->ring, __entry->seqno,
+                      __entry->blocking ?  "yes (NB)" : "no")
 );
 
 DEFINE_EVENT(i915_gem_request, i915_gem_request_wait_end,
diff --git a/drivers/gpu/drm/i915/intel_bios.c b/drivers/gpu/drm/i915/intel_bios.c
index 353459362f6f..8c6074154bf6 100644
--- a/drivers/gpu/drm/i915/intel_bios.c
+++ b/drivers/gpu/drm/i915/intel_bios.c
@@ -692,7 +692,7 @@ static const struct dmi_system_id intel_no_opregion_vbt[] = {
  *
  * Returns 0 on success, nonzero on failure.
  */
-bool
+int
 intel_parse_bios(struct drm_device *dev)
 {
         struct drm_i915_private *dev_priv = dev->dev_private;
diff --git a/drivers/gpu/drm/i915/intel_bios.h b/drivers/gpu/drm/i915/intel_bios.h
index dbda6e3bdf07..31c2107e7825 100644
--- a/drivers/gpu/drm/i915/intel_bios.h
+++ b/drivers/gpu/drm/i915/intel_bios.h
@@ -476,7 +476,7 @@ struct bdb_edp {
 } __attribute__ ((packed));
 
 void intel_setup_bios(struct drm_device *dev);
-bool intel_parse_bios(struct drm_device *dev);
+int intel_parse_bios(struct drm_device *dev);
 
 /*
  * Driver<->VBIOS interaction occurs through scratch bits in
diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c
index 75a70c46ef1b..48e3b76e0ab2 100644
--- a/drivers/gpu/drm/i915/intel_crt.c
+++ b/drivers/gpu/drm/i915/intel_crt.c
@@ -88,6 +88,9 @@ static void gmch_crt_dpms(struct drm_encoder *encoder, int mode)
         temp &= ~(ADPA_HSYNC_CNTL_DISABLE | ADPA_VSYNC_CNTL_DISABLE);
         temp &= ~ADPA_DAC_ENABLE;
 
+        if (IS_VALLEYVIEW(dev) && mode != DRM_MODE_DPMS_ON)
+                mode = DRM_MODE_DPMS_OFF;
+
         switch (mode) {
         case DRM_MODE_DPMS_ON:
                 temp |= ADPA_DAC_ENABLE;
@@ -129,7 +132,7 @@ static int intel_crt_mode_valid(struct drm_connector *connector,
 }
 
 static bool intel_crt_mode_fixup(struct drm_encoder *encoder,
-                                 struct drm_display_mode *mode,
+                                 const struct drm_display_mode *mode,
                                  struct drm_display_mode *adjusted_mode)
 {
         return true;
@@ -230,6 +233,42 @@ static bool intel_ironlake_crt_detect_hotplug(struct drm_connector *connector)
         return ret;
 }
 
+static bool valleyview_crt_detect_hotplug(struct drm_connector *connector)
+{
+        struct drm_device *dev = connector->dev;
+        struct drm_i915_private *dev_priv = dev->dev_private;
+        u32 adpa;
+        bool ret;
+        u32 save_adpa;
+
+        save_adpa = adpa = I915_READ(ADPA);
+        DRM_DEBUG_KMS("trigger hotplug detect cycle: adpa=0x%x\n", adpa);
+
+        adpa |= ADPA_CRT_HOTPLUG_FORCE_TRIGGER;
+
+        I915_WRITE(ADPA, adpa);
+
+        if (wait_for((I915_READ(ADPA) & ADPA_CRT_HOTPLUG_FORCE_TRIGGER) == 0,
+                     1000)) {
+                DRM_DEBUG_KMS("timed out waiting for FORCE_TRIGGER");
+                I915_WRITE(ADPA, save_adpa);
+        }
+
+        /* Check the status to see if both blue and green are on now */
+        adpa = I915_READ(ADPA);
+        if ((adpa & ADPA_CRT_HOTPLUG_MONITOR_MASK) != 0)
+                ret = true;
+        else
+                ret = false;
+
+        DRM_DEBUG_KMS("valleyview hotplug adpa=0x%x, result %d\n", adpa, ret);
+
+        /* FIXME: debug force function and remove */
+        ret = true;
+
+        return ret;
+}
+
 /**
  * Uses CRT_HOTPLUG_EN and CRT_HOTPLUG_STAT to detect CRT presence.
  *
@@ -249,6 +288,9 @@ static bool intel_crt_detect_hotplug(struct drm_connector *connector)
         if (HAS_PCH_SPLIT(dev))
                 return intel_ironlake_crt_detect_hotplug(connector);
 
+        if (IS_VALLEYVIEW(dev))
+                return valleyview_crt_detect_hotplug(connector);
+
         /*
          * On 4 series desktop, CRT detect sequence need to be done twice
          * to get a reliable result.
@@ -453,18 +495,27 @@ intel_crt_detect(struct drm_connector *connector, bool force)
         struct intel_load_detect_pipe tmp;
 
         if (I915_HAS_HOTPLUG(dev)) {
+                /* We can not rely on the HPD pin always being correctly wired
+                 * up, for example many KVM do not pass it through, and so
+                 * only trust an assertion that the monitor is connected.
+                 */
                 if (intel_crt_detect_hotplug(connector)) {
                         DRM_DEBUG_KMS("CRT detected via hotplug\n");
                         return connector_status_connected;
-                } else {
+                } else
                         DRM_DEBUG_KMS("CRT not detected via hotplug\n");
-                        return connector_status_disconnected;
-                }
         }
 
         if (intel_crt_detect_ddc(connector))
                 return connector_status_connected;
 
+        /* Load detection is broken on HPD capable machines. Whoever wants a
+         * broken monitor (without edid) to work behind a broken kvm (that fails
+         * to have the right resistors for HP detection) needs to fix this up.
+         * For now just bail out. */
+        if (I915_HAS_HOTPLUG(dev))
+                return connector_status_disconnected;
+
         if (!force)
                 return connector->status;
 
diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index 46d1e886c692..933c74859172 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -170,6 +170,15 @@ void hsw_fdi_link_train(struct drm_crtc *crtc)
 
                 udelay(600);
 
+                /* We need to program FDI_RX_MISC with the default TP1 to TP2
+                 * values before enabling the receiver, and configure the delay
+                 * for the FDI timing generator to 90h. Luckily, all the other
+                 * bits are supposed to be zeroed, so we can write those values
+                 * directly.
+                 */
+                I915_WRITE(FDI_RX_MISC(pipe), FDI_RX_TP1_TO_TP2_48 |
+                                FDI_RX_FDI_DELAY_90);
+
                 /* Enable CPU FDI Receiver with auto-training */
                 reg = FDI_RX_CTL(pipe);
                 I915_WRITE(reg,
@@ -726,8 +735,7 @@ void intel_ddi_mode_set(struct drm_encoder *encoder,
 
         I915_WRITE(DDI_FUNC_CTL(pipe), temp);
 
-        intel_hdmi_set_avi_infoframe(encoder, adjusted_mode);
-        intel_hdmi_set_spd_infoframe(encoder);
+        intel_hdmi->set_infoframes(encoder, adjusted_mode);
 }
 
 void intel_ddi_dpms(struct drm_encoder *encoder, int mode)
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index a8538ac0299d..bc25a24a08c5 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -98,6 +98,11 @@ intel_find_pll_ironlake_dp(const intel_limit_t *, struct drm_crtc *crtc,
                            int target, int refclk, intel_clock_t *match_clock,
                            intel_clock_t *best_clock);
 
+static bool
+intel_vlv_find_best_pll(const intel_limit_t *limit, struct drm_crtc *crtc,
+                        int target, int refclk, intel_clock_t *match_clock,
+                        intel_clock_t *best_clock);
+
 static inline u32 /* units of 100MHz */
 intel_fdi_link_freq(struct drm_device *dev)
 {
@@ -359,6 +364,48 @@ static const intel_limit_t intel_limits_ironlake_display_port = {
         .find_pll = intel_find_pll_ironlake_dp,
 };
 
+static const intel_limit_t intel_limits_vlv_dac = {
+        .dot = { .min = 25000, .max = 270000 },
+        .vco = { .min = 4000000, .max = 6000000 },
+        .n = { .min = 1, .max = 7 },
+        .m = { .min = 22, .max = 450 }, /* guess */
+        .m1 = { .min = 2, .max = 3 },
+        .m2 = { .min = 11, .max = 156 },
+        .p = { .min = 10, .max = 30 },
+        .p1 = { .min = 2, .max = 3 },
+        .p2 = { .dot_limit = 270000,
+                .p2_slow = 2, .p2_fast = 20 },
+        .find_pll = intel_vlv_find_best_pll,
+};
+
+static const intel_limit_t intel_limits_vlv_hdmi = {
+        .dot = { .min = 20000, .max = 165000 },
+        .vco = { .min = 5994000, .max = 4000000 },
+        .n = { .min = 1, .max = 7 },
+        .m = { .min = 60, .max = 300 }, /* guess */
+        .m1 = { .min = 2, .max = 3 },
+        .m2 = { .min = 11, .max = 156 },
+        .p = { .min = 10, .max = 30 },
+        .p1 = { .min = 2, .max = 3 },
+        .p2 = { .dot_limit = 270000,
+                .p2_slow = 2, .p2_fast = 20 },
+        .find_pll = intel_vlv_find_best_pll,
+};
+
+static const intel_limit_t intel_limits_vlv_dp = {
+        .dot = { .min = 162000, .max = 270000 },
+        .vco = { .min = 5994000, .max = 4000000 },
+        .n = { .min = 1, .max = 7 },
+        .m = { .min = 60, .max = 300 }, /* guess */
+        .m1 = { .min = 2, .max = 3 },
+        .m2 = { .min = 11, .max = 156 },
+        .p = { .min = 10, .max = 30 },
+        .p1 = { .min = 2, .max = 3 },
+        .p2 = { .dot_limit = 270000,
+                .p2_slow = 2, .p2_fast = 20 },
+        .find_pll = intel_vlv_find_best_pll,
+};
+
 u32 intel_dpio_read(struct drm_i915_private *dev_priv, int reg)
 {
         unsigned long flags;
@@ -384,6 +431,28 @@ out_unlock:
         return val;
 }
 
+static void intel_dpio_write(struct drm_i915_private *dev_priv, int reg,
+                             u32 val)
+{
+        unsigned long flags;
+
+        spin_lock_irqsave(&dev_priv->dpio_lock, flags);
+        if (wait_for_atomic_us((I915_READ(DPIO_PKT) & DPIO_BUSY) == 0, 100)) {
+                DRM_ERROR("DPIO idle wait timed out\n");
+                goto out_unlock;
+        }
+
+        I915_WRITE(DPIO_DATA, val);
+        I915_WRITE(DPIO_REG, reg);
+        I915_WRITE(DPIO_PKT, DPIO_RID | DPIO_OP_WRITE | DPIO_PORTID |
+                   DPIO_BYTE);
+        if (wait_for_atomic_us((I915_READ(DPIO_PKT) & DPIO_BUSY) == 0, 100))
+                DRM_ERROR("DPIO write wait timed out\n");
+
+out_unlock:
+       spin_unlock_irqrestore(&dev_priv->dpio_lock, flags);
+}
+
 static void vlv_init_dpio(struct drm_device *dev)
 {
         struct drm_i915_private *dev_priv = dev->dev_private;
@@ -434,7 +503,7 @@ static bool is_dual_link_lvds(struct drm_i915_private *dev_priv,
                  * register is uninitialized.
                  */
                 val = I915_READ(reg);
-                if (!(val & ~LVDS_DETECTED))
+                if (!(val & ~(LVDS_PIPE_MASK | LVDS_DETECTED)))
                         val = dev_priv->bios_lvds_val;
                 dev_priv->lvds_val = val;
         }
@@ -510,6 +579,13 @@ static const intel_limit_t *intel_limit(struct drm_crtc *crtc, int refclk)
                         limit = &intel_limits_pineview_lvds;
                 else
                         limit = &intel_limits_pineview_sdvo;
+        } else if (IS_VALLEYVIEW(dev)) {
+                if (intel_pipe_has_type(crtc, INTEL_OUTPUT_ANALOG))
+                        limit = &intel_limits_vlv_dac;
+                else if (intel_pipe_has_type(crtc, INTEL_OUTPUT_HDMI))
+                        limit = &intel_limits_vlv_hdmi;
+                else
+                        limit = &intel_limits_vlv_dp;
         } else if (!IS_GEN2(dev)) {
                 if (intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS))
                         limit = &intel_limits_i9xx_lvds;
@@ -551,11 +627,10 @@ static void intel_clock(struct drm_device *dev, int refclk, intel_clock_t *clock
 bool intel_pipe_has_type(struct drm_crtc *crtc, int type)
 {
         struct drm_device *dev = crtc->dev;
-        struct drm_mode_config *mode_config = &dev->mode_config;
         struct intel_encoder *encoder;
 
-        list_for_each_entry(encoder, &mode_config->encoder_list, base.head)
-                if (encoder->base.crtc == crtc && encoder->type == type)
+        for_each_encoder_on_crtc(dev, crtc, encoder)
+                if (encoder->type == type)
                         return true;
 
         return false;
@@ -783,6 +858,73 @@ intel_find_pll_g4x_dp(const intel_limit_t *limit, struct drm_crtc *crtc,
         memcpy(best_clock, &clock, sizeof(intel_clock_t));
         return true;
 }
+static bool
+intel_vlv_find_best_pll(const intel_limit_t *limit, struct drm_crtc *crtc,
+                        int target, int refclk, intel_clock_t *match_clock,
+                        intel_clock_t *best_clock)
+{
+        u32 p1, p2, m1, m2, vco, bestn, bestm1, bestm2, bestp1, bestp2;
+        u32 m, n, fastclk;
+        u32 updrate, minupdate, fracbits, p;
+        unsigned long bestppm, ppm, absppm;
+        int dotclk, flag;
+
+        dotclk = target * 1000;
+        bestppm = 1000000;
+        ppm = absppm = 0;
+        fastclk = dotclk / (2*100);
+        updrate = 0;
+        minupdate = 19200;
+        fracbits = 1;
+        n = p = p1 = p2 = m = m1 = m2 = vco = bestn = 0;
+        bestm1 = bestm2 = bestp1 = bestp2 = 0;
+
+        /* based on hardware requirement, prefer smaller n to precision */
+        for (n = limit->n.min; n <= ((refclk) / minupdate); n++) {
+                updrate = refclk / n;
+                for (p1 = limit->p1.max; p1 > limit->p1.min; p1--) {
+                        for (p2 = limit->p2.p2_fast+1; p2 > 0; p2--) {
+                                if (p2 > 10)
+                                        p2 = p2 - 1;
+                                p = p1 * p2;
+                                /* based on hardware requirement, prefer bigger m1,m2 values */
+                                for (m1 = limit->m1.min; m1 <= limit->m1.max; m1++) {
+                                        m2 = (((2*(fastclk * p * n / m1 )) +
+                                               refclk) / (2*refclk));
+                                        m = m1 * m2;
+                                        vco = updrate * m;
+                                        if (vco >= limit->vco.min && vco < limit->vco.max) {
+                                                ppm = 1000000 * ((vco / p) - fastclk) / fastclk;
+                                                absppm = (ppm > 0) ? ppm : (-ppm);
+                                                if (absppm < 100 && ((p1 * p2) > (bestp1 * bestp2))) {
+                                                        bestppm = 0;
+                                                        flag = 1;
+                                                }
+                                                if (absppm < bestppm - 10) {
+                                                        bestppm = absppm;
+                                                        flag = 1;
+                                                }
+                                                if (flag) {
+                                                        bestn = n;
+                                                        bestm1 = m1;
+                                                        bestm2 = m2;
+                                                        bestp1 = p1;
+                                                        bestp2 = p2;
+                                                        flag = 0;
+                                                }
+                                        }
+                                }
+                        }
+                }
+        }
+        best_clock->n = bestn;
+        best_clock->m1 = bestm1;
+        best_clock->m2 = bestm2;
+        best_clock->p1 = bestp1;
+        best_clock->p2 = bestp2;
+
+        return true;
+}
 
 static void ironlake_wait_for_vblank(struct drm_device *dev, int pipe)
 {
@@ -1232,6 +1374,9 @@ static void assert_pch_dp_disabled(struct drm_i915_private *dev_priv,
         WARN(dp_pipe_enabled(dev_priv, pipe, port_sel, val),
              "PCH DP (0x%08x) enabled on transcoder %c, should be disabled\n",
              reg, pipe_name(pipe));
+
+        WARN(HAS_PCH_IBX(dev_priv->dev) && (val & SDVO_PIPE_B_SELECT),
+             "IBX PCH dp port still using transcoder B\n");
 }
 
 static void assert_pch_hdmi_disabled(struct drm_i915_private *dev_priv,
@@ -1241,6 +1386,9 @@ static void assert_pch_hdmi_disabled(struct drm_i915_private *dev_priv,
         WARN(hdmi_pipe_enabled(dev_priv, val, pipe),
              "PCH HDMI (0x%08x) enabled on transcoder %c, should be disabled\n",
              reg, pipe_name(pipe));
+
+        WARN(HAS_PCH_IBX(dev_priv->dev) && (val & SDVO_PIPE_B_SELECT),
+             "IBX PCH hdmi port still using transcoder B\n");
 }
 
 static void assert_pch_ports_disabled(struct drm_i915_private *dev_priv,
@@ -1287,7 +1435,7 @@ static void intel_enable_pll(struct drm_i915_private *dev_priv, enum pipe pipe)
         u32 val;
 
         /* No really, not for ILK+ */
-        BUG_ON(dev_priv->info->gen >= 5);
+        BUG_ON(!IS_VALLEYVIEW(dev_priv->dev) && dev_priv->info->gen >= 5);
 
         /* PLL is protected by panel, make sure we can write it */
         if (IS_MOBILE(dev_priv->dev) && !IS_I830(dev_priv->dev))
@@ -1344,7 +1492,7 @@ intel_sbi_write(struct drm_i915_private *dev_priv, u16 reg, u32 value)
         unsigned long flags;
 
         spin_lock_irqsave(&dev_priv->dpio_lock, flags);
-        if (wait_for((I915_READ(SBI_CTL_STAT) & SBI_READY) == 0,
+        if (wait_for((I915_READ(SBI_CTL_STAT) & SBI_BUSY) == 0,
                                 100)) {
                 DRM_ERROR("timeout waiting for SBI to become ready\n");
                 goto out_unlock;
@@ -1358,7 +1506,7 @@ intel_sbi_write(struct drm_i915_private *dev_priv, u16 reg, u32 value)
                         SBI_BUSY |
                         SBI_CTL_OP_CRWR);
 
-        if (wait_for((I915_READ(SBI_CTL_STAT) & (SBI_READY | SBI_RESPONSE_SUCCESS)) == 0,
+        if (wait_for((I915_READ(SBI_CTL_STAT) & (SBI_BUSY | SBI_RESPONSE_FAIL)) == 0,
                                 100)) {
                 DRM_ERROR("timeout waiting for SBI to complete write transaction\n");
                 goto out_unlock;
@@ -1372,10 +1520,10 @@ static u32
 intel_sbi_read(struct drm_i915_private *dev_priv, u16 reg)
 {
         unsigned long flags;
-        u32 value;
+        u32 value = 0;
 
         spin_lock_irqsave(&dev_priv->dpio_lock, flags);
-        if (wait_for((I915_READ(SBI_CTL_STAT) & SBI_READY) == 0,
+        if (wait_for((I915_READ(SBI_CTL_STAT) & SBI_BUSY) == 0,
                                 100)) {
                 DRM_ERROR("timeout waiting for SBI to become ready\n");
                 goto out_unlock;
@@ -1387,7 +1535,7 @@ intel_sbi_read(struct drm_i915_private *dev_priv, u16 reg)
                         SBI_BUSY |
                         SBI_CTL_OP_CRRD);
 
-        if (wait_for((I915_READ(SBI_CTL_STAT) & (SBI_READY | SBI_RESPONSE_SUCCESS)) == 0,
+        if (wait_for((I915_READ(SBI_CTL_STAT) & (SBI_BUSY | SBI_RESPONSE_FAIL)) == 0,
                                 100)) {
                 DRM_ERROR("timeout waiting for SBI to complete read transaction\n");
                 goto out_unlock;
@@ -1824,6 +1972,22 @@ void intel_unpin_fb_obj(struct drm_i915_gem_object *obj)
         i915_gem_object_unpin(obj);
 }
 
+/* Computes the linear offset to the base tile and adjusts x, y. bytes per pixel
+ * is assumed to be a power-of-two. */
+static unsigned long gen4_compute_dspaddr_offset_xtiled(int *x, int *y,
+                                                        unsigned int bpp,
+                                                        unsigned int pitch)
+{
+        int tile_rows, tiles;
+
+        tile_rows = *y / 8;
+        *y %= 8;
+        tiles = *x / (512/bpp);
+        *x %= 512/bpp;
+
+        return tile_rows * pitch * 8 + tiles * 4096;
+}
+
 static int i9xx_update_plane(struct drm_crtc *crtc, struct drm_framebuffer *fb,
                              int x, int y)
 {
@@ -1833,7 +1997,7 @@ static int i9xx_update_plane(struct drm_crtc *crtc, struct drm_framebuffer *fb,
         struct intel_framebuffer *intel_fb;
         struct drm_i915_gem_object *obj;
         int plane = intel_crtc->plane;
-        unsigned long Start, Offset;
+        unsigned long linear_offset;
         u32 dspcntr;
         u32 reg;
 
@@ -1880,18 +2044,28 @@ static int i9xx_update_plane(struct drm_crtc *crtc, struct drm_framebuffer *fb,
 
         I915_WRITE(reg, dspcntr);
 
-        Start = obj->gtt_offset;
-        Offset = y * fb->pitches[0] + x * (fb->bits_per_pixel / 8);
+        linear_offset = y * fb->pitches[0] + x * (fb->bits_per_pixel / 8);
+
+        if (INTEL_INFO(dev)->gen >= 4) {
+                intel_crtc->dspaddr_offset =
+                        gen4_compute_dspaddr_offset_xtiled(&x, &y,
+                                                           fb->bits_per_pixel / 8,
+                                                           fb->pitches[0]);
+                linear_offset -= intel_crtc->dspaddr_offset;
+        } else {
+                intel_crtc->dspaddr_offset = linear_offset;
+        }
 
-        DRM_DEBUG_KMS("Writing base %08lX %08lX %d %d %d\n",
-                      Start, Offset, x, y, fb->pitches[0]);
+        DRM_DEBUG_KMS("Writing base %08X %08lX %d %d %d\n",
+                      obj->gtt_offset, linear_offset, x, y, fb->pitches[0]);
         I915_WRITE(DSPSTRIDE(plane), fb->pitches[0]);
         if (INTEL_INFO(dev)->gen >= 4) {
-                I915_MODIFY_DISPBASE(DSPSURF(plane), Start);
+                I915_MODIFY_DISPBASE(DSPSURF(plane),
+                                     obj->gtt_offset + intel_crtc->dspaddr_offset);
                 I915_WRITE(DSPTILEOFF(plane), (y << 16) | x);
-                I915_WRITE(DSPADDR(plane), Offset);
+                I915_WRITE(DSPLINOFF(plane), linear_offset);
         } else
-                I915_WRITE(DSPADDR(plane), Start + Offset);
+                I915_WRITE(DSPADDR(plane), obj->gtt_offset + linear_offset);
         POSTING_READ(reg);
 
         return 0;
@@ -1906,7 +2080,7 @@ static int ironlake_update_plane(struct drm_crtc *crtc,
         struct intel_framebuffer *intel_fb;
         struct drm_i915_gem_object *obj;
         int plane = intel_crtc->plane;
-        unsigned long Start, Offset;
+        unsigned long linear_offset;
         u32 dspcntr;
         u32 reg;
 
@@ -1961,15 +2135,20 @@ static int ironlake_update_plane(struct drm_crtc *crtc,
 
         I915_WRITE(reg, dspcntr);
 
-        Start = obj->gtt_offset;
-        Offset = y * fb->pitches[0] + x * (fb->bits_per_pixel / 8);
+        linear_offset = y * fb->pitches[0] + x * (fb->bits_per_pixel / 8);
+        intel_crtc->dspaddr_offset =
+                gen4_compute_dspaddr_offset_xtiled(&x, &y,
+                                                   fb->bits_per_pixel / 8,
+                                                   fb->pitches[0]);
+        linear_offset -= intel_crtc->dspaddr_offset;
 
-        DRM_DEBUG_KMS("Writing base %08lX %08lX %d %d %d\n",
-                      Start, Offset, x, y, fb->pitches[0]);
+        DRM_DEBUG_KMS("Writing base %08X %08lX %d %d %d\n",
+                      obj->gtt_offset, linear_offset, x, y, fb->pitches[0]);
         I915_WRITE(DSPSTRIDE(plane), fb->pitches[0]);
-        I915_MODIFY_DISPBASE(DSPSURF(plane), Start);
+        I915_MODIFY_DISPBASE(DSPSURF(plane),
+                             obj->gtt_offset + intel_crtc->dspaddr_offset);
         I915_WRITE(DSPTILEOFF(plane), (y << 16) | x);
-        I915_WRITE(DSPADDR(plane), Offset);
+        I915_WRITE(DSPLINOFF(plane), linear_offset);
         POSTING_READ(reg);
 
         return 0;
@@ -2656,16 +2835,13 @@ static void intel_crtc_wait_for_pending_flips(struct drm_crtc *crtc)
 static bool intel_crtc_driving_pch(struct drm_crtc *crtc)
 {
         struct drm_device *dev = crtc->dev;
-        struct drm_mode_config *mode_config = &dev->mode_config;
         struct intel_encoder *encoder;
 
         /*
          * If there's a non-PCH eDP on this crtc, it must be DP_A, and that
          * must be driven by its own crtc; no sharing is possible.
          */
-        list_for_each_entry(encoder, &mode_config->encoder_list, base.head) {
-                if (encoder->base.crtc != crtc)
-                        continue;
+        for_each_encoder_on_crtc(dev, crtc, encoder) {
 
                 /* On Haswell, LPT PCH handles the VGA connection via FDI, and Haswell
                  * CPU handles all others */
@@ -3554,16 +3730,12 @@ static bool intel_choose_pipe_bpp_dither(struct drm_crtc *crtc,
 {
         struct drm_device *dev = crtc->dev;
         struct drm_i915_private *dev_priv = dev->dev_private;
-        struct drm_encoder *encoder;
         struct drm_connector *connector;
+        struct intel_encoder *intel_encoder;
         unsigned int display_bpc = UINT_MAX, bpc;
 
         /* Walk the encoders & connectors on this crtc, get min bpc */
-        list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) {
-                struct intel_encoder *intel_encoder = to_intel_encoder(encoder);
-
-                if (encoder->crtc != crtc)
-                        continue;
+        for_each_encoder_on_crtc(dev, crtc, intel_encoder) {
 
                 if (intel_encoder->type == INTEL_OUTPUT_LVDS) {
                         unsigned int lvds_bpc;
@@ -3595,7 +3767,7 @@ static bool intel_choose_pipe_bpp_dither(struct drm_crtc *crtc,
                 /* Not one of the known troublemakers, check the EDID */
                 list_for_each_entry(connector, &dev->mode_config.connector_list,
                                     head) {
-                        if (connector->encoder != encoder)
+                        if (connector->encoder != &intel_encoder->base)
                                 continue;
 
                         /* Don't use an invalid EDID bpc value */
@@ -3666,13 +3838,37 @@ static bool intel_choose_pipe_bpp_dither(struct drm_crtc *crtc,
         return display_bpc != bpc;
 }
 
+static int vlv_get_refclk(struct drm_crtc *crtc)
+{
+        struct drm_device *dev = crtc->dev;
+        struct drm_i915_private *dev_priv = dev->dev_private;
+        int refclk = 27000; /* for DP & HDMI */
+
+        return 100000; /* only one validated so far */
+
+        if (intel_pipe_has_type(crtc, INTEL_OUTPUT_ANALOG)) {
+                refclk = 96000;
+        } else if (intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS)) {
+                if (intel_panel_use_ssc(dev_priv))
+                        refclk = 100000;
+                else
+                        refclk = 96000;
+        } else if (intel_pipe_has_type(crtc, INTEL_OUTPUT_EDP)) {
+                refclk = 100000;
+        }
+
+        return refclk;
+}
+
 static int i9xx_get_refclk(struct drm_crtc *crtc, int num_connectors)
 {
         struct drm_device *dev = crtc->dev;
         struct drm_i915_private *dev_priv = dev->dev_private;
         int refclk;
 
-        if (intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS) &&
+        if (IS_VALLEYVIEW(dev)) {
+                refclk = vlv_get_refclk(crtc);
+        } else if (intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS) &&
             intel_panel_use_ssc(dev_priv) && num_connectors < 2) {
                 refclk = dev_priv->lvds_ssc_freq * 1000;
                 DRM_DEBUG_KMS("using SSC reference clock of %d MHz\n",
@@ -3787,6 +3983,72 @@ static void intel_update_lvds(struct drm_crtc *crtc, intel_clock_t *clock,
         I915_WRITE(LVDS, temp);
 }
 
+static void vlv_update_pll(struct drm_crtc *crtc,
+                           struct drm_display_mode *mode,
+                           struct drm_display_mode *adjusted_mode,
+                           intel_clock_t *clock, intel_clock_t *reduced_clock,
+                           int refclk, int num_connectors)
+{
+        struct drm_device *dev = crtc->dev;
+        struct drm_i915_private *dev_priv = dev->dev_private;
+        struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+        int pipe = intel_crtc->pipe;
+        u32 dpll, mdiv, pdiv;
+        u32 bestn, bestm1, bestm2, bestp1, bestp2;
+        bool is_hdmi;
+
+        is_hdmi = intel_pipe_has_type(crtc, INTEL_OUTPUT_HDMI);
+
+        bestn = clock->n;
+        bestm1 = clock->m1;
+        bestm2 = clock->m2;
+        bestp1 = clock->p1;
+        bestp2 = clock->p2;
+
+        /* Enable DPIO clock input */
+        dpll = DPLL_EXT_BUFFER_ENABLE_VLV | DPLL_REFA_CLK_ENABLE_VLV |
+                DPLL_VGA_MODE_DIS | DPLL_INTEGRATED_CLOCK_VLV;
+        I915_WRITE(DPLL(pipe), dpll);
+        POSTING_READ(DPLL(pipe));
+
+        mdiv = ((bestm1 << DPIO_M1DIV_SHIFT) | (bestm2 & DPIO_M2DIV_MASK));
+        mdiv |= ((bestp1 << DPIO_P1_SHIFT) | (bestp2 << DPIO_P2_SHIFT));
+        mdiv |= ((bestn << DPIO_N_SHIFT));
+        mdiv |= (1 << DPIO_POST_DIV_SHIFT);
+        mdiv |= (1 << DPIO_K_SHIFT);
+        mdiv |= DPIO_ENABLE_CALIBRATION;
+        intel_dpio_write(dev_priv, DPIO_DIV(pipe), mdiv);
+
+        intel_dpio_write(dev_priv, DPIO_CORE_CLK(pipe), 0x01000000);
+
+        pdiv = DPIO_REFSEL_OVERRIDE | (5 << DPIO_PLL_MODESEL_SHIFT) |
+                (3 << DPIO_BIAS_CURRENT_CTL_SHIFT) | (1<<20) |
+                (8 << DPIO_DRIVER_CTL_SHIFT) | (5 << DPIO_CLK_BIAS_CTL_SHIFT);
+        intel_dpio_write(dev_priv, DPIO_REFSFR(pipe), pdiv);
+
+        intel_dpio_write(dev_priv, DPIO_LFP_COEFF(pipe), 0x009f0051);
+
+        dpll |= DPLL_VCO_ENABLE;
+        I915_WRITE(DPLL(pipe), dpll);
+        POSTING_READ(DPLL(pipe));
+        if (wait_for(((I915_READ(DPLL(pipe)) & DPLL_LOCK_VLV) == DPLL_LOCK_VLV), 1))
+                DRM_ERROR("DPLL %d failed to lock\n", pipe);
+
+        if (is_hdmi) {
+                u32 temp = intel_mode_get_pixel_multiplier(adjusted_mode);
+
+                if (temp > 1)
+                        temp = (temp - 1) << DPLL_MD_UDI_MULTIPLIER_SHIFT;
+                else
+                        temp = 0;
+
+                I915_WRITE(DPLL_MD(pipe), temp);
+                POSTING_READ(DPLL_MD(pipe));
+        }
+
+        intel_dpio_write(dev_priv, DPIO_FASTCLK_DISABLE, 0x641); /* ??? */
+}
+
 static void i9xx_update_pll(struct drm_crtc *crtc,
                             struct drm_display_mode *mode,
                             struct drm_display_mode *adjusted_mode,
@@ -3974,15 +4236,11 @@ static int i9xx_crtc_mode_set(struct drm_crtc *crtc,
         u32 dspcntr, pipeconf, vsyncshift;
         bool ok, has_reduced_clock = false, is_sdvo = false;
         bool is_lvds = false, is_tv = false, is_dp = false;
-        struct drm_mode_config *mode_config = &dev->mode_config;
         struct intel_encoder *encoder;
         const intel_limit_t *limit;
         int ret;
 
-        list_for_each_entry(encoder, &mode_config->encoder_list, base.head) {
-                if (encoder->base.crtc != crtc)
-                        continue;
-
+        for_each_encoder_on_crtc(dev, crtc, encoder) {
                 switch (encoder->type) {
                 case INTEL_OUTPUT_LVDS:
                         is_lvds = true;
@@ -4044,6 +4302,9 @@ static int i9xx_crtc_mode_set(struct drm_crtc *crtc,
 
         if (IS_GEN2(dev))
                 i8xx_update_pll(crtc, adjusted_mode, &clock, num_connectors);
+        else if (IS_VALLEYVIEW(dev))
+                vlv_update_pll(crtc, mode,adjusted_mode, &clock, NULL,
+                               refclk, num_connectors);
         else
                 i9xx_update_pll(crtc, mode, adjusted_mode, &clock,
                                 has_reduced_clock ? &reduced_clock : NULL,
@@ -4282,15 +4543,11 @@ static int ironlake_get_refclk(struct drm_crtc *crtc)
         struct drm_device *dev = crtc->dev;
         struct drm_i915_private *dev_priv = dev->dev_private;
         struct intel_encoder *encoder;
-        struct drm_mode_config *mode_config = &dev->mode_config;
         struct intel_encoder *edp_encoder = NULL;
         int num_connectors = 0;
         bool is_lvds = false;
 
-        list_for_each_entry(encoder, &mode_config->encoder_list, base.head) {
-                if (encoder->base.crtc != crtc)
-                        continue;
-
+        for_each_encoder_on_crtc(dev, crtc, encoder) {
                 switch (encoder->type) {
                 case INTEL_OUTPUT_LVDS:
                         is_lvds = true;
@@ -4327,7 +4584,6 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc,
         u32 dpll, fp = 0, fp2 = 0, dspcntr, pipeconf;
         bool ok, has_reduced_clock = false, is_sdvo = false;
         bool is_crt = false, is_lvds = false, is_tv = false, is_dp = false;
-        struct drm_mode_config *mode_config = &dev->mode_config;
         struct intel_encoder *encoder, *edp_encoder = NULL;
         const intel_limit_t *limit;
         int ret;
@@ -4338,10 +4594,7 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc,
         bool dither;
         bool is_cpu_edp = false, is_pch_edp = false;
 
-        list_for_each_entry(encoder, &mode_config->encoder_list, base.head) {
-                if (encoder->base.crtc != crtc)
-                        continue;
-
+        for_each_encoder_on_crtc(dev, crtc, encoder) {
                 switch (encoder->type) {
                 case INTEL_OUTPUT_LVDS:
                         is_lvds = true;
@@ -4405,25 +4658,10 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc,
                                                     &clock,
                                                     &reduced_clock);
         }
-        /* SDVO TV has fixed PLL values depend on its clock range,
-           this mirrors vbios setting. */
-        if (is_sdvo && is_tv) {
-                if (adjusted_mode->clock >= 100000
-                    && adjusted_mode->clock < 140500) {
-                        clock.p1 = 2;
-                        clock.p2 = 10;
-                        clock.n = 3;
-                        clock.m1 = 16;
-                        clock.m2 = 8;
-                } else if (adjusted_mode->clock >= 140500
-                           && adjusted_mode->clock <= 200000) {
-                        clock.p1 = 1;
-                        clock.p2 = 10;
-                        clock.n = 6;
-                        clock.m1 = 12;
-                        clock.m2 = 8;
-                }
-        }
+
+        if (is_sdvo && is_tv)
+                i9xx_adjust_sdvo_tv_clock(adjusted_mode, &clock);
+
 
         /* FDI link */
         pixel_multiplier = intel_mode_get_pixel_multiplier(adjusted_mode);
@@ -4431,16 +4669,8 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc,
         /* CPU eDP doesn't require FDI link, so just set DP M/N
            according to current link config */
         if (is_cpu_edp) {
-                target_clock = mode->clock;
                 intel_edp_link_config(edp_encoder, &lane, &link_bw);
         } else {
-                /* [e]DP over FDI requires target mode clock
-                   instead of link clock */
-                if (is_dp)
-                        target_clock = mode->clock;
-                else
-                        target_clock = adjusted_mode->clock;
-
                 /* FDI is a binary signal running at ~2.7GHz, encoding
                  * each output octet as 10 bits. The actual frequency
                  * is stored as a divider into a 100MHz clock, and the
@@ -4451,6 +4681,14 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc,
                 link_bw = intel_fdi_link_freq(dev) * MHz(100)/KHz(1)/10;
         }
 
+        /* [e]DP over FDI requires target mode clock instead of link clock. */
+        if (edp_encoder)
+                target_clock = intel_edp_target_clock(edp_encoder, mode);
+        else if (is_dp)
+                target_clock = mode->clock;
+        else
+                target_clock = adjusted_mode->clock;
+
         /* determine panel color depth */
         temp = I915_READ(PIPECONF(pipe));
         temp &= ~PIPE_BPC_MASK;
@@ -4662,16 +4900,8 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc,
                 if (is_lvds && has_reduced_clock && i915_powersave) {
                         I915_WRITE(intel_crtc->pch_pll->fp1_reg, fp2);
                         intel_crtc->lowfreq_avail = true;
-                        if (HAS_PIPE_CXSR(dev)) {
-                                DRM_DEBUG_KMS("enabling CxSR downclocking\n");
-                                pipeconf |= PIPECONF_CXSR_DOWNCLOCK;
-                        }
                 } else {
                         I915_WRITE(intel_crtc->pch_pll->fp1_reg, fp);
-                        if (HAS_PIPE_CXSR(dev)) {
-                                DRM_DEBUG_KMS("disabling CxSR downclocking\n");
-                                pipeconf &= ~PIPECONF_CXSR_DOWNCLOCK;
-                        }
                 }
         }
 
@@ -5975,7 +6205,6 @@ static int intel_gen2_queue_flip(struct drm_device *dev,
 {
         struct drm_i915_private *dev_priv = dev->dev_private;
         struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-        unsigned long offset;
         u32 flip_mask;
         struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
         int ret;
@@ -5984,9 +6213,6 @@ static int intel_gen2_queue_flip(struct drm_device *dev,
         if (ret)
                 goto err;
 
-        /* Offset into the new buffer for cases of shared fbs between CRTCs */
-        offset = crtc->y * fb->pitches[0] + crtc->x * fb->bits_per_pixel/8;
-
         ret = intel_ring_begin(ring, 6);
         if (ret)
                 goto err_unpin;
@@ -6003,7 +6229,7 @@ static int intel_gen2_queue_flip(struct drm_device *dev,
         intel_ring_emit(ring, MI_DISPLAY_FLIP |
                         MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
         intel_ring_emit(ring, fb->pitches[0]);
-        intel_ring_emit(ring, obj->gtt_offset + offset);
+        intel_ring_emit(ring, obj->gtt_offset + intel_crtc->dspaddr_offset);
         intel_ring_emit(ring, 0); /* aux display base address, unused */
         intel_ring_advance(ring);
         return 0;
@@ -6021,7 +6247,6 @@ static int intel_gen3_queue_flip(struct drm_device *dev,
 {
         struct drm_i915_private *dev_priv = dev->dev_private;
         struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-        unsigned long offset;
         u32 flip_mask;
         struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
         int ret;
@@ -6030,9 +6255,6 @@ static int intel_gen3_queue_flip(struct drm_device *dev,
         if (ret)
                 goto err;
 
-        /* Offset into the new buffer for cases of shared fbs between CRTCs */
-        offset = crtc->y * fb->pitches[0] + crtc->x * fb->bits_per_pixel/8;
-
         ret = intel_ring_begin(ring, 6);
         if (ret)
                 goto err_unpin;
@@ -6046,7 +6268,7 @@ static int intel_gen3_queue_flip(struct drm_device *dev,
         intel_ring_emit(ring, MI_DISPLAY_FLIP_I915 |
                         MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
         intel_ring_emit(ring, fb->pitches[0]);
-        intel_ring_emit(ring, obj->gtt_offset + offset);
+        intel_ring_emit(ring, obj->gtt_offset + intel_crtc->dspaddr_offset);
         intel_ring_emit(ring, MI_NOOP);
 
         intel_ring_advance(ring);
@@ -6084,7 +6306,9 @@ static int intel_gen4_queue_flip(struct drm_device *dev,
         intel_ring_emit(ring, MI_DISPLAY_FLIP |
                         MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
         intel_ring_emit(ring, fb->pitches[0]);
-        intel_ring_emit(ring, obj->gtt_offset | obj->tiling_mode);
+        intel_ring_emit(ring,
+                        (obj->gtt_offset + intel_crtc->dspaddr_offset) |
+                        obj->tiling_mode);
 
         /* XXX Enabling the panel-fitter across page-flip is so far
          * untested on non-native modes, so ignore it for now.
@@ -6124,7 +6348,7 @@ static int intel_gen6_queue_flip(struct drm_device *dev,
         intel_ring_emit(ring, MI_DISPLAY_FLIP |
                         MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
         intel_ring_emit(ring, fb->pitches[0] | obj->tiling_mode);
-        intel_ring_emit(ring, obj->gtt_offset);
+        intel_ring_emit(ring, obj->gtt_offset + intel_crtc->dspaddr_offset);
 
         /* Contrary to the suggestions in the documentation,
          * "Enable Panel Fitter" does not seem to be required when page
@@ -6187,7 +6411,7 @@ static int intel_gen7_queue_flip(struct drm_device *dev,
 
         intel_ring_emit(ring, MI_DISPLAY_FLIP_I915 | plane_bit);
         intel_ring_emit(ring, (fb->pitches[0] | obj->tiling_mode));
-        intel_ring_emit(ring, (obj->gtt_offset));
+        intel_ring_emit(ring, obj->gtt_offset + intel_crtc->dspaddr_offset);
         intel_ring_emit(ring, (MI_NOOP));
         intel_ring_advance(ring);
         return 0;
@@ -6219,6 +6443,19 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
         unsigned long flags;
         int ret;
 
+        /* Can't change pixel format via MI display flips. */
+        if (fb->pixel_format != crtc->fb->pixel_format)
+                return -EINVAL;
+
+        /*
+         * TILEOFF/LINOFF registers can't be changed via MI display flips.
+         * Note that pitch changes could also affect these register.
+         */
+        if (INTEL_INFO(dev)->gen > 3 &&
+            (fb->offsets[0] != crtc->fb->offsets[0] ||
+             fb->pitches[0] != crtc->fb->pitches[0]))
+                return -EINVAL;
+
         work = kzalloc(sizeof *work, GFP_KERNEL);
         if (work == NULL)
                 return -ENOMEM;
@@ -6566,7 +6803,24 @@ static void intel_setup_outputs(struct drm_device *dev)
 
                 if (!dpd_is_edp && (I915_READ(PCH_DP_D) & DP_DETECTED))
                         intel_dp_init(dev, PCH_DP_D);
+        } else if (IS_VALLEYVIEW(dev)) {
+                int found;
+
+                if (I915_READ(SDVOB) & PORT_DETECTED) {
+                        /* SDVOB multiplex with HDMIB */
+                        found = intel_sdvo_init(dev, SDVOB, true);
+                        if (!found)
+                                intel_hdmi_init(dev, SDVOB);
+                        if (!found && (I915_READ(DP_B) & DP_DETECTED))
+                                intel_dp_init(dev, DP_B);
+                }
+
+                if (I915_READ(SDVOC) & PORT_DETECTED)
+                        intel_hdmi_init(dev, SDVOC);
 
+                /* Shares lanes with HDMI on SDVOC */
+                if (I915_READ(DP_C) & DP_DETECTED)
+                        intel_dp_init(dev, DP_C);
         } else if (SUPPORTS_DIGITAL_OUTPUTS(dev)) {
                 bool found = false;
 
@@ -6623,7 +6877,7 @@ static void intel_setup_outputs(struct drm_device *dev)
         /* disable all the possible outputs/crtcs before entering KMS mode */
         drm_helper_disable_unused_functions(dev);
 
-        if (HAS_PCH_SPLIT(dev))
+        if (HAS_PCH_IBX(dev) || HAS_PCH_CPT(dev))
                 ironlake_init_pch_refclk(dev);
 }
 
@@ -6777,9 +7031,6 @@ static void intel_init_display(struct drm_device *dev)
                         dev_priv->display.write_eld = ironlake_write_eld;
                 } else
                         dev_priv->display.update_wm = NULL;
-        } else if (IS_VALLEYVIEW(dev)) {
-                dev_priv->display.force_wake_get = vlv_force_wake_get;
-                dev_priv->display.force_wake_put = vlv_force_wake_put;
         } else if (IS_G4X(dev)) {
                 dev_priv->display.write_eld = g4x_write_eld;
         }
@@ -6923,20 +7174,13 @@ static void i915_disable_vga(struct drm_device *dev)
 
 void intel_modeset_init_hw(struct drm_device *dev)
 {
-        struct drm_i915_private *dev_priv = dev->dev_private;
+        intel_prepare_ddi(dev);
 
         intel_init_clock_gating(dev);
 
-        if (IS_IRONLAKE_M(dev)) {
-                ironlake_enable_drps(dev);
-                ironlake_enable_rc6(dev);
-                intel_init_emon(dev);
-        }
-
-        if ((IS_GEN6(dev) || IS_GEN7(dev)) && !IS_VALLEYVIEW(dev)) {
-                gen6_enable_rps(dev_priv);
-                gen6_update_ring_freq(dev_priv);
-        }
+        mutex_lock(&dev->struct_mutex);
+        intel_enable_gt_powersave(dev);
+        mutex_unlock(&dev->struct_mutex);
 }
 
 void intel_modeset_init(struct drm_device *dev)
@@ -6958,8 +7202,6 @@ void intel_modeset_init(struct drm_device *dev)
 
         intel_init_pm(dev);
 
-        intel_prepare_ddi(dev);
-
         intel_init_display(dev);
 
         if (IS_GEN2(dev)) {
@@ -6972,7 +7214,7 @@ void intel_modeset_init(struct drm_device *dev)
                 dev->mode_config.max_width = 8192;
                 dev->mode_config.max_height = 8192;
         }
-        dev->mode_config.fb_base = dev->agp->base;
+        dev->mode_config.fb_base = dev_priv->mm.gtt_base_addr;
 
         DRM_DEBUG_KMS("%d display pipe%s available.\n",
                       dev_priv->num_pipe, dev_priv->num_pipe > 1 ? "s" : "");
@@ -7025,13 +7267,9 @@ void intel_modeset_cleanup(struct drm_device *dev)
 
         intel_disable_fbc(dev);
 
-        if (IS_IRONLAKE_M(dev))
-                ironlake_disable_drps(dev);
-        if ((IS_GEN6(dev) || IS_GEN7(dev)) && !IS_VALLEYVIEW(dev))
-                gen6_disable_rps(dev);
+        intel_disable_gt_powersave(dev);
 
-        if (IS_IRONLAKE_M(dev))
-                ironlake_disable_rc6(dev);
+        ironlake_teardown_rc6(dev);
 
         if (IS_VALLEYVIEW(dev))
                 vlv_init_dpio(dev);
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index c0449324143c..2bc1505132c3 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -155,6 +155,18 @@ intel_edp_link_config(struct intel_encoder *intel_encoder,
                 *link_bw = 270000;
 }
 
+int
+intel_edp_target_clock(struct intel_encoder *intel_encoder,
+                       struct drm_display_mode *mode)
+{
+        struct intel_dp *intel_dp = container_of(intel_encoder, struct intel_dp, base);
+
+        if (intel_dp->panel_fixed_mode)
+                return intel_dp->panel_fixed_mode->clock;
+        else
+                return mode->clock;
+}
+
 static int
 intel_dp_max_lane_count(struct intel_dp *intel_dp)
 {
@@ -225,7 +237,7 @@ intel_dp_max_data_rate(int max_link_clock, int max_lanes)
 static bool
 intel_dp_adjust_dithering(struct intel_dp *intel_dp,
                           struct drm_display_mode *mode,
-                          struct drm_display_mode *adjusted_mode)
+                          bool adjust_mode)
 {
         int max_link_clock = intel_dp_link_clock(intel_dp_max_link_bw(intel_dp));
         int max_lanes = intel_dp_max_lane_count(intel_dp);
@@ -239,8 +251,8 @@ intel_dp_adjust_dithering(struct intel_dp *intel_dp,
                 if (mode_rate > max_rate)
                         return false;
 
-                if (adjusted_mode)
-                        adjusted_mode->private_flags
+                if (adjust_mode)
+                        mode->private_flags
                                 |= INTEL_MODE_DP_FORCE_6BPC;
 
                 return true;
@@ -263,7 +275,7 @@ intel_dp_mode_valid(struct drm_connector *connector,
                         return MODE_PANEL;
         }
 
-        if (!intel_dp_adjust_dithering(intel_dp, mode, NULL))
+        if (!intel_dp_adjust_dithering(intel_dp, mode, false))
                 return MODE_CLOCK_HIGH;
 
         if (mode->clock < 10000)
@@ -691,7 +703,8 @@ intel_dp_i2c_init(struct intel_dp *intel_dp,
 }
 
 static bool
-intel_dp_mode_fixup(struct drm_encoder *encoder, struct drm_display_mode *mode,
+intel_dp_mode_fixup(struct drm_encoder *encoder,
+                    const struct drm_display_mode *mode,
                     struct drm_display_mode *adjusted_mode)
 {
         struct drm_device *dev = encoder->dev;
@@ -706,28 +719,23 @@ intel_dp_mode_fixup(struct drm_encoder *encoder, struct drm_display_mode *mode,
                 intel_fixed_panel_mode(intel_dp->panel_fixed_mode, adjusted_mode);
                 intel_pch_panel_fitting(dev, DRM_MODE_SCALE_FULLSCREEN,
                                         mode, adjusted_mode);
-                /*
-                 * the mode->clock is used to calculate the Data&Link M/N
-                 * of the pipe. For the eDP the fixed clock should be used.
-                 */
-                mode->clock = intel_dp->panel_fixed_mode->clock;
         }
 
-        if (mode->flags & DRM_MODE_FLAG_DBLCLK)
+        if (adjusted_mode->flags & DRM_MODE_FLAG_DBLCLK)
                 return false;
 
         DRM_DEBUG_KMS("DP link computation with max lane count %i "
                       "max bw %02x pixel clock %iKHz\n",
-                      max_lane_count, bws[max_clock], mode->clock);
+                      max_lane_count, bws[max_clock], adjusted_mode->clock);
 
-        if (!intel_dp_adjust_dithering(intel_dp, mode, adjusted_mode))
+        if (!intel_dp_adjust_dithering(intel_dp, adjusted_mode, true))
                 return false;
 
         bpp = adjusted_mode->private_flags & INTEL_MODE_DP_FORCE_6BPC ? 18 : 24;
-        mode_rate = intel_dp_link_required(mode->clock, bpp);
+        mode_rate = intel_dp_link_required(adjusted_mode->clock, bpp);
 
-        for (lane_count = 1; lane_count <= max_lane_count; lane_count <<= 1) {
-                for (clock = 0; clock <= max_clock; clock++) {
+        for (clock = 0; clock <= max_clock; clock++) {
+                for (lane_count = 1; lane_count <= max_lane_count; lane_count <<= 1) {
                         int link_avail = intel_dp_max_data_rate(intel_dp_link_clock(bws[clock]), lane_count);
 
                         if (mode_rate <= link_avail) {
@@ -786,8 +794,7 @@ intel_dp_set_m_n(struct drm_crtc *crtc, struct drm_display_mode *mode,
                  struct drm_display_mode *adjusted_mode)
 {
         struct drm_device *dev = crtc->dev;
-        struct drm_mode_config *mode_config = &dev->mode_config;
-        struct drm_encoder *encoder;
+        struct intel_encoder *encoder;
         struct drm_i915_private *dev_priv = dev->dev_private;
         struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
         int lane_count = 4;
@@ -797,13 +804,9 @@ intel_dp_set_m_n(struct drm_crtc *crtc, struct drm_display_mode *mode,
         /*
          * Find the lane count in the intel_encoder private
          */
-        list_for_each_entry(encoder, &mode_config->encoder_list, head) {
-                struct intel_dp *intel_dp;
-
-                if (encoder->crtc != crtc)
-                        continue;
+        for_each_encoder_on_crtc(dev, crtc, encoder) {
+                struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base);
 
-                intel_dp = enc_to_intel_dp(encoder);
                 if (intel_dp->base.type == INTEL_OUTPUT_DISPLAYPORT ||
                     intel_dp->base.type == INTEL_OUTPUT_EDP)
                 {
@@ -1922,7 +1925,7 @@ intel_dp_link_down(struct intel_dp *intel_dp)
                         DP |= DP_LINK_TRAIN_OFF;
         }
 
-        if (!HAS_PCH_CPT(dev) &&
+        if (HAS_PCH_IBX(dev) &&
             I915_READ(intel_dp->output_reg) & DP_PIPEB_SELECT) {
                 struct drm_crtc *crtc = intel_dp->base.base.crtc;
 
@@ -2099,25 +2102,23 @@ g4x_dp_detect(struct intel_dp *intel_dp)
 {
         struct drm_device *dev = intel_dp->base.base.dev;
         struct drm_i915_private *dev_priv = dev->dev_private;
-        uint32_t temp, bit;
+        uint32_t bit;
 
         switch (intel_dp->output_reg) {
         case DP_B:
-                bit = DPB_HOTPLUG_INT_STATUS;
+                bit = DPB_HOTPLUG_LIVE_STATUS;
                 break;
         case DP_C:
-                bit = DPC_HOTPLUG_INT_STATUS;
+                bit = DPC_HOTPLUG_LIVE_STATUS;
                 break;
         case DP_D:
-                bit = DPD_HOTPLUG_INT_STATUS;
+                bit = DPD_HOTPLUG_LIVE_STATUS;
                 break;
         default:
                 return connector_status_unknown;
         }
 
-        temp = I915_READ(PORT_HOTPLUG_STAT);
-
-        if ((temp & bit) == 0)
+        if ((I915_READ(PORT_HOTPLUG_STAT) & bit) == 0)
                 return connector_status_disconnected;
 
         return intel_dp_detect_dpcd(intel_dp);
@@ -2399,16 +2400,11 @@ int
 intel_trans_dp_port_sel(struct drm_crtc *crtc)
 {
         struct drm_device *dev = crtc->dev;
-        struct drm_mode_config *mode_config = &dev->mode_config;
-        struct drm_encoder *encoder;
-
-        list_for_each_entry(encoder, &mode_config->encoder_list, head) {
-                struct intel_dp *intel_dp;
+        struct intel_encoder *encoder;
 
-                if (encoder->crtc != crtc)
-                        continue;
+        for_each_encoder_on_crtc(dev, crtc, encoder) {
+                struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base);
 
-                intel_dp = enc_to_intel_dp(encoder);
                 if (intel_dp->base.type == INTEL_OUTPUT_DISPLAYPORT ||
                     intel_dp->base.type == INTEL_OUTPUT_EDP)
                         return intel_dp->output_reg;
@@ -2520,19 +2516,19 @@ intel_dp_init(struct drm_device *dev, int output_reg)
                 case DP_B:
                 case PCH_DP_B:
                         dev_priv->hotplug_supported_mask |=
-                                HDMIB_HOTPLUG_INT_STATUS;
+                                DPB_HOTPLUG_INT_STATUS;
                         name = "DPDDC-B";
                         break;
                 case DP_C:
                 case PCH_DP_C:
                         dev_priv->hotplug_supported_mask |=
-                                HDMIC_HOTPLUG_INT_STATUS;
+                                DPC_HOTPLUG_INT_STATUS;
                         name = "DPDDC-C";
                         break;
                 case DP_D:
                 case PCH_DP_D:
                         dev_priv->hotplug_supported_mask |=
-                                HDMID_HOTPLUG_INT_STATUS;
+                                DPD_HOTPLUG_INT_STATUS;
                         name = "DPDDC-D";
                         break;
         }
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 3e0918834e7e..b7859e7110a7 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -169,6 +169,7 @@ struct intel_crtc {
         u8 lut_r[256], lut_g[256], lut_b[256];
         int dpms_mode;
         bool active; /* is the crtc on? independent of the dpms mode */
+        bool primary_disabled; /* is the crtc obscured by a plane? */
         bool busy; /* is scanout buffer being updated frequently? */
         struct timer_list idle_timer;
         bool lowfreq_avail;
@@ -176,6 +177,11 @@ struct intel_crtc {
         struct intel_unpin_work *unpin_work;
         int fdi_lanes;
 
+        /* Display surface base address adjustement for pageflips. Note that on
+         * gen4+ this only adjusts up to a tile, offsets within a tile are
+         * handled in the hw itself (with the TILEOFF register). */
+        unsigned long dspaddr_offset;
+
         struct drm_i915_gem_object *cursor_bo;
         uint32_t cursor_addr;
         int16_t cursor_x, cursor_y;
@@ -191,7 +197,6 @@ struct intel_plane {
         struct drm_plane base;
         enum pipe pipe;
         struct drm_i915_gem_object *obj;
-        bool primary_disabled;
         int max_downscale;
         u32 lut_r[1024], lut_g[1024], lut_b[1024];
         void (*update_plane)(struct drm_plane *plane,
@@ -301,6 +306,8 @@ struct intel_hdmi {
         enum hdmi_force_audio force_audio;
         void (*write_infoframe)(struct drm_encoder *encoder,
                                 struct dip_infoframe *frame);
+        void (*set_infoframes)(struct drm_encoder *encoder,
+                               struct drm_display_mode *adjusted_mode);
 };
 
 static inline struct drm_crtc *
@@ -343,9 +350,6 @@ extern void intel_attach_broadcast_rgb_property(struct drm_connector *connector)
 extern void intel_crt_init(struct drm_device *dev);
 extern void intel_hdmi_init(struct drm_device *dev, int sdvox_reg);
 extern struct intel_hdmi *enc_to_intel_hdmi(struct drm_encoder *encoder);
-extern void intel_hdmi_set_avi_infoframe(struct drm_encoder *encoder,
-                            struct drm_display_mode *adjusted_mode);
-extern void intel_hdmi_set_spd_infoframe(struct drm_encoder *encoder);
 extern void intel_dip_infoframe_csum(struct dip_infoframe *avi_if);
 extern bool intel_sdvo_init(struct drm_device *dev, uint32_t sdvo_reg,
                             bool is_sdvob);
@@ -360,6 +364,8 @@ intel_dp_set_m_n(struct drm_crtc *crtc, struct drm_display_mode *mode,
                  struct drm_display_mode *adjusted_mode);
 extern bool intel_dpd_is_edp(struct drm_device *dev);
 extern void intel_edp_link_config(struct intel_encoder *, int *, int *);
+extern int intel_edp_target_clock(struct intel_encoder *,
+                                  struct drm_display_mode *mode);
 extern bool intel_encoder_is_pch_edp(struct drm_encoder *encoder);
 extern int intel_plane_init(struct drm_device *dev, enum pipe pipe);
 extern void intel_flush_display_plane(struct drm_i915_private *dev_priv,
@@ -372,13 +378,14 @@ extern void intel_fixed_panel_mode(struct drm_display_mode *fixed_mode,
                                    struct drm_display_mode *adjusted_mode);
 extern void intel_pch_panel_fitting(struct drm_device *dev,
                                     int fitting_mode,
-                                    struct drm_display_mode *mode,
+                                    const struct drm_display_mode *mode,
                                     struct drm_display_mode *adjusted_mode);
 extern u32 intel_panel_get_max_backlight(struct drm_device *dev);
 extern u32 intel_panel_get_backlight(struct drm_device *dev);
 extern void intel_panel_set_backlight(struct drm_device *dev, u32 level);
 extern int intel_panel_setup_backlight(struct drm_device *dev);
-extern void intel_panel_enable_backlight(struct drm_device *dev);
+extern void intel_panel_enable_backlight(struct drm_device *dev,
+                                         enum pipe pipe);
 extern void intel_panel_disable_backlight(struct drm_device *dev);
 extern void intel_panel_destroy_backlight(struct drm_device *dev);
 extern enum drm_connector_status intel_panel_detect(struct drm_device *dev);
@@ -423,9 +430,6 @@ extern void intel_crtc_fb_gamma_set(struct drm_crtc *crtc, u16 red, u16 green,
 extern void intel_crtc_fb_gamma_get(struct drm_crtc *crtc, u16 *red, u16 *green,
                                     u16 *blue, int regno);
 extern void intel_enable_clock_gating(struct drm_device *dev);
-extern void ironlake_disable_rc6(struct drm_device *dev);
-extern void ironlake_enable_drps(struct drm_device *dev);
-extern void ironlake_disable_drps(struct drm_device *dev);
 
 extern int intel_pin_and_fence_fb_obj(struct drm_device *dev,
                                       struct drm_i915_gem_object *obj,
@@ -492,10 +496,10 @@ extern void intel_update_fbc(struct drm_device *dev);
 extern void intel_gpu_ips_init(struct drm_i915_private *dev_priv);
 extern void intel_gpu_ips_teardown(void);
 
-extern void gen6_enable_rps(struct drm_i915_private *dev_priv);
-extern void gen6_update_ring_freq(struct drm_i915_private *dev_priv);
-extern void gen6_disable_rps(struct drm_device *dev);
-extern void intel_init_emon(struct drm_device *dev);
+extern void intel_enable_gt_powersave(struct drm_device *dev);
+extern void intel_disable_gt_powersave(struct drm_device *dev);
+extern void gen6_gt_check_fifodbg(struct drm_i915_private *dev_priv);
+extern void ironlake_teardown_rc6(struct drm_device *dev);
 
 extern void intel_ddi_dpms(struct drm_encoder *encoder, int mode);
 extern void intel_ddi_mode_set(struct drm_encoder *encoder,
diff --git a/drivers/gpu/drm/i915/intel_dvo.c b/drivers/gpu/drm/i915/intel_dvo.c
index 60ba50b956f2..36c542e5036b 100644
--- a/drivers/gpu/drm/i915/intel_dvo.c
+++ b/drivers/gpu/drm/i915/intel_dvo.c
@@ -136,7 +136,7 @@ static int intel_dvo_mode_valid(struct drm_connector *connector,
 }
 
 static bool intel_dvo_mode_fixup(struct drm_encoder *encoder,
-                                 struct drm_display_mode *mode,
+                                 const struct drm_display_mode *mode,
                                  struct drm_display_mode *adjusted_mode)
 {
         struct intel_dvo *intel_dvo = enc_to_intel_dvo(encoder);
diff --git a/drivers/gpu/drm/i915/intel_fb.c b/drivers/gpu/drm/i915/intel_fb.c
index bf8690720a0c..97f673523b97 100644
--- a/drivers/gpu/drm/i915/intel_fb.c
+++ b/drivers/gpu/drm/i915/intel_fb.c
@@ -65,7 +65,7 @@ static int intelfb_create(struct intel_fbdev *ifbdev,
         struct drm_i915_private *dev_priv = dev->dev_private;
         struct fb_info *info;
         struct drm_framebuffer *fb;
-        struct drm_mode_fb_cmd2 mode_cmd;
+        struct drm_mode_fb_cmd2 mode_cmd = {};
         struct drm_i915_gem_object *obj;
         struct device *device = &dev->pdev->dev;
         int size, ret;
@@ -140,7 +140,9 @@ static int intelfb_create(struct intel_fbdev *ifbdev,
         info->fix.smem_start = dev->mode_config.fb_base + obj->gtt_offset;
         info->fix.smem_len = size;
 
-        info->screen_base = ioremap_wc(dev->agp->base + obj->gtt_offset, size);
+        info->screen_base =
+                ioremap_wc(dev_priv->mm.gtt_base_addr + obj->gtt_offset,
+                           size);
         if (!info->screen_base) {
                 ret = -ENOSPC;
                 goto out_unpin;
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index 2ead3bf7c21d..98f602427eb8 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -37,6 +37,19 @@
 #include "i915_drm.h"
 #include "i915_drv.h"
 
+static void
+assert_hdmi_port_disabled(struct intel_hdmi *intel_hdmi)
+{
+        struct drm_device *dev = intel_hdmi->base.base.dev;
+        struct drm_i915_private *dev_priv = dev->dev_private;
+        uint32_t enabled_bits;
+
+        enabled_bits = IS_HASWELL(dev) ? DDI_BUF_CTL_ENABLE : SDVO_ENABLE;
+
+        WARN(I915_READ(intel_hdmi->sdvox_reg) & enabled_bits,
+             "HDMI port enabled, expecting disabled\n");
+}
+
 struct intel_hdmi *enc_to_intel_hdmi(struct drm_encoder *encoder)
 {
         return container_of(encoder, struct intel_hdmi, base.base);
@@ -121,36 +134,31 @@ static void g4x_write_infoframe(struct drm_encoder *encoder,
         uint32_t *data = (uint32_t *)frame;
         struct drm_device *dev = encoder->dev;
         struct drm_i915_private *dev_priv = dev->dev_private;
-        struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
         u32 val = I915_READ(VIDEO_DIP_CTL);
         unsigned i, len = DIP_HEADER_SIZE + frame->len;
 
-        val &= ~VIDEO_DIP_PORT_MASK;
-        if (intel_hdmi->sdvox_reg == SDVOB)
-                val |= VIDEO_DIP_PORT_B;
-        else if (intel_hdmi->sdvox_reg == SDVOC)
-                val |= VIDEO_DIP_PORT_C;
-        else
-                return;
+        WARN(!(val & VIDEO_DIP_ENABLE), "Writing DIP with CTL reg disabled\n");
 
         val &= ~(VIDEO_DIP_SELECT_MASK | 0xf); /* clear DIP data offset */
         val |= g4x_infoframe_index(frame);
 
         val &= ~g4x_infoframe_enable(frame);
-        val |= VIDEO_DIP_ENABLE;
 
         I915_WRITE(VIDEO_DIP_CTL, val);
 
+        mmiowb();
         for (i = 0; i < len; i += 4) {
                 I915_WRITE(VIDEO_DIP_DATA, *data);
                 data++;
         }
+        mmiowb();
 
         val |= g4x_infoframe_enable(frame);
         val &= ~VIDEO_DIP_FREQ_MASK;
         val |= VIDEO_DIP_FREQ_VSYNC;
 
         I915_WRITE(VIDEO_DIP_CTL, val);
+        POSTING_READ(VIDEO_DIP_CTL);
 }
 
 static void ibx_write_infoframe(struct drm_encoder *encoder,
@@ -160,46 +168,32 @@ static void ibx_write_infoframe(struct drm_encoder *encoder,
         struct drm_device *dev = encoder->dev;
         struct drm_i915_private *dev_priv = dev->dev_private;
         struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
-        struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
         int reg = TVIDEO_DIP_CTL(intel_crtc->pipe);
         unsigned i, len = DIP_HEADER_SIZE + frame->len;
         u32 val = I915_READ(reg);
 
-        val &= ~VIDEO_DIP_PORT_MASK;
-        switch (intel_hdmi->sdvox_reg) {
-        case HDMIB:
-                val |= VIDEO_DIP_PORT_B;
-                break;
-        case HDMIC:
-                val |= VIDEO_DIP_PORT_C;
-                break;
-        case HDMID:
-                val |= VIDEO_DIP_PORT_D;
-                break;
-        default:
-                return;
-        }
-
-        intel_wait_for_vblank(dev, intel_crtc->pipe);
+        WARN(!(val & VIDEO_DIP_ENABLE), "Writing DIP with CTL reg disabled\n");
 
         val &= ~(VIDEO_DIP_SELECT_MASK | 0xf); /* clear DIP data offset */
         val |= g4x_infoframe_index(frame);
 
         val &= ~g4x_infoframe_enable(frame);
-        val |= VIDEO_DIP_ENABLE;
 
         I915_WRITE(reg, val);
 
+        mmiowb();
         for (i = 0; i < len; i += 4) {
                 I915_WRITE(TVIDEO_DIP_DATA(intel_crtc->pipe), *data);
                 data++;
         }
+        mmiowb();
 
         val |= g4x_infoframe_enable(frame);
         val &= ~VIDEO_DIP_FREQ_MASK;
         val |= VIDEO_DIP_FREQ_VSYNC;
 
         I915_WRITE(reg, val);
+        POSTING_READ(reg);
 }
 
 static void cpt_write_infoframe(struct drm_encoder *encoder,
@@ -213,32 +207,31 @@ static void cpt_write_infoframe(struct drm_encoder *encoder,
         unsigned i, len = DIP_HEADER_SIZE + frame->len;
         u32 val = I915_READ(reg);
 
-        intel_wait_for_vblank(dev, intel_crtc->pipe);
+        WARN(!(val & VIDEO_DIP_ENABLE), "Writing DIP with CTL reg disabled\n");
 
         val &= ~(VIDEO_DIP_SELECT_MASK | 0xf); /* clear DIP data offset */
         val |= g4x_infoframe_index(frame);
 
         /* The DIP control register spec says that we need to update the AVI
          * infoframe without clearing its enable bit */
-        if (frame->type == DIP_TYPE_AVI)
-                val |= VIDEO_DIP_ENABLE_AVI;
-        else
+        if (frame->type != DIP_TYPE_AVI)
                 val &= ~g4x_infoframe_enable(frame);
 
-        val |= VIDEO_DIP_ENABLE;
-
         I915_WRITE(reg, val);
 
+        mmiowb();
         for (i = 0; i < len; i += 4) {
                 I915_WRITE(TVIDEO_DIP_DATA(intel_crtc->pipe), *data);
                 data++;
         }
+        mmiowb();
 
         val |= g4x_infoframe_enable(frame);
         val &= ~VIDEO_DIP_FREQ_MASK;
         val |= VIDEO_DIP_FREQ_VSYNC;
 
         I915_WRITE(reg, val);
+        POSTING_READ(reg);
 }
 
 static void vlv_write_infoframe(struct drm_encoder *encoder,
@@ -252,26 +245,28 @@ static void vlv_write_infoframe(struct drm_encoder *encoder,
         unsigned i, len = DIP_HEADER_SIZE + frame->len;
         u32 val = I915_READ(reg);
 
-        intel_wait_for_vblank(dev, intel_crtc->pipe);
+        WARN(!(val & VIDEO_DIP_ENABLE), "Writing DIP with CTL reg disabled\n");
 
         val &= ~(VIDEO_DIP_SELECT_MASK | 0xf); /* clear DIP data offset */
         val |= g4x_infoframe_index(frame);
 
         val &= ~g4x_infoframe_enable(frame);
-        val |= VIDEO_DIP_ENABLE;
 
         I915_WRITE(reg, val);
 
+        mmiowb();
         for (i = 0; i < len; i += 4) {
                 I915_WRITE(VLV_TVIDEO_DIP_DATA(intel_crtc->pipe), *data);
                 data++;
         }
+        mmiowb();
 
         val |= g4x_infoframe_enable(frame);
         val &= ~VIDEO_DIP_FREQ_MASK;
         val |= VIDEO_DIP_FREQ_VSYNC;
 
         I915_WRITE(reg, val);
+        POSTING_READ(reg);
 }
 
 static void hsw_write_infoframe(struct drm_encoder *encoder,
@@ -289,18 +284,19 @@ static void hsw_write_infoframe(struct drm_encoder *encoder,
         if (data_reg == 0)
                 return;
 
-        intel_wait_for_vblank(dev, intel_crtc->pipe);
-
         val &= ~hsw_infoframe_enable(frame);
         I915_WRITE(ctl_reg, val);
 
+        mmiowb();
         for (i = 0; i < len; i += 4) {
                 I915_WRITE(data_reg + i, *data);
                 data++;
         }
+        mmiowb();
 
         val |= hsw_infoframe_enable(frame);
         I915_WRITE(ctl_reg, val);
+        POSTING_READ(ctl_reg);
 }
 
 static void intel_set_infoframe(struct drm_encoder *encoder,
@@ -308,14 +304,11 @@ static void intel_set_infoframe(struct drm_encoder *encoder,
 {
         struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
 
-        if (!intel_hdmi->has_hdmi_sink)
-                return;
-
         intel_dip_infoframe_csum(frame);
         intel_hdmi->write_infoframe(encoder, frame);
 }
 
-void intel_hdmi_set_avi_infoframe(struct drm_encoder *encoder,
+static void intel_hdmi_set_avi_infoframe(struct drm_encoder *encoder,
                                          struct drm_display_mode *adjusted_mode)
 {
         struct dip_infoframe avi_if = {
@@ -330,7 +323,7 @@ void intel_hdmi_set_avi_infoframe(struct drm_encoder *encoder,
         intel_set_infoframe(encoder, &avi_if);
 }
 
-void intel_hdmi_set_spd_infoframe(struct drm_encoder *encoder)
+static void intel_hdmi_set_spd_infoframe(struct drm_encoder *encoder)
 {
         struct dip_infoframe spd_if;
 
@@ -345,6 +338,223 @@ void intel_hdmi_set_spd_infoframe(struct drm_encoder *encoder)
         intel_set_infoframe(encoder, &spd_if);
 }
 
+static void g4x_set_infoframes(struct drm_encoder *encoder,
+                               struct drm_display_mode *adjusted_mode)
+{
+        struct drm_i915_private *dev_priv = encoder->dev->dev_private;
+        struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
+        u32 reg = VIDEO_DIP_CTL;
+        u32 val = I915_READ(reg);
+        u32 port;
+
+        assert_hdmi_port_disabled(intel_hdmi);
+
+        /* If the registers were not initialized yet, they might be zeroes,
+         * which means we're selecting the AVI DIP and we're setting its
+         * frequency to once. This seems to really confuse the HW and make
+         * things stop working (the register spec says the AVI always needs to
+         * be sent every VSync). So here we avoid writing to the register more
+         * than we need and also explicitly select the AVI DIP and explicitly
+         * set its frequency to every VSync. Avoiding to write it twice seems to
+         * be enough to solve the problem, but being defensive shouldn't hurt us
+         * either. */
+        val |= VIDEO_DIP_SELECT_AVI | VIDEO_DIP_FREQ_VSYNC;
+
+        if (!intel_hdmi->has_hdmi_sink) {
+                if (!(val & VIDEO_DIP_ENABLE))
+                        return;
+                val &= ~VIDEO_DIP_ENABLE;
+                I915_WRITE(reg, val);
+                POSTING_READ(reg);
+                return;
+        }
+
+        switch (intel_hdmi->sdvox_reg) {
+        case SDVOB:
+                port = VIDEO_DIP_PORT_B;
+                break;
+        case SDVOC:
+                port = VIDEO_DIP_PORT_C;
+                break;
+        default:
+                return;
+        }
+
+        if (port != (val & VIDEO_DIP_PORT_MASK)) {
+                if (val & VIDEO_DIP_ENABLE) {
+                        val &= ~VIDEO_DIP_ENABLE;
+                        I915_WRITE(reg, val);
+                        POSTING_READ(reg);
+                }
+                val &= ~VIDEO_DIP_PORT_MASK;
+                val |= port;
+        }
+
+        val |= VIDEO_DIP_ENABLE;
+        val &= ~VIDEO_DIP_ENABLE_VENDOR;
+
+        I915_WRITE(reg, val);
+        POSTING_READ(reg);
+
+        intel_hdmi_set_avi_infoframe(encoder, adjusted_mode);
+        intel_hdmi_set_spd_infoframe(encoder);
+}
+
+static void ibx_set_infoframes(struct drm_encoder *encoder,
+                               struct drm_display_mode *adjusted_mode)
+{
+        struct drm_i915_private *dev_priv = encoder->dev->dev_private;
+        struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
+        struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
+        u32 reg = TVIDEO_DIP_CTL(intel_crtc->pipe);
+        u32 val = I915_READ(reg);
+        u32 port;
+
+        assert_hdmi_port_disabled(intel_hdmi);
+
+        /* See the big comment in g4x_set_infoframes() */
+        val |= VIDEO_DIP_SELECT_AVI | VIDEO_DIP_FREQ_VSYNC;
+
+        if (!intel_hdmi->has_hdmi_sink) {
+                if (!(val & VIDEO_DIP_ENABLE))
+                        return;
+                val &= ~VIDEO_DIP_ENABLE;
+                I915_WRITE(reg, val);
+                POSTING_READ(reg);
+                return;
+        }
+
+        switch (intel_hdmi->sdvox_reg) {
+        case HDMIB:
+                port = VIDEO_DIP_PORT_B;
+                break;
+        case HDMIC:
+                port = VIDEO_DIP_PORT_C;
+                break;
+        case HDMID:
+                port = VIDEO_DIP_PORT_D;
+                break;
+        default:
+                return;
+        }
+
+        if (port != (val & VIDEO_DIP_PORT_MASK)) {
+                if (val & VIDEO_DIP_ENABLE) {
+                        val &= ~VIDEO_DIP_ENABLE;
+                        I915_WRITE(reg, val);
+                        POSTING_READ(reg);
+                }
+                val &= ~VIDEO_DIP_PORT_MASK;
+                val |= port;
+        }
+
+        val |= VIDEO_DIP_ENABLE;
+        val &= ~(VIDEO_DIP_ENABLE_VENDOR | VIDEO_DIP_ENABLE_GAMUT |
+                 VIDEO_DIP_ENABLE_GCP);
+
+        I915_WRITE(reg, val);
+        POSTING_READ(reg);
+
+        intel_hdmi_set_avi_infoframe(encoder, adjusted_mode);
+        intel_hdmi_set_spd_infoframe(encoder);
+}
+
+static void cpt_set_infoframes(struct drm_encoder *encoder,
+                               struct drm_display_mode *adjusted_mode)
+{
+        struct drm_i915_private *dev_priv = encoder->dev->dev_private;
+        struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
+        struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
+        u32 reg = TVIDEO_DIP_CTL(intel_crtc->pipe);
+        u32 val = I915_READ(reg);
+
+        assert_hdmi_port_disabled(intel_hdmi);
+
+        /* See the big comment in g4x_set_infoframes() */
+        val |= VIDEO_DIP_SELECT_AVI | VIDEO_DIP_FREQ_VSYNC;
+
+        if (!intel_hdmi->has_hdmi_sink) {
+                if (!(val & VIDEO_DIP_ENABLE))
+                        return;
+                val &= ~(VIDEO_DIP_ENABLE | VIDEO_DIP_ENABLE_AVI);
+                I915_WRITE(reg, val);
+                POSTING_READ(reg);
+                return;
+        }
+
+        /* Set both together, unset both together: see the spec. */
+        val |= VIDEO_DIP_ENABLE | VIDEO_DIP_ENABLE_AVI;
+        val &= ~(VIDEO_DIP_ENABLE_VENDOR | VIDEO_DIP_ENABLE_GAMUT |
+                 VIDEO_DIP_ENABLE_GCP);
+
+        I915_WRITE(reg, val);
+        POSTING_READ(reg);
+
+        intel_hdmi_set_avi_infoframe(encoder, adjusted_mode);
+        intel_hdmi_set_spd_infoframe(encoder);
+}
+
+static void vlv_set_infoframes(struct drm_encoder *encoder,
+                               struct drm_display_mode *adjusted_mode)
+{
+        struct drm_i915_private *dev_priv = encoder->dev->dev_private;
+        struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
+        struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
+        u32 reg = VLV_TVIDEO_DIP_CTL(intel_crtc->pipe);
+        u32 val = I915_READ(reg);
+
+        assert_hdmi_port_disabled(intel_hdmi);
+
+        /* See the big comment in g4x_set_infoframes() */
+        val |= VIDEO_DIP_SELECT_AVI | VIDEO_DIP_FREQ_VSYNC;
+
+        if (!intel_hdmi->has_hdmi_sink) {
+                if (!(val & VIDEO_DIP_ENABLE))
+                        return;
+                val &= ~VIDEO_DIP_ENABLE;
+                I915_WRITE(reg, val);
+                POSTING_READ(reg);
+                return;
+        }
+
+        val |= VIDEO_DIP_ENABLE;
+        val &= ~(VIDEO_DIP_ENABLE_VENDOR | VIDEO_DIP_ENABLE_GAMUT |
+                 VIDEO_DIP_ENABLE_GCP);
+
+        I915_WRITE(reg, val);
+        POSTING_READ(reg);
+
+        intel_hdmi_set_avi_infoframe(encoder, adjusted_mode);
+        intel_hdmi_set_spd_infoframe(encoder);
+}
+
+static void hsw_set_infoframes(struct drm_encoder *encoder,
+                               struct drm_display_mode *adjusted_mode)
+{
+        struct drm_i915_private *dev_priv = encoder->dev->dev_private;
+        struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
+        struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
+        u32 reg = HSW_TVIDEO_DIP_CTL(intel_crtc->pipe);
+        u32 val = I915_READ(reg);
+
+        assert_hdmi_port_disabled(intel_hdmi);
+
+        if (!intel_hdmi->has_hdmi_sink) {
+                I915_WRITE(reg, 0);
+                POSTING_READ(reg);
+                return;
+        }
+
+        val &= ~(VIDEO_DIP_ENABLE_VSC_HSW | VIDEO_DIP_ENABLE_GCP_HSW |
+                 VIDEO_DIP_ENABLE_VS_HSW | VIDEO_DIP_ENABLE_GMP_HSW);
+
+        I915_WRITE(reg, val);
+        POSTING_READ(reg);
+
+        intel_hdmi_set_avi_infoframe(encoder, adjusted_mode);
+        intel_hdmi_set_spd_infoframe(encoder);
+}
+
 static void intel_hdmi_mode_set(struct drm_encoder *encoder,
                                 struct drm_display_mode *mode,
                                 struct drm_display_mode *adjusted_mode)
@@ -355,7 +565,7 @@ static void intel_hdmi_mode_set(struct drm_encoder *encoder,
         struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder);
         u32 sdvox;
 
-        sdvox = SDVO_ENCODING_HDMI | SDVO_BORDER_ENABLE;
+        sdvox = SDVO_ENCODING_HDMI;
         if (!HAS_PCH_SPLIT(dev))
                 sdvox |= intel_hdmi->color_range;
         if (adjusted_mode->flags & DRM_MODE_FLAG_PVSYNC)
@@ -382,14 +592,13 @@ static void intel_hdmi_mode_set(struct drm_encoder *encoder,
 
         if (HAS_PCH_CPT(dev))
                 sdvox |= PORT_TRANS_SEL_CPT(intel_crtc->pipe);
-        else if (intel_crtc->pipe == 1)
+        else if (intel_crtc->pipe == PIPE_B)
                 sdvox |= SDVO_PIPE_B_SELECT;
 
         I915_WRITE(intel_hdmi->sdvox_reg, sdvox);
         POSTING_READ(intel_hdmi->sdvox_reg);
 
-        intel_hdmi_set_avi_infoframe(encoder, adjusted_mode);
-        intel_hdmi_set_spd_infoframe(encoder);
+        intel_hdmi->set_infoframes(encoder, adjusted_mode);
 }
 
 static void intel_hdmi_dpms(struct drm_encoder *encoder, int mode)
@@ -405,6 +614,36 @@ static void intel_hdmi_dpms(struct drm_encoder *encoder, int mode)
 
         temp = I915_READ(intel_hdmi->sdvox_reg);
 
+        /* HW workaround for IBX, we need to move the port to transcoder A
+         * before disabling it. */
+        if (HAS_PCH_IBX(dev)) {
+                struct drm_crtc *crtc = encoder->crtc;
+                int pipe = crtc ? to_intel_crtc(crtc)->pipe : -1;
+
+                if (mode != DRM_MODE_DPMS_ON) {
+                        if (temp & SDVO_PIPE_B_SELECT) {
+                                temp &= ~SDVO_PIPE_B_SELECT;
+                                I915_WRITE(intel_hdmi->sdvox_reg, temp);
+                                POSTING_READ(intel_hdmi->sdvox_reg);
+
+                                /* Again we need to write this twice. */
+                                I915_WRITE(intel_hdmi->sdvox_reg, temp);
+                                POSTING_READ(intel_hdmi->sdvox_reg);
+
+                                /* Transcoder selection bits only update
+                                 * effectively on vblank. */
+                                if (crtc)
+                                        intel_wait_for_vblank(dev, pipe);
+                                else
+                                        msleep(50);
+                        }
+                } else {
+                        /* Restore the transcoder select bit. */
+                        if (pipe == PIPE_B)
+                                enable_bits |= SDVO_PIPE_B_SELECT;
+                }
+        }
+
         /* HW workaround, need to toggle enable bit off and on for 12bpc, but
          * we do this anyway which shows more stable in testing.
          */
@@ -446,12 +685,33 @@ static int intel_hdmi_mode_valid(struct drm_connector *connector,
 }
 
 static bool intel_hdmi_mode_fixup(struct drm_encoder *encoder,
-                                  struct drm_display_mode *mode,
+                                  const struct drm_display_mode *mode,
                                   struct drm_display_mode *adjusted_mode)
 {
         return true;
 }
 
+static bool g4x_hdmi_connected(struct intel_hdmi *intel_hdmi)
+{
+        struct drm_device *dev = intel_hdmi->base.base.dev;
+        struct drm_i915_private *dev_priv = dev->dev_private;
+        uint32_t bit;
+
+        switch (intel_hdmi->sdvox_reg) {
+        case SDVOB:
+                bit = HDMIB_HOTPLUG_LIVE_STATUS;
+                break;
+        case SDVOC:
+                bit = HDMIC_HOTPLUG_LIVE_STATUS;
+                break;
+        default:
+                bit = 0;
+                break;
+        }
+
+        return I915_READ(PORT_HOTPLUG_STAT) & bit;
+}
+
 static enum drm_connector_status
 intel_hdmi_detect(struct drm_connector *connector, bool force)
 {
@@ -460,6 +720,9 @@ intel_hdmi_detect(struct drm_connector *connector, bool force)
         struct edid *edid;
         enum drm_connector_status status = connector_status_disconnected;
 
+        if (IS_G4X(connector->dev) && !g4x_hdmi_connected(intel_hdmi))
+                return status;
+
         intel_hdmi->has_hdmi_sink = false;
         intel_hdmi->has_audio = false;
         edid = drm_get_edid(connector,
@@ -633,7 +896,6 @@ void intel_hdmi_init(struct drm_device *dev, int sdvox_reg)
         struct intel_encoder *intel_encoder;
         struct intel_connector *intel_connector;
         struct intel_hdmi *intel_hdmi;
-        int i;
 
         intel_hdmi = kzalloc(sizeof(struct intel_hdmi), GFP_KERNEL);
         if (!intel_hdmi)
@@ -710,26 +972,19 @@ void intel_hdmi_init(struct drm_device *dev, int sdvox_reg)
 
         if (!HAS_PCH_SPLIT(dev)) {
                 intel_hdmi->write_infoframe = g4x_write_infoframe;
-                I915_WRITE(VIDEO_DIP_CTL, 0);
+                intel_hdmi->set_infoframes = g4x_set_infoframes;
         } else if (IS_VALLEYVIEW(dev)) {
                 intel_hdmi->write_infoframe = vlv_write_infoframe;
-                for_each_pipe(i)
-                        I915_WRITE(VLV_TVIDEO_DIP_CTL(i), 0);
+                intel_hdmi->set_infoframes = vlv_set_infoframes;
         } else if (IS_HASWELL(dev)) {
-                /* FIXME: Haswell has a new set of DIP frame registers, but we are
-                 * just doing the minimal required for HDMI to work at this stage.
-                 */
                 intel_hdmi->write_infoframe = hsw_write_infoframe;
-                for_each_pipe(i)
-                        I915_WRITE(HSW_TVIDEO_DIP_CTL(i), 0);
+                intel_hdmi->set_infoframes = hsw_set_infoframes;
         } else if (HAS_PCH_IBX(dev)) {
                 intel_hdmi->write_infoframe = ibx_write_infoframe;
-                for_each_pipe(i)
-                        I915_WRITE(TVIDEO_DIP_CTL(i), 0);
+                intel_hdmi->set_infoframes = ibx_set_infoframes;
         } else {
                 intel_hdmi->write_infoframe = cpt_write_infoframe;
-                for_each_pipe(i)
-                        I915_WRITE(TVIDEO_DIP_CTL(i), 0);
+                intel_hdmi->set_infoframes = cpt_set_infoframes;
         }
 
         if (IS_HASWELL(dev))
diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c
index 08eb04c787e8..49f09a8b05e9 100644
--- a/drivers/gpu/drm/i915/intel_lvds.c
+++ b/drivers/gpu/drm/i915/intel_lvds.c
@@ -71,6 +71,7 @@ static struct intel_lvds *intel_attached_lvds(struct drm_connector *connector)
 static void intel_lvds_enable(struct intel_lvds *intel_lvds)
 {
         struct drm_device *dev = intel_lvds->base.base.dev;
+        struct intel_crtc *intel_crtc = to_intel_crtc(intel_lvds->base.base.crtc);
         struct drm_i915_private *dev_priv = dev->dev_private;
         u32 ctl_reg, lvds_reg, stat_reg;
 
@@ -107,7 +108,7 @@ static void intel_lvds_enable(struct intel_lvds *intel_lvds)
         if (wait_for((I915_READ(stat_reg) & PP_ON) != 0, 1000))
                 DRM_ERROR("timed out waiting for panel to power on\n");
 
-        intel_panel_enable_backlight(dev);
+        intel_panel_enable_backlight(dev, intel_crtc->pipe);
 }
 
 static void intel_lvds_disable(struct intel_lvds *intel_lvds)
@@ -228,14 +229,14 @@ static inline u32 panel_fitter_scaling(u32 source, u32 target)
 }
 
 static bool intel_lvds_mode_fixup(struct drm_encoder *encoder,
-                                  struct drm_display_mode *mode,
+                                  const struct drm_display_mode *mode,
                                   struct drm_display_mode *adjusted_mode)
 {
         struct drm_device *dev = encoder->dev;
         struct drm_i915_private *dev_priv = dev->dev_private;
         struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
         struct intel_lvds *intel_lvds = to_intel_lvds(encoder);
-        struct drm_encoder *tmp_encoder;
+        struct intel_encoder *tmp_encoder;
         u32 pfit_control = 0, pfit_pgm_ratios = 0, border = 0;
         int pipe;
 
@@ -246,8 +247,8 @@ static bool intel_lvds_mode_fixup(struct drm_encoder *encoder,
         }
 
         /* Should never happen!! */
-        list_for_each_entry(tmp_encoder, &dev->mode_config.encoder_list, head) {
-                if (tmp_encoder != encoder && tmp_encoder->crtc == encoder->crtc) {
+        for_each_encoder_on_crtc(dev, encoder->crtc, tmp_encoder) {
+                if (&tmp_encoder->base != encoder) {
                         DRM_ERROR("Can't enable LVDS and another "
                                "encoder on the same pipe\n");
                         return false;
@@ -777,6 +778,14 @@ static const struct dmi_system_id intel_no_lvds[] = {
                         DMI_MATCH(DMI_BOARD_NAME, "MS-7469"),
                 },
         },
+        {
+                .callback = intel_no_lvds_dmi_callback,
+                .ident = "ZOTAC ZBOXSD-ID12/ID13",
+                .matches = {
+                        DMI_MATCH(DMI_BOARD_VENDOR, "ZOTAC"),
+                        DMI_MATCH(DMI_BOARD_NAME, "ZBOXSD-ID12/ID13"),
+                },
+        },
 
         { }     /* terminating entry */
 };
@@ -967,6 +976,8 @@ bool intel_lvds_init(struct drm_device *dev)
         intel_encoder->clone_mask = (1 << INTEL_LVDS_CLONE_BIT);
         if (HAS_PCH_SPLIT(dev))
                 intel_encoder->crtc_mask = (1 << 0) | (1 << 1) | (1 << 2);
+        else if (IS_GEN4(dev))
+                intel_encoder->crtc_mask = (1 << 0) | (1 << 1);
         else
                 intel_encoder->crtc_mask = (1 << 1);
 
@@ -1074,35 +1085,14 @@ bool intel_lvds_init(struct drm_device *dev)
                 goto failed;
 
 out:
+        /*
+         * Unlock registers and just
+         * leave them unlocked
+         */
         if (HAS_PCH_SPLIT(dev)) {
-                u32 pwm;
-
-                pipe = (I915_READ(PCH_LVDS) & LVDS_PIPEB_SELECT) ? 1 : 0;
-
-                /* make sure PWM is enabled and locked to the LVDS pipe */
-                pwm = I915_READ(BLC_PWM_CPU_CTL2);
-                if (pipe == 0 && (pwm & PWM_PIPE_B))
-                        I915_WRITE(BLC_PWM_CPU_CTL2, pwm & ~PWM_ENABLE);
-                if (pipe)
-                        pwm |= PWM_PIPE_B;
-                else
-                        pwm &= ~PWM_PIPE_B;
-                I915_WRITE(BLC_PWM_CPU_CTL2, pwm | PWM_ENABLE);
-
-                pwm = I915_READ(BLC_PWM_PCH_CTL1);
-                pwm |= PWM_PCH_ENABLE;
-                I915_WRITE(BLC_PWM_PCH_CTL1, pwm);
-                /*
-                 * Unlock registers and just
-                 * leave them unlocked
-                 */
                 I915_WRITE(PCH_PP_CONTROL,
                            I915_READ(PCH_PP_CONTROL) | PANEL_UNLOCK_REGS);
         } else {
-                /*
-                 * Unlock registers and just
-                 * leave them unlocked
-                 */
                 I915_WRITE(PP_CONTROL,
                            I915_READ(PP_CONTROL) | PANEL_UNLOCK_REGS);
         }
diff --git a/drivers/gpu/drm/i915/intel_overlay.c b/drivers/gpu/drm/i915/intel_overlay.c
index 458743da3774..830d0dd610e1 100644
--- a/drivers/gpu/drm/i915/intel_overlay.c
+++ b/drivers/gpu/drm/i915/intel_overlay.c
@@ -226,7 +226,7 @@ static int intel_overlay_do_wait_request(struct intel_overlay *overlay,
         }
         overlay->last_flip_req = request->seqno;
         overlay->flip_tail = tail;
-        ret = i915_wait_request(ring, overlay->last_flip_req);
+        ret = i915_wait_seqno(ring, overlay->last_flip_req);
         if (ret)
                 return ret;
         i915_gem_retire_requests(dev);
@@ -452,7 +452,7 @@ static int intel_overlay_recover_from_interrupt(struct intel_overlay *overlay)
         if (overlay->last_flip_req == 0)
                 return 0;
 
-        ret = i915_wait_request(ring, overlay->last_flip_req);
+        ret = i915_wait_seqno(ring, overlay->last_flip_req);
         if (ret)
                 return ret;
         i915_gem_retire_requests(dev);
diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c
index 2a1625d84a69..58c7ee7238b8 100644
--- a/drivers/gpu/drm/i915/intel_panel.c
+++ b/drivers/gpu/drm/i915/intel_panel.c
@@ -56,7 +56,7 @@ intel_fixed_panel_mode(struct drm_display_mode *fixed_mode,
 void
 intel_pch_panel_fitting(struct drm_device *dev,
                         int fitting_mode,
-                        struct drm_display_mode *mode,
+                        const struct drm_display_mode *mode,
                         struct drm_display_mode *adjusted_mode)
 {
         struct drm_i915_private *dev_priv = dev->dev_private;
@@ -287,9 +287,18 @@ void intel_panel_disable_backlight(struct drm_device *dev)
 
         dev_priv->backlight_enabled = false;
         intel_panel_actually_set_backlight(dev, 0);
+
+        if (INTEL_INFO(dev)->gen >= 4) {
+                uint32_t reg;
+
+                reg = HAS_PCH_SPLIT(dev) ? BLC_PWM_CPU_CTL2 : BLC_PWM_CTL2;
+
+                I915_WRITE(reg, I915_READ(reg) & ~BLM_PWM_ENABLE);
+        }
 }
 
-void intel_panel_enable_backlight(struct drm_device *dev)
+void intel_panel_enable_backlight(struct drm_device *dev,
+                                  enum pipe pipe)
 {
         struct drm_i915_private *dev_priv = dev->dev_private;
 
@@ -298,6 +307,33 @@ void intel_panel_enable_backlight(struct drm_device *dev)
 
         dev_priv->backlight_enabled = true;
         intel_panel_actually_set_backlight(dev, dev_priv->backlight_level);
+
+        if (INTEL_INFO(dev)->gen >= 4) {
+                uint32_t reg, tmp;
+
+                reg = HAS_PCH_SPLIT(dev) ? BLC_PWM_CPU_CTL2 : BLC_PWM_CTL2;
+
+
+                tmp = I915_READ(reg);
+
+                /* Note that this can also get called through dpms changes. And
+                 * we don't track the backlight dpms state, hence check whether
+                 * we have to do anything first. */
+                if (tmp & BLM_PWM_ENABLE)
+                        return;
+
+                if (dev_priv->num_pipe == 3)
+                        tmp &= ~BLM_PIPE_SELECT_IVB;
+                else
+                        tmp &= ~BLM_PIPE_SELECT;
+
+                tmp |= BLM_PIPE(pipe);
+                tmp &= ~BLM_PWM_ENABLE;
+
+                I915_WRITE(reg, tmp);
+                POSTING_READ(reg);
+                I915_WRITE(reg, tmp | BLM_PWM_ENABLE);
+        }
 }
 
 static void intel_panel_init_backlight(struct drm_device *dev)
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index d0ce2a5b1d3f..0bb69fd255a9 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -387,8 +387,6 @@ void intel_update_fbc(struct drm_device *dev)
         struct drm_i915_gem_object *obj;
         int enable_fbc;
 
-        DRM_DEBUG_KMS("\n");
-
         if (!i915_powersave)
                 return;
 
@@ -405,7 +403,9 @@ void intel_update_fbc(struct drm_device *dev)
          *   - going to an unsupported config (interlace, pixel multiply, etc.)
          */
         list_for_each_entry(tmp_crtc, &dev->mode_config.crtc_list, head) {
-                if (tmp_crtc->enabled && tmp_crtc->fb) {
+                if (tmp_crtc->enabled &&
+                    !to_intel_crtc(tmp_crtc)->primary_disabled &&
+                    tmp_crtc->fb) {
                         if (crtc) {
                                 DRM_DEBUG_KMS("more than one pipe active, disabling compression\n");
                                 dev_priv->no_fbc_reason = FBC_MULTIPLE_PIPES;
@@ -2182,7 +2182,7 @@ bool ironlake_set_drps(struct drm_device *dev, u8 val)
         return true;
 }
 
-void ironlake_enable_drps(struct drm_device *dev)
+static void ironlake_enable_drps(struct drm_device *dev)
 {
         struct drm_i915_private *dev_priv = dev->dev_private;
         u32 rgvmodectl = I915_READ(MEMMODECTL);
@@ -2246,7 +2246,7 @@ void ironlake_enable_drps(struct drm_device *dev)
         getrawmonotonic(&dev_priv->last_time2);
 }
 
-void ironlake_disable_drps(struct drm_device *dev)
+static void ironlake_disable_drps(struct drm_device *dev)
 {
         struct drm_i915_private *dev_priv = dev->dev_private;
         u16 rgvswctl = I915_READ16(MEMSWCTL);
@@ -2299,10 +2299,11 @@ void gen6_set_rps(struct drm_device *dev, u8 val)
         dev_priv->cur_delay = val;
 }
 
-void gen6_disable_rps(struct drm_device *dev)
+static void gen6_disable_rps(struct drm_device *dev)
 {
         struct drm_i915_private *dev_priv = dev->dev_private;
 
+        I915_WRITE(GEN6_RC_CONTROL, 0);
         I915_WRITE(GEN6_RPNSWREQ, 1 << 31);
         I915_WRITE(GEN6_PMINTRMSK, 0xffffffff);
         I915_WRITE(GEN6_PMIER, 0);
@@ -2332,9 +2333,11 @@ int intel_enable_rc6(const struct drm_device *dev)
         if (INTEL_INFO(dev)->gen == 5)
                 return 0;
 
-        /* Sorry Haswell, no RC6 for you for now. */
+        /* On Haswell, only RC6 is available. So let's enable it by default to
+         * provide better testing and coverage since the beginning.
+         */
         if (IS_HASWELL(dev))
-                return 0;
+                return INTEL_RC6_ENABLE;
 
         /*
          * Disable rc6 on Sandybridge
@@ -2347,8 +2350,9 @@ int intel_enable_rc6(const struct drm_device *dev)
         return (INTEL_RC6_ENABLE | INTEL_RC6p_ENABLE);
 }
 
-void gen6_enable_rps(struct drm_i915_private *dev_priv)
+static void gen6_enable_rps(struct drm_device *dev)
 {
+        struct drm_i915_private *dev_priv = dev->dev_private;
         struct intel_ring_buffer *ring;
         u32 rp_state_cap;
         u32 gt_perf_status;
@@ -2357,6 +2361,8 @@ void gen6_enable_rps(struct drm_i915_private *dev_priv)
         int rc6_mode;
         int i;
 
+        WARN_ON(!mutex_is_locked(&dev->struct_mutex));
+
         /* Here begins a magic sequence of register writes to enable
          * auto-downclocking.
          *
@@ -2364,7 +2370,6 @@ void gen6_enable_rps(struct drm_i915_private *dev_priv)
          * userspace...
          */
         I915_WRITE(GEN6_RC_STATE, 0);
-        mutex_lock(&dev_priv->dev->struct_mutex);
 
         /* Clear the DBG now so we don't confuse earlier errors */
         if ((gtfifodbg = I915_READ(GTFIFODBG))) {
@@ -2400,20 +2405,24 @@ void gen6_enable_rps(struct drm_i915_private *dev_priv)
         I915_WRITE(GEN6_RC6p_THRESHOLD, 100000);
         I915_WRITE(GEN6_RC6pp_THRESHOLD, 64000); /* unused */
 
+        /* Check if we are enabling RC6 */
         rc6_mode = intel_enable_rc6(dev_priv->dev);
         if (rc6_mode & INTEL_RC6_ENABLE)
                 rc6_mask |= GEN6_RC_CTL_RC6_ENABLE;
 
-        if (rc6_mode & INTEL_RC6p_ENABLE)
-                rc6_mask |= GEN6_RC_CTL_RC6p_ENABLE;
+        /* We don't use those on Haswell */
+        if (!IS_HASWELL(dev)) {
+                if (rc6_mode & INTEL_RC6p_ENABLE)
+                        rc6_mask |= GEN6_RC_CTL_RC6p_ENABLE;
 
-        if (rc6_mode & INTEL_RC6pp_ENABLE)
-                rc6_mask |= GEN6_RC_CTL_RC6pp_ENABLE;
+                if (rc6_mode & INTEL_RC6pp_ENABLE)
+                        rc6_mask |= GEN6_RC_CTL_RC6pp_ENABLE;
+        }
 
         DRM_INFO("Enabling RC6 states: RC6 %s, RC6p %s, RC6pp %s\n",
-                        (rc6_mode & INTEL_RC6_ENABLE) ? "on" : "off",
-                        (rc6_mode & INTEL_RC6p_ENABLE) ? "on" : "off",
-                        (rc6_mode & INTEL_RC6pp_ENABLE) ? "on" : "off");
+                        (rc6_mask & GEN6_RC_CTL_RC6_ENABLE) ? "on" : "off",
+                        (rc6_mask & GEN6_RC_CTL_RC6p_ENABLE) ? "on" : "off",
+                        (rc6_mask & GEN6_RC_CTL_RC6pp_ENABLE) ? "on" : "off");
 
         I915_WRITE(GEN6_RC_CONTROL,
                    rc6_mask |
@@ -2431,10 +2440,19 @@ void gen6_enable_rps(struct drm_i915_private *dev_priv)
         I915_WRITE(GEN6_RP_INTERRUPT_LIMITS,
                    dev_priv->max_delay << 24 |
                    dev_priv->min_delay << 16);
-        I915_WRITE(GEN6_RP_UP_THRESHOLD, 10000);
-        I915_WRITE(GEN6_RP_DOWN_THRESHOLD, 1000000);
-        I915_WRITE(GEN6_RP_UP_EI, 100000);
-        I915_WRITE(GEN6_RP_DOWN_EI, 5000000);
+
+        if (IS_HASWELL(dev)) {
+                I915_WRITE(GEN6_RP_UP_THRESHOLD, 59400);
+                I915_WRITE(GEN6_RP_DOWN_THRESHOLD, 245000);
+                I915_WRITE(GEN6_RP_UP_EI, 66000);
+                I915_WRITE(GEN6_RP_DOWN_EI, 350000);
+        } else {
+                I915_WRITE(GEN6_RP_UP_THRESHOLD, 10000);
+                I915_WRITE(GEN6_RP_DOWN_THRESHOLD, 1000000);
+                I915_WRITE(GEN6_RP_UP_EI, 100000);
+                I915_WRITE(GEN6_RP_DOWN_EI, 5000000);
+        }
+
         I915_WRITE(GEN6_RP_IDLE_HYSTERSIS, 10);
         I915_WRITE(GEN6_RP_CONTROL,
                    GEN6_RP_MEDIA_TURBO |
@@ -2442,7 +2460,7 @@ void gen6_enable_rps(struct drm_i915_private *dev_priv)
                    GEN6_RP_MEDIA_IS_GFX |
                    GEN6_RP_ENABLE |
                    GEN6_RP_UP_BUSY_AVG |
-                   GEN6_RP_DOWN_IDLE_CONT);
+                   (IS_HASWELL(dev) ? GEN7_RP_DOWN_IDLE_AVG : GEN6_RP_DOWN_IDLE_CONT));
 
         if (wait_for((I915_READ(GEN6_PCODE_MAILBOX) & GEN6_PCODE_READY) == 0,
                      500))
@@ -2489,15 +2507,17 @@ void gen6_enable_rps(struct drm_i915_private *dev_priv)
         I915_WRITE(GEN6_PMINTRMSK, 0);
 
         gen6_gt_force_wake_put(dev_priv);
-        mutex_unlock(&dev_priv->dev->struct_mutex);
 }
 
-void gen6_update_ring_freq(struct drm_i915_private *dev_priv)
+static void gen6_update_ring_freq(struct drm_device *dev)
 {
+        struct drm_i915_private *dev_priv = dev->dev_private;
         int min_freq = 15;
         int gpu_freq, ia_freq, max_ia_freq;
         int scaling_factor = 180;
 
+        WARN_ON(!mutex_is_locked(&dev->struct_mutex));
+
         max_ia_freq = cpufreq_quick_get_max(0);
         /*
          * Default to measured freq if none found, PCU will ensure we don't go
@@ -2509,8 +2529,6 @@ void gen6_update_ring_freq(struct drm_i915_private *dev_priv)
         /* Convert from kHz to MHz */
         max_ia_freq /= 1000;
 
-        mutex_lock(&dev_priv->dev->struct_mutex);
-
         /*
          * For each potential GPU frequency, load a ring frequency we'd like
          * to use for memory access.  We do this by specifying the IA frequency
@@ -2541,11 +2559,9 @@ void gen6_update_ring_freq(struct drm_i915_private *dev_priv)
                         continue;
                 }
         }
-
-        mutex_unlock(&dev_priv->dev->struct_mutex);
 }
 
-static void ironlake_teardown_rc6(struct drm_device *dev)
+void ironlake_teardown_rc6(struct drm_device *dev)
 {
         struct drm_i915_private *dev_priv = dev->dev_private;
 
@@ -2562,7 +2578,7 @@ static void ironlake_teardown_rc6(struct drm_device *dev)
         }
 }
 
-void ironlake_disable_rc6(struct drm_device *dev)
+static void ironlake_disable_rc6(struct drm_device *dev)
 {
         struct drm_i915_private *dev_priv = dev->dev_private;
 
@@ -2578,8 +2594,6 @@ void ironlake_disable_rc6(struct drm_device *dev)
                 I915_WRITE(RSTDBYCTL, I915_READ(RSTDBYCTL) & ~RCX_SW_EXIT);
                 POSTING_READ(RSTDBYCTL);
         }
-
-        ironlake_teardown_rc6(dev);
 }
 
 static int ironlake_setup_rc6(struct drm_device *dev)
@@ -2601,7 +2615,7 @@ static int ironlake_setup_rc6(struct drm_device *dev)
         return 0;
 }
 
-void ironlake_enable_rc6(struct drm_device *dev)
+static void ironlake_enable_rc6(struct drm_device *dev)
 {
         struct drm_i915_private *dev_priv = dev->dev_private;
         struct intel_ring_buffer *ring = &dev_priv->ring[RCS];
@@ -2613,12 +2627,11 @@ void ironlake_enable_rc6(struct drm_device *dev)
         if (!intel_enable_rc6(dev))
                 return;
 
-        mutex_lock(&dev->struct_mutex);
+        WARN_ON(!mutex_is_locked(&dev->struct_mutex));
+
         ret = ironlake_setup_rc6(dev);
-        if (ret) {
-                mutex_unlock(&dev->struct_mutex);
+        if (ret)
                 return;
-        }
 
         /*
          * GPU can automatically power down the render unit if given a page
@@ -2627,7 +2640,6 @@ void ironlake_enable_rc6(struct drm_device *dev)
         ret = intel_ring_begin(ring, 6);
         if (ret) {
                 ironlake_teardown_rc6(dev);
-                mutex_unlock(&dev->struct_mutex);
                 return;
         }
 
@@ -2652,13 +2664,11 @@ void ironlake_enable_rc6(struct drm_device *dev)
         if (ret) {
                 DRM_ERROR("failed to enable ironlake power power savings\n");
                 ironlake_teardown_rc6(dev);
-                mutex_unlock(&dev->struct_mutex);
                 return;
         }
 
         I915_WRITE(PWRCTXA, dev_priv->pwrctx->gtt_offset | PWRCTX_EN);
         I915_WRITE(RSTDBYCTL, I915_READ(RSTDBYCTL) & ~RCX_SW_EXIT);
-        mutex_unlock(&dev->struct_mutex);
 }
 
 static unsigned long intel_pxfreq(u32 vidfreq)
@@ -3154,8 +3164,7 @@ void intel_gpu_ips_teardown(void)
         i915_mch_dev = NULL;
         spin_unlock(&mchdev_lock);
 }
-
-void intel_init_emon(struct drm_device *dev)
+static void intel_init_emon(struct drm_device *dev)
 {
         struct drm_i915_private *dev_priv = dev->dev_private;
         u32 lcfuse;
@@ -3226,6 +3235,28 @@ void intel_init_emon(struct drm_device *dev)
         dev_priv->corr = (lcfuse & LCFUSE_HIV_MASK);
 }
 
+void intel_disable_gt_powersave(struct drm_device *dev)
+{
+        if (IS_IRONLAKE_M(dev)) {
+                ironlake_disable_drps(dev);
+                ironlake_disable_rc6(dev);
+        } else if (INTEL_INFO(dev)->gen >= 6 && !IS_VALLEYVIEW(dev)) {
+                gen6_disable_rps(dev);
+        }
+}
+
+void intel_enable_gt_powersave(struct drm_device *dev)
+{
+        if (IS_IRONLAKE_M(dev)) {
+                ironlake_enable_drps(dev);
+                ironlake_enable_rc6(dev);
+                intel_init_emon(dev);
+        } else if ((IS_GEN6(dev) || IS_GEN7(dev)) && !IS_VALLEYVIEW(dev)) {
+                gen6_enable_rps(dev);
+                gen6_update_ring_freq(dev);
+        }
+}
+
 static void ironlake_init_clock_gating(struct drm_device *dev)
 {
         struct drm_i915_private *dev_priv = dev->dev_private;
@@ -3328,8 +3359,12 @@ static void gen6_init_clock_gating(struct drm_device *dev)
          *
          * According to the spec, bit 11 (RCCUNIT) must also be set,
          * but we didn't debug actual testcases to find it out.
+         *
+         * Also apply WaDisableVDSUnitClockGating and
+         * WaDisableRCPBUnitClockGating.
          */
         I915_WRITE(GEN6_UCGCTL2,
+                   GEN7_VDSUNIT_CLOCK_GATE_DISABLE |
                    GEN6_RCPBUNIT_CLOCK_GATE_DISABLE |
                    GEN6_RCCUNIT_CLOCK_GATE_DISABLE);
 
@@ -3357,6 +3392,9 @@ static void gen6_init_clock_gating(struct drm_device *dev)
                    ILK_DPARB_CLK_GATE  |
                    ILK_DPFD_CLK_GATE);
 
+        I915_WRITE(GEN6_MBCTL, I915_READ(GEN6_MBCTL) |
+                   GEN6_MBCTL_ENABLE_BOOT_FETCH);
+
         for_each_pipe(pipe) {
                 I915_WRITE(DSPCNTR(pipe),
                            I915_READ(DSPCNTR(pipe)) |
@@ -3377,7 +3415,7 @@ static void gen7_setup_fixed_func_scheduler(struct drm_i915_private *dev_priv)
         I915_WRITE(GEN7_FF_THREAD_MODE, reg);
 }
 
-static void ivybridge_init_clock_gating(struct drm_device *dev)
+static void haswell_init_clock_gating(struct drm_device *dev)
 {
         struct drm_i915_private *dev_priv = dev->dev_private;
         int pipe;
@@ -3427,13 +3465,24 @@ static void ivybridge_init_clock_gating(struct drm_device *dev)
         /* WaDisable4x2SubspanOptimization */
         I915_WRITE(CACHE_MODE_1,
                    _MASKED_BIT_ENABLE(PIXEL_SUBSPAN_COLLECT_OPT_DISABLE));
+
+        /* XXX: This is a workaround for early silicon revisions and should be
+         * removed later.
+         */
+        I915_WRITE(WM_DBG,
+                        I915_READ(WM_DBG) |
+                        WM_DBG_DISALLOW_MULTIPLE_LP |
+                        WM_DBG_DISALLOW_SPRITE |
+                        WM_DBG_DISALLOW_MAXFIFO);
+
 }
 
-static void valleyview_init_clock_gating(struct drm_device *dev)
+static void ivybridge_init_clock_gating(struct drm_device *dev)
 {
         struct drm_i915_private *dev_priv = dev->dev_private;
         int pipe;
         uint32_t dspclk_gate = VRHUNIT_CLOCK_GATE_DISABLE;
+        uint32_t snpcr;
 
         I915_WRITE(PCH_DSPCLK_GATE_D, dspclk_gate);
 
@@ -3441,10 +3490,77 @@ static void valleyview_init_clock_gating(struct drm_device *dev)
         I915_WRITE(WM2_LP_ILK, 0);
         I915_WRITE(WM1_LP_ILK, 0);
 
-        /* According to the spec, bit 13 (RCZUNIT) must be set on IVB.
+        I915_WRITE(ILK_DSPCLK_GATE, IVB_VRHUNIT_CLK_GATE);
+
+        I915_WRITE(IVB_CHICKEN3,
+                   CHICKEN3_DGMG_REQ_OUT_FIX_DISABLE |
+                   CHICKEN3_DGMG_DONE_FIX_DISABLE);
+
+        /* Apply the WaDisableRHWOOptimizationForRenderHang workaround. */
+        I915_WRITE(GEN7_COMMON_SLICE_CHICKEN1,
+                   GEN7_CSC1_RHWO_OPT_DISABLE_IN_RCC);
+
+        /* WaApplyL3ControlAndL3ChickenMode requires those two on Ivy Bridge */
+        I915_WRITE(GEN7_L3CNTLREG1,
+                        GEN7_WA_FOR_GEN7_L3_CONTROL);
+        I915_WRITE(GEN7_L3_CHICKEN_MODE_REGISTER,
+                        GEN7_WA_L3_CHICKEN_MODE);
+
+        /* According to the BSpec vol1g, bit 12 (RCPBUNIT) clock
+         * gating disable must be set.  Failure to set it results in
+         * flickering pixels due to Z write ordering failures after
+         * some amount of runtime in the Mesa "fire" demo, and Unigine
+         * Sanctuary and Tropics, and apparently anything else with
+         * alpha test or pixel discard.
+         *
+         * According to the spec, bit 11 (RCCUNIT) must also be set,
+         * but we didn't debug actual testcases to find it out.
+         *
+         * According to the spec, bit 13 (RCZUNIT) must be set on IVB.
          * This implements the WaDisableRCZUnitClockGating workaround.
          */
-        I915_WRITE(GEN6_UCGCTL2, GEN6_RCZUNIT_CLOCK_GATE_DISABLE);
+        I915_WRITE(GEN6_UCGCTL2,
+                   GEN6_RCZUNIT_CLOCK_GATE_DISABLE |
+                   GEN6_RCCUNIT_CLOCK_GATE_DISABLE);
+
+        /* This is required by WaCatErrorRejectionIssue */
+        I915_WRITE(GEN7_SQ_CHICKEN_MBCUNIT_CONFIG,
+                        I915_READ(GEN7_SQ_CHICKEN_MBCUNIT_CONFIG) |
+                        GEN7_SQ_CHICKEN_MBCUNIT_SQINTMOB);
+
+        for_each_pipe(pipe) {
+                I915_WRITE(DSPCNTR(pipe),
+                           I915_READ(DSPCNTR(pipe)) |
+                           DISPPLANE_TRICKLE_FEED_DISABLE);
+                intel_flush_display_plane(dev_priv, pipe);
+        }
+
+        I915_WRITE(GEN6_MBCTL, I915_READ(GEN6_MBCTL) |
+                   GEN6_MBCTL_ENABLE_BOOT_FETCH);
+
+        gen7_setup_fixed_func_scheduler(dev_priv);
+
+        /* WaDisable4x2SubspanOptimization */
+        I915_WRITE(CACHE_MODE_1,
+                   _MASKED_BIT_ENABLE(PIXEL_SUBSPAN_COLLECT_OPT_DISABLE));
+
+        snpcr = I915_READ(GEN6_MBCUNIT_SNPCR);
+        snpcr &= ~GEN6_MBC_SNPCR_MASK;
+        snpcr |= GEN6_MBC_SNPCR_MED;
+        I915_WRITE(GEN6_MBCUNIT_SNPCR, snpcr);
+}
+
+static void valleyview_init_clock_gating(struct drm_device *dev)
+{
+        struct drm_i915_private *dev_priv = dev->dev_private;
+        int pipe;
+        uint32_t dspclk_gate = VRHUNIT_CLOCK_GATE_DISABLE;
+
+        I915_WRITE(PCH_DSPCLK_GATE_D, dspclk_gate);
+
+        I915_WRITE(WM3_LP_ILK, 0);
+        I915_WRITE(WM2_LP_ILK, 0);
+        I915_WRITE(WM1_LP_ILK, 0);
 
         I915_WRITE(ILK_DSPCLK_GATE, IVB_VRHUNIT_CLK_GATE);
 
@@ -3465,6 +3581,35 @@ static void valleyview_init_clock_gating(struct drm_device *dev)
                    I915_READ(GEN7_SQ_CHICKEN_MBCUNIT_CONFIG) |
                    GEN7_SQ_CHICKEN_MBCUNIT_SQINTMOB);
 
+        I915_WRITE(GEN6_MBCTL, I915_READ(GEN6_MBCTL) |
+                   GEN6_MBCTL_ENABLE_BOOT_FETCH);
+
+
+        /* According to the BSpec vol1g, bit 12 (RCPBUNIT) clock
+         * gating disable must be set.  Failure to set it results in
+         * flickering pixels due to Z write ordering failures after
+         * some amount of runtime in the Mesa "fire" demo, and Unigine
+         * Sanctuary and Tropics, and apparently anything else with
+         * alpha test or pixel discard.
+         *
+         * According to the spec, bit 11 (RCCUNIT) must also be set,
+         * but we didn't debug actual testcases to find it out.
+         *
+         * According to the spec, bit 13 (RCZUNIT) must be set on IVB.
+         * This implements the WaDisableRCZUnitClockGating workaround.
+         *
+         * Also apply WaDisableVDSUnitClockGating and
+         * WaDisableRCPBUnitClockGating.
+         */
+        I915_WRITE(GEN6_UCGCTL2,
+                   GEN7_VDSUNIT_CLOCK_GATE_DISABLE |
+                   GEN7_TDLUNIT_CLOCK_GATE_DISABLE |
+                   GEN6_RCZUNIT_CLOCK_GATE_DISABLE |
+                   GEN6_RCPBUNIT_CLOCK_GATE_DISABLE |
+                   GEN6_RCCUNIT_CLOCK_GATE_DISABLE);
+
+        I915_WRITE(GEN7_UCGCTL4, GEN7_L3BANK2X_CLOCK_GATE_DISABLE);
+
         for_each_pipe(pipe) {
                 I915_WRITE(DSPCNTR(pipe),
                            I915_READ(DSPCNTR(pipe)) |
@@ -3474,6 +3619,19 @@ static void valleyview_init_clock_gating(struct drm_device *dev)
 
         I915_WRITE(CACHE_MODE_1,
                    _MASKED_BIT_ENABLE(PIXEL_SUBSPAN_COLLECT_OPT_DISABLE));
+
+        /*
+         * On ValleyView, the GUnit needs to signal the GT
+         * when flip and other events complete.  So enable
+         * all the GUnit->GT interrupts here
+         */
+        I915_WRITE(VLV_DPFLIPSTAT, PIPEB_LINE_COMPARE_INT_EN |
+                   PIPEB_HLINE_INT_EN | PIPEB_VBLANK_INT_EN |
+                   SPRITED_FLIPDONE_INT_EN | SPRITEC_FLIPDONE_INT_EN |
+                   PLANEB_FLIPDONE_INT_EN | PIPEA_LINE_COMPARE_INT_EN |
+                   PIPEA_HLINE_INT_EN | PIPEA_VBLANK_INT_EN |
+                   SPRITEB_FLIPDONE_INT_EN | SPRITEA_FLIPDONE_INT_EN |
+                   PLANEA_FLIPDONE_INT_EN);
 }
 
 static void g4x_init_clock_gating(struct drm_device *dev)
@@ -3681,34 +3839,6 @@ void intel_init_pm(struct drm_device *dev)
 
         /* For FIFO watermark updates */
         if (HAS_PCH_SPLIT(dev)) {
-                dev_priv->display.force_wake_get = __gen6_gt_force_wake_get;
-                dev_priv->display.force_wake_put = __gen6_gt_force_wake_put;
-
-                /* IVB configs may use multi-threaded forcewake */
-                if (IS_IVYBRIDGE(dev) || IS_HASWELL(dev)) {
-                        u32     ecobus;
-
-                        /* A small trick here - if the bios hasn't configured MT forcewake,
-                         * and if the device is in RC6, then force_wake_mt_get will not wake
-                         * the device and the ECOBUS read will return zero. Which will be
-                         * (correctly) interpreted by the test below as MT forcewake being
-                         * disabled.
-                         */
-                        mutex_lock(&dev->struct_mutex);
-                        __gen6_gt_force_wake_mt_get(dev_priv);
-                        ecobus = I915_READ_NOTRACE(ECOBUS);
-                        __gen6_gt_force_wake_mt_put(dev_priv);
-                        mutex_unlock(&dev->struct_mutex);
-
-                        if (ecobus & FORCEWAKE_MT_ENABLE) {
-                                DRM_DEBUG_KMS("Using MT version of forcewake\n");
-                                dev_priv->display.force_wake_get =
-                                        __gen6_gt_force_wake_mt_get;
-                                dev_priv->display.force_wake_put =
-                                        __gen6_gt_force_wake_mt_put;
-                        }
-                }
-
                 if (HAS_PCH_IBX(dev))
                         dev_priv->display.init_pch_clock_gating = ibx_init_clock_gating;
                 else if (HAS_PCH_CPT(dev))
@@ -3756,7 +3886,7 @@ void intel_init_pm(struct drm_device *dev)
                                               "Disable CxSR\n");
                                 dev_priv->display.update_wm = NULL;
                         }
-                        dev_priv->display.init_clock_gating = ivybridge_init_clock_gating;
+                        dev_priv->display.init_clock_gating = haswell_init_clock_gating;
                         dev_priv->display.sanitize_pm = gen6_sanitize_pm;
                 } else
                         dev_priv->display.update_wm = NULL;
@@ -3764,8 +3894,6 @@ void intel_init_pm(struct drm_device *dev)
                 dev_priv->display.update_wm = valleyview_update_wm;
                 dev_priv->display.init_clock_gating =
                         valleyview_init_clock_gating;
-                dev_priv->display.force_wake_get = vlv_force_wake_get;
-                dev_priv->display.force_wake_put = vlv_force_wake_put;
         } else if (IS_PINEVIEW(dev)) {
                 if (!intel_get_cxsr_latency(IS_PINEVIEW_G(dev),
                                             dev_priv->is_ddr3,
@@ -3818,3 +3946,194 @@ void intel_init_pm(struct drm_device *dev)
         intel_init_power_wells(dev);
 }
 
+static void __gen6_gt_wait_for_thread_c0(struct drm_i915_private *dev_priv)
+{
+        u32 gt_thread_status_mask;
+
+        if (IS_HASWELL(dev_priv->dev))
+                gt_thread_status_mask = GEN6_GT_THREAD_STATUS_CORE_MASK_HSW;
+        else
+                gt_thread_status_mask = GEN6_GT_THREAD_STATUS_CORE_MASK;
+
+        /* w/a for a sporadic read returning 0 by waiting for the GT
+         * thread to wake up.
+         */
+        if (wait_for_atomic_us((I915_READ_NOTRACE(GEN6_GT_THREAD_STATUS_REG) & gt_thread_status_mask) == 0, 500))
+                DRM_ERROR("GT thread status wait timed out\n");
+}
+
+static void __gen6_gt_force_wake_get(struct drm_i915_private *dev_priv)
+{
+        u32 forcewake_ack;
+
+        if (IS_HASWELL(dev_priv->dev))
+                forcewake_ack = FORCEWAKE_ACK_HSW;
+        else
+                forcewake_ack = FORCEWAKE_ACK;
+
+        if (wait_for_atomic_us((I915_READ_NOTRACE(forcewake_ack) & 1) == 0, 500))
+                DRM_ERROR("Force wake wait timed out\n");
+
+        I915_WRITE_NOTRACE(FORCEWAKE, 1);
+
+        if (wait_for_atomic_us((I915_READ_NOTRACE(forcewake_ack) & 1), 500))
+                DRM_ERROR("Force wake wait timed out\n");
+
+        __gen6_gt_wait_for_thread_c0(dev_priv);
+}
+
+static void __gen6_gt_force_wake_mt_get(struct drm_i915_private *dev_priv)
+{
+        u32 forcewake_ack;
+
+        if (IS_HASWELL(dev_priv->dev))
+                forcewake_ack = FORCEWAKE_ACK_HSW;
+        else
+                forcewake_ack = FORCEWAKE_MT_ACK;
+
+        if (wait_for_atomic_us((I915_READ_NOTRACE(forcewake_ack) & 1) == 0, 500))
+                DRM_ERROR("Force wake wait timed out\n");
+
+        I915_WRITE_NOTRACE(FORCEWAKE_MT, _MASKED_BIT_ENABLE(1));
+
+        if (wait_for_atomic_us((I915_READ_NOTRACE(forcewake_ack) & 1), 500))
+                DRM_ERROR("Force wake wait timed out\n");
+
+        __gen6_gt_wait_for_thread_c0(dev_priv);
+}
+
+/*
+ * Generally this is called implicitly by the register read function. However,
+ * if some sequence requires the GT to not power down then this function should
+ * be called at the beginning of the sequence followed by a call to
+ * gen6_gt_force_wake_put() at the end of the sequence.
+ */
+void gen6_gt_force_wake_get(struct drm_i915_private *dev_priv)
+{
+        unsigned long irqflags;
+
+        spin_lock_irqsave(&dev_priv->gt_lock, irqflags);
+        if (dev_priv->forcewake_count++ == 0)
+                dev_priv->gt.force_wake_get(dev_priv);
+        spin_unlock_irqrestore(&dev_priv->gt_lock, irqflags);
+}
+
+void gen6_gt_check_fifodbg(struct drm_i915_private *dev_priv)
+{
+        u32 gtfifodbg;
+        gtfifodbg = I915_READ_NOTRACE(GTFIFODBG);
+        if (WARN(gtfifodbg & GT_FIFO_CPU_ERROR_MASK,
+             "MMIO read or write has been dropped %x\n", gtfifodbg))
+                I915_WRITE_NOTRACE(GTFIFODBG, GT_FIFO_CPU_ERROR_MASK);
+}
+
+static void __gen6_gt_force_wake_put(struct drm_i915_private *dev_priv)
+{
+        I915_WRITE_NOTRACE(FORCEWAKE, 0);
+        /* The below doubles as a POSTING_READ */
+        gen6_gt_check_fifodbg(dev_priv);
+}
+
+static void __gen6_gt_force_wake_mt_put(struct drm_i915_private *dev_priv)
+{
+        I915_WRITE_NOTRACE(FORCEWAKE_MT, _MASKED_BIT_DISABLE(1));
+        /* The below doubles as a POSTING_READ */
+        gen6_gt_check_fifodbg(dev_priv);
+}
+
+/*
+ * see gen6_gt_force_wake_get()
+ */
+void gen6_gt_force_wake_put(struct drm_i915_private *dev_priv)
+{
+        unsigned long irqflags;
+
+        spin_lock_irqsave(&dev_priv->gt_lock, irqflags);
+        if (--dev_priv->forcewake_count == 0)
+                dev_priv->gt.force_wake_put(dev_priv);
+        spin_unlock_irqrestore(&dev_priv->gt_lock, irqflags);
+}
+
+int __gen6_gt_wait_for_fifo(struct drm_i915_private *dev_priv)
+{
+        int ret = 0;
+
+        if (dev_priv->gt_fifo_count < GT_FIFO_NUM_RESERVED_ENTRIES) {
+                int loop = 500;
+                u32 fifo = I915_READ_NOTRACE(GT_FIFO_FREE_ENTRIES);
+                while (fifo <= GT_FIFO_NUM_RESERVED_ENTRIES && loop--) {
+                        udelay(10);
+                        fifo = I915_READ_NOTRACE(GT_FIFO_FREE_ENTRIES);
+                }
+                if (WARN_ON(loop < 0 && fifo <= GT_FIFO_NUM_RESERVED_ENTRIES))
+                        ++ret;
+                dev_priv->gt_fifo_count = fifo;
+        }
+        dev_priv->gt_fifo_count--;
+
+        return ret;
+}
+
+static void vlv_force_wake_get(struct drm_i915_private *dev_priv)
+{
+        /* Already awake? */
+        if ((I915_READ(0x130094) & 0xa1) == 0xa1)
+                return;
+
+        I915_WRITE_NOTRACE(FORCEWAKE_VLV, 0xffffffff);
+        POSTING_READ(FORCEWAKE_VLV);
+
+        if (wait_for_atomic_us((I915_READ_NOTRACE(FORCEWAKE_ACK_VLV) & 1), 500))
+                DRM_ERROR("Force wake wait timed out\n");
+
+        __gen6_gt_wait_for_thread_c0(dev_priv);
+}
+
+static void vlv_force_wake_put(struct drm_i915_private *dev_priv)
+{
+        I915_WRITE_NOTRACE(FORCEWAKE_VLV, 0xffff0000);
+        /* FIXME: confirm VLV behavior with Punit folks */
+        POSTING_READ(FORCEWAKE_VLV);
+}
+
+void intel_gt_init(struct drm_device *dev)
+{
+        struct drm_i915_private *dev_priv = dev->dev_private;
+
+        spin_lock_init(&dev_priv->gt_lock);
+
+        if (IS_VALLEYVIEW(dev)) {
+                dev_priv->gt.force_wake_get = vlv_force_wake_get;
+                dev_priv->gt.force_wake_put = vlv_force_wake_put;
+        } else if (INTEL_INFO(dev)->gen >= 6) {
+                dev_priv->gt.force_wake_get = __gen6_gt_force_wake_get;
+                dev_priv->gt.force_wake_put = __gen6_gt_force_wake_put;
+
+                /* IVB configs may use multi-threaded forcewake */
+                if (IS_IVYBRIDGE(dev) || IS_HASWELL(dev)) {
+                        u32 ecobus;
+
+                        /* A small trick here - if the bios hasn't configured
+                         * MT forcewake, and if the device is in RC6, then
+                         * force_wake_mt_get will not wake the device and the
+                         * ECOBUS read will return zero. Which will be
+                         * (correctly) interpreted by the test below as MT
+                         * forcewake being disabled.
+                         */
+                        mutex_lock(&dev->struct_mutex);
+                        __gen6_gt_force_wake_mt_get(dev_priv);
+                        ecobus = I915_READ_NOTRACE(ECOBUS);
+                        __gen6_gt_force_wake_mt_put(dev_priv);
+                        mutex_unlock(&dev->struct_mutex);
+
+                        if (ecobus & FORCEWAKE_MT_ENABLE) {
+                                DRM_DEBUG_KMS("Using MT version of forcewake\n");
+                                dev_priv->gt.force_wake_get =
+                                        __gen6_gt_force_wake_mt_get;
+                                dev_priv->gt.force_wake_put =
+                                        __gen6_gt_force_wake_mt_put;
+                        }
+                }
+        }
+}
+
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index e5b84ff89ca5..d42d821c64d6 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -219,19 +219,28 @@ gen6_render_ring_flush(struct intel_ring_buffer *ring,
         int ret;
 
         /* Force SNB workarounds for PIPE_CONTROL flushes */
-        intel_emit_post_sync_nonzero_flush(ring);
+        ret = intel_emit_post_sync_nonzero_flush(ring);
+        if (ret)
+                return ret;
 
         /* Just flush everything.  Experiments have shown that reducing the
          * number of bits based on the write domains has little performance
          * impact.
          */
         flags |= PIPE_CONTROL_RENDER_TARGET_CACHE_FLUSH;
+        flags |= PIPE_CONTROL_TLB_INVALIDATE;
         flags |= PIPE_CONTROL_INSTRUCTION_CACHE_INVALIDATE;
         flags |= PIPE_CONTROL_TEXTURE_CACHE_INVALIDATE;
         flags |= PIPE_CONTROL_DEPTH_CACHE_FLUSH;
         flags |= PIPE_CONTROL_VF_CACHE_INVALIDATE;
         flags |= PIPE_CONTROL_CONST_CACHE_INVALIDATE;
         flags |= PIPE_CONTROL_STATE_CACHE_INVALIDATE;
+        /*
+         * Ensure that any following seqno writes only happen when the render
+         * cache is indeed flushed (but only if the caller actually wants that).
+         */
+        if (flush_domains)
+                flags |= PIPE_CONTROL_CS_STALL;
 
         ret = intel_ring_begin(ring, 6);
         if (ret)
@@ -433,11 +442,21 @@ static int init_render_ring(struct intel_ring_buffer *ring)
                  */
                 I915_WRITE(CACHE_MODE_0,
                            _MASKED_BIT_DISABLE(CM0_STC_EVICT_DISABLE_LRA_SNB));
+
+                /* This is not explicitly set for GEN6, so read the register.
+                 * see intel_ring_mi_set_context() for why we care.
+                 * TODO: consider explicitly setting the bit for GEN5
+                 */
+                ring->itlb_before_ctx_switch =
+                        !!(I915_READ(GFX_MODE) & GFX_TLB_INVALIDATE_ALWAYS);
         }
 
         if (INTEL_INFO(dev)->gen >= 6)
                 I915_WRITE(INSTPM, _MASKED_BIT_ENABLE(INSTPM_FORCE_ORDERING));
 
+        if (IS_IVYBRIDGE(dev))
+                I915_WRITE_IMR(ring, ~GEN6_RENDER_L3_PARITY_ERROR);
+
         return ret;
 }
 
@@ -825,7 +844,11 @@ gen6_ring_get_irq(struct intel_ring_buffer *ring)
 
         spin_lock_irqsave(&dev_priv->irq_lock, flags);
         if (ring->irq_refcount++ == 0) {
-                I915_WRITE_IMR(ring, ~ring->irq_enable_mask);
+                if (IS_IVYBRIDGE(dev) && ring->id == RCS)
+                        I915_WRITE_IMR(ring, ~(ring->irq_enable_mask |
+                                                GEN6_RENDER_L3_PARITY_ERROR));
+                else
+                        I915_WRITE_IMR(ring, ~ring->irq_enable_mask);
                 dev_priv->gt_irq_mask &= ~ring->irq_enable_mask;
                 I915_WRITE(GTIMR, dev_priv->gt_irq_mask);
                 POSTING_READ(GTIMR);
@@ -844,7 +867,10 @@ gen6_ring_put_irq(struct intel_ring_buffer *ring)
 
         spin_lock_irqsave(&dev_priv->irq_lock, flags);
         if (--ring->irq_refcount == 0) {
-                I915_WRITE_IMR(ring, ~0);
+                if (IS_IVYBRIDGE(dev) && ring->id == RCS)
+                        I915_WRITE_IMR(ring, ~GEN6_RENDER_L3_PARITY_ERROR);
+                else
+                        I915_WRITE_IMR(ring, ~0);
                 dev_priv->gt_irq_mask |= ring->irq_enable_mask;
                 I915_WRITE(GTIMR, dev_priv->gt_irq_mask);
                 POSTING_READ(GTIMR);
@@ -969,6 +995,7 @@ static int intel_init_ring_buffer(struct drm_device *dev,
                                   struct intel_ring_buffer *ring)
 {
         struct drm_i915_gem_object *obj;
+        struct drm_i915_private *dev_priv = dev->dev_private;
         int ret;
 
         ring->dev = dev;
@@ -1002,8 +1029,9 @@ static int intel_init_ring_buffer(struct drm_device *dev,
         if (ret)
                 goto err_unpin;
 
-        ring->virtual_start = ioremap_wc(dev->agp->base + obj->gtt_offset,
-                                         ring->size);
+        ring->virtual_start =
+                ioremap_wc(dev_priv->mm.gtt->gma_bus_addr + obj->gtt_offset,
+                           ring->size);
         if (ring->virtual_start == NULL) {
                 DRM_ERROR("Failed to map ringbuffer.\n");
                 ret = -EINVAL;
@@ -1089,20 +1117,9 @@ static int intel_wrap_ring_buffer(struct intel_ring_buffer *ring)
 
 static int intel_ring_wait_seqno(struct intel_ring_buffer *ring, u32 seqno)
 {
-        struct drm_i915_private *dev_priv = ring->dev->dev_private;
-        bool was_interruptible;
         int ret;
 
-        /* XXX As we have not yet audited all the paths to check that
-         * they are ready for ERESTARTSYS from intel_ring_begin, do not
-         * allow us to be interruptible by a signal.
-         */
-        was_interruptible = dev_priv->mm.interruptible;
-        dev_priv->mm.interruptible = false;
-
-        ret = i915_wait_request(ring, seqno);
-
-        dev_priv->mm.interruptible = was_interruptible;
+        ret = i915_wait_seqno(ring, seqno);
         if (!ret)
                 i915_gem_retire_requests_ring(ring);
 
@@ -1200,8 +1217,10 @@ int intel_wait_ring_buffer(struct intel_ring_buffer *ring, int n)
                 }
 
                 msleep(1);
-                if (atomic_read(&dev_priv->mm.wedged))
-                        return -EAGAIN;
+
+                ret = i915_gem_check_wedge(dev_priv, dev_priv->mm.interruptible);
+                if (ret)
+                        return ret;
         } while (!time_after(jiffies, end));
         trace_i915_ring_wait_end(ring);
         return -EBUSY;
@@ -1210,12 +1229,13 @@ int intel_wait_ring_buffer(struct intel_ring_buffer *ring, int n)
 int intel_ring_begin(struct intel_ring_buffer *ring,
                      int num_dwords)
 {
-        struct drm_i915_private *dev_priv = ring->dev->dev_private;
+        drm_i915_private_t *dev_priv = ring->dev->dev_private;
         int n = 4*num_dwords;
         int ret;
 
-        if (unlikely(atomic_read(&dev_priv->mm.wedged)))
-                return -EIO;
+        ret = i915_gem_check_wedge(dev_priv, dev_priv->mm.interruptible);
+        if (ret)
+                return ret;
 
         if (unlikely(ring->tail + n > ring->effective_size)) {
                 ret = intel_wrap_ring_buffer(ring);
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h
index 55d3da26bae7..1d3c81fdad92 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.h
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.h
@@ -113,9 +113,17 @@ struct  intel_ring_buffer {
          * Do we have some not yet emitted requests outstanding?
          */
         u32 outstanding_lazy_request;
+        bool gpu_caches_dirty;
 
         wait_queue_head_t irq_queue;
 
+        /**
+         * Do an explicit TLB flush before MI_SET_CONTEXT
+         */
+        bool itlb_before_ctx_switch;
+        struct i915_hw_context *default_context;
+        struct drm_i915_gem_object *last_context_obj;
+
         void *private;
 };
 
diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index b6a9d45fc3c6..26a6a4d0d078 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -140,9 +140,6 @@ struct intel_sdvo {
 
         /* DDC bus used by this SDVO encoder */
         uint8_t ddc_bus;
-
-        /* Input timings for adjusted_mode */
-        struct intel_sdvo_dtd input_dtd;
 };
 
 struct intel_sdvo_connector {
@@ -938,7 +935,7 @@ static bool intel_sdvo_set_tv_format(struct intel_sdvo *intel_sdvo)
 
 static bool
 intel_sdvo_set_output_timings_from_mode(struct intel_sdvo *intel_sdvo,
-                                        struct drm_display_mode *mode)
+                                        const struct drm_display_mode *mode)
 {
         struct intel_sdvo_dtd output_dtd;
 
@@ -953,11 +950,15 @@ intel_sdvo_set_output_timings_from_mode(struct intel_sdvo *intel_sdvo,
         return true;
 }
 
+/* Asks the sdvo controller for the preferred input mode given the output mode.
+ * Unfortunately we have to set up the full output mode to do that. */
 static bool
-intel_sdvo_set_input_timings_for_mode(struct intel_sdvo *intel_sdvo,
-                                        struct drm_display_mode *mode,
-                                        struct drm_display_mode *adjusted_mode)
+intel_sdvo_get_preferred_input_mode(struct intel_sdvo *intel_sdvo,
+                                    const struct drm_display_mode *mode,
+                                    struct drm_display_mode *adjusted_mode)
 {
+        struct intel_sdvo_dtd input_dtd;
+
         /* Reset the input timing to the screen. Assume always input 0. */
         if (!intel_sdvo_set_target_input(intel_sdvo))
                 return false;
@@ -969,16 +970,16 @@ intel_sdvo_set_input_timings_for_mode(struct intel_sdvo *intel_sdvo,
                 return false;
 
         if (!intel_sdvo_get_preferred_input_timing(intel_sdvo,
-                                                   &intel_sdvo->input_dtd))
+                                                   &input_dtd))
                 return false;
 
-        intel_sdvo_get_mode_from_dtd(adjusted_mode, &intel_sdvo->input_dtd);
+        intel_sdvo_get_mode_from_dtd(adjusted_mode, &input_dtd);
 
         return true;
 }
 
 static bool intel_sdvo_mode_fixup(struct drm_encoder *encoder,
-                                  struct drm_display_mode *mode,
+                                  const struct drm_display_mode *mode,
                                   struct drm_display_mode *adjusted_mode)
 {
         struct intel_sdvo *intel_sdvo = to_intel_sdvo(encoder);
@@ -993,17 +994,17 @@ static bool intel_sdvo_mode_fixup(struct drm_encoder *encoder,
                 if (!intel_sdvo_set_output_timings_from_mode(intel_sdvo, mode))
                         return false;
 
-                (void) intel_sdvo_set_input_timings_for_mode(intel_sdvo,
-                                                             mode,
-                                                             adjusted_mode);
+                (void) intel_sdvo_get_preferred_input_mode(intel_sdvo,
+                                                           mode,
+                                                           adjusted_mode);
         } else if (intel_sdvo->is_lvds) {
                 if (!intel_sdvo_set_output_timings_from_mode(intel_sdvo,
                                                              intel_sdvo->sdvo_lvds_fixed_mode))
                         return false;
 
-                (void) intel_sdvo_set_input_timings_for_mode(intel_sdvo,
-                                                             mode,
-                                                             adjusted_mode);
+                (void) intel_sdvo_get_preferred_input_mode(intel_sdvo,
+                                                           mode,
+                                                           adjusted_mode);
         }
 
         /* Make the CRTC code factor in the SDVO pixel multiplier.  The
@@ -1057,7 +1058,9 @@ static void intel_sdvo_mode_set(struct drm_encoder *encoder,
                                              intel_sdvo->sdvo_lvds_fixed_mode);
         else
                 intel_sdvo_get_dtd_from_mode(&output_dtd, mode);
-        (void) intel_sdvo_set_output_timing(intel_sdvo, &output_dtd);
+        if (!intel_sdvo_set_output_timing(intel_sdvo, &output_dtd))
+                DRM_INFO("Setting output timings on %s failed\n",
+                         SDVO_NAME(intel_sdvo));
 
         /* Set the input timing to the screen. Assume always input 0. */
         if (!intel_sdvo_set_target_input(intel_sdvo))
@@ -1079,7 +1082,9 @@ static void intel_sdvo_mode_set(struct drm_encoder *encoder,
          * adjusted_mode.
          */
         intel_sdvo_get_dtd_from_mode(&input_dtd, adjusted_mode);
-        (void) intel_sdvo_set_input_timing(intel_sdvo, &input_dtd);
+        if (!intel_sdvo_set_input_timing(intel_sdvo, &input_dtd))
+                DRM_INFO("Setting input timings on %s failed\n",
+                         SDVO_NAME(intel_sdvo));
 
         switch (pixel_multiplier) {
         default:
@@ -1376,7 +1381,7 @@ intel_sdvo_detect(struct drm_connector *connector, bool force)
 
         /* add 30ms delay when the output type might be TV */
         if (intel_sdvo->caps.output_flags & SDVO_TV_MASK)
-                mdelay(30);
+                msleep(30);
 
         if (!intel_sdvo_read_response(intel_sdvo, &response, 2))
                 return connector_status_unknown;
@@ -2521,6 +2526,7 @@ bool intel_sdvo_init(struct drm_device *dev, uint32_t sdvo_reg, bool is_sdvob)
         struct drm_i915_private *dev_priv = dev->dev_private;
         struct intel_encoder *intel_encoder;
         struct intel_sdvo *intel_sdvo;
+        u32 hotplug_mask;
         int i;
 
         intel_sdvo = kzalloc(sizeof(struct intel_sdvo), GFP_KERNEL);
@@ -2552,10 +2558,18 @@ bool intel_sdvo_init(struct drm_device *dev, uint32_t sdvo_reg, bool is_sdvob)
                 }
         }
 
-        if (intel_sdvo->is_sdvob)
-                dev_priv->hotplug_supported_mask |= SDVOB_HOTPLUG_INT_STATUS;
-        else
-                dev_priv->hotplug_supported_mask |= SDVOC_HOTPLUG_INT_STATUS;
+        hotplug_mask = 0;
+        if (IS_G4X(dev)) {
+                hotplug_mask = intel_sdvo->is_sdvob ?
+                        SDVOB_HOTPLUG_INT_STATUS_G4X : SDVOC_HOTPLUG_INT_STATUS_G4X;
+        } else if (IS_GEN4(dev)) {
+                hotplug_mask = intel_sdvo->is_sdvob ?
+                        SDVOB_HOTPLUG_INT_STATUS_I965 : SDVOC_HOTPLUG_INT_STATUS_I965;
+        } else {
+                hotplug_mask = intel_sdvo->is_sdvob ?
+                        SDVOB_HOTPLUG_INT_STATUS_I915 : SDVOC_HOTPLUG_INT_STATUS_I915;
+        }
+        dev_priv->hotplug_supported_mask |= hotplug_mask;
 
         drm_encoder_helper_add(&intel_encoder->base, &intel_sdvo_helper_funcs);
 
diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
index 2a20fb0781d7..1a1483b924d0 100644
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -56,6 +56,7 @@ ivb_update_plane(struct drm_plane *plane, struct drm_framebuffer *fb,
         sprctl &= ~SPRITE_PIXFORMAT_MASK;
         sprctl &= ~SPRITE_RGB_ORDER_RGBX;
         sprctl &= ~SPRITE_YUV_BYTE_ORDER_MASK;
+        sprctl &= ~SPRITE_TILED;
 
         switch (fb->pixel_format) {
         case DRM_FORMAT_XBGR8888:
@@ -84,7 +85,7 @@ ivb_update_plane(struct drm_plane *plane, struct drm_framebuffer *fb,
                 break;
         default:
                 DRM_DEBUG_DRIVER("bad pixel format, assuming RGBX888\n");
-                sprctl |= DVS_FORMAT_RGBX888;
+                sprctl |= SPRITE_FORMAT_RGBX888;
                 pixel_size = 4;
                 break;
         }
@@ -326,6 +327,12 @@ intel_enable_primary(struct drm_crtc *crtc)
         struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
         int reg = DSPCNTR(intel_crtc->plane);
 
+        if (!intel_crtc->primary_disabled)
+                return;
+
+        intel_crtc->primary_disabled = false;
+        intel_update_fbc(dev);
+
         I915_WRITE(reg, I915_READ(reg) | DISPLAY_PLANE_ENABLE);
 }
 
@@ -337,7 +344,13 @@ intel_disable_primary(struct drm_crtc *crtc)
         struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
         int reg = DSPCNTR(intel_crtc->plane);
 
+        if (intel_crtc->primary_disabled)
+                return;
+
         I915_WRITE(reg, I915_READ(reg) & ~DISPLAY_PLANE_ENABLE);
+
+        intel_crtc->primary_disabled = true;
+        intel_update_fbc(dev);
 }
 
 static int
@@ -485,18 +498,14 @@ intel_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
          * Be sure to re-enable the primary before the sprite is no longer
          * covering it fully.
          */
-        if (!disable_primary && intel_plane->primary_disabled) {
+        if (!disable_primary)
                 intel_enable_primary(crtc);
-                intel_plane->primary_disabled = false;
-        }
 
         intel_plane->update_plane(plane, fb, obj, crtc_x, crtc_y,
                                   crtc_w, crtc_h, x, y, src_w, src_h);
 
-        if (disable_primary) {
+        if (disable_primary)
                 intel_disable_primary(crtc);
-                intel_plane->primary_disabled = true;
-        }
 
         /* Unpin old obj after new one is active to avoid ugliness */
         if (old_obj) {
@@ -527,11 +536,8 @@ intel_disable_plane(struct drm_plane *plane)
         struct intel_plane *intel_plane = to_intel_plane(plane);
         int ret = 0;
 
-        if (intel_plane->primary_disabled) {
+        if (plane->crtc)
                 intel_enable_primary(plane->crtc);
-                intel_plane->primary_disabled = false;
-        }
-
         intel_plane->disable_plane(plane);
 
         if (!intel_plane->obj)
@@ -685,6 +691,7 @@ intel_plane_init(struct drm_device *dev, enum pipe pipe)
                 break;
 
         default:
+                kfree(intel_plane);
                 return -ENODEV;
         }
 
@@ -699,4 +706,3 @@ intel_plane_init(struct drm_device *dev, enum pipe pipe)
 
         return ret;
 }
-
diff --git a/drivers/gpu/drm/i915/intel_tv.c b/drivers/gpu/drm/i915/intel_tv.c
index a233a51fd7e6..befce6c49704 100644
--- a/drivers/gpu/drm/i915/intel_tv.c
+++ b/drivers/gpu/drm/i915/intel_tv.c
@@ -891,24 +891,21 @@ intel_tv_mode_valid(struct drm_connector *connector,
 
 
 static bool
-intel_tv_mode_fixup(struct drm_encoder *encoder, struct drm_display_mode *mode,
+intel_tv_mode_fixup(struct drm_encoder *encoder,
+                    const struct drm_display_mode *mode,
                     struct drm_display_mode *adjusted_mode)
 {
         struct drm_device *dev = encoder->dev;
-        struct drm_mode_config *drm_config = &dev->mode_config;
         struct intel_tv *intel_tv = enc_to_intel_tv(encoder);
         const struct tv_mode *tv_mode = intel_tv_mode_find(intel_tv);
-        struct drm_encoder *other_encoder;
+        struct intel_encoder *other_encoder;
 
         if (!tv_mode)
                 return false;
 
-        /* FIXME: lock encoder list */
-        list_for_each_entry(other_encoder, &drm_config->encoder_list, head) {
-                if (other_encoder != encoder &&
-                    other_encoder->crtc == encoder->crtc)
+        for_each_encoder_on_crtc(dev, encoder->crtc, other_encoder)
+                if (&other_encoder->base != encoder)
                         return false;
-        }
 
         adjusted_mode->clock = tv_mode->clock;
         return true;
diff --git a/drivers/gpu/drm/mga/mga_drv.c b/drivers/gpu/drm/mga/mga_drv.c
index f9a925d58819..b1bb46de3f5a 100644
--- a/drivers/gpu/drm/mga/mga_drv.c
+++ b/drivers/gpu/drm/mga/mga_drv.c
@@ -75,7 +75,6 @@ static struct drm_driver driver = {
         .irq_postinstall = mga_driver_irq_postinstall,
         .irq_uninstall = mga_driver_irq_uninstall,
         .irq_handler = mga_driver_irq_handler,
-        .reclaim_buffers = drm_core_reclaim_buffers,
         .ioctls = mga_ioctls,
         .dma_ioctl = mga_dma_buffers,
         .fops = &mga_driver_fops,
diff --git a/drivers/gpu/drm/mgag200/mgag200_drv.c b/drivers/gpu/drm/mgag200/mgag200_drv.c
index 93e832d6c328..ea1024d79974 100644
--- a/drivers/gpu/drm/mgag200/mgag200_drv.c
+++ b/drivers/gpu/drm/mgag200/mgag200_drv.c
@@ -47,6 +47,9 @@ static void mgag200_kick_out_firmware_fb(struct pci_dev *pdev)
         bool primary = false;
 
         ap = alloc_apertures(1);
+        if (!ap)
+                return;
+
         ap->ranges[0].base = pci_resource_start(pdev, 0);
         ap->ranges[0].size = pci_resource_len(pdev, 0);
 
diff --git a/drivers/gpu/drm/mgag200/mgag200_mode.c b/drivers/gpu/drm/mgag200/mgag200_mode.c
index d303061b251e..a4d7c500c97b 100644
--- a/drivers/gpu/drm/mgag200/mgag200_mode.c
+++ b/drivers/gpu/drm/mgag200/mgag200_mode.c
@@ -78,8 +78,8 @@ static inline void mga_wait_busy(struct mga_device *mdev)
  * to just pass that straight through, so this does nothing
  */
 static bool mga_crtc_mode_fixup(struct drm_crtc *crtc,
-                                   struct drm_display_mode *mode,
-                                   struct drm_display_mode *adjusted_mode)
+                                const struct drm_display_mode *mode,
+                                struct drm_display_mode *adjusted_mode)
 {
         return true;
 }
@@ -1322,8 +1322,8 @@ void mga_crtc_fb_gamma_get(struct drm_crtc *crtc, u16 *red, u16 *green,
  * to handle any encoder-specific limitations
  */
 static bool mga_encoder_mode_fixup(struct drm_encoder *encoder,
-                                  struct drm_display_mode *mode,
-                                  struct drm_display_mode *adjusted_mode)
+                                   const struct drm_display_mode *mode,
+                                   struct drm_display_mode *adjusted_mode)
 {
         return true;
 }
diff --git a/drivers/gpu/drm/nouveau/nouveau_drv.c b/drivers/gpu/drm/nouveau/nouveau_drv.c
index 727447d296f1..9a36f5f39b06 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drv.c
+++ b/drivers/gpu/drm/nouveau/nouveau_drv.c
@@ -438,7 +438,6 @@ static struct drm_driver driver = {
         .get_vblank_counter = drm_vblank_count,
         .enable_vblank = nouveau_vblank_enable,
         .disable_vblank = nouveau_vblank_disable,
-        .reclaim_buffers = drm_core_reclaim_buffers,
         .ioctls = nouveau_ioctls,
         .fops = &nouveau_driver_fops,
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_gem.c b/drivers/gpu/drm/nouveau/nouveau_gem.c
index 34d0bc596213..af7cfb825716 100644
--- a/drivers/gpu/drm/nouveau/nouveau_gem.c
+++ b/drivers/gpu/drm/nouveau/nouveau_gem.c
@@ -207,8 +207,7 @@ nouveau_gem_ioctl_new(struct drm_device *dev, void *data,
         struct nouveau_bo *nvbo = NULL;
         int ret = 0;
 
-        if (unlikely(dev_priv->ttm.bdev.dev_mapping == NULL))
-                dev_priv->ttm.bdev.dev_mapping = dev_priv->dev->dev_mapping;
+        dev_priv->ttm.bdev.dev_mapping = dev->dev_mapping;
 
         if (!dev_priv->engine.vram.flags_valid(dev, req->info.tile_flags)) {
                 NV_ERROR(dev, "bad page flags: 0x%08x\n", req->info.tile_flags);
diff --git a/drivers/gpu/drm/nouveau/nv04_crtc.c b/drivers/gpu/drm/nouveau/nv04_crtc.c
index 4c31c63e5528..43accc11102f 100644
--- a/drivers/gpu/drm/nouveau/nv04_crtc.c
+++ b/drivers/gpu/drm/nouveau/nv04_crtc.c
@@ -215,7 +215,7 @@ nv_crtc_dpms(struct drm_crtc *crtc, int mode)
 }
 
 static bool
-nv_crtc_mode_fixup(struct drm_crtc *crtc, struct drm_display_mode *mode,
+nv_crtc_mode_fixup(struct drm_crtc *crtc, const struct drm_display_mode *mode,
                    struct drm_display_mode *adjusted_mode)
 {
         return true;
diff --git a/drivers/gpu/drm/nouveau/nv04_dac.c b/drivers/gpu/drm/nouveau/nv04_dac.c
index 8300266ffaea..38f19479417c 100644
--- a/drivers/gpu/drm/nouveau/nv04_dac.c
+++ b/drivers/gpu/drm/nouveau/nv04_dac.c
@@ -332,7 +332,7 @@ nv17_dac_detect(struct drm_encoder *encoder, struct drm_connector *connector)
 }
 
 static bool nv04_dac_mode_fixup(struct drm_encoder *encoder,
-                                struct drm_display_mode *mode,
+                                const struct drm_display_mode *mode,
                                 struct drm_display_mode *adjusted_mode)
 {
         if (nv04_dac_in_use(encoder))
diff --git a/drivers/gpu/drm/nouveau/nv04_dfp.c b/drivers/gpu/drm/nouveau/nv04_dfp.c
index 2258746016f8..c2675623b7cd 100644
--- a/drivers/gpu/drm/nouveau/nv04_dfp.c
+++ b/drivers/gpu/drm/nouveau/nv04_dfp.c
@@ -179,7 +179,7 @@ static struct drm_encoder *get_tmds_slave(struct drm_encoder *encoder)
 }
 
 static bool nv04_dfp_mode_fixup(struct drm_encoder *encoder,
-                                struct drm_display_mode *mode,
+                                const struct drm_display_mode *mode,
                                 struct drm_display_mode *adjusted_mode)
 {
         struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
diff --git a/drivers/gpu/drm/nouveau/nv17_tv.c b/drivers/gpu/drm/nouveau/nv17_tv.c
index 696d7e7dc2a0..67be5db021f5 100644
--- a/drivers/gpu/drm/nouveau/nv17_tv.c
+++ b/drivers/gpu/drm/nouveau/nv17_tv.c
@@ -338,7 +338,7 @@ static int nv17_tv_mode_valid(struct drm_encoder *encoder,
 }
 
 static bool nv17_tv_mode_fixup(struct drm_encoder *encoder,
-                               struct drm_display_mode *mode,
+                               const struct drm_display_mode *mode,
                                struct drm_display_mode *adjusted_mode)
 {
         struct nv17_tv_norm_params *tv_norm = get_tv_norm(encoder);
diff --git a/drivers/gpu/drm/nouveau/nv50_crtc.c b/drivers/gpu/drm/nouveau/nv50_crtc.c
index 97a477b3d52d..22cebd5dd694 100644
--- a/drivers/gpu/drm/nouveau/nv50_crtc.c
+++ b/drivers/gpu/drm/nouveau/nv50_crtc.c
@@ -527,7 +527,7 @@ nv50_crtc_commit(struct drm_crtc *crtc)
 }
 
 static bool
-nv50_crtc_mode_fixup(struct drm_crtc *crtc, struct drm_display_mode *mode,
+nv50_crtc_mode_fixup(struct drm_crtc *crtc, const struct drm_display_mode *mode,
                      struct drm_display_mode *adjusted_mode)
 {
         return true;
diff --git a/drivers/gpu/drm/nouveau/nv50_dac.c b/drivers/gpu/drm/nouveau/nv50_dac.c
index eb216a446b89..2c36a6b92c53 100644
--- a/drivers/gpu/drm/nouveau/nv50_dac.c
+++ b/drivers/gpu/drm/nouveau/nv50_dac.c
@@ -175,7 +175,8 @@ nv50_dac_restore(struct drm_encoder *encoder)
 }
 
 static bool
-nv50_dac_mode_fixup(struct drm_encoder *encoder, struct drm_display_mode *mode,
+nv50_dac_mode_fixup(struct drm_encoder *encoder,
+                    const struct drm_display_mode *mode,
                     struct drm_display_mode *adjusted_mode)
 {
         struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
diff --git a/drivers/gpu/drm/nouveau/nv50_sor.c b/drivers/gpu/drm/nouveau/nv50_sor.c
index a9514eaa74c1..93240bde891b 100644
--- a/drivers/gpu/drm/nouveau/nv50_sor.c
+++ b/drivers/gpu/drm/nouveau/nv50_sor.c
@@ -327,7 +327,8 @@ nv50_sor_restore(struct drm_encoder *encoder)
 }
 
 static bool
-nv50_sor_mode_fixup(struct drm_encoder *encoder, struct drm_display_mode *mode,
+nv50_sor_mode_fixup(struct drm_encoder *encoder,
+                    const struct drm_display_mode *mode,
                     struct drm_display_mode *adjusted_mode)
 {
         struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
diff --git a/drivers/gpu/drm/nouveau/nvd0_display.c b/drivers/gpu/drm/nouveau/nvd0_display.c
index c486d3ce3c2c..d0d60e1e7f95 100644
--- a/drivers/gpu/drm/nouveau/nvd0_display.c
+++ b/drivers/gpu/drm/nouveau/nvd0_display.c
@@ -607,7 +607,7 @@ nvd0_crtc_commit(struct drm_crtc *crtc)
 }
 
 static bool
-nvd0_crtc_mode_fixup(struct drm_crtc *crtc, struct drm_display_mode *mode,
+nvd0_crtc_mode_fixup(struct drm_crtc *crtc, const struct drm_display_mode *mode,
                      struct drm_display_mode *adjusted_mode)
 {
         return true;
@@ -938,7 +938,8 @@ nvd0_dac_dpms(struct drm_encoder *encoder, int mode)
 }
 
 static bool
-nvd0_dac_mode_fixup(struct drm_encoder *encoder, struct drm_display_mode *mode,
+nvd0_dac_mode_fixup(struct drm_encoder *encoder,
+                    const struct drm_display_mode *mode,
                     struct drm_display_mode *adjusted_mode)
 {
         struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
@@ -1377,7 +1378,8 @@ nvd0_sor_dpms(struct drm_encoder *encoder, int mode)
 }
 
 static bool
-nvd0_sor_mode_fixup(struct drm_encoder *encoder, struct drm_display_mode *mode,
+nvd0_sor_mode_fixup(struct drm_encoder *encoder,
+                    const struct drm_display_mode *mode,
                     struct drm_display_mode *adjusted_mode)
 {
         struct nouveau_encoder *nv_encoder = nouveau_encoder(encoder);
diff --git a/drivers/gpu/drm/r128/r128_drv.c b/drivers/gpu/drm/r128/r128_drv.c
index 88718fad5d6d..2666a5308ab9 100644
--- a/drivers/gpu/drm/r128/r128_drv.c
+++ b/drivers/gpu/drm/r128/r128_drv.c
@@ -71,7 +71,6 @@ static struct drm_driver driver = {
         .irq_postinstall = r128_driver_irq_postinstall,
         .irq_uninstall = r128_driver_irq_uninstall,
         .irq_handler = r128_driver_irq_handler,
-        .reclaim_buffers = drm_core_reclaim_buffers,
         .ioctls = r128_ioctls,
         .dma_ioctl = r128_cce_buffers,
         .fops = &r128_driver_fops,
diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c
index 3904d7964a4b..9e6f76fec527 100644
--- a/drivers/gpu/drm/radeon/atombios_crtc.c
+++ b/drivers/gpu/drm/radeon/atombios_crtc.c
@@ -457,22 +457,18 @@ static void atombios_crtc_program_ss(struct radeon_device *rdev,
                 switch (pll_id) {
                 case ATOM_PPLL1:
                         args.v3.ucSpreadSpectrumType |= ATOM_PPLL_SS_TYPE_V3_P1PLL;
-                        args.v3.usSpreadSpectrumAmount = cpu_to_le16(ss->amount);
-                        args.v3.usSpreadSpectrumStep = cpu_to_le16(ss->step);
                         break;
                 case ATOM_PPLL2:
                         args.v3.ucSpreadSpectrumType |= ATOM_PPLL_SS_TYPE_V3_P2PLL;
-                        args.v3.usSpreadSpectrumAmount = cpu_to_le16(ss->amount);
-                        args.v3.usSpreadSpectrumStep = cpu_to_le16(ss->step);
                         break;
                 case ATOM_DCPLL:
                         args.v3.ucSpreadSpectrumType |= ATOM_PPLL_SS_TYPE_V3_DCPLL;
-                        args.v3.usSpreadSpectrumAmount = cpu_to_le16(0);
-                        args.v3.usSpreadSpectrumStep = cpu_to_le16(0);
                         break;
                 case ATOM_PPLL_INVALID:
                         return;
                 }
+                args.v3.usSpreadSpectrumAmount = cpu_to_le16(ss->amount);
+                args.v3.usSpreadSpectrumStep = cpu_to_le16(ss->step);
                 args.v3.ucEnable = enable;
                 if ((ss->percentage == 0) || (ss->type & ATOM_EXTERNAL_SS_MASK) || ASIC_IS_DCE61(rdev))
                         args.v3.ucEnable = ATOM_DISABLE;
@@ -482,22 +478,18 @@ static void atombios_crtc_program_ss(struct radeon_device *rdev,
                 switch (pll_id) {
                 case ATOM_PPLL1:
                         args.v2.ucSpreadSpectrumType |= ATOM_PPLL_SS_TYPE_V2_P1PLL;
-                        args.v2.usSpreadSpectrumAmount = cpu_to_le16(ss->amount);
-                        args.v2.usSpreadSpectrumStep = cpu_to_le16(ss->step);
                         break;
                 case ATOM_PPLL2:
                         args.v2.ucSpreadSpectrumType |= ATOM_PPLL_SS_TYPE_V2_P2PLL;
-                        args.v2.usSpreadSpectrumAmount = cpu_to_le16(ss->amount);
-                        args.v2.usSpreadSpectrumStep = cpu_to_le16(ss->step);
                         break;
                 case ATOM_DCPLL:
                         args.v2.ucSpreadSpectrumType |= ATOM_PPLL_SS_TYPE_V2_DCPLL;
-                        args.v2.usSpreadSpectrumAmount = cpu_to_le16(0);
-                        args.v2.usSpreadSpectrumStep = cpu_to_le16(0);
                         break;
                 case ATOM_PPLL_INVALID:
                         return;
                 }
+                args.v2.usSpreadSpectrumAmount = cpu_to_le16(ss->amount);
+                args.v2.usSpreadSpectrumStep = cpu_to_le16(ss->step);
                 args.v2.ucEnable = enable;
                 if ((ss->percentage == 0) || (ss->type & ATOM_EXTERNAL_SS_MASK) || ASIC_IS_DCE41(rdev))
                         args.v2.ucEnable = ATOM_DISABLE;
@@ -1539,7 +1531,11 @@ static int radeon_atom_pick_pll(struct drm_crtc *crtc)
                                  * crtc virtual pixel clock.
                                  */
                                 if (ENCODER_MODE_IS_DP(atombios_get_encoder_mode(test_encoder))) {
-                                        if (ASIC_IS_DCE5(rdev) || rdev->clock.dp_extclk)
+                                        if (ASIC_IS_DCE5(rdev))
+                                                return ATOM_DCPLL;
+                                        else if (ASIC_IS_DCE6(rdev))
+                                                return ATOM_PPLL0;
+                                        else if (rdev->clock.dp_extclk)
                                                 return ATOM_PPLL_INVALID;
                                 }
                         }
@@ -1628,7 +1624,7 @@ int atombios_crtc_mode_set(struct drm_crtc *crtc,
 }
 
 static bool atombios_crtc_mode_fixup(struct drm_crtc *crtc,
-                                     struct drm_display_mode *mode,
+                                     const struct drm_display_mode *mode,
                                      struct drm_display_mode *adjusted_mode)
 {
         if (!radeon_crtc_scaling_mode_fixup(crtc, mode, adjusted_mode))
diff --git a/drivers/gpu/drm/radeon/atombios_dp.c b/drivers/gpu/drm/radeon/atombios_dp.c
index 5131b3b0f7d2..7712cf5ab33b 100644
--- a/drivers/gpu/drm/radeon/atombios_dp.c
+++ b/drivers/gpu/drm/radeon/atombios_dp.c
@@ -22,6 +22,7 @@
  *
  * Authors: Dave Airlie
  *          Alex Deucher
+ *          Jerome Glisse
  */
 #include "drmP.h"
 #include "radeon_drm.h"
@@ -608,7 +609,7 @@ int radeon_dp_get_panel_mode(struct drm_encoder *encoder,
 }
 
 void radeon_dp_set_link_config(struct drm_connector *connector,
-                               struct drm_display_mode *mode)
+                               const struct drm_display_mode *mode)
 {
         struct radeon_connector *radeon_connector = to_radeon_connector(connector);
         struct radeon_connector_atom_dig *dig_connector;
@@ -654,7 +655,6 @@ static bool radeon_dp_get_link_status(struct radeon_connector *radeon_connector,
         ret = radeon_dp_aux_native_read(radeon_connector, DP_LANE0_1_STATUS,
                                         link_status, DP_LINK_STATUS_SIZE, 100);
         if (ret <= 0) {
-                DRM_ERROR("displayport link status failed\n");
                 return false;
         }
 
@@ -833,8 +833,10 @@ static int radeon_dp_link_train_cr(struct radeon_dp_link_train_info *dp_info)
                 else
                         mdelay(dp_info->rd_interval * 4);
 
-                if (!radeon_dp_get_link_status(dp_info->radeon_connector, dp_info->link_status))
+                if (!radeon_dp_get_link_status(dp_info->radeon_connector, dp_info->link_status)) {
+                        DRM_ERROR("displayport link status failed\n");
                         break;
+                }
 
                 if (dp_clock_recovery_ok(dp_info->link_status, dp_info->dp_lane_count)) {
                         clock_recovery = true;
@@ -896,8 +898,10 @@ static int radeon_dp_link_train_ce(struct radeon_dp_link_train_info *dp_info)
                 else
                         mdelay(dp_info->rd_interval * 4);
 
-                if (!radeon_dp_get_link_status(dp_info->radeon_connector, dp_info->link_status))
+                if (!radeon_dp_get_link_status(dp_info->radeon_connector, dp_info->link_status)) {
+                        DRM_ERROR("displayport link status failed\n");
                         break;
+                }
 
                 if (dp_channel_eq_ok(dp_info->link_status, dp_info->dp_lane_count)) {
                         channel_eq = true;
diff --git a/drivers/gpu/drm/radeon/atombios_encoders.c b/drivers/gpu/drm/radeon/atombios_encoders.c
index 486ccdf4aacd..f9bc27fe269a 100644
--- a/drivers/gpu/drm/radeon/atombios_encoders.c
+++ b/drivers/gpu/drm/radeon/atombios_encoders.c
@@ -58,7 +58,7 @@ static inline bool radeon_encoder_is_digital(struct drm_encoder *encoder)
 }
 
 static bool radeon_atom_mode_fixup(struct drm_encoder *encoder,
-                                   struct drm_display_mode *mode,
+                                   const struct drm_display_mode *mode,
                                    struct drm_display_mode *adjusted_mode)
 {
         struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
@@ -1392,10 +1392,18 @@ radeon_atom_encoder_dpms_dig(struct drm_encoder *encoder, int mode)
         case DRM_MODE_DPMS_ON:
                 /* some early dce3.2 boards have a bug in their transmitter control table */
                 if ((rdev->family == CHIP_RV710) || (rdev->family == CHIP_RV730) ||
-                    ASIC_IS_DCE41(rdev) || ASIC_IS_DCE5(rdev))
+                    ASIC_IS_DCE41(rdev) || ASIC_IS_DCE5(rdev)) {
+                        if (ASIC_IS_DCE6(rdev)) {
+                                /* It seems we need to call ATOM_ENCODER_CMD_SETUP again
+                                 * before reenabling encoder on DPMS ON, otherwise we never
+                                 * get picture
+                                 */
+                                atombios_dig_encoder_setup(encoder, ATOM_ENCODER_CMD_SETUP, 0);
+                        }
                         atombios_dig_transmitter_setup(encoder, ATOM_TRANSMITTER_ACTION_ENABLE, 0, 0);
-                else
+                } else {
                         atombios_dig_transmitter_setup(encoder, ATOM_TRANSMITTER_ACTION_ENABLE_OUTPUT, 0, 0);
+                }
                 if (ENCODER_MODE_IS_DP(atombios_get_encoder_mode(encoder)) && connector) {
                         if (connector->connector_type == DRM_MODE_CONNECTOR_eDP) {
                                 atombios_set_edp_panel_power(connector,
@@ -2234,7 +2242,7 @@ radeon_atom_ext_dpms(struct drm_encoder *encoder, int mode)
 }
 
 static bool radeon_atom_ext_mode_fixup(struct drm_encoder *encoder,
-                                       struct drm_display_mode *mode,
+                                       const struct drm_display_mode *mode,
                                        struct drm_display_mode *adjusted_mode)
 {
         return true;
diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index 7fb3d2e0434c..e585a3b947eb 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -99,6 +99,14 @@ void evergreen_fix_pci_max_read_req_size(struct radeon_device *rdev)
         }
 }
 
+/**
+ * dce4_wait_for_vblank - vblank wait asic callback.
+ *
+ * @rdev: radeon_device pointer
+ * @crtc: crtc to wait for vblank on
+ *
+ * Wait for vblank on the requested crtc (evergreen+).
+ */
 void dce4_wait_for_vblank(struct radeon_device *rdev, int crtc)
 {
         struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc];
@@ -118,18 +126,49 @@ void dce4_wait_for_vblank(struct radeon_device *rdev, int crtc)
         }
 }
 
+/**
+ * radeon_irq_kms_pflip_irq_get - pre-pageflip callback.
+ *
+ * @rdev: radeon_device pointer
+ * @crtc: crtc to prepare for pageflip on
+ *
+ * Pre-pageflip callback (evergreen+).
+ * Enables the pageflip irq (vblank irq).
+ */
 void evergreen_pre_page_flip(struct radeon_device *rdev, int crtc)
 {
         /* enable the pflip int */
         radeon_irq_kms_pflip_irq_get(rdev, crtc);
 }
 
+/**
+ * evergreen_post_page_flip - pos-pageflip callback.
+ *
+ * @rdev: radeon_device pointer
+ * @crtc: crtc to cleanup pageflip on
+ *
+ * Post-pageflip callback (evergreen+).
+ * Disables the pageflip irq (vblank irq).
+ */
 void evergreen_post_page_flip(struct radeon_device *rdev, int crtc)
 {
         /* disable the pflip int */
         radeon_irq_kms_pflip_irq_put(rdev, crtc);
 }
 
+/**
+ * evergreen_page_flip - pageflip callback.
+ *
+ * @rdev: radeon_device pointer
+ * @crtc_id: crtc to cleanup pageflip on
+ * @crtc_base: new address of the crtc (GPU MC address)
+ *
+ * Does the actual pageflip (evergreen+).
+ * During vblank we take the crtc lock and wait for the update_pending
+ * bit to go high, when it does, we release the lock, and allow the
+ * double buffered update to take place.
+ * Returns the current update pending status.
+ */
 u32 evergreen_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 {
         struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
@@ -214,6 +253,15 @@ int sumo_get_temp(struct radeon_device *rdev)
         return actual_temp * 1000;
 }
 
+/**
+ * sumo_pm_init_profile - Initialize power profiles callback.
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Initialize the power states used in profile mode
+ * (sumo, trinity, SI).
+ * Used for profile mode only.
+ */
 void sumo_pm_init_profile(struct radeon_device *rdev)
 {
         int idx;
@@ -265,6 +313,14 @@ void sumo_pm_init_profile(struct radeon_device *rdev)
                 rdev->pm.power_state[idx].num_clock_modes - 1;
 }
 
+/**
+ * evergreen_pm_misc - set additional pm hw parameters callback.
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Set non-clock parameters associated with a power state
+ * (voltage, etc.) (evergreen+).
+ */
 void evergreen_pm_misc(struct radeon_device *rdev)
 {
         int req_ps_idx = rdev->pm.requested_power_state_index;
@@ -292,6 +348,13 @@ void evergreen_pm_misc(struct radeon_device *rdev)
         }
 }
 
+/**
+ * evergreen_pm_prepare - pre-power state change callback.
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Prepare for a power state change (evergreen+).
+ */
 void evergreen_pm_prepare(struct radeon_device *rdev)
 {
         struct drm_device *ddev = rdev->ddev;
@@ -310,6 +373,13 @@ void evergreen_pm_prepare(struct radeon_device *rdev)
         }
 }
 
+/**
+ * evergreen_pm_finish - post-power state change callback.
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Clean up after a power state change (evergreen+).
+ */
 void evergreen_pm_finish(struct radeon_device *rdev)
 {
         struct drm_device *ddev = rdev->ddev;
@@ -328,6 +398,15 @@ void evergreen_pm_finish(struct radeon_device *rdev)
         }
 }
 
+/**
+ * evergreen_hpd_sense - hpd sense callback.
+ *
+ * @rdev: radeon_device pointer
+ * @hpd: hpd (hotplug detect) pin
+ *
+ * Checks if a digital monitor is connected (evergreen+).
+ * Returns true if connected, false if not connected.
+ */
 bool evergreen_hpd_sense(struct radeon_device *rdev, enum radeon_hpd_id hpd)
 {
         bool connected = false;
@@ -364,6 +443,14 @@ bool evergreen_hpd_sense(struct radeon_device *rdev, enum radeon_hpd_id hpd)
         return connected;
 }
 
+/**
+ * evergreen_hpd_set_polarity - hpd set polarity callback.
+ *
+ * @rdev: radeon_device pointer
+ * @hpd: hpd (hotplug detect) pin
+ *
+ * Set the polarity of the hpd pin (evergreen+).
+ */
 void evergreen_hpd_set_polarity(struct radeon_device *rdev,
                                 enum radeon_hpd_id hpd)
 {
@@ -424,10 +511,19 @@ void evergreen_hpd_set_polarity(struct radeon_device *rdev,
         }
 }
 
+/**
+ * evergreen_hpd_init - hpd setup callback.
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Setup the hpd pins used by the card (evergreen+).
+ * Enable the pin, set the polarity, and enable the hpd interrupts.
+ */
 void evergreen_hpd_init(struct radeon_device *rdev)
 {
         struct drm_device *dev = rdev->ddev;
         struct drm_connector *connector;
+        unsigned enabled = 0;
         u32 tmp = DC_HPDx_CONNECTION_TIMER(0x9c4) |
                 DC_HPDx_RX_INT_TIMER(0xfa) | DC_HPDx_EN;
 
@@ -436,73 +532,72 @@ void evergreen_hpd_init(struct radeon_device *rdev)
                 switch (radeon_connector->hpd.hpd) {
                 case RADEON_HPD_1:
                         WREG32(DC_HPD1_CONTROL, tmp);
-                        rdev->irq.hpd[0] = true;
                         break;
                 case RADEON_HPD_2:
                         WREG32(DC_HPD2_CONTROL, tmp);
-                        rdev->irq.hpd[1] = true;
                         break;
                 case RADEON_HPD_3:
                         WREG32(DC_HPD3_CONTROL, tmp);
-                        rdev->irq.hpd[2] = true;
                         break;
                 case RADEON_HPD_4:
                         WREG32(DC_HPD4_CONTROL, tmp);
-                        rdev->irq.hpd[3] = true;
                         break;
                 case RADEON_HPD_5:
                         WREG32(DC_HPD5_CONTROL, tmp);
-                        rdev->irq.hpd[4] = true;
                         break;
                 case RADEON_HPD_6:
                         WREG32(DC_HPD6_CONTROL, tmp);
-                        rdev->irq.hpd[5] = true;
                         break;
                 default:
                         break;
                 }
                 radeon_hpd_set_polarity(rdev, radeon_connector->hpd.hpd);
+                enabled |= 1 << radeon_connector->hpd.hpd;
         }
-        if (rdev->irq.installed)
-                evergreen_irq_set(rdev);
+        radeon_irq_kms_enable_hpd(rdev, enabled);
 }
 
+/**
+ * evergreen_hpd_fini - hpd tear down callback.
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Tear down the hpd pins used by the card (evergreen+).
+ * Disable the hpd interrupts.
+ */
 void evergreen_hpd_fini(struct radeon_device *rdev)
 {
         struct drm_device *dev = rdev->ddev;
         struct drm_connector *connector;
+        unsigned disabled = 0;
 
         list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
                 struct radeon_connector *radeon_connector = to_radeon_connector(connector);
                 switch (radeon_connector->hpd.hpd) {
                 case RADEON_HPD_1:
                         WREG32(DC_HPD1_CONTROL, 0);
-                        rdev->irq.hpd[0] = false;
                         break;
                 case RADEON_HPD_2:
                         WREG32(DC_HPD2_CONTROL, 0);
-                        rdev->irq.hpd[1] = false;
                         break;
                 case RADEON_HPD_3:
                         WREG32(DC_HPD3_CONTROL, 0);
-                        rdev->irq.hpd[2] = false;
                         break;
                 case RADEON_HPD_4:
                         WREG32(DC_HPD4_CONTROL, 0);
-                        rdev->irq.hpd[3] = false;
                         break;
                 case RADEON_HPD_5:
                         WREG32(DC_HPD5_CONTROL, 0);
-                        rdev->irq.hpd[4] = false;
                         break;
                 case RADEON_HPD_6:
                         WREG32(DC_HPD6_CONTROL, 0);
-                        rdev->irq.hpd[5] = false;
                         break;
                 default:
                         break;
                 }
+                disabled |= 1 << radeon_connector->hpd.hpd;
         }
+        radeon_irq_kms_disable_hpd(rdev, disabled);
 }
 
 /* watermark setup */
@@ -933,6 +1028,14 @@ static void evergreen_program_watermarks(struct radeon_device *rdev,
 
 }
 
+/**
+ * evergreen_bandwidth_update - update display watermarks callback.
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Update the display watermarks based on the requested mode(s)
+ * (evergreen+).
+ */
 void evergreen_bandwidth_update(struct radeon_device *rdev)
 {
         struct drm_display_mode *mode0 = NULL;
@@ -956,6 +1059,15 @@ void evergreen_bandwidth_update(struct radeon_device *rdev)
         }
 }
 
+/**
+ * evergreen_mc_wait_for_idle - wait for MC idle callback.
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Wait for the MC (memory controller) to be idle.
+ * (evergreen+).
+ * Returns 0 if the MC is idle, -1 if not.
+ */
 int evergreen_mc_wait_for_idle(struct radeon_device *rdev)
 {
         unsigned i;
@@ -1371,12 +1483,28 @@ void evergreen_mc_program(struct radeon_device *rdev)
  */
 void evergreen_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib)
 {
-        struct radeon_ring *ring = &rdev->ring[ib->fence->ring];
+        struct radeon_ring *ring = &rdev->ring[ib->ring];
+        u32 next_rptr;
 
         /* set to DX10/11 mode */
         radeon_ring_write(ring, PACKET3(PACKET3_MODE_CONTROL, 0));
         radeon_ring_write(ring, 1);
-        /* FIXME: implement */
+
+        if (ring->rptr_save_reg) {
+                next_rptr = ring->wptr + 3 + 4;
+                radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1));
+                radeon_ring_write(ring, ((ring->rptr_save_reg - 
+                                          PACKET3_SET_CONFIG_REG_START) >> 2));
+                radeon_ring_write(ring, next_rptr);
+        } else if (rdev->wb.enabled) {
+                next_rptr = ring->wptr + 5 + 4;
+                radeon_ring_write(ring, PACKET3(PACKET3_MEM_WRITE, 3));
+                radeon_ring_write(ring, ring->next_rptr_gpu_addr & 0xfffffffc);
+                radeon_ring_write(ring, (upper_32_bits(ring->next_rptr_gpu_addr) & 0xff) | (1 << 18));
+                radeon_ring_write(ring, next_rptr);
+                radeon_ring_write(ring, 0);
+        }
+
         radeon_ring_write(ring, PACKET3(PACKET3_INDIRECT_BUFFER, 2));
         radeon_ring_write(ring,
 #ifdef __BIG_ENDIAN
@@ -2188,6 +2316,14 @@ static int evergreen_gpu_soft_reset(struct radeon_device *rdev)
                 RREG32(GRBM_STATUS_SE1));
         dev_info(rdev->dev, "  SRBM_STATUS=0x%08X\n",
                 RREG32(SRBM_STATUS));
+        dev_info(rdev->dev, "  R_008674_CP_STALLED_STAT1 = 0x%08X\n",
+                RREG32(CP_STALLED_STAT1));
+        dev_info(rdev->dev, "  R_008678_CP_STALLED_STAT2 = 0x%08X\n",
+                RREG32(CP_STALLED_STAT2));
+        dev_info(rdev->dev, "  R_00867C_CP_BUSY_STAT     = 0x%08X\n",
+                RREG32(CP_BUSY_STAT));
+        dev_info(rdev->dev, "  R_008680_CP_STAT          = 0x%08X\n",
+                RREG32(CP_STAT));
         evergreen_mc_stop(rdev, &save);
         if (evergreen_mc_wait_for_idle(rdev)) {
                 dev_warn(rdev->dev, "Wait for MC idle timedout !\n");
@@ -2225,6 +2361,14 @@ static int evergreen_gpu_soft_reset(struct radeon_device *rdev)
                 RREG32(GRBM_STATUS_SE1));
         dev_info(rdev->dev, "  SRBM_STATUS=0x%08X\n",
                 RREG32(SRBM_STATUS));
+        dev_info(rdev->dev, "  R_008674_CP_STALLED_STAT1 = 0x%08X\n",
+                RREG32(CP_STALLED_STAT1));
+        dev_info(rdev->dev, "  R_008678_CP_STALLED_STAT2 = 0x%08X\n",
+                RREG32(CP_STALLED_STAT2));
+        dev_info(rdev->dev, "  R_00867C_CP_BUSY_STAT     = 0x%08X\n",
+                RREG32(CP_BUSY_STAT));
+        dev_info(rdev->dev, "  R_008680_CP_STAT          = 0x%08X\n",
+                RREG32(CP_STAT));
         evergreen_mc_resume(rdev, &save);
         return 0;
 }
@@ -2348,20 +2492,20 @@ int evergreen_irq_set(struct radeon_device *rdev)
 
         if (rdev->family >= CHIP_CAYMAN) {
                 /* enable CP interrupts on all rings */
-                if (rdev->irq.sw_int[RADEON_RING_TYPE_GFX_INDEX]) {
+                if (atomic_read(&rdev->irq.ring_int[RADEON_RING_TYPE_GFX_INDEX])) {
                         DRM_DEBUG("evergreen_irq_set: sw int gfx\n");
                         cp_int_cntl |= TIME_STAMP_INT_ENABLE;
                 }
-                if (rdev->irq.sw_int[CAYMAN_RING_TYPE_CP1_INDEX]) {
+                if (atomic_read(&rdev->irq.ring_int[CAYMAN_RING_TYPE_CP1_INDEX])) {
                         DRM_DEBUG("evergreen_irq_set: sw int cp1\n");
                         cp_int_cntl1 |= TIME_STAMP_INT_ENABLE;
                 }
-                if (rdev->irq.sw_int[CAYMAN_RING_TYPE_CP2_INDEX]) {
+                if (atomic_read(&rdev->irq.ring_int[CAYMAN_RING_TYPE_CP2_INDEX])) {
                         DRM_DEBUG("evergreen_irq_set: sw int cp2\n");
                         cp_int_cntl2 |= TIME_STAMP_INT_ENABLE;
                 }
         } else {
-                if (rdev->irq.sw_int[RADEON_RING_TYPE_GFX_INDEX]) {
+                if (atomic_read(&rdev->irq.ring_int[RADEON_RING_TYPE_GFX_INDEX])) {
                         DRM_DEBUG("evergreen_irq_set: sw int gfx\n");
                         cp_int_cntl |= RB_INT_ENABLE;
                         cp_int_cntl |= TIME_STAMP_INT_ENABLE;
@@ -2369,32 +2513,32 @@ int evergreen_irq_set(struct radeon_device *rdev)
         }
 
         if (rdev->irq.crtc_vblank_int[0] ||
-            rdev->irq.pflip[0]) {
+            atomic_read(&rdev->irq.pflip[0])) {
                 DRM_DEBUG("evergreen_irq_set: vblank 0\n");
                 crtc1 |= VBLANK_INT_MASK;
         }
         if (rdev->irq.crtc_vblank_int[1] ||
-            rdev->irq.pflip[1]) {
+            atomic_read(&rdev->irq.pflip[1])) {
                 DRM_DEBUG("evergreen_irq_set: vblank 1\n");
                 crtc2 |= VBLANK_INT_MASK;
         }
         if (rdev->irq.crtc_vblank_int[2] ||
-            rdev->irq.pflip[2]) {
+            atomic_read(&rdev->irq.pflip[2])) {
                 DRM_DEBUG("evergreen_irq_set: vblank 2\n");
                 crtc3 |= VBLANK_INT_MASK;
         }
         if (rdev->irq.crtc_vblank_int[3] ||
-            rdev->irq.pflip[3]) {
+            atomic_read(&rdev->irq.pflip[3])) {
                 DRM_DEBUG("evergreen_irq_set: vblank 3\n");
                 crtc4 |= VBLANK_INT_MASK;
         }
         if (rdev->irq.crtc_vblank_int[4] ||
-            rdev->irq.pflip[4]) {
+            atomic_read(&rdev->irq.pflip[4])) {
                 DRM_DEBUG("evergreen_irq_set: vblank 4\n");
                 crtc5 |= VBLANK_INT_MASK;
         }
         if (rdev->irq.crtc_vblank_int[5] ||
-            rdev->irq.pflip[5]) {
+            atomic_read(&rdev->irq.pflip[5])) {
                 DRM_DEBUG("evergreen_irq_set: vblank 5\n");
                 crtc6 |= VBLANK_INT_MASK;
         }
@@ -2676,7 +2820,6 @@ int evergreen_irq_process(struct radeon_device *rdev)
         u32 rptr;
         u32 src_id, src_data;
         u32 ring_index;
-        unsigned long flags;
         bool queue_hotplug = false;
         bool queue_hdmi = false;
 
@@ -2684,22 +2827,21 @@ int evergreen_irq_process(struct radeon_device *rdev)
                 return IRQ_NONE;
 
         wptr = evergreen_get_ih_wptr(rdev);
+
+restart_ih:
+        /* is somebody else already processing irqs? */
+        if (atomic_xchg(&rdev->ih.lock, 1))
+                return IRQ_NONE;
+
         rptr = rdev->ih.rptr;
         DRM_DEBUG("r600_irq_process start: rptr %d, wptr %d\n", rptr, wptr);
 
-        spin_lock_irqsave(&rdev->ih.lock, flags);
-        if (rptr == wptr) {
-                spin_unlock_irqrestore(&rdev->ih.lock, flags);
-                return IRQ_NONE;
-        }
-restart_ih:
         /* Order reading of wptr vs. reading of IH ring data */
         rmb();
 
         /* display interrupts */
         evergreen_irq_ack(rdev);
 
-        rdev->ih.wptr = wptr;
         while (rptr != wptr) {
                 /* wptr/rptr are in bytes! */
                 ring_index = rptr / 4;
@@ -2716,7 +2858,7 @@ restart_ih:
                                                 rdev->pm.vblank_sync = true;
                                                 wake_up(&rdev->irq.vblank_queue);
                                         }
-                                        if (rdev->irq.pflip[0])
+                                        if (atomic_read(&rdev->irq.pflip[0]))
                                                 radeon_crtc_handle_flip(rdev, 0);
                                         rdev->irq.stat_regs.evergreen.disp_int &= ~LB_D1_VBLANK_INTERRUPT;
                                         DRM_DEBUG("IH: D1 vblank\n");
@@ -2742,7 +2884,7 @@ restart_ih:
                                                 rdev->pm.vblank_sync = true;
                                                 wake_up(&rdev->irq.vblank_queue);
                                         }
-                                        if (rdev->irq.pflip[1])
+                                        if (atomic_read(&rdev->irq.pflip[1]))
                                                 radeon_crtc_handle_flip(rdev, 1);
                                         rdev->irq.stat_regs.evergreen.disp_int_cont &= ~LB_D2_VBLANK_INTERRUPT;
                                         DRM_DEBUG("IH: D2 vblank\n");
@@ -2768,7 +2910,7 @@ restart_ih:
                                                 rdev->pm.vblank_sync = true;
                                                 wake_up(&rdev->irq.vblank_queue);
                                         }
-                                        if (rdev->irq.pflip[2])
+                                        if (atomic_read(&rdev->irq.pflip[2]))
                                                 radeon_crtc_handle_flip(rdev, 2);
                                         rdev->irq.stat_regs.evergreen.disp_int_cont2 &= ~LB_D3_VBLANK_INTERRUPT;
                                         DRM_DEBUG("IH: D3 vblank\n");
@@ -2794,7 +2936,7 @@ restart_ih:
                                                 rdev->pm.vblank_sync = true;
                                                 wake_up(&rdev->irq.vblank_queue);
                                         }
-                                        if (rdev->irq.pflip[3])
+                                        if (atomic_read(&rdev->irq.pflip[3]))
                                                 radeon_crtc_handle_flip(rdev, 3);
                                         rdev->irq.stat_regs.evergreen.disp_int_cont3 &= ~LB_D4_VBLANK_INTERRUPT;
                                         DRM_DEBUG("IH: D4 vblank\n");
@@ -2820,7 +2962,7 @@ restart_ih:
                                                 rdev->pm.vblank_sync = true;
                                                 wake_up(&rdev->irq.vblank_queue);
                                         }
-                                        if (rdev->irq.pflip[4])
+                                        if (atomic_read(&rdev->irq.pflip[4]))
                                                 radeon_crtc_handle_flip(rdev, 4);
                                         rdev->irq.stat_regs.evergreen.disp_int_cont4 &= ~LB_D5_VBLANK_INTERRUPT;
                                         DRM_DEBUG("IH: D5 vblank\n");
@@ -2846,7 +2988,7 @@ restart_ih:
                                                 rdev->pm.vblank_sync = true;
                                                 wake_up(&rdev->irq.vblank_queue);
                                         }
-                                        if (rdev->irq.pflip[5])
+                                        if (atomic_read(&rdev->irq.pflip[5]))
                                                 radeon_crtc_handle_flip(rdev, 5);
                                         rdev->irq.stat_regs.evergreen.disp_int_cont5 &= ~LB_D6_VBLANK_INTERRUPT;
                                         DRM_DEBUG("IH: D6 vblank\n");
@@ -2986,7 +3128,6 @@ restart_ih:
                         break;
                 case 233: /* GUI IDLE */
                         DRM_DEBUG("IH: GUI idle\n");
-                        rdev->pm.gui_idle = true;
                         wake_up(&rdev->irq.idle_queue);
                         break;
                 default:
@@ -2998,17 +3139,19 @@ restart_ih:
                 rptr += 16;
                 rptr &= rdev->ih.ptr_mask;
         }
-        /* make sure wptr hasn't changed while processing */
-        wptr = evergreen_get_ih_wptr(rdev);
-        if (wptr != rdev->ih.wptr)
-                goto restart_ih;
         if (queue_hotplug)
                 schedule_work(&rdev->hotplug_work);
         if (queue_hdmi)
                 schedule_work(&rdev->audio_work);
         rdev->ih.rptr = rptr;
         WREG32(IH_RB_RPTR, rdev->ih.rptr);
-        spin_unlock_irqrestore(&rdev->ih.lock, flags);
+        atomic_set(&rdev->ih.lock, 0);
+
+        /* make sure wptr hasn't changed while processing */
+        wptr = evergreen_get_ih_wptr(rdev);
+        if (wptr != rptr)
+                goto restart_ih;
+
         return IRQ_HANDLED;
 }
 
@@ -3096,13 +3239,11 @@ static int evergreen_startup(struct radeon_device *rdev)
         if (r)
                 return r;
 
-        r = radeon_ib_pool_start(rdev);
-        if (r)
-                return r;
-
-        r = radeon_ib_ring_tests(rdev);
-        if (r)
+        r = radeon_ib_pool_init(rdev);
+        if (r) {
+                dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
                 return r;
+        }
 
         r = r600_audio_init(rdev);
         if (r) {
@@ -3146,9 +3287,6 @@ int evergreen_suspend(struct radeon_device *rdev)
         struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 
         r600_audio_fini(rdev);
-        /* FIXME: we should wait for ring to be empty */
-        radeon_ib_pool_suspend(rdev);
-        r600_blit_suspend(rdev);
         r700_cp_stop(rdev);
         ring->ready = false;
         evergreen_irq_suspend(rdev);
@@ -3234,20 +3372,14 @@ int evergreen_init(struct radeon_device *rdev)
         if (r)
                 return r;
 
-        r = radeon_ib_pool_init(rdev);
         rdev->accel_working = true;
-        if (r) {
-                dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
-                rdev->accel_working = false;
-        }
-
         r = evergreen_startup(rdev);
         if (r) {
                 dev_err(rdev->dev, "disabling GPU acceleration\n");
                 r700_cp_fini(rdev);
                 r600_irq_fini(rdev);
                 radeon_wb_fini(rdev);
-                r100_ib_fini(rdev);
+                radeon_ib_pool_fini(rdev);
                 radeon_irq_kms_fini(rdev);
                 evergreen_pcie_gart_fini(rdev);
                 rdev->accel_working = false;
@@ -3274,7 +3406,7 @@ void evergreen_fini(struct radeon_device *rdev)
         r700_cp_fini(rdev);
         r600_irq_fini(rdev);
         radeon_wb_fini(rdev);
-        r100_ib_fini(rdev);
+        radeon_ib_pool_fini(rdev);
         radeon_irq_kms_fini(rdev);
         evergreen_pcie_gart_fini(rdev);
         r600_vram_scratch_fini(rdev);
@@ -3289,7 +3421,8 @@ void evergreen_fini(struct radeon_device *rdev)
 
 void evergreen_pcie_gen2_enable(struct radeon_device *rdev)
 {
-        u32 link_width_cntl, speed_cntl;
+        u32 link_width_cntl, speed_cntl, mask;
+        int ret;
 
         if (radeon_pcie_gen2 == 0)
                 return;
@@ -3304,6 +3437,15 @@ void evergreen_pcie_gen2_enable(struct radeon_device *rdev)
         if (ASIC_IS_X2(rdev))
                 return;
 
+        ret = drm_pcie_get_speed_cap_mask(rdev->ddev, &mask);
+        if (ret != 0)
+                return;
+
+        if (!(mask & DRM_PCIE_SPEED_50))
+                return;
+
+        DRM_INFO("enabling PCIE gen 2 link speeds, disable with radeon.pcie_gen2=0\n");
+
         speed_cntl = RREG32_PCIE_P(PCIE_LC_SPEED_CNTL);
         if ((speed_cntl & LC_OTHER_SIDE_EVER_SENT_GEN2) ||
             (speed_cntl & LC_OTHER_SIDE_SUPPORTS_GEN2)) {
diff --git a/drivers/gpu/drm/radeon/evergreen_blit_kms.c b/drivers/gpu/drm/radeon/evergreen_blit_kms.c
index 1e96bd458cfd..89cb9feb5653 100644
--- a/drivers/gpu/drm/radeon/evergreen_blit_kms.c
+++ b/drivers/gpu/drm/radeon/evergreen_blit_kms.c
@@ -622,7 +622,8 @@ int evergreen_blit_init(struct radeon_device *rdev)
         rdev->r600_blit.primitives.draw_auto = draw_auto;
         rdev->r600_blit.primitives.set_default_state = set_default_state;
 
-        rdev->r600_blit.ring_size_common = 55; /* shaders + def state */
+        rdev->r600_blit.ring_size_common = 8; /* sync semaphore */
+        rdev->r600_blit.ring_size_common += 55; /* shaders + def state */
         rdev->r600_blit.ring_size_common += 16; /* fence emit for VB IB */
         rdev->r600_blit.ring_size_common += 5; /* done copy */
         rdev->r600_blit.ring_size_common += 16; /* fence emit for done copy */
@@ -633,10 +634,6 @@ int evergreen_blit_init(struct radeon_device *rdev)
 
         rdev->r600_blit.max_dim = 16384;
 
-        /* pin copy shader into vram if already initialized */
-        if (rdev->r600_blit.shader_obj)
-                goto done;
-
         rdev->r600_blit.state_offset = 0;
 
         if (rdev->family < CHIP_CAYMAN)
@@ -667,11 +664,26 @@ int evergreen_blit_init(struct radeon_device *rdev)
                 obj_size += cayman_ps_size * 4;
         obj_size = ALIGN(obj_size, 256);
 
-        r = radeon_bo_create(rdev, obj_size, PAGE_SIZE, true, RADEON_GEM_DOMAIN_VRAM,
-                             NULL, &rdev->r600_blit.shader_obj);
-        if (r) {
-                DRM_ERROR("evergreen failed to allocate shader\n");
-                return r;
+        /* pin copy shader into vram if not already initialized */
+        if (!rdev->r600_blit.shader_obj) {
+                r = radeon_bo_create(rdev, obj_size, PAGE_SIZE, true,
+                                     RADEON_GEM_DOMAIN_VRAM,
+                                     NULL, &rdev->r600_blit.shader_obj);
+                if (r) {
+                        DRM_ERROR("evergreen failed to allocate shader\n");
+                        return r;
+                }
+
+                r = radeon_bo_reserve(rdev->r600_blit.shader_obj, false);
+                if (unlikely(r != 0))
+                        return r;
+                r = radeon_bo_pin(rdev->r600_blit.shader_obj, RADEON_GEM_DOMAIN_VRAM,
+                                  &rdev->r600_blit.shader_gpu_addr);
+                radeon_bo_unreserve(rdev->r600_blit.shader_obj);
+                if (r) {
+                        dev_err(rdev->dev, "(%d) pin blit object failed\n", r);
+                        return r;
+                }
         }
 
         DRM_DEBUG("evergreen blit allocated bo %08x vs %08x ps %08x\n",
@@ -713,17 +725,6 @@ int evergreen_blit_init(struct radeon_device *rdev)
         radeon_bo_kunmap(rdev->r600_blit.shader_obj);
         radeon_bo_unreserve(rdev->r600_blit.shader_obj);
 
-done:
-        r = radeon_bo_reserve(rdev->r600_blit.shader_obj, false);
-        if (unlikely(r != 0))
-                return r;
-        r = radeon_bo_pin(rdev->r600_blit.shader_obj, RADEON_GEM_DOMAIN_VRAM,
-                          &rdev->r600_blit.shader_gpu_addr);
-        radeon_bo_unreserve(rdev->r600_blit.shader_obj);
-        if (r) {
-                dev_err(rdev->dev, "(%d) pin blit object failed\n", r);
-                return r;
-        }
         radeon_ttm_set_active_vram_size(rdev, rdev->mc.real_vram_size);
         return 0;
 }
diff --git a/drivers/gpu/drm/radeon/evergreend.h b/drivers/gpu/drm/radeon/evergreend.h
index b50b15c70498..d3bd098e4e19 100644
--- a/drivers/gpu/drm/radeon/evergreend.h
+++ b/drivers/gpu/drm/radeon/evergreend.h
@@ -88,6 +88,10 @@
 #define CONFIG_MEMSIZE                                  0x5428
 
 #define CP_COHER_BASE                                   0x85F8
+#define CP_STALLED_STAT1                        0x8674
+#define CP_STALLED_STAT2                        0x8678
+#define CP_BUSY_STAT                            0x867C
+#define CP_STAT                                         0x8680
 #define CP_ME_CNTL                                      0x86D8
 #define         CP_ME_HALT                                      (1 << 28)
 #define         CP_PFP_HALT                                     (1 << 26)
diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c
index b7bf18e40215..9945d86d9001 100644
--- a/drivers/gpu/drm/radeon/ni.c
+++ b/drivers/gpu/drm/radeon/ni.c
@@ -850,11 +850,20 @@ void cayman_fence_ring_emit(struct radeon_device *rdev,
 
 void cayman_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib)
 {
-        struct radeon_ring *ring = &rdev->ring[ib->fence->ring];
+        struct radeon_ring *ring = &rdev->ring[ib->ring];
 
         /* set to DX10/11 mode */
         radeon_ring_write(ring, PACKET3(PACKET3_MODE_CONTROL, 0));
         radeon_ring_write(ring, 1);
+
+        if (ring->rptr_save_reg) {
+                uint32_t next_rptr = ring->wptr + 3 + 4 + 8;
+                radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1));
+                radeon_ring_write(ring, ((ring->rptr_save_reg - 
+                                          PACKET3_SET_CONFIG_REG_START) >> 2));
+                radeon_ring_write(ring, next_rptr);
+        }
+
         radeon_ring_write(ring, PACKET3(PACKET3_INDIRECT_BUFFER, 2));
         radeon_ring_write(ring,
 #ifdef __BIG_ENDIAN
@@ -981,16 +990,41 @@ static int cayman_cp_start(struct radeon_device *rdev)
 
 static void cayman_cp_fini(struct radeon_device *rdev)
 {
+        struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
         cayman_cp_enable(rdev, false);
-        radeon_ring_fini(rdev, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]);
+        radeon_ring_fini(rdev, ring);
+        radeon_scratch_free(rdev, ring->rptr_save_reg);
 }
 
 int cayman_cp_resume(struct radeon_device *rdev)
 {
+        static const int ridx[] = {
+                RADEON_RING_TYPE_GFX_INDEX,
+                CAYMAN_RING_TYPE_CP1_INDEX,
+                CAYMAN_RING_TYPE_CP2_INDEX
+        };
+        static const unsigned cp_rb_cntl[] = {
+                CP_RB0_CNTL,
+                CP_RB1_CNTL,
+                CP_RB2_CNTL,
+        };
+        static const unsigned cp_rb_rptr_addr[] = {
+                CP_RB0_RPTR_ADDR,
+                CP_RB1_RPTR_ADDR,
+                CP_RB2_RPTR_ADDR
+        };
+        static const unsigned cp_rb_rptr_addr_hi[] = {
+                CP_RB0_RPTR_ADDR_HI,
+                CP_RB1_RPTR_ADDR_HI,
+                CP_RB2_RPTR_ADDR_HI
+        };
+        static const unsigned cp_rb_base[] = {
+                CP_RB0_BASE,
+                CP_RB1_BASE,
+                CP_RB2_BASE
+        };
         struct radeon_ring *ring;
-        u32 tmp;
-        u32 rb_bufsz;
-        int r;
+        int i, r;
 
         /* Reset cp; if cp is reset, then PA, SH, VGT also need to be reset */
         WREG32(GRBM_SOFT_RESET, (SOFT_RESET_CP |
@@ -1012,91 +1046,47 @@ int cayman_cp_resume(struct radeon_device *rdev)
 
         WREG32(CP_DEBUG, (1 << 27));
 
-        /* ring 0 - compute and gfx */
-        /* Set ring buffer size */
-        ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
-        rb_bufsz = drm_order(ring->ring_size / 8);
-        tmp = (drm_order(RADEON_GPU_PAGE_SIZE/8) << 8) | rb_bufsz;
-#ifdef __BIG_ENDIAN
-        tmp |= BUF_SWAP_32BIT;
-#endif
-        WREG32(CP_RB0_CNTL, tmp);
-
-        /* Initialize the ring buffer's read and write pointers */
-        WREG32(CP_RB0_CNTL, tmp | RB_RPTR_WR_ENA);
-        ring->wptr = 0;
-        WREG32(CP_RB0_WPTR, ring->wptr);
-
         /* set the wb address wether it's enabled or not */
-        WREG32(CP_RB0_RPTR_ADDR, (rdev->wb.gpu_addr + RADEON_WB_CP_RPTR_OFFSET) & 0xFFFFFFFC);
-        WREG32(CP_RB0_RPTR_ADDR_HI, upper_32_bits(rdev->wb.gpu_addr + RADEON_WB_CP_RPTR_OFFSET) & 0xFF);
         WREG32(SCRATCH_ADDR, ((rdev->wb.gpu_addr + RADEON_WB_SCRATCH_OFFSET) >> 8) & 0xFFFFFFFF);
+        WREG32(SCRATCH_UMSK, 0xff);
 
-        if (rdev->wb.enabled)
-                WREG32(SCRATCH_UMSK, 0xff);
-        else {
-                tmp |= RB_NO_UPDATE;
-                WREG32(SCRATCH_UMSK, 0);
-        }
+        for (i = 0; i < 3; ++i) {
+                uint32_t rb_cntl;
+                uint64_t addr;
 
-        mdelay(1);
-        WREG32(CP_RB0_CNTL, tmp);
-
-        WREG32(CP_RB0_BASE, ring->gpu_addr >> 8);
-
-        ring->rptr = RREG32(CP_RB0_RPTR);
-
-        /* ring1  - compute only */
-        /* Set ring buffer size */
-        ring = &rdev->ring[CAYMAN_RING_TYPE_CP1_INDEX];
-        rb_bufsz = drm_order(ring->ring_size / 8);
-        tmp = (drm_order(RADEON_GPU_PAGE_SIZE/8) << 8) | rb_bufsz;
+                /* Set ring buffer size */
+                ring = &rdev->ring[ridx[i]];
+                rb_cntl = drm_order(ring->ring_size / 8);
+                rb_cntl |= drm_order(RADEON_GPU_PAGE_SIZE/8) << 8;
 #ifdef __BIG_ENDIAN
-        tmp |= BUF_SWAP_32BIT;
+                rb_cntl |= BUF_SWAP_32BIT;
 #endif
-        WREG32(CP_RB1_CNTL, tmp);
-
-        /* Initialize the ring buffer's read and write pointers */
-        WREG32(CP_RB1_CNTL, tmp | RB_RPTR_WR_ENA);
-        ring->wptr = 0;
-        WREG32(CP_RB1_WPTR, ring->wptr);
-
-        /* set the wb address wether it's enabled or not */
-        WREG32(CP_RB1_RPTR_ADDR, (rdev->wb.gpu_addr + RADEON_WB_CP1_RPTR_OFFSET) & 0xFFFFFFFC);
-        WREG32(CP_RB1_RPTR_ADDR_HI, upper_32_bits(rdev->wb.gpu_addr + RADEON_WB_CP1_RPTR_OFFSET) & 0xFF);
-
-        mdelay(1);
-        WREG32(CP_RB1_CNTL, tmp);
+                WREG32(cp_rb_cntl[i], rb_cntl);
 
-        WREG32(CP_RB1_BASE, ring->gpu_addr >> 8);
-
-        ring->rptr = RREG32(CP_RB1_RPTR);
-
-        /* ring2 - compute only */
-        /* Set ring buffer size */
-        ring = &rdev->ring[CAYMAN_RING_TYPE_CP2_INDEX];
-        rb_bufsz = drm_order(ring->ring_size / 8);
-        tmp = (drm_order(RADEON_GPU_PAGE_SIZE/8) << 8) | rb_bufsz;
-#ifdef __BIG_ENDIAN
-        tmp |= BUF_SWAP_32BIT;
-#endif
-        WREG32(CP_RB2_CNTL, tmp);
-
-        /* Initialize the ring buffer's read and write pointers */
-        WREG32(CP_RB2_CNTL, tmp | RB_RPTR_WR_ENA);
-        ring->wptr = 0;
-        WREG32(CP_RB2_WPTR, ring->wptr);
+                /* set the wb address wether it's enabled or not */
+                addr = rdev->wb.gpu_addr + RADEON_WB_CP_RPTR_OFFSET;
+                WREG32(cp_rb_rptr_addr[i], addr & 0xFFFFFFFC);
+                WREG32(cp_rb_rptr_addr_hi[i], upper_32_bits(addr) & 0xFF);
+        }
 
-        /* set the wb address wether it's enabled or not */
-        WREG32(CP_RB2_RPTR_ADDR, (rdev->wb.gpu_addr + RADEON_WB_CP2_RPTR_OFFSET) & 0xFFFFFFFC);
-        WREG32(CP_RB2_RPTR_ADDR_HI, upper_32_bits(rdev->wb.gpu_addr + RADEON_WB_CP2_RPTR_OFFSET) & 0xFF);
+        /* set the rb base addr, this causes an internal reset of ALL rings */
+        for (i = 0; i < 3; ++i) {
+                ring = &rdev->ring[ridx[i]];
+                WREG32(cp_rb_base[i], ring->gpu_addr >> 8);
+        }
 
-        mdelay(1);
-        WREG32(CP_RB2_CNTL, tmp);
+        for (i = 0; i < 3; ++i) {
+                /* Initialize the ring buffer's read and write pointers */
+                ring = &rdev->ring[ridx[i]];
+                WREG32_P(cp_rb_cntl[i], RB_RPTR_WR_ENA, ~RB_RPTR_WR_ENA);
 
-        WREG32(CP_RB2_BASE, ring->gpu_addr >> 8);
+                ring->rptr = ring->wptr = 0;
+                WREG32(ring->rptr_reg, ring->rptr);
+                WREG32(ring->wptr_reg, ring->wptr);
 
-        ring->rptr = RREG32(CP_RB2_RPTR);
+                mdelay(1);
+                WREG32_P(cp_rb_cntl[i], 0, ~RB_RPTR_WR_ENA);
+        }
 
         /* start the rings */
         cayman_cp_start(rdev);
@@ -1132,6 +1122,14 @@ static int cayman_gpu_soft_reset(struct radeon_device *rdev)
                 RREG32(GRBM_STATUS_SE1));
         dev_info(rdev->dev, "  SRBM_STATUS=0x%08X\n",
                 RREG32(SRBM_STATUS));
+        dev_info(rdev->dev, "  R_008674_CP_STALLED_STAT1 = 0x%08X\n",
+                RREG32(CP_STALLED_STAT1));
+        dev_info(rdev->dev, "  R_008678_CP_STALLED_STAT2 = 0x%08X\n",
+                RREG32(CP_STALLED_STAT2));
+        dev_info(rdev->dev, "  R_00867C_CP_BUSY_STAT     = 0x%08X\n",
+                RREG32(CP_BUSY_STAT));
+        dev_info(rdev->dev, "  R_008680_CP_STAT          = 0x%08X\n",
+                RREG32(CP_STAT));
         dev_info(rdev->dev, "  VM_CONTEXT0_PROTECTION_FAULT_ADDR   0x%08X\n",
                  RREG32(0x14F8));
         dev_info(rdev->dev, "  VM_CONTEXT0_PROTECTION_FAULT_STATUS 0x%08X\n",
@@ -1180,6 +1178,14 @@ static int cayman_gpu_soft_reset(struct radeon_device *rdev)
                 RREG32(GRBM_STATUS_SE1));
         dev_info(rdev->dev, "  SRBM_STATUS=0x%08X\n",
                 RREG32(SRBM_STATUS));
+        dev_info(rdev->dev, "  R_008674_CP_STALLED_STAT1 = 0x%08X\n",
+                RREG32(CP_STALLED_STAT1));
+        dev_info(rdev->dev, "  R_008678_CP_STALLED_STAT2 = 0x%08X\n",
+                RREG32(CP_STALLED_STAT2));
+        dev_info(rdev->dev, "  R_00867C_CP_BUSY_STAT     = 0x%08X\n",
+                RREG32(CP_BUSY_STAT));
+        dev_info(rdev->dev, "  R_008680_CP_STAT          = 0x%08X\n",
+                RREG32(CP_STAT));
         evergreen_mc_resume(rdev, &save);
         return 0;
 }
@@ -1291,17 +1297,17 @@ static int cayman_startup(struct radeon_device *rdev)
         if (r)
                 return r;
 
-        r = radeon_ib_pool_start(rdev);
-        if (r)
-                return r;
-
-        r = radeon_ib_ring_tests(rdev);
-        if (r)
+        r = radeon_ib_pool_init(rdev);
+        if (r) {
+                dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
                 return r;
+        }
 
-        r = radeon_vm_manager_start(rdev);
-        if (r)
+        r = radeon_vm_manager_init(rdev);
+        if (r) {
+                dev_err(rdev->dev, "vm manager initialization failed (%d).\n", r);
                 return r;
+        }
 
         r = r600_audio_init(rdev);
         if (r)
@@ -1334,10 +1340,6 @@ int cayman_resume(struct radeon_device *rdev)
 int cayman_suspend(struct radeon_device *rdev)
 {
         r600_audio_fini(rdev);
-        /* FIXME: we should wait for ring to be empty */
-        radeon_ib_pool_suspend(rdev);
-        radeon_vm_manager_suspend(rdev);
-        r600_blit_suspend(rdev);
         cayman_cp_enable(rdev, false);
         rdev->ring[RADEON_RING_TYPE_GFX_INDEX].ready = false;
         evergreen_irq_suspend(rdev);
@@ -1413,17 +1415,7 @@ int cayman_init(struct radeon_device *rdev)
         if (r)
                 return r;
 
-        r = radeon_ib_pool_init(rdev);
         rdev->accel_working = true;
-        if (r) {
-                dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
-                rdev->accel_working = false;
-        }
-        r = radeon_vm_manager_init(rdev);
-        if (r) {
-                dev_err(rdev->dev, "vm manager initialization failed (%d).\n", r);
-        }
-
         r = cayman_startup(rdev);
         if (r) {
                 dev_err(rdev->dev, "disabling GPU acceleration\n");
@@ -1432,7 +1424,7 @@ int cayman_init(struct radeon_device *rdev)
                 if (rdev->flags & RADEON_IS_IGP)
                         si_rlc_fini(rdev);
                 radeon_wb_fini(rdev);
-                r100_ib_fini(rdev);
+                radeon_ib_pool_fini(rdev);
                 radeon_vm_manager_fini(rdev);
                 radeon_irq_kms_fini(rdev);
                 cayman_pcie_gart_fini(rdev);
@@ -1463,7 +1455,7 @@ void cayman_fini(struct radeon_device *rdev)
                 si_rlc_fini(rdev);
         radeon_wb_fini(rdev);
         radeon_vm_manager_fini(rdev);
-        r100_ib_fini(rdev);
+        radeon_ib_pool_fini(rdev);
         radeon_irq_kms_fini(rdev);
         cayman_pcie_gart_fini(rdev);
         r600_vram_scratch_fini(rdev);
diff --git a/drivers/gpu/drm/radeon/nid.h b/drivers/gpu/drm/radeon/nid.h
index a0b98066e207..870db340d377 100644
--- a/drivers/gpu/drm/radeon/nid.h
+++ b/drivers/gpu/drm/radeon/nid.h
@@ -236,6 +236,10 @@
 #define CP_SEM_WAIT_TIMER                               0x85BC
 #define CP_SEM_INCOMPLETE_TIMER_CNTL                    0x85C8
 #define CP_COHER_CNTL2                                  0x85E8
+#define CP_STALLED_STAT1                        0x8674
+#define CP_STALLED_STAT2                        0x8678
+#define CP_BUSY_STAT                            0x867C
+#define CP_STAT                                         0x8680
 #define CP_ME_CNTL                                      0x86D8
 #define         CP_ME_HALT                                      (1 << 28)
 #define         CP_PFP_HALT                                     (1 << 26)
diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c
index fb44e7e49083..8acb34fd3fd5 100644
--- a/drivers/gpu/drm/radeon/r100.c
+++ b/drivers/gpu/drm/radeon/r100.c
@@ -65,6 +65,19 @@ MODULE_FIRMWARE(FIRMWARE_R520);
 
 #include "r100_track.h"
 
+/* This files gather functions specifics to:
+ * r100,rv100,rs100,rv200,rs200,r200,rv250,rs300,rv280
+ * and others in some cases.
+ */
+
+/**
+ * r100_wait_for_vblank - vblank wait asic callback.
+ *
+ * @rdev: radeon_device pointer
+ * @crtc: crtc to wait for vblank on
+ *
+ * Wait for vblank on the requested crtc (r1xx-r4xx).
+ */
 void r100_wait_for_vblank(struct radeon_device *rdev, int crtc)
 {
         struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc];
@@ -99,128 +112,49 @@ void r100_wait_for_vblank(struct radeon_device *rdev, int crtc)
         }
 }
 
-/* This files gather functions specifics to:
- * r100,rv100,rs100,rv200,rs200,r200,rv250,rs300,rv280
+/**
+ * r100_pre_page_flip - pre-pageflip callback.
+ *
+ * @rdev: radeon_device pointer
+ * @crtc: crtc to prepare for pageflip on
+ *
+ * Pre-pageflip callback (r1xx-r4xx).
+ * Enables the pageflip irq (vblank irq).
  */
-
-int r100_reloc_pitch_offset(struct radeon_cs_parser *p,
-                            struct radeon_cs_packet *pkt,
-                            unsigned idx,
-                            unsigned reg)
-{
-        int r;
-        u32 tile_flags = 0;
-        u32 tmp;
-        struct radeon_cs_reloc *reloc;
-        u32 value;
-
-        r = r100_cs_packet_next_reloc(p, &reloc);
-        if (r) {
-                DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-                          idx, reg);
-                r100_cs_dump_packet(p, pkt);
-                return r;
-        }
-
-        value = radeon_get_ib_value(p, idx);
-        tmp = value & 0x003fffff;
-        tmp += (((u32)reloc->lobj.gpu_offset) >> 10);
-
-        if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS)) {
-                if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO)
-                        tile_flags |= RADEON_DST_TILE_MACRO;
-                if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO) {
-                        if (reg == RADEON_SRC_PITCH_OFFSET) {
-                                DRM_ERROR("Cannot src blit from microtiled surface\n");
-                                r100_cs_dump_packet(p, pkt);
-                                return -EINVAL;
-                        }
-                        tile_flags |= RADEON_DST_TILE_MICRO;
-                }
-
-                tmp |= tile_flags;
-                p->ib.ptr[idx] = (value & 0x3fc00000) | tmp;
-        } else
-                p->ib.ptr[idx] = (value & 0xffc00000) | tmp;
-        return 0;
-}
-
-int r100_packet3_load_vbpntr(struct radeon_cs_parser *p,
-                             struct radeon_cs_packet *pkt,
-                             int idx)
-{
-        unsigned c, i;
-        struct radeon_cs_reloc *reloc;
-        struct r100_cs_track *track;
-        int r = 0;
-        volatile uint32_t *ib;
-        u32 idx_value;
-
-        ib = p->ib.ptr;
-        track = (struct r100_cs_track *)p->track;
-        c = radeon_get_ib_value(p, idx++) & 0x1F;
-        if (c > 16) {
-            DRM_ERROR("Only 16 vertex buffers are allowed %d\n",
-                      pkt->opcode);
-            r100_cs_dump_packet(p, pkt);
-            return -EINVAL;
-        }
-        track->num_arrays = c;
-        for (i = 0; i < (c - 1); i+=2, idx+=3) {
-                r = r100_cs_packet_next_reloc(p, &reloc);
-                if (r) {
-                        DRM_ERROR("No reloc for packet3 %d\n",
-                                  pkt->opcode);
-                        r100_cs_dump_packet(p, pkt);
-                        return r;
-                }
-                idx_value = radeon_get_ib_value(p, idx);
-                ib[idx+1] = radeon_get_ib_value(p, idx + 1) + ((u32)reloc->lobj.gpu_offset);
-
-                track->arrays[i + 0].esize = idx_value >> 8;
-                track->arrays[i + 0].robj = reloc->robj;
-                track->arrays[i + 0].esize &= 0x7F;
-                r = r100_cs_packet_next_reloc(p, &reloc);
-                if (r) {
-                        DRM_ERROR("No reloc for packet3 %d\n",
-                                  pkt->opcode);
-                        r100_cs_dump_packet(p, pkt);
-                        return r;
-                }
-                ib[idx+2] = radeon_get_ib_value(p, idx + 2) + ((u32)reloc->lobj.gpu_offset);
-                track->arrays[i + 1].robj = reloc->robj;
-                track->arrays[i + 1].esize = idx_value >> 24;
-                track->arrays[i + 1].esize &= 0x7F;
-        }
-        if (c & 1) {
-                r = r100_cs_packet_next_reloc(p, &reloc);
-                if (r) {
-                        DRM_ERROR("No reloc for packet3 %d\n",
-                                          pkt->opcode);
-                        r100_cs_dump_packet(p, pkt);
-                        return r;
-                }
-                idx_value = radeon_get_ib_value(p, idx);
-                ib[idx+1] = radeon_get_ib_value(p, idx + 1) + ((u32)reloc->lobj.gpu_offset);
-                track->arrays[i + 0].robj = reloc->robj;
-                track->arrays[i + 0].esize = idx_value >> 8;
-                track->arrays[i + 0].esize &= 0x7F;
-        }
-        return r;
-}
-
 void r100_pre_page_flip(struct radeon_device *rdev, int crtc)
 {
         /* enable the pflip int */
         radeon_irq_kms_pflip_irq_get(rdev, crtc);
 }
 
+/**
+ * r100_post_page_flip - pos-pageflip callback.
+ *
+ * @rdev: radeon_device pointer
+ * @crtc: crtc to cleanup pageflip on
+ *
+ * Post-pageflip callback (r1xx-r4xx).
+ * Disables the pageflip irq (vblank irq).
+ */
 void r100_post_page_flip(struct radeon_device *rdev, int crtc)
 {
         /* disable the pflip int */
         radeon_irq_kms_pflip_irq_put(rdev, crtc);
 }
 
+/**
+ * r100_page_flip - pageflip callback.
+ *
+ * @rdev: radeon_device pointer
+ * @crtc_id: crtc to cleanup pageflip on
+ * @crtc_base: new address of the crtc (GPU MC address)
+ *
+ * Does the actual pageflip (r1xx-r4xx).
+ * During vblank we take the crtc lock and wait for the update_pending
+ * bit to go high, when it does, we release the lock, and allow the
+ * double buffered update to take place.
+ * Returns the current update pending status.
+ */
 u32 r100_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
 {
         struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc_id];
@@ -247,6 +181,15 @@ u32 r100_page_flip(struct radeon_device *rdev, int crtc_id, u64 crtc_base)
         return RREG32(RADEON_CRTC_OFFSET + radeon_crtc->crtc_offset) & RADEON_CRTC_OFFSET__GUI_TRIG_OFFSET;
 }
 
+/**
+ * r100_pm_get_dynpm_state - look up dynpm power state callback.
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Look up the optimal power state based on the
+ * current state of the GPU (r1xx-r5xx).
+ * Used for dynpm only.
+ */
 void r100_pm_get_dynpm_state(struct radeon_device *rdev)
 {
         int i;
@@ -329,6 +272,15 @@ void r100_pm_get_dynpm_state(struct radeon_device *rdev)
                   pcie_lanes);
 }
 
+/**
+ * r100_pm_init_profile - Initialize power profiles callback.
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Initialize the power states used in profile mode
+ * (r1xx-r3xx).
+ * Used for profile mode only.
+ */
 void r100_pm_init_profile(struct radeon_device *rdev)
 {
         /* default */
@@ -368,6 +320,14 @@ void r100_pm_init_profile(struct radeon_device *rdev)
         rdev->pm.profiles[PM_PROFILE_HIGH_MH_IDX].dpms_on_cm_idx = 0;
 }
 
+/**
+ * r100_pm_misc - set additional pm hw parameters callback.
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Set non-clock parameters associated with a power state
+ * (voltage, pcie lanes, etc.) (r1xx-r4xx).
+ */
 void r100_pm_misc(struct radeon_device *rdev)
 {
         int requested_index = rdev->pm.requested_power_state_index;
@@ -459,6 +419,13 @@ void r100_pm_misc(struct radeon_device *rdev)
         }
 }
 
+/**
+ * r100_pm_prepare - pre-power state change callback.
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Prepare for a power state change (r1xx-r4xx).
+ */
 void r100_pm_prepare(struct radeon_device *rdev)
 {
         struct drm_device *ddev = rdev->ddev;
@@ -483,6 +450,13 @@ void r100_pm_prepare(struct radeon_device *rdev)
         }
 }
 
+/**
+ * r100_pm_finish - post-power state change callback.
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Clean up after a power state change (r1xx-r4xx).
+ */
 void r100_pm_finish(struct radeon_device *rdev)
 {
         struct drm_device *ddev = rdev->ddev;
@@ -507,6 +481,14 @@ void r100_pm_finish(struct radeon_device *rdev)
         }
 }
 
+/**
+ * r100_gui_idle - gui idle callback.
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Check of the GUI (2D/3D engines) are idle (r1xx-r5xx).
+ * Returns true if idle, false if not.
+ */
 bool r100_gui_idle(struct radeon_device *rdev)
 {
         if (RREG32(RADEON_RBBM_STATUS) & RADEON_RBBM_ACTIVE)
@@ -516,6 +498,15 @@ bool r100_gui_idle(struct radeon_device *rdev)
 }
 
 /* hpd for digital panel detect/disconnect */
+/**
+ * r100_hpd_sense - hpd sense callback.
+ *
+ * @rdev: radeon_device pointer
+ * @hpd: hpd (hotplug detect) pin
+ *
+ * Checks if a digital monitor is connected (r1xx-r4xx).
+ * Returns true if connected, false if not connected.
+ */
 bool r100_hpd_sense(struct radeon_device *rdev, enum radeon_hpd_id hpd)
 {
         bool connected = false;
@@ -535,6 +526,14 @@ bool r100_hpd_sense(struct radeon_device *rdev, enum radeon_hpd_id hpd)
         return connected;
 }
 
+/**
+ * r100_hpd_set_polarity - hpd set polarity callback.
+ *
+ * @rdev: radeon_device pointer
+ * @hpd: hpd (hotplug detect) pin
+ *
+ * Set the polarity of the hpd pin (r1xx-r4xx).
+ */
 void r100_hpd_set_polarity(struct radeon_device *rdev,
                            enum radeon_hpd_id hpd)
 {
@@ -563,47 +562,47 @@ void r100_hpd_set_polarity(struct radeon_device *rdev,
         }
 }
 
+/**
+ * r100_hpd_init - hpd setup callback.
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Setup the hpd pins used by the card (r1xx-r4xx).
+ * Set the polarity, and enable the hpd interrupts.
+ */
 void r100_hpd_init(struct radeon_device *rdev)
 {
         struct drm_device *dev = rdev->ddev;
         struct drm_connector *connector;
+        unsigned enable = 0;
 
         list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
                 struct radeon_connector *radeon_connector = to_radeon_connector(connector);
-                switch (radeon_connector->hpd.hpd) {
-                case RADEON_HPD_1:
-                        rdev->irq.hpd[0] = true;
-                        break;
-                case RADEON_HPD_2:
-                        rdev->irq.hpd[1] = true;
-                        break;
-                default:
-                        break;
-                }
+                enable |= 1 << radeon_connector->hpd.hpd;
                 radeon_hpd_set_polarity(rdev, radeon_connector->hpd.hpd);
         }
-        if (rdev->irq.installed)
-                r100_irq_set(rdev);
+        radeon_irq_kms_enable_hpd(rdev, enable);
 }
 
+/**
+ * r100_hpd_fini - hpd tear down callback.
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Tear down the hpd pins used by the card (r1xx-r4xx).
+ * Disable the hpd interrupts.
+ */
 void r100_hpd_fini(struct radeon_device *rdev)
 {
         struct drm_device *dev = rdev->ddev;
         struct drm_connector *connector;
+        unsigned disable = 0;
 
         list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
                 struct radeon_connector *radeon_connector = to_radeon_connector(connector);
-                switch (radeon_connector->hpd.hpd) {
-                case RADEON_HPD_1:
-                        rdev->irq.hpd[0] = false;
-                        break;
-                case RADEON_HPD_2:
-                        rdev->irq.hpd[1] = false;
-                        break;
-                default:
-                        break;
-                }
+                disable |= 1 << radeon_connector->hpd.hpd;
         }
+        radeon_irq_kms_disable_hpd(rdev, disable);
 }
 
 /*
@@ -635,15 +634,6 @@ int r100_pci_gart_init(struct radeon_device *rdev)
         return radeon_gart_table_ram_alloc(rdev);
 }
 
-/* required on r1xx, r2xx, r300, r(v)350, r420/r481, rs400/rs480 */
-void r100_enable_bm(struct radeon_device *rdev)
-{
-        uint32_t tmp;
-        /* Enable bus mastering */
-        tmp = RREG32(RADEON_BUS_CNTL) & ~RADEON_BUS_MASTER_DIS;
-        WREG32(RADEON_BUS_CNTL, tmp);
-}
-
 int r100_pci_gart_enable(struct radeon_device *rdev)
 {
         uint32_t tmp;
@@ -705,18 +695,18 @@ int r100_irq_set(struct radeon_device *rdev)
                 WREG32(R_000040_GEN_INT_CNTL, 0);
                 return -EINVAL;
         }
-        if (rdev->irq.sw_int[RADEON_RING_TYPE_GFX_INDEX]) {
+        if (atomic_read(&rdev->irq.ring_int[RADEON_RING_TYPE_GFX_INDEX])) {
                 tmp |= RADEON_SW_INT_ENABLE;
         }
         if (rdev->irq.gui_idle) {
                 tmp |= RADEON_GUI_IDLE_MASK;
         }
         if (rdev->irq.crtc_vblank_int[0] ||
-            rdev->irq.pflip[0]) {
+            atomic_read(&rdev->irq.pflip[0])) {
                 tmp |= RADEON_CRTC_VBLANK_MASK;
         }
         if (rdev->irq.crtc_vblank_int[1] ||
-            rdev->irq.pflip[1]) {
+            atomic_read(&rdev->irq.pflip[1])) {
                 tmp |= RADEON_CRTC2_VBLANK_MASK;
         }
         if (rdev->irq.hpd[0]) {
@@ -782,7 +772,6 @@ int r100_irq_process(struct radeon_device *rdev)
                 /* gui idle interrupt */
                 if (status & RADEON_GUI_IDLE_STAT) {
                         rdev->irq.gui_idle_acked = true;
-                        rdev->pm.gui_idle = true;
                         wake_up(&rdev->irq.idle_queue);
                 }
                 /* Vertical blank interrupts */
@@ -792,7 +781,7 @@ int r100_irq_process(struct radeon_device *rdev)
                                 rdev->pm.vblank_sync = true;
                                 wake_up(&rdev->irq.vblank_queue);
                         }
-                        if (rdev->irq.pflip[0])
+                        if (atomic_read(&rdev->irq.pflip[0]))
                                 radeon_crtc_handle_flip(rdev, 0);
                 }
                 if (status & RADEON_CRTC2_VBLANK_STAT) {
@@ -801,7 +790,7 @@ int r100_irq_process(struct radeon_device *rdev)
                                 rdev->pm.vblank_sync = true;
                                 wake_up(&rdev->irq.vblank_queue);
                         }
-                        if (rdev->irq.pflip[1])
+                        if (atomic_read(&rdev->irq.pflip[1]))
                                 radeon_crtc_handle_flip(rdev, 1);
                 }
                 if (status & RADEON_FP_DETECT_STAT) {
@@ -883,7 +872,7 @@ int r100_copy_blit(struct radeon_device *rdev,
                    uint64_t src_offset,
                    uint64_t dst_offset,
                    unsigned num_gpu_pages,
-                   struct radeon_fence *fence)
+                   struct radeon_fence **fence)
 {
         struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
         uint32_t cur_pages;
@@ -947,7 +936,7 @@ int r100_copy_blit(struct radeon_device *rdev,
                           RADEON_WAIT_HOST_IDLECLEAN |
                           RADEON_WAIT_DMA_GUI_IDLE);
         if (fence) {
-                r = radeon_fence_emit(rdev, fence);
+                r = radeon_fence_emit(rdev, fence, RADEON_RING_TYPE_GFX_INDEX);
         }
         radeon_ring_unlock_commit(rdev, ring);
         return r;
@@ -1192,6 +1181,14 @@ int r100_cp_init(struct radeon_device *rdev, unsigned ring_size)
         }
         ring->ready = true;
         radeon_ttm_set_active_vram_size(rdev, rdev->mc.real_vram_size);
+
+        if (radeon_ring_supports_scratch_reg(rdev, ring)) {
+                r = radeon_scratch_get(rdev, &ring->rptr_save_reg);
+                if (r) {
+                        DRM_ERROR("failed to get scratch reg for rptr save (%d).\n", r);
+                        ring->rptr_save_reg = 0;
+                }
+        }
         return 0;
 }
 
@@ -1202,6 +1199,7 @@ void r100_cp_fini(struct radeon_device *rdev)
         }
         /* Disable ring */
         r100_cp_disable(rdev);
+        radeon_scratch_free(rdev, rdev->ring[RADEON_RING_TYPE_GFX_INDEX].rptr_save_reg);
         radeon_ring_fini(rdev, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]);
         DRM_INFO("radeon: cp finalized\n");
 }
@@ -1223,6 +1221,112 @@ void r100_cp_disable(struct radeon_device *rdev)
 /*
  * CS functions
  */
+int r100_reloc_pitch_offset(struct radeon_cs_parser *p,
+                            struct radeon_cs_packet *pkt,
+                            unsigned idx,
+                            unsigned reg)
+{
+        int r;
+        u32 tile_flags = 0;
+        u32 tmp;
+        struct radeon_cs_reloc *reloc;
+        u32 value;
+
+        r = r100_cs_packet_next_reloc(p, &reloc);
+        if (r) {
+                DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
+                          idx, reg);
+                r100_cs_dump_packet(p, pkt);
+                return r;
+        }
+
+        value = radeon_get_ib_value(p, idx);
+        tmp = value & 0x003fffff;
+        tmp += (((u32)reloc->lobj.gpu_offset) >> 10);
+
+        if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS)) {
+                if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO)
+                        tile_flags |= RADEON_DST_TILE_MACRO;
+                if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO) {
+                        if (reg == RADEON_SRC_PITCH_OFFSET) {
+                                DRM_ERROR("Cannot src blit from microtiled surface\n");
+                                r100_cs_dump_packet(p, pkt);
+                                return -EINVAL;
+                        }
+                        tile_flags |= RADEON_DST_TILE_MICRO;
+                }
+
+                tmp |= tile_flags;
+                p->ib.ptr[idx] = (value & 0x3fc00000) | tmp;
+        } else
+                p->ib.ptr[idx] = (value & 0xffc00000) | tmp;
+        return 0;
+}
+
+int r100_packet3_load_vbpntr(struct radeon_cs_parser *p,
+                             struct radeon_cs_packet *pkt,
+                             int idx)
+{
+        unsigned c, i;
+        struct radeon_cs_reloc *reloc;
+        struct r100_cs_track *track;
+        int r = 0;
+        volatile uint32_t *ib;
+        u32 idx_value;
+
+        ib = p->ib.ptr;
+        track = (struct r100_cs_track *)p->track;
+        c = radeon_get_ib_value(p, idx++) & 0x1F;
+        if (c > 16) {
+            DRM_ERROR("Only 16 vertex buffers are allowed %d\n",
+                      pkt->opcode);
+            r100_cs_dump_packet(p, pkt);
+            return -EINVAL;
+        }
+        track->num_arrays = c;
+        for (i = 0; i < (c - 1); i+=2, idx+=3) {
+                r = r100_cs_packet_next_reloc(p, &reloc);
+                if (r) {
+                        DRM_ERROR("No reloc for packet3 %d\n",
+                                  pkt->opcode);
+                        r100_cs_dump_packet(p, pkt);
+                        return r;
+                }
+                idx_value = radeon_get_ib_value(p, idx);
+                ib[idx+1] = radeon_get_ib_value(p, idx + 1) + ((u32)reloc->lobj.gpu_offset);
+
+                track->arrays[i + 0].esize = idx_value >> 8;
+                track->arrays[i + 0].robj = reloc->robj;
+                track->arrays[i + 0].esize &= 0x7F;
+                r = r100_cs_packet_next_reloc(p, &reloc);
+                if (r) {
+                        DRM_ERROR("No reloc for packet3 %d\n",
+                                  pkt->opcode);
+                        r100_cs_dump_packet(p, pkt);
+                        return r;
+                }
+                ib[idx+2] = radeon_get_ib_value(p, idx + 2) + ((u32)reloc->lobj.gpu_offset);
+                track->arrays[i + 1].robj = reloc->robj;
+                track->arrays[i + 1].esize = idx_value >> 24;
+                track->arrays[i + 1].esize &= 0x7F;
+        }
+        if (c & 1) {
+                r = r100_cs_packet_next_reloc(p, &reloc);
+                if (r) {
+                        DRM_ERROR("No reloc for packet3 %d\n",
+                                          pkt->opcode);
+                        r100_cs_dump_packet(p, pkt);
+                        return r;
+                }
+                idx_value = radeon_get_ib_value(p, idx);
+                ib[idx+1] = radeon_get_ib_value(p, idx + 1) + ((u32)reloc->lobj.gpu_offset);
+                track->arrays[i + 0].robj = reloc->robj;
+                track->arrays[i + 0].esize = idx_value >> 8;
+                track->arrays[i + 0].esize &= 0x7F;
+        }
+        return r;
+}
+
 int r100_cs_parse_packet0(struct radeon_cs_parser *p,
                           struct radeon_cs_packet *pkt,
                           const unsigned *auth, unsigned n,
@@ -2048,6 +2152,379 @@ int r100_cs_parse(struct radeon_cs_parser *p)
         return 0;
 }
 
+static void r100_cs_track_texture_print(struct r100_cs_track_texture *t)
+{
+        DRM_ERROR("pitch                      %d\n", t->pitch);
+        DRM_ERROR("use_pitch                  %d\n", t->use_pitch);
+        DRM_ERROR("width                      %d\n", t->width);
+        DRM_ERROR("width_11                   %d\n", t->width_11);
+        DRM_ERROR("height                     %d\n", t->height);
+        DRM_ERROR("height_11                  %d\n", t->height_11);
+        DRM_ERROR("num levels                 %d\n", t->num_levels);
+        DRM_ERROR("depth                      %d\n", t->txdepth);
+        DRM_ERROR("bpp                        %d\n", t->cpp);
+        DRM_ERROR("coordinate type            %d\n", t->tex_coord_type);
+        DRM_ERROR("width round to power of 2  %d\n", t->roundup_w);
+        DRM_ERROR("height round to power of 2 %d\n", t->roundup_h);
+        DRM_ERROR("compress format            %d\n", t->compress_format);
+}
+
+static int r100_track_compress_size(int compress_format, int w, int h)
+{
+        int block_width, block_height, block_bytes;
+        int wblocks, hblocks;
+        int min_wblocks;
+        int sz;
+
+        block_width = 4;
+        block_height = 4;
+
+        switch (compress_format) {
+        case R100_TRACK_COMP_DXT1:
+                block_bytes = 8;
+                min_wblocks = 4;
+                break;
+        default:
+        case R100_TRACK_COMP_DXT35:
+                block_bytes = 16;
+                min_wblocks = 2;
+                break;
+        }
+
+        hblocks = (h + block_height - 1) / block_height;
+        wblocks = (w + block_width - 1) / block_width;
+        if (wblocks < min_wblocks)
+                wblocks = min_wblocks;
+        sz = wblocks * hblocks * block_bytes;
+        return sz;
+}
+
+static int r100_cs_track_cube(struct radeon_device *rdev,
+                              struct r100_cs_track *track, unsigned idx)
+{
+        unsigned face, w, h;
+        struct radeon_bo *cube_robj;
+        unsigned long size;
+        unsigned compress_format = track->textures[idx].compress_format;
+
+        for (face = 0; face < 5; face++) {
+                cube_robj = track->textures[idx].cube_info[face].robj;
+                w = track->textures[idx].cube_info[face].width;
+                h = track->textures[idx].cube_info[face].height;
+
+                if (compress_format) {
+                        size = r100_track_compress_size(compress_format, w, h);
+                } else
+                        size = w * h;
+                size *= track->textures[idx].cpp;
+
+                size += track->textures[idx].cube_info[face].offset;
+
+                if (size > radeon_bo_size(cube_robj)) {
+                        DRM_ERROR("Cube texture offset greater than object size %lu %lu\n",
+                                  size, radeon_bo_size(cube_robj));
+                        r100_cs_track_texture_print(&track->textures[idx]);
+                        return -1;
+                }
+        }
+        return 0;
+}
+
+static int r100_cs_track_texture_check(struct radeon_device *rdev,
+                                       struct r100_cs_track *track)
+{
+        struct radeon_bo *robj;
+        unsigned long size;
+        unsigned u, i, w, h, d;
+        int ret;
+
+        for (u = 0; u < track->num_texture; u++) {
+                if (!track->textures[u].enabled)
+                        continue;
+                if (track->textures[u].lookup_disable)
+                        continue;
+                robj = track->textures[u].robj;
+                if (robj == NULL) {
+                        DRM_ERROR("No texture bound to unit %u\n", u);
+                        return -EINVAL;
+                }
+                size = 0;
+                for (i = 0; i <= track->textures[u].num_levels; i++) {
+                        if (track->textures[u].use_pitch) {
+                                if (rdev->family < CHIP_R300)
+                                        w = (track->textures[u].pitch / track->textures[u].cpp) / (1 << i);
+                                else
+                                        w = track->textures[u].pitch / (1 << i);
+                        } else {
+                                w = track->textures[u].width;
+                                if (rdev->family >= CHIP_RV515)
+                                        w |= track->textures[u].width_11;
+                                w = w / (1 << i);
+                                if (track->textures[u].roundup_w)
+                                        w = roundup_pow_of_two(w);
+                        }
+                        h = track->textures[u].height;
+                        if (rdev->family >= CHIP_RV515)
+                                h |= track->textures[u].height_11;
+                        h = h / (1 << i);
+                        if (track->textures[u].roundup_h)
+                                h = roundup_pow_of_two(h);
+                        if (track->textures[u].tex_coord_type == 1) {
+                                d = (1 << track->textures[u].txdepth) / (1 << i);
+                                if (!d)
+                                        d = 1;
+                        } else {
+                                d = 1;
+                        }
+                        if (track->textures[u].compress_format) {
+
+                                size += r100_track_compress_size(track->textures[u].compress_format, w, h) * d;
+                                /* compressed textures are block based */
+                        } else
+                                size += w * h * d;
+                }
+                size *= track->textures[u].cpp;
+
+                switch (track->textures[u].tex_coord_type) {
+                case 0:
+                case 1:
+                        break;
+                case 2:
+                        if (track->separate_cube) {
+                                ret = r100_cs_track_cube(rdev, track, u);
+                                if (ret)
+                                        return ret;
+                        } else
+                                size *= 6;
+                        break;
+                default:
+                        DRM_ERROR("Invalid texture coordinate type %u for unit "
+                                  "%u\n", track->textures[u].tex_coord_type, u);
+                        return -EINVAL;
+                }
+                if (size > radeon_bo_size(robj)) {
+                        DRM_ERROR("Texture of unit %u needs %lu bytes but is "
+                                  "%lu\n", u, size, radeon_bo_size(robj));
+                        r100_cs_track_texture_print(&track->textures[u]);
+                        return -EINVAL;
+                }
+        }
+        return 0;
+}
+
+int r100_cs_track_check(struct radeon_device *rdev, struct r100_cs_track *track)
+{
+        unsigned i;
+        unsigned long size;
+        unsigned prim_walk;
+        unsigned nverts;
+        unsigned num_cb = track->cb_dirty ? track->num_cb : 0;
+
+        if (num_cb && !track->zb_cb_clear && !track->color_channel_mask &&
+            !track->blend_read_enable)
+                num_cb = 0;
+
+        for (i = 0; i < num_cb; i++) {
+                if (track->cb[i].robj == NULL) {
+                        DRM_ERROR("[drm] No buffer for color buffer %d !\n", i);
+                        return -EINVAL;
+                }
+                size = track->cb[i].pitch * track->cb[i].cpp * track->maxy;
+                size += track->cb[i].offset;
+                if (size > radeon_bo_size(track->cb[i].robj)) {
+                        DRM_ERROR("[drm] Buffer too small for color buffer %d "
+                                  "(need %lu have %lu) !\n", i, size,
+                                  radeon_bo_size(track->cb[i].robj));
+                        DRM_ERROR("[drm] color buffer %d (%u %u %u %u)\n",
+                                  i, track->cb[i].pitch, track->cb[i].cpp,
+                                  track->cb[i].offset, track->maxy);
+                        return -EINVAL;
+                }
+        }
+        track->cb_dirty = false;
+
+        if (track->zb_dirty && track->z_enabled) {
+                if (track->zb.robj == NULL) {
+                        DRM_ERROR("[drm] No buffer for z buffer !\n");
+                        return -EINVAL;
+                }
+                size = track->zb.pitch * track->zb.cpp * track->maxy;
+                size += track->zb.offset;
+                if (size > radeon_bo_size(track->zb.robj)) {
+                        DRM_ERROR("[drm] Buffer too small for z buffer "
+                                  "(need %lu have %lu) !\n", size,
+                                  radeon_bo_size(track->zb.robj));
+                        DRM_ERROR("[drm] zbuffer (%u %u %u %u)\n",
+                                  track->zb.pitch, track->zb.cpp,
+                                  track->zb.offset, track->maxy);
+                        return -EINVAL;
+                }
+        }
+        track->zb_dirty = false;
+
+        if (track->aa_dirty && track->aaresolve) {
+                if (track->aa.robj == NULL) {
+                        DRM_ERROR("[drm] No buffer for AA resolve buffer %d !\n", i);
+                        return -EINVAL;
+                }
+                /* I believe the format comes from colorbuffer0. */
+                size = track->aa.pitch * track->cb[0].cpp * track->maxy;
+                size += track->aa.offset;
+                if (size > radeon_bo_size(track->aa.robj)) {
+                        DRM_ERROR("[drm] Buffer too small for AA resolve buffer %d "
+                                  "(need %lu have %lu) !\n", i, size,
+                                  radeon_bo_size(track->aa.robj));
+                        DRM_ERROR("[drm] AA resolve buffer %d (%u %u %u %u)\n",
+                                  i, track->aa.pitch, track->cb[0].cpp,
+                                  track->aa.offset, track->maxy);
+                        return -EINVAL;
+                }
+        }
+        track->aa_dirty = false;
+
+        prim_walk = (track->vap_vf_cntl >> 4) & 0x3;
+        if (track->vap_vf_cntl & (1 << 14)) {
+                nverts = track->vap_alt_nverts;
+        } else {
+                nverts = (track->vap_vf_cntl >> 16) & 0xFFFF;
+        }
+        switch (prim_walk) {
+        case 1:
+                for (i = 0; i < track->num_arrays; i++) {
+                        size = track->arrays[i].esize * track->max_indx * 4;
+                        if (track->arrays[i].robj == NULL) {
+                                DRM_ERROR("(PW %u) Vertex array %u no buffer "
+                                          "bound\n", prim_walk, i);
+                                return -EINVAL;
+                        }
+                        if (size > radeon_bo_size(track->arrays[i].robj)) {
+                                dev_err(rdev->dev, "(PW %u) Vertex array %u "
+                                        "need %lu dwords have %lu dwords\n",
+                                        prim_walk, i, size >> 2,
+                                        radeon_bo_size(track->arrays[i].robj)
+                                        >> 2);
+                                DRM_ERROR("Max indices %u\n", track->max_indx);
+                                return -EINVAL;
+                        }
+                }
+                break;
+        case 2:
+                for (i = 0; i < track->num_arrays; i++) {
+                        size = track->arrays[i].esize * (nverts - 1) * 4;
+                        if (track->arrays[i].robj == NULL) {
+                                DRM_ERROR("(PW %u) Vertex array %u no buffer "
+                                          "bound\n", prim_walk, i);
+                                return -EINVAL;
+                        }
+                        if (size > radeon_bo_size(track->arrays[i].robj)) {
+                                dev_err(rdev->dev, "(PW %u) Vertex array %u "
+                                        "need %lu dwords have %lu dwords\n",
+                                        prim_walk, i, size >> 2,
+                                        radeon_bo_size(track->arrays[i].robj)
+                                        >> 2);
+                                return -EINVAL;
+                        }
+                }
+                break;
+        case 3:
+                size = track->vtx_size * nverts;
+                if (size != track->immd_dwords) {
+                        DRM_ERROR("IMMD draw %u dwors but needs %lu dwords\n",
+                                  track->immd_dwords, size);
+                        DRM_ERROR("VAP_VF_CNTL.NUM_VERTICES %u, VTX_SIZE %u\n",
+                                  nverts, track->vtx_size);
+                        return -EINVAL;
+                }
+                break;
+        default:
+                DRM_ERROR("[drm] Invalid primitive walk %d for VAP_VF_CNTL\n",
+                          prim_walk);
+                return -EINVAL;
+        }
+
+        if (track->tex_dirty) {
+                track->tex_dirty = false;
+                return r100_cs_track_texture_check(rdev, track);
+        }
+        return 0;
+}
+
+void r100_cs_track_clear(struct radeon_device *rdev, struct r100_cs_track *track)
+{
+        unsigned i, face;
+
+        track->cb_dirty = true;
+        track->zb_dirty = true;
+        track->tex_dirty = true;
+        track->aa_dirty = true;
+
+        if (rdev->family < CHIP_R300) {
+                track->num_cb = 1;
+                if (rdev->family <= CHIP_RS200)
+                        track->num_texture = 3;
+                else
+                        track->num_texture = 6;
+                track->maxy = 2048;
+                track->separate_cube = 1;
+        } else {
+                track->num_cb = 4;
+                track->num_texture = 16;
+                track->maxy = 4096;
+                track->separate_cube = 0;
+                track->aaresolve = false;
+                track->aa.robj = NULL;
+        }
+
+        for (i = 0; i < track->num_cb; i++) {
+                track->cb[i].robj = NULL;
+                track->cb[i].pitch = 8192;
+                track->cb[i].cpp = 16;
+                track->cb[i].offset = 0;
+        }
+        track->z_enabled = true;
+        track->zb.robj = NULL;
+        track->zb.pitch = 8192;
+        track->zb.cpp = 4;
+        track->zb.offset = 0;
+        track->vtx_size = 0x7F;
+        track->immd_dwords = 0xFFFFFFFFUL;
+        track->num_arrays = 11;
+        track->max_indx = 0x00FFFFFFUL;
+        for (i = 0; i < track->num_arrays; i++) {
+                track->arrays[i].robj = NULL;
+                track->arrays[i].esize = 0x7F;
+        }
+        for (i = 0; i < track->num_texture; i++) {
+                track->textures[i].compress_format = R100_TRACK_COMP_NONE;
+                track->textures[i].pitch = 16536;
+                track->textures[i].width = 16536;
+                track->textures[i].height = 16536;
+                track->textures[i].width_11 = 1 << 11;
+                track->textures[i].height_11 = 1 << 11;
+                track->textures[i].num_levels = 12;
+                if (rdev->family <= CHIP_RS200) {
+                        track->textures[i].tex_coord_type = 0;
+                        track->textures[i].txdepth = 0;
+                } else {
+                        track->textures[i].txdepth = 16;
+                        track->textures[i].tex_coord_type = 1;
+                }
+                track->textures[i].cpp = 64;
+                track->textures[i].robj = NULL;
+                /* CS IB emission code makes sure texture unit are disabled */
+                track->textures[i].enabled = false;
+                track->textures[i].lookup_disable = false;
+                track->textures[i].roundup_w = true;
+                track->textures[i].roundup_h = true;
+                if (track->separate_cube)
+                        for (face = 0; face < 5; face++) {
+                                track->textures[i].cube_info[face].robj = NULL;
+                                track->textures[i].cube_info[face].width = 16536;
+                                track->textures[i].cube_info[face].height = 16536;
+                                track->textures[i].cube_info[face].offset = 0;
+                        }
+        }
+}
 
 /*
  * Global GPU functions
@@ -2175,6 +2652,15 @@ bool r100_gpu_is_lockup(struct radeon_device *rdev, struct radeon_ring *ring)
         return radeon_ring_test_lockup(rdev, ring);
 }
 
+/* required on r1xx, r2xx, r300, r(v)350, r420/r481, rs400/rs480 */
+void r100_enable_bm(struct radeon_device *rdev)
+{
+        uint32_t tmp;
+        /* Enable bus mastering */
+        tmp = RREG32(RADEON_BUS_CNTL) & ~RADEON_BUS_MASTER_DIS;
+        WREG32(RADEON_BUS_CNTL, tmp);
+}
+
 void r100_bm_disable(struct radeon_device *rdev)
 {
         u32 tmp;
@@ -3261,380 +3747,6 @@ void r100_bandwidth_update(struct radeon_device *rdev)
         }
 }
 
-static void r100_cs_track_texture_print(struct r100_cs_track_texture *t)
-{
-        DRM_ERROR("pitch                      %d\n", t->pitch);
-        DRM_ERROR("use_pitch                  %d\n", t->use_pitch);
-        DRM_ERROR("width                      %d\n", t->width);
-        DRM_ERROR("width_11                   %d\n", t->width_11);
-        DRM_ERROR("height                     %d\n", t->height);
-        DRM_ERROR("height_11                  %d\n", t->height_11);
-        DRM_ERROR("num levels                 %d\n", t->num_levels);
-        DRM_ERROR("depth                      %d\n", t->txdepth);
-        DRM_ERROR("bpp                        %d\n", t->cpp);
-        DRM_ERROR("coordinate type            %d\n", t->tex_coord_type);
-        DRM_ERROR("width round to power of 2  %d\n", t->roundup_w);
-        DRM_ERROR("height round to power of 2 %d\n", t->roundup_h);
-        DRM_ERROR("compress format            %d\n", t->compress_format);
-}
-
-static int r100_track_compress_size(int compress_format, int w, int h)
-{
-        int block_width, block_height, block_bytes;
-        int wblocks, hblocks;
-        int min_wblocks;
-        int sz;
-
-        block_width = 4;
-        block_height = 4;
-
-        switch (compress_format) {
-        case R100_TRACK_COMP_DXT1:
-                block_bytes = 8;
-                min_wblocks = 4;
-                break;
-        default:
-        case R100_TRACK_COMP_DXT35:
-                block_bytes = 16;
-                min_wblocks = 2;
-                break;
-        }
-
-        hblocks = (h + block_height - 1) / block_height;
-        wblocks = (w + block_width - 1) / block_width;
-        if (wblocks < min_wblocks)
-                wblocks = min_wblocks;
-        sz = wblocks * hblocks * block_bytes;
-        return sz;
-}
-
-static int r100_cs_track_cube(struct radeon_device *rdev,
-                              struct r100_cs_track *track, unsigned idx)
-{
-        unsigned face, w, h;
-        struct radeon_bo *cube_robj;
-        unsigned long size;
-        unsigned compress_format = track->textures[idx].compress_format;
-
-        for (face = 0; face < 5; face++) {
-                cube_robj = track->textures[idx].cube_info[face].robj;
-                w = track->textures[idx].cube_info[face].width;
-                h = track->textures[idx].cube_info[face].height;
-
-                if (compress_format) {
-                        size = r100_track_compress_size(compress_format, w, h);
-                } else
-                        size = w * h;
-                size *= track->textures[idx].cpp;
-
-                size += track->textures[idx].cube_info[face].offset;
-
-                if (size > radeon_bo_size(cube_robj)) {
-                        DRM_ERROR("Cube texture offset greater than object size %lu %lu\n",
-                                  size, radeon_bo_size(cube_robj));
-                        r100_cs_track_texture_print(&track->textures[idx]);
-                        return -1;
-                }
-        }
-        return 0;
-}
-
-static int r100_cs_track_texture_check(struct radeon_device *rdev,
-                                       struct r100_cs_track *track)
-{
-        struct radeon_bo *robj;
-        unsigned long size;
-        unsigned u, i, w, h, d;
-        int ret;
-
-        for (u = 0; u < track->num_texture; u++) {
-                if (!track->textures[u].enabled)
-                        continue;
-                if (track->textures[u].lookup_disable)
-                        continue;
-                robj = track->textures[u].robj;
-                if (robj == NULL) {
-                        DRM_ERROR("No texture bound to unit %u\n", u);
-                        return -EINVAL;
-                }
-                size = 0;
-                for (i = 0; i <= track->textures[u].num_levels; i++) {
-                        if (track->textures[u].use_pitch) {
-                                if (rdev->family < CHIP_R300)
-                                        w = (track->textures[u].pitch / track->textures[u].cpp) / (1 << i);
-                                else
-                                        w = track->textures[u].pitch / (1 << i);
-                        } else {
-                                w = track->textures[u].width;
-                                if (rdev->family >= CHIP_RV515)
-                                        w |= track->textures[u].width_11;
-                                w = w / (1 << i);
-                                if (track->textures[u].roundup_w)
-                                        w = roundup_pow_of_two(w);
-                        }
-                        h = track->textures[u].height;
-                        if (rdev->family >= CHIP_RV515)
-                                h |= track->textures[u].height_11;
-                        h = h / (1 << i);
-                        if (track->textures[u].roundup_h)
-                                h = roundup_pow_of_two(h);
-                        if (track->textures[u].tex_coord_type == 1) {
-                                d = (1 << track->textures[u].txdepth) / (1 << i);
-                                if (!d)
-                                        d = 1;
-                        } else {
-                                d = 1;
-                        }
-                        if (track->textures[u].compress_format) {
-
-                                size += r100_track_compress_size(track->textures[u].compress_format, w, h) * d;
-                                /* compressed textures are block based */
-                        } else
-                                size += w * h * d;
-                }
-                size *= track->textures[u].cpp;
-
-                switch (track->textures[u].tex_coord_type) {
-                case 0:
-                case 1:
-                        break;
-                case 2:
-                        if (track->separate_cube) {
-                                ret = r100_cs_track_cube(rdev, track, u);
-                                if (ret)
-                                        return ret;
-                        } else
-                                size *= 6;
-                        break;
-                default:
-                        DRM_ERROR("Invalid texture coordinate type %u for unit "
-                                  "%u\n", track->textures[u].tex_coord_type, u);
-                        return -EINVAL;
-                }
-                if (size > radeon_bo_size(robj)) {
-                        DRM_ERROR("Texture of unit %u needs %lu bytes but is "
-                                  "%lu\n", u, size, radeon_bo_size(robj));
-                        r100_cs_track_texture_print(&track->textures[u]);
-                        return -EINVAL;
-                }
-        }
-        return 0;
-}
-
-int r100_cs_track_check(struct radeon_device *rdev, struct r100_cs_track *track)
-{
-        unsigned i;
-        unsigned long size;
-        unsigned prim_walk;
-        unsigned nverts;
-        unsigned num_cb = track->cb_dirty ? track->num_cb : 0;
-
-        if (num_cb && !track->zb_cb_clear && !track->color_channel_mask &&
-            !track->blend_read_enable)
-                num_cb = 0;
-
-        for (i = 0; i < num_cb; i++) {
-                if (track->cb[i].robj == NULL) {
-                        DRM_ERROR("[drm] No buffer for color buffer %d !\n", i);
-                        return -EINVAL;
-                }
-                size = track->cb[i].pitch * track->cb[i].cpp * track->maxy;
-                size += track->cb[i].offset;
-                if (size > radeon_bo_size(track->cb[i].robj)) {
-                        DRM_ERROR("[drm] Buffer too small for color buffer %d "
-                                  "(need %lu have %lu) !\n", i, size,
-                                  radeon_bo_size(track->cb[i].robj));
-                        DRM_ERROR("[drm] color buffer %d (%u %u %u %u)\n",
-                                  i, track->cb[i].pitch, track->cb[i].cpp,
-                                  track->cb[i].offset, track->maxy);
-                        return -EINVAL;
-                }
-        }
-        track->cb_dirty = false;
-
-        if (track->zb_dirty && track->z_enabled) {
-                if (track->zb.robj == NULL) {
-                        DRM_ERROR("[drm] No buffer for z buffer !\n");
-                        return -EINVAL;
-                }
-                size = track->zb.pitch * track->zb.cpp * track->maxy;
-                size += track->zb.offset;
-                if (size > radeon_bo_size(track->zb.robj)) {
-                        DRM_ERROR("[drm] Buffer too small for z buffer "
-                                  "(need %lu have %lu) !\n", size,
-                                  radeon_bo_size(track->zb.robj));
-                        DRM_ERROR("[drm] zbuffer (%u %u %u %u)\n",
-                                  track->zb.pitch, track->zb.cpp,
-                                  track->zb.offset, track->maxy);
-                        return -EINVAL;
-                }
-        }
-        track->zb_dirty = false;
-
-        if (track->aa_dirty && track->aaresolve) {
-                if (track->aa.robj == NULL) {
-                        DRM_ERROR("[drm] No buffer for AA resolve buffer %d !\n", i);
-                        return -EINVAL;
-                }
-                /* I believe the format comes from colorbuffer0. */
-                size = track->aa.pitch * track->cb[0].cpp * track->maxy;
-                size += track->aa.offset;
-                if (size > radeon_bo_size(track->aa.robj)) {
-                        DRM_ERROR("[drm] Buffer too small for AA resolve buffer %d "
-                                  "(need %lu have %lu) !\n", i, size,
-                                  radeon_bo_size(track->aa.robj));
-                        DRM_ERROR("[drm] AA resolve buffer %d (%u %u %u %u)\n",
-                                  i, track->aa.pitch, track->cb[0].cpp,
-                                  track->aa.offset, track->maxy);
-                        return -EINVAL;
-                }
-        }
-        track->aa_dirty = false;
-
-        prim_walk = (track->vap_vf_cntl >> 4) & 0x3;
-        if (track->vap_vf_cntl & (1 << 14)) {
-                nverts = track->vap_alt_nverts;
-        } else {
-                nverts = (track->vap_vf_cntl >> 16) & 0xFFFF;
-        }
-        switch (prim_walk) {
-        case 1:
-                for (i = 0; i < track->num_arrays; i++) {
-                        size = track->arrays[i].esize * track->max_indx * 4;
-                        if (track->arrays[i].robj == NULL) {
-                                DRM_ERROR("(PW %u) Vertex array %u no buffer "
-                                          "bound\n", prim_walk, i);
-                                return -EINVAL;
-                        }
-                        if (size > radeon_bo_size(track->arrays[i].robj)) {
-                                dev_err(rdev->dev, "(PW %u) Vertex array %u "
-                                        "need %lu dwords have %lu dwords\n",
-                                        prim_walk, i, size >> 2,
-                                        radeon_bo_size(track->arrays[i].robj)
-                                        >> 2);
-                                DRM_ERROR("Max indices %u\n", track->max_indx);
-                                return -EINVAL;
-                        }
-                }
-                break;
-        case 2:
-                for (i = 0; i < track->num_arrays; i++) {
-                        size = track->arrays[i].esize * (nverts - 1) * 4;
-                        if (track->arrays[i].robj == NULL) {
-                                DRM_ERROR("(PW %u) Vertex array %u no buffer "
-                                          "bound\n", prim_walk, i);
-                                return -EINVAL;
-                        }
-                        if (size > radeon_bo_size(track->arrays[i].robj)) {
-                                dev_err(rdev->dev, "(PW %u) Vertex array %u "
-                                        "need %lu dwords have %lu dwords\n",
-                                        prim_walk, i, size >> 2,
-                                        radeon_bo_size(track->arrays[i].robj)
-                                        >> 2);
-                                return -EINVAL;
-                        }
-                }
-                break;
-        case 3:
-                size = track->vtx_size * nverts;
-                if (size != track->immd_dwords) {
-                        DRM_ERROR("IMMD draw %u dwors but needs %lu dwords\n",
-                                  track->immd_dwords, size);
-                        DRM_ERROR("VAP_VF_CNTL.NUM_VERTICES %u, VTX_SIZE %u\n",
-                                  nverts, track->vtx_size);
-                        return -EINVAL;
-                }
-                break;
-        default:
-                DRM_ERROR("[drm] Invalid primitive walk %d for VAP_VF_CNTL\n",
-                          prim_walk);
-                return -EINVAL;
-        }
-
-        if (track->tex_dirty) {
-                track->tex_dirty = false;
-                return r100_cs_track_texture_check(rdev, track);
-        }
-        return 0;
-}
-
-void r100_cs_track_clear(struct radeon_device *rdev, struct r100_cs_track *track)
-{
-        unsigned i, face;
-
-        track->cb_dirty = true;
-        track->zb_dirty = true;
-        track->tex_dirty = true;
-        track->aa_dirty = true;
-
-        if (rdev->family < CHIP_R300) {
-                track->num_cb = 1;
-                if (rdev->family <= CHIP_RS200)
-                        track->num_texture = 3;
-                else
-                        track->num_texture = 6;
-                track->maxy = 2048;
-                track->separate_cube = 1;
-        } else {
-                track->num_cb = 4;
-                track->num_texture = 16;
-                track->maxy = 4096;
-                track->separate_cube = 0;
-                track->aaresolve = false;
-                track->aa.robj = NULL;
-        }
-
-        for (i = 0; i < track->num_cb; i++) {
-                track->cb[i].robj = NULL;
-                track->cb[i].pitch = 8192;
-                track->cb[i].cpp = 16;
-                track->cb[i].offset = 0;
-        }
-        track->z_enabled = true;
-        track->zb.robj = NULL;
-        track->zb.pitch = 8192;
-        track->zb.cpp = 4;
-        track->zb.offset = 0;
-        track->vtx_size = 0x7F;
-        track->immd_dwords = 0xFFFFFFFFUL;
-        track->num_arrays = 11;
-        track->max_indx = 0x00FFFFFFUL;
-        for (i = 0; i < track->num_arrays; i++) {
-                track->arrays[i].robj = NULL;
-                track->arrays[i].esize = 0x7F;
-        }
-        for (i = 0; i < track->num_texture; i++) {
-                track->textures[i].compress_format = R100_TRACK_COMP_NONE;
-                track->textures[i].pitch = 16536;
-                track->textures[i].width = 16536;
-                track->textures[i].height = 16536;
-                track->textures[i].width_11 = 1 << 11;
-                track->textures[i].height_11 = 1 << 11;
-                track->textures[i].num_levels = 12;
-                if (rdev->family <= CHIP_RS200) {
-                        track->textures[i].tex_coord_type = 0;
-                        track->textures[i].txdepth = 0;
-                } else {
-                        track->textures[i].txdepth = 16;
-                        track->textures[i].tex_coord_type = 1;
-                }
-                track->textures[i].cpp = 64;
-                track->textures[i].robj = NULL;
-                /* CS IB emission code makes sure texture unit are disabled */
-                track->textures[i].enabled = false;
-                track->textures[i].lookup_disable = false;
-                track->textures[i].roundup_w = true;
-                track->textures[i].roundup_h = true;
-                if (track->separate_cube)
-                        for (face = 0; face < 5; face++) {
-                                track->textures[i].cube_info[face].robj = NULL;
-                                track->textures[i].cube_info[face].width = 16536;
-                                track->textures[i].cube_info[face].height = 16536;
-                                track->textures[i].cube_info[face].offset = 0;
-                        }
-        }
-}
-
 int r100_ring_test(struct radeon_device *rdev, struct radeon_ring *ring)
 {
         uint32_t scratch;
@@ -3679,6 +3791,12 @@ void r100_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib)
 {
         struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
 
+        if (ring->rptr_save_reg) {
+                u32 next_rptr = ring->wptr + 2 + 3;
+                radeon_ring_write(ring, PACKET0(ring->rptr_save_reg, 0));
+                radeon_ring_write(ring, next_rptr);
+        }
+
         radeon_ring_write(ring, PACKET0(RADEON_CP_IB_BASE, 1));
         radeon_ring_write(ring, ib->gpu_addr);
         radeon_ring_write(ring, ib->length_dw);
@@ -3711,7 +3829,7 @@ int r100_ib_test(struct radeon_device *rdev, struct radeon_ring *ring)
         ib.ptr[6] = PACKET2(0);
         ib.ptr[7] = PACKET2(0);
         ib.length_dw = 8;
-        r = radeon_ib_schedule(rdev, &ib);
+        r = radeon_ib_schedule(rdev, &ib, NULL);
         if (r) {
                 radeon_scratch_free(rdev, scratch);
                 radeon_ib_free(rdev, &ib);
@@ -3740,12 +3858,6 @@ int r100_ib_test(struct radeon_device *rdev, struct radeon_ring *ring)
         return r;
 }
 
-void r100_ib_fini(struct radeon_device *rdev)
-{
-        radeon_ib_pool_suspend(rdev);
-        radeon_ib_pool_fini(rdev);
-}
-
 void r100_mc_stop(struct radeon_device *rdev, struct r100_mc_save *save)
 {
         /* Shutdown CP we shouldn't need to do that but better be safe than
@@ -3905,13 +4017,11 @@ static int r100_startup(struct radeon_device *rdev)
                 return r;
         }
 
-        r = radeon_ib_pool_start(rdev);
-        if (r)
-                return r;
-
-        r = radeon_ib_ring_tests(rdev);
-        if (r)
+        r = radeon_ib_pool_init(rdev);
+        if (r) {
+                dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
                 return r;
+        }
 
         return 0;
 }
@@ -3948,7 +4058,6 @@ int r100_resume(struct radeon_device *rdev)
 
 int r100_suspend(struct radeon_device *rdev)
 {
-        radeon_ib_pool_suspend(rdev);
         r100_cp_disable(rdev);
         radeon_wb_disable(rdev);
         r100_irq_disable(rdev);
@@ -3961,7 +4070,7 @@ void r100_fini(struct radeon_device *rdev)
 {
         r100_cp_fini(rdev);
         radeon_wb_fini(rdev);
-        r100_ib_fini(rdev);
+        radeon_ib_pool_fini(rdev);
         radeon_gem_fini(rdev);
         if (rdev->flags & RADEON_IS_PCI)
                 r100_pci_gart_fini(rdev);
@@ -4068,20 +4177,14 @@ int r100_init(struct radeon_device *rdev)
         }
         r100_set_safe_registers(rdev);
 
-        r = radeon_ib_pool_init(rdev);
         rdev->accel_working = true;
-        if (r) {
-                dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
-                rdev->accel_working = false;
-        }
-
         r = r100_startup(rdev);
         if (r) {
                 /* Somethings want wront with the accel init stop accel */
                 dev_err(rdev->dev, "Disabling GPU acceleration\n");
                 r100_cp_fini(rdev);
                 radeon_wb_fini(rdev);
-                r100_ib_fini(rdev);
+                radeon_ib_pool_fini(rdev);
                 radeon_irq_kms_fini(rdev);
                 if (rdev->flags & RADEON_IS_PCI)
                         r100_pci_gart_fini(rdev);
diff --git a/drivers/gpu/drm/radeon/r200.c b/drivers/gpu/drm/radeon/r200.c
index a26144d01207..f0889259eb08 100644
--- a/drivers/gpu/drm/radeon/r200.c
+++ b/drivers/gpu/drm/radeon/r200.c
@@ -85,7 +85,7 @@ int r200_copy_dma(struct radeon_device *rdev,
                   uint64_t src_offset,
                   uint64_t dst_offset,
                   unsigned num_gpu_pages,
-                  struct radeon_fence *fence)
+                  struct radeon_fence **fence)
 {
         struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
         uint32_t size;
@@ -120,7 +120,7 @@ int r200_copy_dma(struct radeon_device *rdev,
         radeon_ring_write(ring, PACKET0(RADEON_WAIT_UNTIL, 0));
         radeon_ring_write(ring, RADEON_WAIT_DMA_GUI_IDLE);
         if (fence) {
-                r = radeon_fence_emit(rdev, fence);
+                r = radeon_fence_emit(rdev, fence, RADEON_RING_TYPE_GFX_INDEX);
         }
         radeon_ring_unlock_commit(rdev, ring);
         return r;
diff --git a/drivers/gpu/drm/radeon/r300.c b/drivers/gpu/drm/radeon/r300.c
index 97722a33e513..646a1927dda7 100644
--- a/drivers/gpu/drm/radeon/r300.c
+++ b/drivers/gpu/drm/radeon/r300.c
@@ -1391,13 +1391,11 @@ static int r300_startup(struct radeon_device *rdev)
                 return r;
         }
 
-        r = radeon_ib_pool_start(rdev);
-        if (r)
-                return r;
-
-        r = radeon_ib_ring_tests(rdev);
-        if (r)
+        r = radeon_ib_pool_init(rdev);
+        if (r) {
+                dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
                 return r;
+        }
 
         return 0;
 }
@@ -1436,7 +1434,6 @@ int r300_resume(struct radeon_device *rdev)
 
 int r300_suspend(struct radeon_device *rdev)
 {
-        radeon_ib_pool_suspend(rdev);
         r100_cp_disable(rdev);
         radeon_wb_disable(rdev);
         r100_irq_disable(rdev);
@@ -1451,7 +1448,7 @@ void r300_fini(struct radeon_device *rdev)
 {
         r100_cp_fini(rdev);
         radeon_wb_fini(rdev);
-        r100_ib_fini(rdev);
+        radeon_ib_pool_fini(rdev);
         radeon_gem_fini(rdev);
         if (rdev->flags & RADEON_IS_PCIE)
                 rv370_pcie_gart_fini(rdev);
@@ -1538,20 +1535,14 @@ int r300_init(struct radeon_device *rdev)
         }
         r300_set_reg_safe(rdev);
 
-        r = radeon_ib_pool_init(rdev);
         rdev->accel_working = true;
-        if (r) {
-                dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
-                rdev->accel_working = false;
-        }
-
         r = r300_startup(rdev);
         if (r) {
                 /* Somethings want wront with the accel init stop accel */
                 dev_err(rdev->dev, "Disabling GPU acceleration\n");
                 r100_cp_fini(rdev);
                 radeon_wb_fini(rdev);
-                r100_ib_fini(rdev);
+                radeon_ib_pool_fini(rdev);
                 radeon_irq_kms_fini(rdev);
                 if (rdev->flags & RADEON_IS_PCIE)
                         rv370_pcie_gart_fini(rdev);
diff --git a/drivers/gpu/drm/radeon/r420.c b/drivers/gpu/drm/radeon/r420.c
index 99137be7a300..f2f5bf6d339f 100644
--- a/drivers/gpu/drm/radeon/r420.c
+++ b/drivers/gpu/drm/radeon/r420.c
@@ -275,13 +275,11 @@ static int r420_startup(struct radeon_device *rdev)
         }
         r420_cp_errata_init(rdev);
 
-        r = radeon_ib_pool_start(rdev);
-        if (r)
-                return r;
-
-        r = radeon_ib_ring_tests(rdev);
-        if (r)
+        r = radeon_ib_pool_init(rdev);
+        if (r) {
+                dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
                 return r;
+        }
 
         return 0;
 }
@@ -324,7 +322,6 @@ int r420_resume(struct radeon_device *rdev)
 
 int r420_suspend(struct radeon_device *rdev)
 {
-        radeon_ib_pool_suspend(rdev);
         r420_cp_errata_fini(rdev);
         r100_cp_disable(rdev);
         radeon_wb_disable(rdev);
@@ -340,7 +337,7 @@ void r420_fini(struct radeon_device *rdev)
 {
         r100_cp_fini(rdev);
         radeon_wb_fini(rdev);
-        r100_ib_fini(rdev);
+        radeon_ib_pool_fini(rdev);
         radeon_gem_fini(rdev);
         if (rdev->flags & RADEON_IS_PCIE)
                 rv370_pcie_gart_fini(rdev);
@@ -438,20 +435,14 @@ int r420_init(struct radeon_device *rdev)
         }
         r420_set_reg_safe(rdev);
 
-        r = radeon_ib_pool_init(rdev);
         rdev->accel_working = true;
-        if (r) {
-                dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
-                rdev->accel_working = false;
-        }
-
         r = r420_startup(rdev);
         if (r) {
                 /* Somethings want wront with the accel init stop accel */
                 dev_err(rdev->dev, "Disabling GPU acceleration\n");
                 r100_cp_fini(rdev);
                 radeon_wb_fini(rdev);
-                r100_ib_fini(rdev);
+                radeon_ib_pool_fini(rdev);
                 radeon_irq_kms_fini(rdev);
                 if (rdev->flags & RADEON_IS_PCIE)
                         rv370_pcie_gart_fini(rdev);
diff --git a/drivers/gpu/drm/radeon/r520.c b/drivers/gpu/drm/radeon/r520.c
index b5cf8375cd25..079d3c52c08a 100644
--- a/drivers/gpu/drm/radeon/r520.c
+++ b/drivers/gpu/drm/radeon/r520.c
@@ -203,13 +203,11 @@ static int r520_startup(struct radeon_device *rdev)
                 return r;
         }
 
-        r = radeon_ib_pool_start(rdev);
-        if (r)
-                return r;
-
-        r = radeon_ib_ring_tests(rdev);
-        if (r)
+        r = radeon_ib_pool_init(rdev);
+        if (r) {
+                dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
                 return r;
+        }
 
         return 0;
 }
@@ -311,20 +309,14 @@ int r520_init(struct radeon_device *rdev)
                 return r;
         rv515_set_safe_registers(rdev);
 
-        r = radeon_ib_pool_init(rdev);
         rdev->accel_working = true;
-        if (r) {
-                dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
-                rdev->accel_working = false;
-        }
-
         r = r520_startup(rdev);
         if (r) {
                 /* Somethings want wront with the accel init stop accel */
                 dev_err(rdev->dev, "Disabling GPU acceleration\n");
                 r100_cp_fini(rdev);
                 radeon_wb_fini(rdev);
-                r100_ib_fini(rdev);
+                radeon_ib_pool_fini(rdev);
                 radeon_irq_kms_fini(rdev);
                 rv370_pcie_gart_fini(rdev);
                 radeon_agp_fini(rdev);
diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c
index bff627293812..637280f541a3 100644
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -709,6 +709,7 @@ void r600_hpd_init(struct radeon_device *rdev)
 {
         struct drm_device *dev = rdev->ddev;
         struct drm_connector *connector;
+        unsigned enable = 0;
 
         list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
                 struct radeon_connector *radeon_connector = to_radeon_connector(connector);
@@ -729,28 +730,22 @@ void r600_hpd_init(struct radeon_device *rdev)
                         switch (radeon_connector->hpd.hpd) {
                         case RADEON_HPD_1:
                                 WREG32(DC_HPD1_CONTROL, tmp);
-                                rdev->irq.hpd[0] = true;
                                 break;
                         case RADEON_HPD_2:
                                 WREG32(DC_HPD2_CONTROL, tmp);
-                                rdev->irq.hpd[1] = true;
                                 break;
                         case RADEON_HPD_3:
                                 WREG32(DC_HPD3_CONTROL, tmp);
-                                rdev->irq.hpd[2] = true;
                                 break;
                         case RADEON_HPD_4:
                                 WREG32(DC_HPD4_CONTROL, tmp);
-                                rdev->irq.hpd[3] = true;
                                 break;
                                 /* DCE 3.2 */
                         case RADEON_HPD_5:
                                 WREG32(DC_HPD5_CONTROL, tmp);
-                                rdev->irq.hpd[4] = true;
                                 break;
                         case RADEON_HPD_6:
                                 WREG32(DC_HPD6_CONTROL, tmp);
-                                rdev->irq.hpd[5] = true;
                                 break;
                         default:
                                 break;
@@ -759,85 +754,73 @@ void r600_hpd_init(struct radeon_device *rdev)
                         switch (radeon_connector->hpd.hpd) {
                         case RADEON_HPD_1:
                                 WREG32(DC_HOT_PLUG_DETECT1_CONTROL, DC_HOT_PLUG_DETECTx_EN);
-                                rdev->irq.hpd[0] = true;
                                 break;
                         case RADEON_HPD_2:
                                 WREG32(DC_HOT_PLUG_DETECT2_CONTROL, DC_HOT_PLUG_DETECTx_EN);
-                                rdev->irq.hpd[1] = true;
                                 break;
                         case RADEON_HPD_3:
                                 WREG32(DC_HOT_PLUG_DETECT3_CONTROL, DC_HOT_PLUG_DETECTx_EN);
-                                rdev->irq.hpd[2] = true;
                                 break;
                         default:
                                 break;
                         }
                 }
+                enable |= 1 << radeon_connector->hpd.hpd;
                 radeon_hpd_set_polarity(rdev, radeon_connector->hpd.hpd);
         }
-        if (rdev->irq.installed)
-                r600_irq_set(rdev);
+        radeon_irq_kms_enable_hpd(rdev, enable);
 }
 
 void r600_hpd_fini(struct radeon_device *rdev)
 {
         struct drm_device *dev = rdev->ddev;
         struct drm_connector *connector;
+        unsigned disable = 0;
 
-        if (ASIC_IS_DCE3(rdev)) {
-                list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
-                        struct radeon_connector *radeon_connector = to_radeon_connector(connector);
+        list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
+                struct radeon_connector *radeon_connector = to_radeon_connector(connector);
+                if (ASIC_IS_DCE3(rdev)) {
                         switch (radeon_connector->hpd.hpd) {
                         case RADEON_HPD_1:
                                 WREG32(DC_HPD1_CONTROL, 0);
-                                rdev->irq.hpd[0] = false;
                                 break;
                         case RADEON_HPD_2:
                                 WREG32(DC_HPD2_CONTROL, 0);
-                                rdev->irq.hpd[1] = false;
                                 break;
                         case RADEON_HPD_3:
                                 WREG32(DC_HPD3_CONTROL, 0);
-                                rdev->irq.hpd[2] = false;
                                 break;
                         case RADEON_HPD_4:
                                 WREG32(DC_HPD4_CONTROL, 0);
-                                rdev->irq.hpd[3] = false;
                                 break;
                                 /* DCE 3.2 */
                         case RADEON_HPD_5:
                                 WREG32(DC_HPD5_CONTROL, 0);
-                                rdev->irq.hpd[4] = false;
                                 break;
                         case RADEON_HPD_6:
                                 WREG32(DC_HPD6_CONTROL, 0);
-                                rdev->irq.hpd[5] = false;
                                 break;
                         default:
                                 break;
                         }
-                }
-        } else {
-                list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
-                        struct radeon_connector *radeon_connector = to_radeon_connector(connector);
+                } else {
                         switch (radeon_connector->hpd.hpd) {
                         case RADEON_HPD_1:
                                 WREG32(DC_HOT_PLUG_DETECT1_CONTROL, 0);
-                                rdev->irq.hpd[0] = false;
                                 break;
                         case RADEON_HPD_2:
                                 WREG32(DC_HOT_PLUG_DETECT2_CONTROL, 0);
-                                rdev->irq.hpd[1] = false;
                                 break;
                         case RADEON_HPD_3:
                                 WREG32(DC_HOT_PLUG_DETECT3_CONTROL, 0);
-                                rdev->irq.hpd[2] = false;
                                 break;
                         default:
                                 break;
                         }
                 }
+                disable |= 1 << radeon_connector->hpd.hpd;
         }
+        radeon_irq_kms_disable_hpd(rdev, disable);
 }
 
 /*
@@ -1306,6 +1289,14 @@ int r600_gpu_soft_reset(struct radeon_device *rdev)
                 RREG32(R_008014_GRBM_STATUS2));
         dev_info(rdev->dev, "  R_000E50_SRBM_STATUS=0x%08X\n",
                 RREG32(R_000E50_SRBM_STATUS));
+        dev_info(rdev->dev, "  R_008674_CP_STALLED_STAT1 = 0x%08X\n",
+                RREG32(CP_STALLED_STAT1));
+        dev_info(rdev->dev, "  R_008678_CP_STALLED_STAT2 = 0x%08X\n",
+                RREG32(CP_STALLED_STAT2));
+        dev_info(rdev->dev, "  R_00867C_CP_BUSY_STAT     = 0x%08X\n",
+                RREG32(CP_BUSY_STAT));
+        dev_info(rdev->dev, "  R_008680_CP_STAT          = 0x%08X\n",
+                RREG32(CP_STAT));
         rv515_mc_stop(rdev, &save);
         if (r600_mc_wait_for_idle(rdev)) {
                 dev_warn(rdev->dev, "Wait for MC idle timedout !\n");
@@ -1349,6 +1340,14 @@ int r600_gpu_soft_reset(struct radeon_device *rdev)
                 RREG32(R_008014_GRBM_STATUS2));
         dev_info(rdev->dev, "  R_000E50_SRBM_STATUS=0x%08X\n",
                 RREG32(R_000E50_SRBM_STATUS));
+        dev_info(rdev->dev, "  R_008674_CP_STALLED_STAT1 = 0x%08X\n",
+                RREG32(CP_STALLED_STAT1));
+        dev_info(rdev->dev, "  R_008678_CP_STALLED_STAT2 = 0x%08X\n",
+                RREG32(CP_STALLED_STAT2));
+        dev_info(rdev->dev, "  R_00867C_CP_BUSY_STAT     = 0x%08X\n",
+                RREG32(CP_BUSY_STAT));
+        dev_info(rdev->dev, "  R_008680_CP_STAT          = 0x%08X\n",
+                RREG32(CP_STAT));
         rv515_mc_resume(rdev, &save);
         return 0;
 }
@@ -2172,18 +2171,29 @@ int r600_cp_resume(struct radeon_device *rdev)
 void r600_ring_init(struct radeon_device *rdev, struct radeon_ring *ring, unsigned ring_size)
 {
         u32 rb_bufsz;
+        int r;
 
         /* Align ring size */
         rb_bufsz = drm_order(ring_size / 8);
         ring_size = (1 << (rb_bufsz + 1)) * 4;
         ring->ring_size = ring_size;
         ring->align_mask = 16 - 1;
+
+        if (radeon_ring_supports_scratch_reg(rdev, ring)) {
+                r = radeon_scratch_get(rdev, &ring->rptr_save_reg);
+                if (r) {
+                        DRM_ERROR("failed to get scratch reg for rptr save (%d).\n", r);
+                        ring->rptr_save_reg = 0;
+                }
+        }
 }
 
 void r600_cp_fini(struct radeon_device *rdev)
 {
+        struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
         r600_cp_stop(rdev);
-        radeon_ring_fini(rdev, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]);
+        radeon_ring_fini(rdev, ring);
+        radeon_scratch_free(rdev, ring->rptr_save_reg);
 }
 
 
@@ -2206,7 +2216,7 @@ int r600_ring_test(struct radeon_device *rdev, struct radeon_ring *ring)
 {
         uint32_t scratch;
         uint32_t tmp = 0;
-        unsigned i, ridx = radeon_ring_index(rdev, ring);
+        unsigned i;
         int r;
 
         r = radeon_scratch_get(rdev, &scratch);
@@ -2217,7 +2227,7 @@ int r600_ring_test(struct radeon_device *rdev, struct radeon_ring *ring)
         WREG32(scratch, 0xCAFEDEAD);
         r = radeon_ring_lock(rdev, ring, 3);
         if (r) {
-                DRM_ERROR("radeon: cp failed to lock ring %d (%d).\n", ridx, r);
+                DRM_ERROR("radeon: cp failed to lock ring %d (%d).\n", ring->idx, r);
                 radeon_scratch_free(rdev, scratch);
                 return r;
         }
@@ -2232,10 +2242,10 @@ int r600_ring_test(struct radeon_device *rdev, struct radeon_ring *ring)
                 DRM_UDELAY(1);
         }
         if (i < rdev->usec_timeout) {
-                DRM_INFO("ring test on %d succeeded in %d usecs\n", ridx, i);
+                DRM_INFO("ring test on %d succeeded in %d usecs\n", ring->idx, i);
         } else {
                 DRM_ERROR("radeon: ring %d test failed (scratch(0x%04X)=0x%08X)\n",
-                          ridx, scratch, tmp);
+                          ring->idx, scratch, tmp);
                 r = -EINVAL;
         }
         radeon_scratch_free(rdev, scratch);
@@ -2309,34 +2319,21 @@ int r600_copy_blit(struct radeon_device *rdev,
                    uint64_t src_offset,
                    uint64_t dst_offset,
                    unsigned num_gpu_pages,
-                   struct radeon_fence *fence)
+                   struct radeon_fence **fence)
 {
+        struct radeon_semaphore *sem = NULL;
         struct radeon_sa_bo *vb = NULL;
         int r;
 
-        r = r600_blit_prepare_copy(rdev, num_gpu_pages, &vb);
+        r = r600_blit_prepare_copy(rdev, num_gpu_pages, fence, &vb, &sem);
         if (r) {
                 return r;
         }
         r600_kms_blit_copy(rdev, src_offset, dst_offset, num_gpu_pages, vb);
-        r600_blit_done_copy(rdev, fence, vb);
+        r600_blit_done_copy(rdev, fence, vb, sem);
         return 0;
 }
 
-void r600_blit_suspend(struct radeon_device *rdev)
-{
-        int r;
-
-        /* unpin shaders bo */
-        if (rdev->r600_blit.shader_obj) {
-                r = radeon_bo_reserve(rdev->r600_blit.shader_obj, false);
-                if (!r) {
-                        radeon_bo_unpin(rdev->r600_blit.shader_obj);
-                        radeon_bo_unreserve(rdev->r600_blit.shader_obj);
-                }
-        }
-}
-
 int r600_set_surface_reg(struct radeon_device *rdev, int reg,
                          uint32_t tiling_flags, uint32_t pitch,
                          uint32_t offset, uint32_t obj_size)
@@ -2419,13 +2416,11 @@ int r600_startup(struct radeon_device *rdev)
         if (r)
                 return r;
 
-        r = radeon_ib_pool_start(rdev);
-        if (r)
-                return r;
-
-        r = radeon_ib_ring_tests(rdev);
-        if (r)
+        r = radeon_ib_pool_init(rdev);
+        if (r) {
+                dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
                 return r;
+        }
 
         r = r600_audio_init(rdev);
         if (r) {
@@ -2475,9 +2470,6 @@ int r600_resume(struct radeon_device *rdev)
 int r600_suspend(struct radeon_device *rdev)
 {
         r600_audio_fini(rdev);
-        radeon_ib_pool_suspend(rdev);
-        r600_blit_suspend(rdev);
-        /* FIXME: we should wait for ring to be empty */
         r600_cp_stop(rdev);
         rdev->ring[RADEON_RING_TYPE_GFX_INDEX].ready = false;
         r600_irq_suspend(rdev);
@@ -2559,20 +2551,14 @@ int r600_init(struct radeon_device *rdev)
         if (r)
                 return r;
 
-        r = radeon_ib_pool_init(rdev);
         rdev->accel_working = true;
-        if (r) {
-                dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
-                rdev->accel_working = false;
-        }
-
         r = r600_startup(rdev);
         if (r) {
                 dev_err(rdev->dev, "disabling GPU acceleration\n");
                 r600_cp_fini(rdev);
                 r600_irq_fini(rdev);
                 radeon_wb_fini(rdev);
-                r100_ib_fini(rdev);
+                radeon_ib_pool_fini(rdev);
                 radeon_irq_kms_fini(rdev);
                 r600_pcie_gart_fini(rdev);
                 rdev->accel_working = false;
@@ -2588,7 +2574,7 @@ void r600_fini(struct radeon_device *rdev)
         r600_cp_fini(rdev);
         r600_irq_fini(rdev);
         radeon_wb_fini(rdev);
-        r100_ib_fini(rdev);
+        radeon_ib_pool_fini(rdev);
         radeon_irq_kms_fini(rdev);
         r600_pcie_gart_fini(rdev);
         r600_vram_scratch_fini(rdev);
@@ -2607,9 +2593,24 @@ void r600_fini(struct radeon_device *rdev)
  */
 void r600_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib)
 {
-        struct radeon_ring *ring = &rdev->ring[ib->fence->ring];
+        struct radeon_ring *ring = &rdev->ring[ib->ring];
+        u32 next_rptr;
+
+        if (ring->rptr_save_reg) {
+                next_rptr = ring->wptr + 3 + 4;
+                radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1));
+                radeon_ring_write(ring, ((ring->rptr_save_reg -
+                                         PACKET3_SET_CONFIG_REG_OFFSET) >> 2));
+                radeon_ring_write(ring, next_rptr);
+        } else if (rdev->wb.enabled) {
+                next_rptr = ring->wptr + 5 + 4;
+                radeon_ring_write(ring, PACKET3(PACKET3_MEM_WRITE, 3));
+                radeon_ring_write(ring, ring->next_rptr_gpu_addr & 0xfffffffc);
+                radeon_ring_write(ring, (upper_32_bits(ring->next_rptr_gpu_addr) & 0xff) | (1 << 18));
+                radeon_ring_write(ring, next_rptr);
+                radeon_ring_write(ring, 0);
+        }
 
-        /* FIXME: implement */
         radeon_ring_write(ring, PACKET3(PACKET3_INDIRECT_BUFFER, 2));
         radeon_ring_write(ring,
 #ifdef __BIG_ENDIAN
@@ -2627,7 +2628,6 @@ int r600_ib_test(struct radeon_device *rdev, struct radeon_ring *ring)
         uint32_t tmp = 0;
         unsigned i;
         int r;
-        int ring_index = radeon_ring_index(rdev, ring);
 
         r = radeon_scratch_get(rdev, &scratch);
         if (r) {
@@ -2635,7 +2635,7 @@ int r600_ib_test(struct radeon_device *rdev, struct radeon_ring *ring)
                 return r;
         }
         WREG32(scratch, 0xCAFEDEAD);
-        r = radeon_ib_get(rdev, ring_index, &ib, 256);
+        r = radeon_ib_get(rdev, ring->idx, &ib, 256);
         if (r) {
                 DRM_ERROR("radeon: failed to get ib (%d).\n", r);
                 return r;
@@ -2644,7 +2644,7 @@ int r600_ib_test(struct radeon_device *rdev, struct radeon_ring *ring)
         ib.ptr[1] = ((scratch - PACKET3_SET_CONFIG_REG_OFFSET) >> 2);
         ib.ptr[2] = 0xDEADBEEF;
         ib.length_dw = 3;
-        r = radeon_ib_schedule(rdev, &ib);
+        r = radeon_ib_schedule(rdev, &ib, NULL);
         if (r) {
                 radeon_scratch_free(rdev, scratch);
                 radeon_ib_free(rdev, &ib);
@@ -2857,7 +2857,6 @@ void r600_disable_interrupts(struct radeon_device *rdev)
         WREG32(IH_RB_RPTR, 0);
         WREG32(IH_RB_WPTR, 0);
         rdev->ih.enabled = false;
-        rdev->ih.wptr = 0;
         rdev->ih.rptr = 0;
 }
 
@@ -3042,18 +3041,18 @@ int r600_irq_set(struct radeon_device *rdev)
                 hdmi1 = RREG32(HDMI1_AUDIO_PACKET_CONTROL) & ~HDMI0_AZ_FORMAT_WTRIG_MASK;
         }
 
-        if (rdev->irq.sw_int[RADEON_RING_TYPE_GFX_INDEX]) {
+        if (atomic_read(&rdev->irq.ring_int[RADEON_RING_TYPE_GFX_INDEX])) {
                 DRM_DEBUG("r600_irq_set: sw int\n");
                 cp_int_cntl |= RB_INT_ENABLE;
                 cp_int_cntl |= TIME_STAMP_INT_ENABLE;
         }
         if (rdev->irq.crtc_vblank_int[0] ||
-            rdev->irq.pflip[0]) {
+            atomic_read(&rdev->irq.pflip[0])) {
                 DRM_DEBUG("r600_irq_set: vblank 0\n");
                 mode_int |= D1MODE_VBLANK_INT_MASK;
         }
         if (rdev->irq.crtc_vblank_int[1] ||
-            rdev->irq.pflip[1]) {
+            atomic_read(&rdev->irq.pflip[1])) {
                 DRM_DEBUG("r600_irq_set: vblank 1\n");
                 mode_int |= D2MODE_VBLANK_INT_MASK;
         }
@@ -3309,7 +3308,6 @@ int r600_irq_process(struct radeon_device *rdev)
         u32 rptr;
         u32 src_id, src_data;
         u32 ring_index;
-        unsigned long flags;
         bool queue_hotplug = false;
         bool queue_hdmi = false;
 
@@ -3321,24 +3319,21 @@ int r600_irq_process(struct radeon_device *rdev)
                 RREG32(IH_RB_WPTR);
 
         wptr = r600_get_ih_wptr(rdev);
-        rptr = rdev->ih.rptr;
-        DRM_DEBUG("r600_irq_process start: rptr %d, wptr %d\n", rptr, wptr);
-
-        spin_lock_irqsave(&rdev->ih.lock, flags);
 
-        if (rptr == wptr) {
-                spin_unlock_irqrestore(&rdev->ih.lock, flags);
+restart_ih:
+        /* is somebody else already processing irqs? */
+        if (atomic_xchg(&rdev->ih.lock, 1))
                 return IRQ_NONE;
-        }
 
-restart_ih:
+        rptr = rdev->ih.rptr;
+        DRM_DEBUG("r600_irq_process start: rptr %d, wptr %d\n", rptr, wptr);
+
         /* Order reading of wptr vs. reading of IH ring data */
         rmb();
 
         /* display interrupts */
         r600_irq_ack(rdev);
 
-        rdev->ih.wptr = wptr;
         while (rptr != wptr) {
                 /* wptr/rptr are in bytes! */
                 ring_index = rptr / 4;
@@ -3355,7 +3350,7 @@ restart_ih:
                                                 rdev->pm.vblank_sync = true;
                                                 wake_up(&rdev->irq.vblank_queue);
                                         }
-                                        if (rdev->irq.pflip[0])
+                                        if (atomic_read(&rdev->irq.pflip[0]))
                                                 radeon_crtc_handle_flip(rdev, 0);
                                         rdev->irq.stat_regs.r600.disp_int &= ~LB_D1_VBLANK_INTERRUPT;
                                         DRM_DEBUG("IH: D1 vblank\n");
@@ -3381,7 +3376,7 @@ restart_ih:
                                                 rdev->pm.vblank_sync = true;
                                                 wake_up(&rdev->irq.vblank_queue);
                                         }
-                                        if (rdev->irq.pflip[1])
+                                        if (atomic_read(&rdev->irq.pflip[1]))
                                                 radeon_crtc_handle_flip(rdev, 1);
                                         rdev->irq.stat_regs.r600.disp_int &= ~LB_D2_VBLANK_INTERRUPT;
                                         DRM_DEBUG("IH: D2 vblank\n");
@@ -3480,7 +3475,6 @@ restart_ih:
                         break;
                 case 233: /* GUI IDLE */
                         DRM_DEBUG("IH: GUI idle\n");
-                        rdev->pm.gui_idle = true;
                         wake_up(&rdev->irq.idle_queue);
                         break;
                 default:
@@ -3492,17 +3486,19 @@ restart_ih:
                 rptr += 16;
                 rptr &= rdev->ih.ptr_mask;
         }
-        /* make sure wptr hasn't changed while processing */
-        wptr = r600_get_ih_wptr(rdev);
-        if (wptr != rdev->ih.wptr)
-                goto restart_ih;
         if (queue_hotplug)
                 schedule_work(&rdev->hotplug_work);
         if (queue_hdmi)
                 schedule_work(&rdev->audio_work);
         rdev->ih.rptr = rptr;
         WREG32(IH_RB_RPTR, rdev->ih.rptr);
-        spin_unlock_irqrestore(&rdev->ih.lock, flags);
+        atomic_set(&rdev->ih.lock, 0);
+
+        /* make sure wptr hasn't changed while processing */
+        wptr = r600_get_ih_wptr(rdev);
+        if (wptr != rptr)
+                goto restart_ih;
+
         return IRQ_HANDLED;
 }
 
@@ -3685,6 +3681,8 @@ static void r600_pcie_gen2_enable(struct radeon_device *rdev)
 {
         u32 link_width_cntl, lanes, speed_cntl, training_cntl, tmp;
         u16 link_cntl2;
+        u32 mask;
+        int ret;
 
         if (radeon_pcie_gen2 == 0)
                 return;
@@ -3703,6 +3701,15 @@ static void r600_pcie_gen2_enable(struct radeon_device *rdev)
         if (rdev->family <= CHIP_R600)
                 return;
 
+        ret = drm_pcie_get_speed_cap_mask(rdev->ddev, &mask);
+        if (ret != 0)
+                return;
+
+        if (!(mask & DRM_PCIE_SPEED_50))
+                return;
+
+        DRM_INFO("enabling PCIE gen 2 link speeds, disable with radeon.pcie_gen2=0\n");
+
         /* 55 nm r6xx asics */
         if ((rdev->family == CHIP_RV670) ||
             (rdev->family == CHIP_RV620) ||
diff --git a/drivers/gpu/drm/radeon/r600_blit_kms.c b/drivers/gpu/drm/radeon/r600_blit_kms.c
index 03b6e0d3d503..2bef8549ddfe 100644
--- a/drivers/gpu/drm/radeon/r600_blit_kms.c
+++ b/drivers/gpu/drm/radeon/r600_blit_kms.c
@@ -512,7 +512,8 @@ int r600_blit_init(struct radeon_device *rdev)
         rdev->r600_blit.primitives.draw_auto = draw_auto;
         rdev->r600_blit.primitives.set_default_state = set_default_state;
 
-        rdev->r600_blit.ring_size_common = 40; /* shaders + def state */
+        rdev->r600_blit.ring_size_common = 8; /* sync semaphore */
+        rdev->r600_blit.ring_size_common += 40; /* shaders + def state */
         rdev->r600_blit.ring_size_common += 5; /* done copy */
         rdev->r600_blit.ring_size_common += 16; /* fence emit for done copy */
 
@@ -523,10 +524,6 @@ int r600_blit_init(struct radeon_device *rdev)
 
         rdev->r600_blit.max_dim = 8192;
 
-        /* pin copy shader into vram if already initialized */
-        if (rdev->r600_blit.shader_obj)
-                goto done;
-
         rdev->r600_blit.state_offset = 0;
 
         if (rdev->family >= CHIP_RV770)
@@ -551,11 +548,26 @@ int r600_blit_init(struct radeon_device *rdev)
         obj_size += r6xx_ps_size * 4;
         obj_size = ALIGN(obj_size, 256);
 
-        r = radeon_bo_create(rdev, obj_size, PAGE_SIZE, true, RADEON_GEM_DOMAIN_VRAM,
-                             NULL, &rdev->r600_blit.shader_obj);
-        if (r) {
-                DRM_ERROR("r600 failed to allocate shader\n");
-                return r;
+        /* pin copy shader into vram if not already initialized */
+        if (rdev->r600_blit.shader_obj == NULL) {
+                r = radeon_bo_create(rdev, obj_size, PAGE_SIZE, true,
+                                     RADEON_GEM_DOMAIN_VRAM,
+                                     NULL, &rdev->r600_blit.shader_obj);
+                if (r) {
+                        DRM_ERROR("r600 failed to allocate shader\n");
+                        return r;
+                }
+
+                r = radeon_bo_reserve(rdev->r600_blit.shader_obj, false);
+                if (unlikely(r != 0))
+                        return r;
+                r = radeon_bo_pin(rdev->r600_blit.shader_obj, RADEON_GEM_DOMAIN_VRAM,
+                                  &rdev->r600_blit.shader_gpu_addr);
+                radeon_bo_unreserve(rdev->r600_blit.shader_obj);
+                if (r) {
+                        dev_err(rdev->dev, "(%d) pin blit object failed\n", r);
+                        return r;
+                }
         }
 
         DRM_DEBUG("r6xx blit allocated bo %08x vs %08x ps %08x\n",
@@ -586,17 +598,6 @@ int r600_blit_init(struct radeon_device *rdev)
         radeon_bo_kunmap(rdev->r600_blit.shader_obj);
         radeon_bo_unreserve(rdev->r600_blit.shader_obj);
 
-done:
-        r = radeon_bo_reserve(rdev->r600_blit.shader_obj, false);
-        if (unlikely(r != 0))
-                return r;
-        r = radeon_bo_pin(rdev->r600_blit.shader_obj, RADEON_GEM_DOMAIN_VRAM,
-                          &rdev->r600_blit.shader_gpu_addr);
-        radeon_bo_unreserve(rdev->r600_blit.shader_obj);
-        if (r) {
-                dev_err(rdev->dev, "(%d) pin blit object failed\n", r);
-                return r;
-        }
         radeon_ttm_set_active_vram_size(rdev, rdev->mc.real_vram_size);
         return 0;
 }
@@ -666,7 +667,8 @@ static unsigned r600_blit_create_rect(unsigned num_gpu_pages,
 
 
 int r600_blit_prepare_copy(struct radeon_device *rdev, unsigned num_gpu_pages,
-                           struct radeon_sa_bo **vb)
+                           struct radeon_fence **fence, struct radeon_sa_bo **vb,
+                           struct radeon_semaphore **sem)
 {
         struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
         int r;
@@ -689,34 +691,50 @@ int r600_blit_prepare_copy(struct radeon_device *rdev, unsigned num_gpu_pages,
                 return r;
         }
 
+        r = radeon_semaphore_create(rdev, sem);
+        if (r) {
+                radeon_sa_bo_free(rdev, vb, NULL);
+                return r;
+        }
+
         /* calculate number of loops correctly */
         ring_size = num_loops * dwords_per_loop;
         ring_size += rdev->r600_blit.ring_size_common;
         r = radeon_ring_lock(rdev, ring, ring_size);
         if (r) {
                 radeon_sa_bo_free(rdev, vb, NULL);
+                radeon_semaphore_free(rdev, sem, NULL);
                 return r;
         }
 
+        if (radeon_fence_need_sync(*fence, RADEON_RING_TYPE_GFX_INDEX)) {
+                radeon_semaphore_sync_rings(rdev, *sem, (*fence)->ring,
+                                            RADEON_RING_TYPE_GFX_INDEX);
+                radeon_fence_note_sync(*fence, RADEON_RING_TYPE_GFX_INDEX);
+        } else {
+                radeon_semaphore_free(rdev, sem, NULL);
+        }
+
         rdev->r600_blit.primitives.set_default_state(rdev);
         rdev->r600_blit.primitives.set_shaders(rdev);
         return 0;
 }
 
-void r600_blit_done_copy(struct radeon_device *rdev, struct radeon_fence *fence,
-                         struct radeon_sa_bo *vb)
+void r600_blit_done_copy(struct radeon_device *rdev, struct radeon_fence **fence,
+                         struct radeon_sa_bo *vb, struct radeon_semaphore *sem)
 {
         struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
         int r;
 
-        r = radeon_fence_emit(rdev, fence);
+        r = radeon_fence_emit(rdev, fence, RADEON_RING_TYPE_GFX_INDEX);
         if (r) {
                 radeon_ring_unlock_undo(rdev, ring);
                 return;
         }
 
         radeon_ring_unlock_commit(rdev, ring);
-        radeon_sa_bo_free(rdev, &vb, fence);
+        radeon_sa_bo_free(rdev, &vb, *fence);
+        radeon_semaphore_free(rdev, &sem, *fence);
 }
 
 void r600_kms_blit_copy(struct radeon_device *rdev,
diff --git a/drivers/gpu/drm/radeon/r600_hdmi.c b/drivers/gpu/drm/radeon/r600_hdmi.c
index 82a0a4c919c0..e3558c3ef24a 100644
--- a/drivers/gpu/drm/radeon/r600_hdmi.c
+++ b/drivers/gpu/drm/radeon/r600_hdmi.c
@@ -519,8 +519,7 @@ void r600_hdmi_enable(struct drm_encoder *encoder)
 
         if (rdev->irq.installed) {
                 /* if irq is available use it */
-                rdev->irq.afmt[dig->afmt->id] = true;
-                radeon_irq_set(rdev);
+                radeon_irq_kms_enable_afmt(rdev, dig->afmt->id);
         }
 
         dig->afmt->enabled = true;
@@ -556,8 +555,7 @@ void r600_hdmi_disable(struct drm_encoder *encoder)
                   offset, radeon_encoder->encoder_id);
 
         /* disable irq */
-        rdev->irq.afmt[dig->afmt->id] = false;
-        radeon_irq_set(rdev);
+        radeon_irq_kms_disable_afmt(rdev, dig->afmt->id);
 
         /* Older chipsets not handled by AtomBIOS */
         if (rdev->family >= CHIP_R600 && !ASIC_IS_DCE3(rdev)) {
diff --git a/drivers/gpu/drm/radeon/r600d.h b/drivers/gpu/drm/radeon/r600d.h
index 025fd5b6c08c..4b116ae75fc2 100644
--- a/drivers/gpu/drm/radeon/r600d.h
+++ b/drivers/gpu/drm/radeon/r600d.h
@@ -153,6 +153,9 @@
 
 #define CONFIG_MEMSIZE                                  0x5428
 #define CONFIG_CNTL                                     0x5424
+#define CP_STALLED_STAT1                        0x8674
+#define CP_STALLED_STAT2                        0x8678
+#define CP_BUSY_STAT                            0x867C
 #define CP_STAT                                         0x8680
 #define CP_COHER_BASE                                   0x85F8
 #define CP_DEBUG                                        0xC1FC
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index fefcca55c1eb..5431af292408 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -113,7 +113,6 @@ extern int radeon_lockup_timeout;
 
 /* fence seq are set to this number when signaled */
 #define RADEON_FENCE_SIGNALED_SEQ               0LL
-#define RADEON_FENCE_NOTEMITED_SEQ              (~0LL)
 
 /* internal ring indices */
 /* r1xx+ has gfx CP ring */
@@ -160,48 +159,6 @@ static inline int radeon_atrm_get_bios_chunk(uint8_t *bios, int offset, int len)
 #endif
 bool radeon_get_bios(struct radeon_device *rdev);
 
-
-/*
- * Mutex which allows recursive locking from the same process.
- */
-struct radeon_mutex {
-        struct mutex            mutex;
-        struct task_struct      *owner;
-        int                     level;
-};
-
-static inline void radeon_mutex_init(struct radeon_mutex *mutex)
-{
-        mutex_init(&mutex->mutex);
-        mutex->owner = NULL;
-        mutex->level = 0;
-}
-
-static inline void radeon_mutex_lock(struct radeon_mutex *mutex)
-{
-        if (mutex_trylock(&mutex->mutex)) {
-                /* The mutex was unlocked before, so it's ours now */
-                mutex->owner = current;
-        } else if (mutex->owner != current) {
-                /* Another process locked the mutex, take it */
-                mutex_lock(&mutex->mutex);
-                mutex->owner = current;
-        }
-        /* Otherwise the mutex was already locked by this process */
-
-        mutex->level++;
-}
-
-static inline void radeon_mutex_unlock(struct radeon_mutex *mutex)
-{
-        if (--mutex->level > 0)
-                return;
-
-        mutex->owner = NULL;
-        mutex_unlock(&mutex->mutex);
-}
-
-
 /*
  * Dummy page
  */
@@ -258,8 +215,8 @@ struct radeon_fence_driver {
         uint32_t                        scratch_reg;
         uint64_t                        gpu_addr;
         volatile uint32_t               *cpu_addr;
-        /* seq is protected by ring emission lock */
-        uint64_t                        seq;
+        /* sync_seq is protected by ring emission lock */
+        uint64_t                        sync_seq[RADEON_NUM_RINGS];
         atomic64_t                      last_seq;
         unsigned long                   last_activity;
         bool                            initialized;
@@ -277,19 +234,39 @@ struct radeon_fence {
 int radeon_fence_driver_start_ring(struct radeon_device *rdev, int ring);
 int radeon_fence_driver_init(struct radeon_device *rdev);
 void radeon_fence_driver_fini(struct radeon_device *rdev);
-int radeon_fence_create(struct radeon_device *rdev, struct radeon_fence **fence, int ring);
-int radeon_fence_emit(struct radeon_device *rdev, struct radeon_fence *fence);
+int radeon_fence_emit(struct radeon_device *rdev, struct radeon_fence **fence, int ring);
 void radeon_fence_process(struct radeon_device *rdev, int ring);
 bool radeon_fence_signaled(struct radeon_fence *fence);
 int radeon_fence_wait(struct radeon_fence *fence, bool interruptible);
 int radeon_fence_wait_next_locked(struct radeon_device *rdev, int ring);
-int radeon_fence_wait_empty_locked(struct radeon_device *rdev, int ring);
+void radeon_fence_wait_empty_locked(struct radeon_device *rdev, int ring);
 int radeon_fence_wait_any(struct radeon_device *rdev,
                           struct radeon_fence **fences,
                           bool intr);
 struct radeon_fence *radeon_fence_ref(struct radeon_fence *fence);
 void radeon_fence_unref(struct radeon_fence **fence);
 unsigned radeon_fence_count_emitted(struct radeon_device *rdev, int ring);
+bool radeon_fence_need_sync(struct radeon_fence *fence, int ring);
+void radeon_fence_note_sync(struct radeon_fence *fence, int ring);
+static inline struct radeon_fence *radeon_fence_later(struct radeon_fence *a,
+                                                      struct radeon_fence *b)
+{
+        if (!a) {
+                return b;
+        }
+
+        if (!b) {
+                return a;
+        }
+
+        BUG_ON(a->ring != b->ring);
+
+        if (a->seq > b->seq) {
+                return a;
+        } else {
+                return b;
+        }
+}
 
 /*
  * Tiling registers
@@ -385,7 +362,7 @@ struct radeon_bo_list {
  * alignment).
  */
 struct radeon_sa_manager {
-        spinlock_t              lock;
+        wait_queue_head_t       wq;
         struct radeon_bo        *bo;
         struct list_head        *hole;
         struct list_head        flist[RADEON_NUM_RINGS];
@@ -451,10 +428,9 @@ void radeon_semaphore_emit_wait(struct radeon_device *rdev, int ring,
                                 struct radeon_semaphore *semaphore);
 int radeon_semaphore_sync_rings(struct radeon_device *rdev,
                                 struct radeon_semaphore *semaphore,
-                                bool sync_to[RADEON_NUM_RINGS],
-                                int dst_ring);
+                                int signaler, int waiter);
 void radeon_semaphore_free(struct radeon_device *rdev,
-                           struct radeon_semaphore *semaphore,
+                           struct radeon_semaphore **semaphore,
                            struct radeon_fence *fence);
 
 /*
@@ -597,21 +573,18 @@ union radeon_irq_stat_regs {
 #define RADEON_MAX_AFMT_BLOCKS 6
 
 struct radeon_irq {
-        bool            installed;
-        bool            sw_int[RADEON_NUM_RINGS];
-        bool            crtc_vblank_int[RADEON_MAX_CRTCS];
-        bool            pflip[RADEON_MAX_CRTCS];
-        wait_queue_head_t       vblank_queue;
-        bool            hpd[RADEON_MAX_HPD_PINS];
-        bool            gui_idle;
-        bool            gui_idle_acked;
-        wait_queue_head_t       idle_queue;
-        bool            afmt[RADEON_MAX_AFMT_BLOCKS];
-        spinlock_t sw_lock;
-        int sw_refcount[RADEON_NUM_RINGS];
-        union radeon_irq_stat_regs stat_regs;
-        spinlock_t pflip_lock[RADEON_MAX_CRTCS];
-        int pflip_refcount[RADEON_MAX_CRTCS];
+        bool                            installed;
+        spinlock_t                      lock;
+        atomic_t                        ring_int[RADEON_NUM_RINGS];
+        bool                            crtc_vblank_int[RADEON_MAX_CRTCS];
+        atomic_t                        pflip[RADEON_MAX_CRTCS];
+        wait_queue_head_t               vblank_queue;
+        bool                            hpd[RADEON_MAX_HPD_PINS];
+        bool                            gui_idle;
+        bool                            gui_idle_acked;
+        wait_queue_head_t               idle_queue;
+        bool                            afmt[RADEON_MAX_AFMT_BLOCKS];
+        union radeon_irq_stat_regs      stat_regs;
 };
 
 int radeon_irq_kms_init(struct radeon_device *rdev);
@@ -620,6 +593,11 @@ void radeon_irq_kms_sw_irq_get(struct radeon_device *rdev, int ring);
 void radeon_irq_kms_sw_irq_put(struct radeon_device *rdev, int ring);
 void radeon_irq_kms_pflip_irq_get(struct radeon_device *rdev, int crtc);
 void radeon_irq_kms_pflip_irq_put(struct radeon_device *rdev, int crtc);
+void radeon_irq_kms_enable_afmt(struct radeon_device *rdev, int block);
+void radeon_irq_kms_disable_afmt(struct radeon_device *rdev, int block);
+void radeon_irq_kms_enable_hpd(struct radeon_device *rdev, unsigned hpd_mask);
+void radeon_irq_kms_disable_hpd(struct radeon_device *rdev, unsigned hpd_mask);
+int radeon_irq_kms_wait_gui_idle(struct radeon_device *rdev);
 
 /*
  * CP & rings.
@@ -630,9 +608,11 @@ struct radeon_ib {
         uint32_t                        length_dw;
         uint64_t                        gpu_addr;
         uint32_t                        *ptr;
+        int                             ring;
         struct radeon_fence             *fence;
         unsigned                        vm_id;
         bool                            is_const_ib;
+        struct radeon_fence             *sync_to[RADEON_NUM_RINGS];
         struct radeon_semaphore         *semaphore;
 };
 
@@ -642,6 +622,9 @@ struct radeon_ring {
         unsigned                rptr;
         unsigned                rptr_offs;
         unsigned                rptr_reg;
+        unsigned                rptr_save_reg;
+        u64                     next_rptr_gpu_addr;
+        volatile u32            *next_rptr_cpu_addr;
         unsigned                wptr;
         unsigned                wptr_old;
         unsigned                wptr_reg;
@@ -657,6 +640,7 @@ struct radeon_ring {
         u32                     ptr_reg_shift;
         u32                     ptr_reg_mask;
         u32                     nop;
+        u32                     idx;
 };
 
 /*
@@ -690,6 +674,7 @@ struct radeon_vm_funcs {
 };
 
 struct radeon_vm_manager {
+        struct mutex                    lock;
         struct list_head                lru_vm;
         uint32_t                        use_bitmap;
         struct radeon_sa_manager        sa_manager;
@@ -718,13 +703,10 @@ struct r600_ih {
         struct radeon_bo        *ring_obj;
         volatile uint32_t       *ring;
         unsigned                rptr;
-        unsigned                rptr_offs;
-        unsigned                wptr;
-        unsigned                wptr_old;
         unsigned                ring_size;
         uint64_t                gpu_addr;
         uint32_t                ptr_mask;
-        spinlock_t              lock;
+        atomic_t                lock;
         bool                    enabled;
 };
 
@@ -757,8 +739,6 @@ struct r600_blit {
         u32 state_len;
 };
 
-void r600_blit_suspend(struct radeon_device *rdev);
-
 /*
  * SI RLC stuff
  */
@@ -774,14 +754,14 @@ struct si_rlc {
 int radeon_ib_get(struct radeon_device *rdev, int ring,
                   struct radeon_ib *ib, unsigned size);
 void radeon_ib_free(struct radeon_device *rdev, struct radeon_ib *ib);
-int radeon_ib_schedule(struct radeon_device *rdev, struct radeon_ib *ib);
+int radeon_ib_schedule(struct radeon_device *rdev, struct radeon_ib *ib,
+                       struct radeon_ib *const_ib);
 int radeon_ib_pool_init(struct radeon_device *rdev);
 void radeon_ib_pool_fini(struct radeon_device *rdev);
-int radeon_ib_pool_start(struct radeon_device *rdev);
-int radeon_ib_pool_suspend(struct radeon_device *rdev);
 int radeon_ib_ring_tests(struct radeon_device *rdev);
 /* Ring access between begin & end cannot sleep */
-int radeon_ring_index(struct radeon_device *rdev, struct radeon_ring *cp);
+bool radeon_ring_supports_scratch_reg(struct radeon_device *rdev,
+                                      struct radeon_ring *ring);
 void radeon_ring_free_size(struct radeon_device *rdev, struct radeon_ring *cp);
 int radeon_ring_alloc(struct radeon_device *rdev, struct radeon_ring *cp, unsigned ndw);
 int radeon_ring_lock(struct radeon_device *rdev, struct radeon_ring *cp, unsigned ndw);
@@ -793,6 +773,10 @@ int radeon_ring_test(struct radeon_device *rdev, struct radeon_ring *cp);
 void radeon_ring_force_activity(struct radeon_device *rdev, struct radeon_ring *ring);
 void radeon_ring_lockup_update(struct radeon_ring *ring);
 bool radeon_ring_test_lockup(struct radeon_device *rdev, struct radeon_ring *ring);
+unsigned radeon_ring_backup(struct radeon_device *rdev, struct radeon_ring *ring,
+                            uint32_t **data);
+int radeon_ring_restore(struct radeon_device *rdev, struct radeon_ring *ring,
+                        unsigned size, uint32_t *data);
 int radeon_ring_init(struct radeon_device *rdev, struct radeon_ring *cp, unsigned ring_size,
                      unsigned rptr_offs, unsigned rptr_reg, unsigned wptr_reg,
                      u32 ptr_reg_shift, u32 ptr_reg_mask, u32 nop);
@@ -891,6 +875,7 @@ struct radeon_wb {
 };
 
 #define RADEON_WB_SCRATCH_OFFSET 0
+#define RADEON_WB_RING0_NEXT_RPTR 256
 #define RADEON_WB_CP_RPTR_OFFSET 1024
 #define RADEON_WB_CP1_RPTR_OFFSET 1280
 #define RADEON_WB_CP2_RPTR_OFFSET 1536
@@ -1039,11 +1024,12 @@ struct radeon_power_state {
 
 struct radeon_pm {
         struct mutex            mutex;
+        /* write locked while reprogramming mclk */
+        struct rw_semaphore     mclk_lock;
         u32                     active_crtcs;
         int                     active_crtc_count;
         int                     req_vblank;
         bool                    vblank_sync;
-        bool                    gui_idle;
         fixed20_12              max_bandwidth;
         fixed20_12              igp_sideport_mclk;
         fixed20_12              igp_system_mclk;
@@ -1192,20 +1178,20 @@ struct radeon_asic {
                             uint64_t src_offset,
                             uint64_t dst_offset,
                             unsigned num_gpu_pages,
-                            struct radeon_fence *fence);
+                            struct radeon_fence **fence);
                 u32 blit_ring_index;
                 int (*dma)(struct radeon_device *rdev,
                            uint64_t src_offset,
                            uint64_t dst_offset,
                            unsigned num_gpu_pages,
-                           struct radeon_fence *fence);
+                           struct radeon_fence **fence);
                 u32 dma_ring_index;
                 /* method used for bo copy */
                 int (*copy)(struct radeon_device *rdev,
                             uint64_t src_offset,
                             uint64_t dst_offset,
                             unsigned num_gpu_pages,
-                            struct radeon_fence *fence);
+                            struct radeon_fence **fence);
                 /* ring used for bo copies */
                 u32 copy_ring_index;
         } copy;
@@ -1467,6 +1453,7 @@ struct radeon_device {
         struct device                   *dev;
         struct drm_device               *ddev;
         struct pci_dev                  *pdev;
+        struct rw_semaphore             exclusive_lock;
         /* ASIC */
         union radeon_asic_config        config;
         enum radeon_family              family;
@@ -1512,7 +1499,6 @@ struct radeon_device {
         struct radeon_gem               gem;
         struct radeon_pm                pm;
         uint32_t                        bios_scratch[RADEON_BIOS_NUM_SCRATCH];
-        struct radeon_mutex             cs_mutex;
         struct radeon_wb                wb;
         struct radeon_dummy_page        dummy_page;
         bool                            shutdown;
@@ -1534,7 +1520,6 @@ struct radeon_device {
         struct work_struct audio_work;
         int num_crtc; /* number of crtcs */
         struct mutex dc_hw_i2c_mutex; /* display controller hw i2c mutex */
-        struct mutex vram_mutex;
         bool audio_enabled;
         struct r600_audio audio_status; /* audio stuff */
         struct notifier_block acpi_nb;
@@ -1785,8 +1770,6 @@ extern void radeon_ttm_set_active_vram_size(struct radeon_device *rdev, u64 size
  */
 int radeon_vm_manager_init(struct radeon_device *rdev);
 void radeon_vm_manager_fini(struct radeon_device *rdev);
-int radeon_vm_manager_start(struct radeon_device *rdev);
-int radeon_vm_manager_suspend(struct radeon_device *rdev);
 int radeon_vm_init(struct radeon_device *rdev, struct radeon_vm *vm);
 void radeon_vm_fini(struct radeon_device *rdev, struct radeon_vm *vm);
 int radeon_vm_bind(struct radeon_device *rdev, struct radeon_vm *vm);
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index f533df5f7d50..973417c4b014 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -40,6 +40,16 @@
 /*
  * Registers accessors functions.
  */
+/**
+ * radeon_invalid_rreg - dummy reg read function
+ *
+ * @rdev: radeon device pointer
+ * @reg: offset of register
+ *
+ * Dummy register read function.  Used for register blocks
+ * that certain asics don't have (all asics).
+ * Returns the value in the register.
+ */
 static uint32_t radeon_invalid_rreg(struct radeon_device *rdev, uint32_t reg)
 {
         DRM_ERROR("Invalid callback to read register 0x%04X\n", reg);
@@ -47,6 +57,16 @@ static uint32_t radeon_invalid_rreg(struct radeon_device *rdev, uint32_t reg)
         return 0;
 }
 
+/**
+ * radeon_invalid_wreg - dummy reg write function
+ *
+ * @rdev: radeon device pointer
+ * @reg: offset of register
+ * @v: value to write to the register
+ *
+ * Dummy register read function.  Used for register blocks
+ * that certain asics don't have (all asics).
+ */
 static void radeon_invalid_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v)
 {
         DRM_ERROR("Invalid callback to write register 0x%04X with 0x%08X\n",
@@ -54,6 +74,14 @@ static void radeon_invalid_wreg(struct radeon_device *rdev, uint32_t reg, uint32
         BUG_ON(1);
 }
 
+/**
+ * radeon_register_accessor_init - sets up the register accessor callbacks
+ *
+ * @rdev: radeon device pointer
+ *
+ * Sets up the register accessor callbacks for various register
+ * apertures.  Not all asics have all apertures (all asics).
+ */
 static void radeon_register_accessor_init(struct radeon_device *rdev)
 {
         rdev->mc_rreg = &radeon_invalid_rreg;
@@ -102,6 +130,14 @@ static void radeon_register_accessor_init(struct radeon_device *rdev)
 
 
 /* helper to disable agp */
+/**
+ * radeon_agp_disable - AGP disable helper function
+ *
+ * @rdev: radeon device pointer
+ *
+ * Removes AGP flags and changes the gart callbacks on AGP
+ * cards when using the internal gart rather than AGP (all asics).
+ */
 void radeon_agp_disable(struct radeon_device *rdev)
 {
         rdev->flags &= ~RADEON_IS_AGP;
@@ -1608,6 +1644,16 @@ static struct radeon_asic si_asic = {
         },
 };
 
+/**
+ * radeon_asic_init - register asic specific callbacks
+ *
+ * @rdev: radeon device pointer
+ *
+ * Registers the appropriate asic specific callbacks for each
+ * chip family.  Also sets other asics specific info like the number
+ * of crtcs and the register aperture accessors (all asics).
+ * Returns 0 for success.
+ */
 int radeon_asic_init(struct radeon_device *rdev)
 {
         radeon_register_accessor_init(rdev);
diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h
index e76a941ef14e..f4af24310438 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.h
+++ b/drivers/gpu/drm/radeon/radeon_asic.h
@@ -79,7 +79,7 @@ int r100_copy_blit(struct radeon_device *rdev,
                    uint64_t src_offset,
                    uint64_t dst_offset,
                    unsigned num_gpu_pages,
-                   struct radeon_fence *fence);
+                   struct radeon_fence **fence);
 int r100_set_surface_reg(struct radeon_device *rdev, int reg,
                          uint32_t tiling_flags, uint32_t pitch,
                          uint32_t offset, uint32_t obj_size);
@@ -103,7 +103,6 @@ int r100_pci_gart_enable(struct radeon_device *rdev);
 void r100_pci_gart_disable(struct radeon_device *rdev);
 int r100_debugfs_mc_info_init(struct radeon_device *rdev);
 int r100_gui_wait_for_idle(struct radeon_device *rdev);
-void r100_ib_fini(struct radeon_device *rdev);
 int r100_ib_test(struct radeon_device *rdev, struct radeon_ring *ring);
 void r100_irq_disable(struct radeon_device *rdev);
 void r100_mc_stop(struct radeon_device *rdev, struct r100_mc_save *save);
@@ -144,7 +143,7 @@ extern int r200_copy_dma(struct radeon_device *rdev,
                          uint64_t src_offset,
                          uint64_t dst_offset,
                          unsigned num_gpu_pages,
-                         struct radeon_fence *fence);
+                         struct radeon_fence **fence);
 void r200_set_safe_registers(struct radeon_device *rdev);
 
 /*
@@ -318,7 +317,7 @@ void r600_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib);
 int r600_ring_test(struct radeon_device *rdev, struct radeon_ring *cp);
 int r600_copy_blit(struct radeon_device *rdev,
                    uint64_t src_offset, uint64_t dst_offset,
-                   unsigned num_gpu_pages, struct radeon_fence *fence);
+                   unsigned num_gpu_pages, struct radeon_fence **fence);
 void r600_hpd_init(struct radeon_device *rdev);
 void r600_hpd_fini(struct radeon_device *rdev);
 bool r600_hpd_sense(struct radeon_device *rdev, enum radeon_hpd_id hpd);
@@ -363,9 +362,10 @@ int r600_hdmi_buffer_status_changed(struct drm_encoder *encoder);
 void r600_hdmi_update_audio_settings(struct drm_encoder *encoder);
 /* r600 blit */
 int r600_blit_prepare_copy(struct radeon_device *rdev, unsigned num_gpu_pages,
-                           struct radeon_sa_bo **vb);
-void r600_blit_done_copy(struct radeon_device *rdev, struct radeon_fence *fence,
-                         struct radeon_sa_bo *vb);
+                           struct radeon_fence **fence, struct radeon_sa_bo **vb,
+                           struct radeon_semaphore **sem);
+void r600_blit_done_copy(struct radeon_device *rdev, struct radeon_fence **fence,
+                         struct radeon_sa_bo *vb, struct radeon_semaphore *sem);
 void r600_kms_blit_copy(struct radeon_device *rdev,
                         u64 src_gpu_addr, u64 dst_gpu_addr,
                         unsigned num_gpu_pages,
diff --git a/drivers/gpu/drm/radeon/radeon_benchmark.c b/drivers/gpu/drm/radeon/radeon_benchmark.c
index 364f5b1a04b9..bedda9caadd9 100644
--- a/drivers/gpu/drm/radeon/radeon_benchmark.c
+++ b/drivers/gpu/drm/radeon/radeon_benchmark.c
@@ -45,20 +45,14 @@ static int radeon_benchmark_do_move(struct radeon_device *rdev, unsigned size,
         for (i = 0; i < n; i++) {
                 switch (flag) {
                 case RADEON_BENCHMARK_COPY_DMA:
-                        r = radeon_fence_create(rdev, &fence, radeon_copy_dma_ring_index(rdev));
-                        if (r)
-                                return r;
                         r = radeon_copy_dma(rdev, saddr, daddr,
                                             size / RADEON_GPU_PAGE_SIZE,
-                                            fence);
+                                            &fence);
                         break;
                 case RADEON_BENCHMARK_COPY_BLIT:
-                        r = radeon_fence_create(rdev, &fence, radeon_copy_blit_ring_index(rdev));
-                        if (r)
-                                return r;
                         r = radeon_copy_blit(rdev, saddr, daddr,
                                              size / RADEON_GPU_PAGE_SIZE,
-                                             fence);
+                                             &fence);
                         break;
                 default:
                         DRM_ERROR("Unknown copy method\n");
diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c
index 2914c5761cfc..895e628b60f8 100644
--- a/drivers/gpu/drm/radeon/radeon_connectors.c
+++ b/drivers/gpu/drm/radeon/radeon_connectors.c
@@ -64,14 +64,33 @@ void radeon_connector_hotplug(struct drm_connector *connector)
 
         /* just deal with DP (not eDP) here. */
         if (connector->connector_type == DRM_MODE_CONNECTOR_DisplayPort) {
-                int saved_dpms = connector->dpms;
-
-                /* Only turn off the display it it's physically disconnected */
-                if (!radeon_hpd_sense(rdev, radeon_connector->hpd.hpd))
-                        drm_helper_connector_dpms(connector, DRM_MODE_DPMS_OFF);
-                else if (radeon_dp_needs_link_train(radeon_connector))
-                        drm_helper_connector_dpms(connector, DRM_MODE_DPMS_ON);
-                connector->dpms = saved_dpms;
+                struct radeon_connector_atom_dig *dig_connector =
+                        radeon_connector->con_priv;
+
+                /* if existing sink type was not DP no need to retrain */
+                if (dig_connector->dp_sink_type != CONNECTOR_OBJECT_ID_DISPLAYPORT)
+                        return;
+
+                /* first get sink type as it may be reset after (un)plug */
+                dig_connector->dp_sink_type = radeon_dp_getsinktype(radeon_connector);
+                /* don't do anything if sink is not display port, i.e.,
+                 * passive dp->(dvi|hdmi) adaptor
+                 */
+                if (dig_connector->dp_sink_type == CONNECTOR_OBJECT_ID_DISPLAYPORT) {
+                        int saved_dpms = connector->dpms;
+                        /* Only turn off the display if it's physically disconnected */
+                        if (!radeon_hpd_sense(rdev, radeon_connector->hpd.hpd)) {
+                                drm_helper_connector_dpms(connector, DRM_MODE_DPMS_OFF);
+                        } else if (radeon_dp_needs_link_train(radeon_connector)) {
+                                /* set it to OFF so that drm_helper_connector_dpms()
+                                 * won't return immediately since the current state
+                                 * is ON at this point.
+                                 */
+                                connector->dpms = DRM_MODE_DPMS_OFF;
+                                drm_helper_connector_dpms(connector, DRM_MODE_DPMS_ON);
+                        }
+                        connector->dpms = saved_dpms;
+                }
         }
 }
 
diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c
index 142f89462aa4..8a4c49ef0cc4 100644
--- a/drivers/gpu/drm/radeon/radeon_cs.c
+++ b/drivers/gpu/drm/radeon/radeon_cs.c
@@ -115,36 +115,20 @@ static int radeon_cs_get_ring(struct radeon_cs_parser *p, u32 ring, s32 priority
         return 0;
 }
 
-static int radeon_cs_sync_rings(struct radeon_cs_parser *p)
+static void radeon_cs_sync_rings(struct radeon_cs_parser *p)
 {
-        bool sync_to_ring[RADEON_NUM_RINGS] = { };
-        bool need_sync = false;
-        int i, r;
+        int i;
 
         for (i = 0; i < p->nrelocs; i++) {
-                struct radeon_fence *fence;
+                struct radeon_fence *a, *b;
 
                 if (!p->relocs[i].robj || !p->relocs[i].robj->tbo.sync_obj)
                         continue;
 
-                fence = p->relocs[i].robj->tbo.sync_obj;
-                if (fence->ring != p->ring && !radeon_fence_signaled(fence)) {
-                        sync_to_ring[fence->ring] = true;
-                        need_sync = true;
-                }
-        }
-
-        if (!need_sync) {
-                return 0;
-        }
-
-        r = radeon_semaphore_create(p->rdev, &p->ib.semaphore);
-        if (r) {
-                return r;
+                a = p->relocs[i].robj->tbo.sync_obj;
+                b = p->ib.sync_to[a->ring];
+                p->ib.sync_to[a->ring] = radeon_fence_later(a, b);
         }
-
-        return radeon_semaphore_sync_rings(p->rdev, p->ib.semaphore,
-                                           sync_to_ring, p->ring);
 }
 
 /* XXX: note that this is called from the legacy UMS CS ioctl as well */
@@ -368,16 +352,13 @@ static int radeon_cs_ib_chunk(struct radeon_device *rdev,
                 DRM_ERROR("Invalid command stream !\n");
                 return r;
         }
-        r = radeon_cs_sync_rings(parser);
-        if (r) {
-                DRM_ERROR("Failed to synchronize rings !\n");
-        }
+        radeon_cs_sync_rings(parser);
         parser->ib.vm_id = 0;
-        r = radeon_ib_schedule(rdev, &parser->ib);
+        r = radeon_ib_schedule(rdev, &parser->ib, NULL);
         if (r) {
                 DRM_ERROR("Failed to schedule IB !\n");
         }
-        return 0;
+        return r;
 }
 
 static int radeon_bo_vm_update_pte(struct radeon_cs_parser *parser,
@@ -459,6 +440,7 @@ static int radeon_cs_ib_vm_chunk(struct radeon_device *rdev,
                 return r;
         }
 
+        mutex_lock(&rdev->vm_manager.lock);
         mutex_lock(&vm->mutex);
         r = radeon_vm_bind(rdev, vm);
         if (r) {
@@ -468,30 +450,26 @@ static int radeon_cs_ib_vm_chunk(struct radeon_device *rdev,
         if (r) {
                 goto out;
         }
-        r = radeon_cs_sync_rings(parser);
-        if (r) {
-                DRM_ERROR("Failed to synchronize rings !\n");
-        }
+        radeon_cs_sync_rings(parser);
+
+        parser->ib.vm_id = vm->id;
+        /* ib pool is bind at 0 in virtual address space,
+         * so gpu_addr is the offset inside the pool bo
+         */
+        parser->ib.gpu_addr = parser->ib.sa_bo->soffset;
 
         if ((rdev->family >= CHIP_TAHITI) &&
             (parser->chunk_const_ib_idx != -1)) {
                 parser->const_ib.vm_id = vm->id;
-                /* ib pool is bind at 0 in virtual address space to gpu_addr is the
-                 * offset inside the pool bo
+                /* ib pool is bind at 0 in virtual address space,
+                 * so gpu_addr is the offset inside the pool bo
                  */
                 parser->const_ib.gpu_addr = parser->const_ib.sa_bo->soffset;
-                r = radeon_ib_schedule(rdev, &parser->const_ib);
-                if (r)
-                        goto out;
+                r = radeon_ib_schedule(rdev, &parser->ib, &parser->const_ib);
+        } else {
+                r = radeon_ib_schedule(rdev, &parser->ib, NULL);
         }
 
-        parser->ib.vm_id = vm->id;
-        /* ib pool is bind at 0 in virtual address space to gpu_addr is the
-         * offset inside the pool bo
-         */
-        parser->ib.gpu_addr = parser->ib.sa_bo->soffset;
-        parser->ib.is_const_ib = false;
-        r = radeon_ib_schedule(rdev, &parser->ib);
 out:
         if (!r) {
                 if (vm->fence) {
@@ -499,7 +477,8 @@ out:
                 }
                 vm->fence = radeon_fence_ref(parser->ib.fence);
         }
-        mutex_unlock(&fpriv->vm.mutex);
+        mutex_unlock(&vm->mutex);
+        mutex_unlock(&rdev->vm_manager.lock);
         return r;
 }
 
@@ -519,9 +498,9 @@ int radeon_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
         struct radeon_cs_parser parser;
         int r;
 
-        radeon_mutex_lock(&rdev->cs_mutex);
+        down_read(&rdev->exclusive_lock);
         if (!rdev->accel_working) {
-                radeon_mutex_unlock(&rdev->cs_mutex);
+                up_read(&rdev->exclusive_lock);
                 return -EBUSY;
         }
         /* initialize parser */
@@ -534,8 +513,8 @@ int radeon_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
         if (r) {
                 DRM_ERROR("Failed to initialize parser !\n");
                 radeon_cs_parser_fini(&parser, r);
+                up_read(&rdev->exclusive_lock);
                 r = radeon_cs_handle_lockup(rdev, r);
-                radeon_mutex_unlock(&rdev->cs_mutex);
                 return r;
         }
         r = radeon_cs_parser_relocs(&parser);
@@ -543,8 +522,8 @@ int radeon_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
                 if (r != -ERESTARTSYS)
                         DRM_ERROR("Failed to parse relocation %d!\n", r);
                 radeon_cs_parser_fini(&parser, r);
+                up_read(&rdev->exclusive_lock);
                 r = radeon_cs_handle_lockup(rdev, r);
-                radeon_mutex_unlock(&rdev->cs_mutex);
                 return r;
         }
         r = radeon_cs_ib_chunk(rdev, &parser);
@@ -557,8 +536,8 @@ int radeon_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
         }
 out:
         radeon_cs_parser_fini(&parser, r);
+        up_read(&rdev->exclusive_lock);
         r = radeon_cs_handle_lockup(rdev, r);
-        radeon_mutex_unlock(&rdev->cs_mutex);
         return r;
 }
 
diff --git a/drivers/gpu/drm/radeon/radeon_cursor.c b/drivers/gpu/drm/radeon/radeon_cursor.c
index 42acc6449dd6..711e95ad39bf 100644
--- a/drivers/gpu/drm/radeon/radeon_cursor.c
+++ b/drivers/gpu/drm/radeon/radeon_cursor.c
@@ -262,8 +262,14 @@ int radeon_crtc_cursor_move(struct drm_crtc *crtc,
                                 if (!(cursor_end & 0x7f))
                                         w--;
                         }
-                        if (w <= 0)
+                        if (w <= 0) {
                                 w = 1;
+                                cursor_end = x - xorigin + w;
+                                if (!(cursor_end & 0x7f)) {
+                                        x--;
+                                        WARN_ON_ONCE(x < 0);
+                                }
+                        }
                 }
         }
 
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index 066c98b888a5..742af8244e89 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -96,8 +96,12 @@ static const char radeon_family_name[][16] = {
         "LAST",
 };
 
-/*
- * Clear GPU surface registers.
+/**
+ * radeon_surface_init - Clear GPU surface registers.
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Clear GPU surface registers (r1xx-r5xx).
  */
 void radeon_surface_init(struct radeon_device *rdev)
 {
@@ -119,6 +123,13 @@ void radeon_surface_init(struct radeon_device *rdev)
 /*
  * GPU scratch registers helpers function.
  */
+/**
+ * radeon_scratch_init - Init scratch register driver information.
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Init CP scratch register driver information (r1xx-r5xx)
+ */
 void radeon_scratch_init(struct radeon_device *rdev)
 {
         int i;
@@ -136,6 +147,15 @@ void radeon_scratch_init(struct radeon_device *rdev)
         }
 }
 
+/**
+ * radeon_scratch_get - Allocate a scratch register
+ *
+ * @rdev: radeon_device pointer
+ * @reg: scratch register mmio offset
+ *
+ * Allocate a CP scratch register for use by the driver (all asics).
+ * Returns 0 on success or -EINVAL on failure.
+ */
 int radeon_scratch_get(struct radeon_device *rdev, uint32_t *reg)
 {
         int i;
@@ -150,6 +170,14 @@ int radeon_scratch_get(struct radeon_device *rdev, uint32_t *reg)
         return -EINVAL;
 }
 
+/**
+ * radeon_scratch_free - Free a scratch register
+ *
+ * @rdev: radeon_device pointer
+ * @reg: scratch register mmio offset
+ *
+ * Free a CP scratch register allocated for use by the driver (all asics)
+ */
 void radeon_scratch_free(struct radeon_device *rdev, uint32_t reg)
 {
         int i;
@@ -162,6 +190,20 @@ void radeon_scratch_free(struct radeon_device *rdev, uint32_t reg)
         }
 }
 
+/*
+ * radeon_wb_*()
+ * Writeback is the the method by which the the GPU updates special pages
+ * in memory with the status of certain GPU events (fences, ring pointers,
+ * etc.).
+ */
+
+/**
+ * radeon_wb_disable - Disable Writeback
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Disables Writeback (all asics).  Used for suspend.
+ */
 void radeon_wb_disable(struct radeon_device *rdev)
 {
         int r;
@@ -177,6 +219,14 @@ void radeon_wb_disable(struct radeon_device *rdev)
         rdev->wb.enabled = false;
 }
 
+/**
+ * radeon_wb_fini - Disable Writeback and free memory
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Disables Writeback and frees the Writeback memory (all asics).
+ * Used at driver shutdown.
+ */
 void radeon_wb_fini(struct radeon_device *rdev)
 {
         radeon_wb_disable(rdev);
@@ -187,6 +237,15 @@ void radeon_wb_fini(struct radeon_device *rdev)
         }
 }
 
+/**
+ * radeon_wb_init- Init Writeback driver info and allocate memory
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Disables Writeback and frees the Writeback memory (all asics).
+ * Used at driver startup.
+ * Returns 0 on success or an -error on failure.
+ */
 int radeon_wb_init(struct radeon_device *rdev)
 {
         int r;
@@ -355,6 +414,15 @@ void radeon_gtt_location(struct radeon_device *rdev, struct radeon_mc *mc)
 /*
  * GPU helpers function.
  */
+/**
+ * radeon_card_posted - check if the hw has already been initialized
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Check if the asic has been initialized (all asics).
+ * Used at driver startup.
+ * Returns true if initialized or false if not.
+ */
 bool radeon_card_posted(struct radeon_device *rdev)
 {
         uint32_t reg;
@@ -404,6 +472,14 @@ bool radeon_card_posted(struct radeon_device *rdev)
 
 }
 
+/**
+ * radeon_update_bandwidth_info - update display bandwidth params
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Used when sclk/mclk are switched or display modes are set.
+ * params are used to calculate display watermarks (all asics)
+ */
 void radeon_update_bandwidth_info(struct radeon_device *rdev)
 {
         fixed20_12 a;
@@ -424,6 +500,15 @@ void radeon_update_bandwidth_info(struct radeon_device *rdev)
         }
 }
 
+/**
+ * radeon_boot_test_post_card - check and possibly initialize the hw
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Check if the asic is initialized and if not, attempt to initialize
+ * it (all asics).
+ * Returns true if initialized or false if not.
+ */
 bool radeon_boot_test_post_card(struct radeon_device *rdev)
 {
         if (radeon_card_posted(rdev))
@@ -442,6 +527,16 @@ bool radeon_boot_test_post_card(struct radeon_device *rdev)
         }
 }
 
+/**
+ * radeon_dummy_page_init - init dummy page used by the driver
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Allocate the dummy page used by the driver (all asics).
+ * This dummy page is used by the driver as a filler for gart entries
+ * when pages are taken out of the GART
+ * Returns 0 on sucess, -ENOMEM on failure.
+ */
 int radeon_dummy_page_init(struct radeon_device *rdev)
 {
         if (rdev->dummy_page.page)
@@ -460,6 +555,13 @@ int radeon_dummy_page_init(struct radeon_device *rdev)
         return 0;
 }
 
+/**
+ * radeon_dummy_page_fini - free dummy page used by the driver
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Frees the dummy page used by the driver (all asics).
+ */
 void radeon_dummy_page_fini(struct radeon_device *rdev)
 {
         if (rdev->dummy_page.page == NULL)
@@ -472,6 +574,23 @@ void radeon_dummy_page_fini(struct radeon_device *rdev)
 
 
 /* ATOM accessor methods */
+/*
+ * ATOM is an interpreted byte code stored in tables in the vbios.  The
+ * driver registers callbacks to access registers and the interpreter
+ * in the driver parses the tables and executes then to program specific
+ * actions (set display modes, asic init, etc.).  See radeon_atombios.c,
+ * atombios.h, and atom.c
+ */
+
+/**
+ * cail_pll_read - read PLL register
+ *
+ * @info: atom card_info pointer
+ * @reg: PLL register offset
+ *
+ * Provides a PLL register accessor for the atom interpreter (r4xx+).
+ * Returns the value of the PLL register.
+ */
 static uint32_t cail_pll_read(struct card_info *info, uint32_t reg)
 {
         struct radeon_device *rdev = info->dev->dev_private;
@@ -481,6 +600,15 @@ static uint32_t cail_pll_read(struct card_info *info, uint32_t reg)
         return r;
 }
 
+/**
+ * cail_pll_write - write PLL register
+ *
+ * @info: atom card_info pointer
+ * @reg: PLL register offset
+ * @val: value to write to the pll register
+ *
+ * Provides a PLL register accessor for the atom interpreter (r4xx+).
+ */
 static void cail_pll_write(struct card_info *info, uint32_t reg, uint32_t val)
 {
         struct radeon_device *rdev = info->dev->dev_private;
@@ -488,6 +616,15 @@ static void cail_pll_write(struct card_info *info, uint32_t reg, uint32_t val)
         rdev->pll_wreg(rdev, reg, val);
 }
 
+/**
+ * cail_mc_read - read MC (Memory Controller) register
+ *
+ * @info: atom card_info pointer
+ * @reg: MC register offset
+ *
+ * Provides an MC register accessor for the atom interpreter (r4xx+).
+ * Returns the value of the MC register.
+ */
 static uint32_t cail_mc_read(struct card_info *info, uint32_t reg)
 {
         struct radeon_device *rdev = info->dev->dev_private;
@@ -497,6 +634,15 @@ static uint32_t cail_mc_read(struct card_info *info, uint32_t reg)
         return r;
 }
 
+/**
+ * cail_mc_write - write MC (Memory Controller) register
+ *
+ * @info: atom card_info pointer
+ * @reg: MC register offset
+ * @val: value to write to the pll register
+ *
+ * Provides a MC register accessor for the atom interpreter (r4xx+).
+ */
 static void cail_mc_write(struct card_info *info, uint32_t reg, uint32_t val)
 {
         struct radeon_device *rdev = info->dev->dev_private;
@@ -504,6 +650,15 @@ static void cail_mc_write(struct card_info *info, uint32_t reg, uint32_t val)
         rdev->mc_wreg(rdev, reg, val);
 }
 
+/**
+ * cail_reg_write - write MMIO register
+ *
+ * @info: atom card_info pointer
+ * @reg: MMIO register offset
+ * @val: value to write to the pll register
+ *
+ * Provides a MMIO register accessor for the atom interpreter (r4xx+).
+ */
 static void cail_reg_write(struct card_info *info, uint32_t reg, uint32_t val)
 {
         struct radeon_device *rdev = info->dev->dev_private;
@@ -511,6 +666,15 @@ static void cail_reg_write(struct card_info *info, uint32_t reg, uint32_t val)
         WREG32(reg*4, val);
 }
 
+/**
+ * cail_reg_read - read MMIO register
+ *
+ * @info: atom card_info pointer
+ * @reg: MMIO register offset
+ *
+ * Provides an MMIO register accessor for the atom interpreter (r4xx+).
+ * Returns the value of the MMIO register.
+ */
 static uint32_t cail_reg_read(struct card_info *info, uint32_t reg)
 {
         struct radeon_device *rdev = info->dev->dev_private;
@@ -520,6 +684,15 @@ static uint32_t cail_reg_read(struct card_info *info, uint32_t reg)
         return r;
 }
 
+/**
+ * cail_ioreg_write - write IO register
+ *
+ * @info: atom card_info pointer
+ * @reg: IO register offset
+ * @val: value to write to the pll register
+ *
+ * Provides a IO register accessor for the atom interpreter (r4xx+).
+ */
 static void cail_ioreg_write(struct card_info *info, uint32_t reg, uint32_t val)
 {
         struct radeon_device *rdev = info->dev->dev_private;
@@ -527,6 +700,15 @@ static void cail_ioreg_write(struct card_info *info, uint32_t reg, uint32_t val)
         WREG32_IO(reg*4, val);
 }
 
+/**
+ * cail_ioreg_read - read IO register
+ *
+ * @info: atom card_info pointer
+ * @reg: IO register offset
+ *
+ * Provides an IO register accessor for the atom interpreter (r4xx+).
+ * Returns the value of the IO register.
+ */
 static uint32_t cail_ioreg_read(struct card_info *info, uint32_t reg)
 {
         struct radeon_device *rdev = info->dev->dev_private;
@@ -536,6 +718,16 @@ static uint32_t cail_ioreg_read(struct card_info *info, uint32_t reg)
         return r;
 }
 
+/**
+ * radeon_atombios_init - init the driver info and callbacks for atombios
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Initializes the driver info and register access callbacks for the
+ * ATOM interpreter (r4xx+).
+ * Returns 0 on sucess, -ENOMEM on failure.
+ * Called at driver startup.
+ */
 int radeon_atombios_init(struct radeon_device *rdev)
 {
         struct card_info *atom_card_info =
@@ -569,6 +761,15 @@ int radeon_atombios_init(struct radeon_device *rdev)
         return 0;
 }
 
+/**
+ * radeon_atombios_fini - free the driver info and callbacks for atombios
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Frees the driver info and register access callbacks for the ATOM
+ * interpreter (r4xx+).
+ * Called at driver shutdown.
+ */
 void radeon_atombios_fini(struct radeon_device *rdev)
 {
         if (rdev->mode_info.atom_context) {
@@ -578,17 +779,50 @@ void radeon_atombios_fini(struct radeon_device *rdev)
         kfree(rdev->mode_info.atom_card_info);
 }
 
+/* COMBIOS */
+/*
+ * COMBIOS is the bios format prior to ATOM. It provides
+ * command tables similar to ATOM, but doesn't have a unified
+ * parser.  See radeon_combios.c
+ */
+
+/**
+ * radeon_combios_init - init the driver info for combios
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Initializes the driver info for combios (r1xx-r3xx).
+ * Returns 0 on sucess.
+ * Called at driver startup.
+ */
 int radeon_combios_init(struct radeon_device *rdev)
 {
         radeon_combios_initialize_bios_scratch_regs(rdev->ddev);
         return 0;
 }
 
+/**
+ * radeon_combios_fini - free the driver info for combios
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Frees the driver info for combios (r1xx-r3xx).
+ * Called at driver shutdown.
+ */
 void radeon_combios_fini(struct radeon_device *rdev)
 {
 }
 
-/* if we get transitioned to only one device, tak VGA back */
+/* if we get transitioned to only one device, take VGA back */
+/**
+ * radeon_vga_set_decode - enable/disable vga decode
+ *
+ * @cookie: radeon_device pointer
+ * @state: enable/disable vga decode
+ *
+ * Enable/disable vga decode (all asics).
+ * Returns VGA resource flags.
+ */
 static unsigned int radeon_vga_set_decode(void *cookie, bool state)
 {
         struct radeon_device *rdev = cookie;
@@ -600,6 +834,14 @@ static unsigned int radeon_vga_set_decode(void *cookie, bool state)
                 return VGA_RSRC_NORMAL_IO | VGA_RSRC_NORMAL_MEM;
 }
 
+/**
+ * radeon_check_arguments - validate module params
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Validates certain module parameters and updates
+ * the associated values used by the driver (all asics).
+ */
 void radeon_check_arguments(struct radeon_device *rdev)
 {
         /* vramlimit must be a power of two */
@@ -666,6 +908,15 @@ void radeon_check_arguments(struct radeon_device *rdev)
         }
 }
 
+/**
+ * radeon_switcheroo_set_state - set switcheroo state
+ *
+ * @pdev: pci dev pointer
+ * @state: vga switcheroo state
+ *
+ * Callback for the switcheroo driver.  Suspends or resumes the
+ * the asics before or after it is powered up using ACPI methods.
+ */
 static void radeon_switcheroo_set_state(struct pci_dev *pdev, enum vga_switcheroo_state state)
 {
         struct drm_device *dev = pci_get_drvdata(pdev);
@@ -686,6 +937,15 @@ static void radeon_switcheroo_set_state(struct pci_dev *pdev, enum vga_switchero
         }
 }
 
+/**
+ * radeon_switcheroo_can_switch - see if switcheroo state can change
+ *
+ * @pdev: pci dev pointer
+ *
+ * Callback for the switcheroo driver.  Check of the switcheroo
+ * state can be changed.
+ * Returns true if the state can be changed, false if not.
+ */
 static bool radeon_switcheroo_can_switch(struct pci_dev *pdev)
 {
         struct drm_device *dev = pci_get_drvdata(pdev);
@@ -703,6 +963,18 @@ static const struct vga_switcheroo_client_ops radeon_switcheroo_ops = {
         .can_switch = radeon_switcheroo_can_switch,
 };
 
+/**
+ * radeon_device_init - initialize the driver
+ *
+ * @rdev: radeon_device pointer
+ * @pdev: drm dev pointer
+ * @pdev: pci dev pointer
+ * @flags: driver flags
+ *
+ * Initializes the driver info and hw (all asics).
+ * Returns 0 for success or an error on failure.
+ * Called at driver startup.
+ */
 int radeon_device_init(struct radeon_device *rdev,
                        struct drm_device *ddev,
                        struct pci_dev *pdev,
@@ -721,6 +993,10 @@ int radeon_device_init(struct radeon_device *rdev,
         rdev->usec_timeout = RADEON_MAX_USEC_TIMEOUT;
         rdev->mc.gtt_size = radeon_gart_size * 1024 * 1024;
         rdev->accel_working = false;
+        /* set up ring ids */
+        for (i = 0; i < RADEON_NUM_RINGS; i++) {
+                rdev->ring[i].idx = i;
+        }
 
         DRM_INFO("initializing kernel modesetting (%s 0x%04X:0x%04X 0x%04X:0x%04X).\n",
                 radeon_family_name[rdev->family], pdev->vendor, pdev->device,
@@ -728,20 +1004,20 @@ int radeon_device_init(struct radeon_device *rdev,
 
         /* mutex initialization are all done here so we
          * can recall function without having locking issues */
-        radeon_mutex_init(&rdev->cs_mutex);
         mutex_init(&rdev->ring_lock);
         mutex_init(&rdev->dc_hw_i2c_mutex);
-        if (rdev->family >= CHIP_R600)
-                spin_lock_init(&rdev->ih.lock);
+        atomic_set(&rdev->ih.lock, 0);
         mutex_init(&rdev->gem.mutex);
         mutex_init(&rdev->pm.mutex);
-        mutex_init(&rdev->vram_mutex);
+        init_rwsem(&rdev->pm.mclk_lock);
+        init_rwsem(&rdev->exclusive_lock);
         init_waitqueue_head(&rdev->irq.vblank_queue);
         init_waitqueue_head(&rdev->irq.idle_queue);
         r = radeon_gem_init(rdev);
         if (r)
                 return r;
         /* initialize vm here */
+        mutex_init(&rdev->vm_manager.lock);
         rdev->vm_manager.use_bitmap = 1;
         rdev->vm_manager.max_pfn = 1 << 20;
         INIT_LIST_HEAD(&rdev->vm_manager.lru_vm);
@@ -822,6 +1098,10 @@ int radeon_device_init(struct radeon_device *rdev,
         if (r)
                 return r;
 
+        r = radeon_ib_ring_tests(rdev);
+        if (r)
+                DRM_ERROR("ib ring test failed (%d).\n", r);
+
         if (rdev->flags & RADEON_IS_AGP && !rdev->accel_working) {
                 /* Acceleration not working on AGP card try again
                  * with fallback to PCI or PCIE GART
@@ -847,6 +1127,14 @@ int radeon_device_init(struct radeon_device *rdev,
 
 static void radeon_debugfs_remove_files(struct radeon_device *rdev);
 
+/**
+ * radeon_device_fini - tear down the driver
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Tear down the driver info (all asics).
+ * Called at driver shutdown.
+ */
 void radeon_device_fini(struct radeon_device *rdev)
 {
         DRM_INFO("radeon: finishing device.\n");
@@ -868,6 +1156,16 @@ void radeon_device_fini(struct radeon_device *rdev)
 /*
  * Suspend & resume.
  */
+/**
+ * radeon_suspend_kms - initiate device suspend
+ *
+ * @pdev: drm dev pointer
+ * @state: suspend state
+ *
+ * Puts the hw in the suspend state (all asics).
+ * Returns 0 for success or an error on failure.
+ * Called at driver suspend.
+ */
 int radeon_suspend_kms(struct drm_device *dev, pm_message_t state)
 {
         struct radeon_device *rdev;
@@ -942,10 +1240,20 @@ int radeon_suspend_kms(struct drm_device *dev, pm_message_t state)
         return 0;
 }
 
+/**
+ * radeon_resume_kms - initiate device resume
+ *
+ * @pdev: drm dev pointer
+ *
+ * Bring the hw back to operating state (all asics).
+ * Returns 0 for success or an error on failure.
+ * Called at driver resume.
+ */
 int radeon_resume_kms(struct drm_device *dev)
 {
         struct drm_connector *connector;
         struct radeon_device *rdev = dev->dev_private;
+        int r;
 
         if (dev->switch_power_state == DRM_SWITCH_POWER_OFF)
                 return 0;
@@ -960,6 +1268,11 @@ int radeon_resume_kms(struct drm_device *dev)
         /* resume AGP if in use */
         radeon_agp_resume(rdev);
         radeon_resume(rdev);
+
+        r = radeon_ib_ring_tests(rdev);
+        if (r)
+                DRM_ERROR("ib ring test failed (%d).\n", r);
+
         radeon_pm_resume(rdev);
         radeon_restore_bios_scratch_regs(rdev);
 
@@ -984,30 +1297,77 @@ int radeon_resume_kms(struct drm_device *dev)
         return 0;
 }
 
+/**
+ * radeon_gpu_reset - reset the asic
+ *
+ * @rdev: radeon device pointer
+ *
+ * Attempt the reset the GPU if it has hung (all asics).
+ * Returns 0 for success or an error on failure.
+ */
 int radeon_gpu_reset(struct radeon_device *rdev)
 {
-        int r;
+        unsigned ring_sizes[RADEON_NUM_RINGS];
+        uint32_t *ring_data[RADEON_NUM_RINGS];
+
+        bool saved = false;
+
+        int i, r;
         int resched;
 
+        down_write(&rdev->exclusive_lock);
         radeon_save_bios_scratch_regs(rdev);
         /* block TTM */
         resched = ttm_bo_lock_delayed_workqueue(&rdev->mman.bdev);
         radeon_suspend(rdev);
 
+        for (i = 0; i < RADEON_NUM_RINGS; ++i) {
+                ring_sizes[i] = radeon_ring_backup(rdev, &rdev->ring[i],
+                                                   &ring_data[i]);
+                if (ring_sizes[i]) {
+                        saved = true;
+                        dev_info(rdev->dev, "Saved %d dwords of commands "
+                                 "on ring %d.\n", ring_sizes[i], i);
+                }
+        }
+
+retry:
         r = radeon_asic_reset(rdev);
         if (!r) {
-                dev_info(rdev->dev, "GPU reset succeed\n");
+                dev_info(rdev->dev, "GPU reset succeeded, trying to resume\n");
                 radeon_resume(rdev);
-                radeon_restore_bios_scratch_regs(rdev);
-                drm_helper_resume_force_mode(rdev->ddev);
-                ttm_bo_unlock_delayed_workqueue(&rdev->mman.bdev, resched);
         }
 
+        radeon_restore_bios_scratch_regs(rdev);
+        drm_helper_resume_force_mode(rdev->ddev);
+
+        if (!r) {
+                for (i = 0; i < RADEON_NUM_RINGS; ++i) {
+                        radeon_ring_restore(rdev, &rdev->ring[i],
+                                            ring_sizes[i], ring_data[i]);
+                }
+
+                r = radeon_ib_ring_tests(rdev);
+                if (r) {
+                        dev_err(rdev->dev, "ib ring test failed (%d).\n", r);
+                        if (saved) {
+                                radeon_suspend(rdev);
+                                goto retry;
+                        }
+                }
+        } else {
+                for (i = 0; i < RADEON_NUM_RINGS; ++i) {
+                        kfree(ring_data[i]);
+                }
+        }
+
+        ttm_bo_unlock_delayed_workqueue(&rdev->mman.bdev, resched);
         if (r) {
                 /* bad news, how to tell it to userspace ? */
                 dev_info(rdev->dev, "GPU reset failed\n");
         }
 
+        up_write(&rdev->exclusive_lock);
         return r;
 }
 
diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c
index 64a008d14493..7ddef8f30d0e 100644
--- a/drivers/gpu/drm/radeon/radeon_display.c
+++ b/drivers/gpu/drm/radeon/radeon_display.c
@@ -1401,7 +1401,7 @@ void radeon_modeset_fini(struct radeon_device *rdev)
         radeon_i2c_fini(rdev);
 }
 
-static bool is_hdtv_mode(struct drm_display_mode *mode)
+static bool is_hdtv_mode(const struct drm_display_mode *mode)
 {
         /* try and guess if this is a tv or a monitor */
         if ((mode->vdisplay == 480 && mode->hdisplay == 720) || /* 480p */
@@ -1414,7 +1414,7 @@ static bool is_hdtv_mode(struct drm_display_mode *mode)
 }
 
 bool radeon_crtc_scaling_mode_fixup(struct drm_crtc *crtc,
-                                struct drm_display_mode *mode,
+                                const struct drm_display_mode *mode,
                                 struct drm_display_mode *adjusted_mode)
 {
         struct drm_device *dev = crtc->dev;
diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c
index 2c4d53fd20c5..dcea6f01ae4e 100644
--- a/drivers/gpu/drm/radeon/radeon_drv.c
+++ b/drivers/gpu/drm/radeon/radeon_drv.c
@@ -133,7 +133,7 @@ int radeon_tv = 1;
 int radeon_audio = 0;
 int radeon_disp_priority = 0;
 int radeon_hw_i2c = 0;
-int radeon_pcie_gen2 = 0;
+int radeon_pcie_gen2 = -1;
 int radeon_msi = -1;
 int radeon_lockup_timeout = 10000;
 
@@ -179,7 +179,7 @@ module_param_named(disp_priority, radeon_disp_priority, int, 0444);
 MODULE_PARM_DESC(hw_i2c, "hw i2c engine enable (0 = disable)");
 module_param_named(hw_i2c, radeon_hw_i2c, int, 0444);
 
-MODULE_PARM_DESC(pcie_gen2, "PCIE Gen2 mode (1 = enable)");
+MODULE_PARM_DESC(pcie_gen2, "PCIE Gen2 mode (-1 = auto, 0 = disable, 1 = enable)");
 module_param_named(pcie_gen2, radeon_pcie_gen2, int, 0444);
 
 MODULE_PARM_DESC(msi, "MSI support (1 = enable, 0 = disable, -1 = auto)");
@@ -262,7 +262,6 @@ static struct drm_driver driver_old = {
         .irq_postinstall = radeon_driver_irq_postinstall,
         .irq_uninstall = radeon_driver_irq_uninstall,
         .irq_handler = radeon_driver_irq_handler,
-        .reclaim_buffers = drm_core_reclaim_buffers,
         .ioctls = radeon_ioctls,
         .dma_ioctl = radeon_cp_buffers,
         .fops = &radeon_driver_old_fops,
@@ -365,7 +364,6 @@ static struct drm_driver kms_driver = {
         .irq_postinstall = radeon_driver_irq_postinstall_kms,
         .irq_uninstall = radeon_driver_irq_uninstall_kms,
         .irq_handler = radeon_driver_irq_handler_kms,
-        .reclaim_buffers = drm_core_reclaim_buffers,
         .ioctls = radeon_ioctls_kms,
         .gem_init_object = radeon_gem_object_init,
         .gem_free_object = radeon_gem_object_free,
diff --git a/drivers/gpu/drm/radeon/radeon_fence.c b/drivers/gpu/drm/radeon/radeon_fence.c
index 11f5f402d22c..7b737b9339ad 100644
--- a/drivers/gpu/drm/radeon/radeon_fence.c
+++ b/drivers/gpu/drm/radeon/radeon_fence.c
@@ -40,39 +40,95 @@
 #include "radeon.h"
 #include "radeon_trace.h"
 
+/*
+ * Fences
+ * Fences mark an event in the GPUs pipeline and are used
+ * for GPU/CPU synchronization.  When the fence is written,
+ * it is expected that all buffers associated with that fence
+ * are no longer in use by the associated ring on the GPU and
+ * that the the relevant GPU caches have been flushed.  Whether
+ * we use a scratch register or memory location depends on the asic
+ * and whether writeback is enabled.
+ */
+
+/**
+ * radeon_fence_write - write a fence value
+ *
+ * @rdev: radeon_device pointer
+ * @seq: sequence number to write
+ * @ring: ring index the fence is associated with
+ *
+ * Writes a fence value to memory or a scratch register (all asics).
+ */
 static void radeon_fence_write(struct radeon_device *rdev, u32 seq, int ring)
 {
-        if (rdev->wb.enabled) {
-                *rdev->fence_drv[ring].cpu_addr = cpu_to_le32(seq);
+        struct radeon_fence_driver *drv = &rdev->fence_drv[ring];
+        if (likely(rdev->wb.enabled || !drv->scratch_reg)) {
+                *drv->cpu_addr = cpu_to_le32(seq);
         } else {
-                WREG32(rdev->fence_drv[ring].scratch_reg, seq);
+                WREG32(drv->scratch_reg, seq);
         }
 }
 
+/**
+ * radeon_fence_read - read a fence value
+ *
+ * @rdev: radeon_device pointer
+ * @ring: ring index the fence is associated with
+ *
+ * Reads a fence value from memory or a scratch register (all asics).
+ * Returns the value of the fence read from memory or register.
+ */
 static u32 radeon_fence_read(struct radeon_device *rdev, int ring)
 {
+        struct radeon_fence_driver *drv = &rdev->fence_drv[ring];
         u32 seq = 0;
 
-        if (rdev->wb.enabled) {
-                seq = le32_to_cpu(*rdev->fence_drv[ring].cpu_addr);
+        if (likely(rdev->wb.enabled || !drv->scratch_reg)) {
+                seq = le32_to_cpu(*drv->cpu_addr);
         } else {
-                seq = RREG32(rdev->fence_drv[ring].scratch_reg);
+                seq = RREG32(drv->scratch_reg);
         }
         return seq;
 }
 
-int radeon_fence_emit(struct radeon_device *rdev, struct radeon_fence *fence)
+/**
+ * radeon_fence_emit - emit a fence on the requested ring
+ *
+ * @rdev: radeon_device pointer
+ * @fence: radeon fence object
+ * @ring: ring index the fence is associated with
+ *
+ * Emits a fence command on the requested ring (all asics).
+ * Returns 0 on success, -ENOMEM on failure.
+ */
+int radeon_fence_emit(struct radeon_device *rdev,
+                      struct radeon_fence **fence,
+                      int ring)
 {
         /* we are protected by the ring emission mutex */
-        if (fence->seq && fence->seq < RADEON_FENCE_NOTEMITED_SEQ) {
-                return 0;
+        *fence = kmalloc(sizeof(struct radeon_fence), GFP_KERNEL);
+        if ((*fence) == NULL) {
+                return -ENOMEM;
         }
-        fence->seq = ++rdev->fence_drv[fence->ring].seq;
-        radeon_fence_ring_emit(rdev, fence->ring, fence);
-        trace_radeon_fence_emit(rdev->ddev, fence->seq);
+        kref_init(&((*fence)->kref));
+        (*fence)->rdev = rdev;
+        (*fence)->seq = ++rdev->fence_drv[ring].sync_seq[ring];
+        (*fence)->ring = ring;
+        radeon_fence_ring_emit(rdev, ring, *fence);
+        trace_radeon_fence_emit(rdev->ddev, (*fence)->seq);
         return 0;
 }
 
+/**
+ * radeon_fence_process - process a fence
+ *
+ * @rdev: radeon_device pointer
+ * @ring: ring index the fence is associated with
+ *
+ * Checks the current fence value and wakes the fence queue
+ * if the sequence number has increased (all asics).
+ */
 void radeon_fence_process(struct radeon_device *rdev, int ring)
 {
         uint64_t seq, last_seq;
@@ -133,30 +189,35 @@ void radeon_fence_process(struct radeon_device *rdev, int ring)
         }
 }
 
+/**
+ * radeon_fence_destroy - destroy a fence
+ *
+ * @kref: fence kref
+ *
+ * Frees the fence object (all asics).
+ */
 static void radeon_fence_destroy(struct kref *kref)
 {
         struct radeon_fence *fence;
 
         fence = container_of(kref, struct radeon_fence, kref);
-        fence->seq = RADEON_FENCE_NOTEMITED_SEQ;
         kfree(fence);
 }
 
-int radeon_fence_create(struct radeon_device *rdev,
-                        struct radeon_fence **fence,
-                        int ring)
-{
-        *fence = kmalloc(sizeof(struct radeon_fence), GFP_KERNEL);
-        if ((*fence) == NULL) {
-                return -ENOMEM;
-        }
-        kref_init(&((*fence)->kref));
-        (*fence)->rdev = rdev;
-        (*fence)->seq = RADEON_FENCE_NOTEMITED_SEQ;
-        (*fence)->ring = ring;
-        return 0;
-}
-
+/**
+ * radeon_fence_seq_signaled - check if a fence sequeuce number has signaled
+ *
+ * @rdev: radeon device pointer
+ * @seq: sequence number
+ * @ring: ring index the fence is associated with
+ *
+ * Check if the last singled fence sequnce number is >= the requested
+ * sequence number (all asics).
+ * Returns true if the fence has signaled (current fence value
+ * is >= requested value) or false if it has not (current fence
+ * value is < the requested value.  Helper function for
+ * radeon_fence_signaled().
+ */
 static bool radeon_fence_seq_signaled(struct radeon_device *rdev,
                                       u64 seq, unsigned ring)
 {
@@ -171,15 +232,19 @@ static bool radeon_fence_seq_signaled(struct radeon_device *rdev,
         return false;
 }
 
+/**
+ * radeon_fence_signaled - check if a fence has signaled
+ *
+ * @fence: radeon fence object
+ *
+ * Check if the requested fence has signaled (all asics).
+ * Returns true if the fence has signaled or false if it has not.
+ */
 bool radeon_fence_signaled(struct radeon_fence *fence)
 {
         if (!fence) {
                 return true;
         }
-        if (fence->seq == RADEON_FENCE_NOTEMITED_SEQ) {
-                WARN(1, "Querying an unemitted fence : %p !\n", fence);
-                return true;
-        }
         if (fence->seq == RADEON_FENCE_SIGNALED_SEQ) {
                 return true;
         }
@@ -190,6 +255,24 @@ bool radeon_fence_signaled(struct radeon_fence *fence)
         return false;
 }
 
+/**
+ * radeon_fence_wait_seq - wait for a specific sequence number
+ *
+ * @rdev: radeon device pointer
+ * @target_seq: sequence number we want to wait for
+ * @ring: ring index the fence is associated with
+ * @intr: use interruptable sleep
+ * @lock_ring: whether the ring should be locked or not
+ *
+ * Wait for the requested sequence number to be written (all asics).
+ * @intr selects whether to use interruptable (true) or non-interruptable
+ * (false) sleep when waiting for the sequence number.  Helper function
+ * for radeon_fence_wait(), et al.
+ * Returns 0 if the sequence number has passed, error for all other cases.
+ * -EDEADLK is returned when a GPU lockup has been detected and the ring is
+ * marked as not ready so no further jobs get scheduled until a successful
+ * reset.
+ */
 static int radeon_fence_wait_seq(struct radeon_device *rdev, u64 target_seq,
                                  unsigned ring, bool intr, bool lock_ring)
 {
@@ -285,6 +368,17 @@ static int radeon_fence_wait_seq(struct radeon_device *rdev, u64 target_seq,
         return 0;
 }
 
+/**
+ * radeon_fence_wait - wait for a fence to signal
+ *
+ * @fence: radeon fence object
+ * @intr: use interruptable sleep
+ *
+ * Wait for the requested fence to signal (all asics).
+ * @intr selects whether to use interruptable (true) or non-interruptable
+ * (false) sleep when waiting for the fence.
+ * Returns 0 if the fence has passed, error for all other cases.
+ */
 int radeon_fence_wait(struct radeon_fence *fence, bool intr)
 {
         int r;
@@ -315,6 +409,20 @@ bool radeon_fence_any_seq_signaled(struct radeon_device *rdev, u64 *seq)
         return false;
 }
 
+/**
+ * radeon_fence_wait_any_seq - wait for a sequence number on any ring
+ *
+ * @rdev: radeon device pointer
+ * @target_seq: sequence number(s) we want to wait for
+ * @intr: use interruptable sleep
+ *
+ * Wait for the requested sequence number(s) to be written by any ring
+ * (all asics).  Sequnce number array is indexed by ring id.
+ * @intr selects whether to use interruptable (true) or non-interruptable
+ * (false) sleep when waiting for the sequence number.  Helper function
+ * for radeon_fence_wait_any(), et al.
+ * Returns 0 if the sequence number has passed, error for all other cases.
+ */
 static int radeon_fence_wait_any_seq(struct radeon_device *rdev,
                                      u64 *target_seq, bool intr)
 {
@@ -343,7 +451,7 @@ static int radeon_fence_wait_any_seq(struct radeon_device *rdev,
 
         /* nothing to wait for ? */
         if (ring == RADEON_NUM_RINGS) {
-                return 0;
+                return -ENOENT;
         }
 
         while (!radeon_fence_any_seq_signaled(rdev, target_seq)) {
@@ -424,6 +532,19 @@ static int radeon_fence_wait_any_seq(struct radeon_device *rdev,
         return 0;
 }
 
+/**
+ * radeon_fence_wait_any - wait for a fence to signal on any ring
+ *
+ * @rdev: radeon device pointer
+ * @fences: radeon fence object(s)
+ * @intr: use interruptable sleep
+ *
+ * Wait for any requested fence to signal (all asics).  Fence
+ * array is indexed by ring id.  @intr selects whether to use
+ * interruptable (true) or non-interruptable (false) sleep when
+ * waiting for the fences. Used by the suballocator.
+ * Returns 0 if any fence has passed, error for all other cases.
+ */
 int radeon_fence_wait_any(struct radeon_device *rdev,
                           struct radeon_fence **fences,
                           bool intr)
@@ -444,9 +565,7 @@ int radeon_fence_wait_any(struct radeon_device *rdev,
                         return 0;
                 }
 
-                if (fences[i]->seq < RADEON_FENCE_NOTEMITED_SEQ) {
-                        seq[i] = fences[i]->seq;
-                }
+                seq[i] = fences[i]->seq;
         }
 
         r = radeon_fence_wait_any_seq(rdev, seq, intr);
@@ -456,16 +575,22 @@ int radeon_fence_wait_any(struct radeon_device *rdev,
         return 0;
 }
 
+/**
+ * radeon_fence_wait_next_locked - wait for the next fence to signal
+ *
+ * @rdev: radeon device pointer
+ * @ring: ring index the fence is associated with
+ *
+ * Wait for the next fence on the requested ring to signal (all asics).
+ * Returns 0 if the next fence has passed, error for all other cases.
+ * Caller must hold ring lock.
+ */
 int radeon_fence_wait_next_locked(struct radeon_device *rdev, int ring)
 {
         uint64_t seq;
 
-        /* We are not protected by ring lock when reading current seq but
-         * it's ok as worst case is we return to early while we could have
-         * wait.
-         */
         seq = atomic64_read(&rdev->fence_drv[ring].last_seq) + 1ULL;
-        if (seq >= rdev->fence_drv[ring].seq) {
+        if (seq >= rdev->fence_drv[ring].sync_seq[ring]) {
                 /* nothing to wait for, last_seq is
                    already the last emited fence */
                 return -ENOENT;
@@ -473,23 +598,59 @@ int radeon_fence_wait_next_locked(struct radeon_device *rdev, int ring)
         return radeon_fence_wait_seq(rdev, seq, ring, false, false);
 }
 
-int radeon_fence_wait_empty_locked(struct radeon_device *rdev, int ring)
+/**
+ * radeon_fence_wait_empty_locked - wait for all fences to signal
+ *
+ * @rdev: radeon device pointer
+ * @ring: ring index the fence is associated with
+ *
+ * Wait for all fences on the requested ring to signal (all asics).
+ * Returns 0 if the fences have passed, error for all other cases.
+ * Caller must hold ring lock.
+ */
+void radeon_fence_wait_empty_locked(struct radeon_device *rdev, int ring)
 {
-        /* We are not protected by ring lock when reading current seq
-         * but it's ok as wait empty is call from place where no more
-         * activity can be scheduled so there won't be concurrent access
-         * to seq value.
-         */
-        return radeon_fence_wait_seq(rdev, rdev->fence_drv[ring].seq,
-                                     ring, false, false);
+        uint64_t seq = rdev->fence_drv[ring].sync_seq[ring];
+
+        while(1) {
+                int r;
+                r = radeon_fence_wait_seq(rdev, seq, ring, false, false);
+                if (r == -EDEADLK) {
+                        mutex_unlock(&rdev->ring_lock);
+                        r = radeon_gpu_reset(rdev);
+                        mutex_lock(&rdev->ring_lock);
+                        if (!r)
+                                continue;
+                }
+                if (r) {
+                        dev_err(rdev->dev, "error waiting for ring to become"
+                                " idle (%d)\n", r);
+                }
+                return;
+        }
 }
 
+/**
+ * radeon_fence_ref - take a ref on a fence
+ *
+ * @fence: radeon fence object
+ *
+ * Take a reference on a fence (all asics).
+ * Returns the fence.
+ */
 struct radeon_fence *radeon_fence_ref(struct radeon_fence *fence)
 {
         kref_get(&fence->kref);
         return fence;
 }
 
+/**
+ * radeon_fence_unref - remove a ref on a fence
+ *
+ * @fence: radeon fence object
+ *
+ * Remove a reference on a fence (all asics).
+ */
 void radeon_fence_unref(struct radeon_fence **fence)
 {
         struct radeon_fence *tmp = *fence;
@@ -500,6 +661,16 @@ void radeon_fence_unref(struct radeon_fence **fence)
         }
 }
 
+/**
+ * radeon_fence_count_emitted - get the count of emitted fences
+ *
+ * @rdev: radeon device pointer
+ * @ring: ring index the fence is associated with
+ *
+ * Get the number of fences emitted on the requested ring (all asics).
+ * Returns the number of emitted fences on the ring.  Used by the
+ * dynpm code to ring track activity.
+ */
 unsigned radeon_fence_count_emitted(struct radeon_device *rdev, int ring)
 {
         uint64_t emitted;
@@ -508,7 +679,8 @@ unsigned radeon_fence_count_emitted(struct radeon_device *rdev, int ring)
          * but it's ok to report slightly wrong fence count here.
          */
         radeon_fence_process(rdev, ring);
-        emitted = rdev->fence_drv[ring].seq - atomic64_read(&rdev->fence_drv[ring].last_seq);
+        emitted = rdev->fence_drv[ring].sync_seq[ring]
+                - atomic64_read(&rdev->fence_drv[ring].last_seq);
         /* to avoid 32bits warp around */
         if (emitted > 0x10000000) {
                 emitted = 0x10000000;
@@ -516,6 +688,83 @@ unsigned radeon_fence_count_emitted(struct radeon_device *rdev, int ring)
         return (unsigned)emitted;
 }
 
+/**
+ * radeon_fence_need_sync - do we need a semaphore
+ *
+ * @fence: radeon fence object
+ * @dst_ring: which ring to check against
+ *
+ * Check if the fence needs to be synced against another ring
+ * (all asics).  If so, we need to emit a semaphore.
+ * Returns true if we need to sync with another ring, false if
+ * not.
+ */
+bool radeon_fence_need_sync(struct radeon_fence *fence, int dst_ring)
+{
+        struct radeon_fence_driver *fdrv;
+
+        if (!fence) {
+                return false;
+        }
+
+        if (fence->ring == dst_ring) {
+                return false;
+        }
+
+        /* we are protected by the ring mutex */
+        fdrv = &fence->rdev->fence_drv[dst_ring];
+        if (fence->seq <= fdrv->sync_seq[fence->ring]) {
+                return false;
+        }
+
+        return true;
+}
+
+/**
+ * radeon_fence_note_sync - record the sync point
+ *
+ * @fence: radeon fence object
+ * @dst_ring: which ring to check against
+ *
+ * Note the sequence number at which point the fence will
+ * be synced with the requested ring (all asics).
+ */
+void radeon_fence_note_sync(struct radeon_fence *fence, int dst_ring)
+{
+        struct radeon_fence_driver *dst, *src;
+        unsigned i;
+
+        if (!fence) {
+                return;
+        }
+
+        if (fence->ring == dst_ring) {
+                return;
+        }
+
+        /* we are protected by the ring mutex */
+        src = &fence->rdev->fence_drv[fence->ring];
+        dst = &fence->rdev->fence_drv[dst_ring];
+        for (i = 0; i < RADEON_NUM_RINGS; ++i) {
+                if (i == dst_ring) {
+                        continue;
+                }
+                dst->sync_seq[i] = max(dst->sync_seq[i], src->sync_seq[i]);
+        }
+}
+
+/**
+ * radeon_fence_driver_start_ring - make the fence driver
+ * ready for use on the requested ring.
+ *
+ * @rdev: radeon device pointer
+ * @ring: ring index to start the fence driver on
+ *
+ * Make the fence driver ready for processing (all asics).
+ * Not all asics have all rings, so each asic will only
+ * start the fence driver on the rings it has.
+ * Returns 0 for success, errors for failure.
+ */
 int radeon_fence_driver_start_ring(struct radeon_device *rdev, int ring)
 {
         uint64_t index;
@@ -537,24 +786,49 @@ int radeon_fence_driver_start_ring(struct radeon_device *rdev, int ring)
         }
         rdev->fence_drv[ring].cpu_addr = &rdev->wb.wb[index/4];
         rdev->fence_drv[ring].gpu_addr = rdev->wb.gpu_addr + index;
-        radeon_fence_write(rdev, rdev->fence_drv[ring].seq, ring);
+        radeon_fence_write(rdev, atomic64_read(&rdev->fence_drv[ring].last_seq), ring);
         rdev->fence_drv[ring].initialized = true;
         dev_info(rdev->dev, "fence driver on ring %d use gpu addr 0x%016llx and cpu addr 0x%p\n",
                  ring, rdev->fence_drv[ring].gpu_addr, rdev->fence_drv[ring].cpu_addr);
         return 0;
 }
 
+/**
+ * radeon_fence_driver_init_ring - init the fence driver
+ * for the requested ring.
+ *
+ * @rdev: radeon device pointer
+ * @ring: ring index to start the fence driver on
+ *
+ * Init the fence driver for the requested ring (all asics).
+ * Helper function for radeon_fence_driver_init().
+ */
 static void radeon_fence_driver_init_ring(struct radeon_device *rdev, int ring)
 {
+        int i;
+
         rdev->fence_drv[ring].scratch_reg = -1;
         rdev->fence_drv[ring].cpu_addr = NULL;
         rdev->fence_drv[ring].gpu_addr = 0;
-        rdev->fence_drv[ring].seq = 0;
+        for (i = 0; i < RADEON_NUM_RINGS; ++i)
+                rdev->fence_drv[ring].sync_seq[i] = 0;
         atomic64_set(&rdev->fence_drv[ring].last_seq, 0);
         rdev->fence_drv[ring].last_activity = jiffies;
         rdev->fence_drv[ring].initialized = false;
 }
 
+/**
+ * radeon_fence_driver_init - init the fence driver
+ * for all possible rings.
+ *
+ * @rdev: radeon device pointer
+ *
+ * Init the fence driver for all possible rings (all asics).
+ * Not all asics have all rings, so each asic will only
+ * start the fence driver on the rings it has using
+ * radeon_fence_driver_start_ring().
+ * Returns 0 for success.
+ */
 int radeon_fence_driver_init(struct radeon_device *rdev)
 {
         int ring;
@@ -569,6 +843,14 @@ int radeon_fence_driver_init(struct radeon_device *rdev)
         return 0;
 }
 
+/**
+ * radeon_fence_driver_fini - tear down the fence driver
+ * for all possible rings.
+ *
+ * @rdev: radeon device pointer
+ *
+ * Tear down the fence driver for all possible rings (all asics).
+ */
 void radeon_fence_driver_fini(struct radeon_device *rdev)
 {
         int ring;
@@ -595,7 +877,7 @@ static int radeon_debugfs_fence_info(struct seq_file *m, void *data)
         struct drm_info_node *node = (struct drm_info_node *)m->private;
         struct drm_device *dev = node->minor->dev;
         struct radeon_device *rdev = dev->dev_private;
-        int i;
+        int i, j;
 
         for (i = 0; i < RADEON_NUM_RINGS; ++i) {
                 if (!rdev->fence_drv[i].initialized)
@@ -604,8 +886,14 @@ static int radeon_debugfs_fence_info(struct seq_file *m, void *data)
                 seq_printf(m, "--- ring %d ---\n", i);
                 seq_printf(m, "Last signaled fence 0x%016llx\n",
                            (unsigned long long)atomic64_read(&rdev->fence_drv[i].last_seq));
-                seq_printf(m, "Last emitted  0x%016llx\n",
-                           rdev->fence_drv[i].seq);
+                seq_printf(m, "Last emitted        0x%016llx\n",
+                           rdev->fence_drv[i].sync_seq[i]);
+
+                for (j = 0; j < RADEON_NUM_RINGS; ++j) {
+                        if (i != j && rdev->fence_drv[j].initialized)
+                                seq_printf(m, "Last sync to ring %d 0x%016llx\n",
+                                           j, rdev->fence_drv[i].sync_seq[j]);
+                }
         }
         return 0;
 }
diff --git a/drivers/gpu/drm/radeon/radeon_gart.c b/drivers/gpu/drm/radeon/radeon_gart.c
index 84b648a7ddd8..b3720054614d 100644
--- a/drivers/gpu/drm/radeon/radeon_gart.c
+++ b/drivers/gpu/drm/radeon/radeon_gart.c
@@ -31,8 +31,38 @@
 #include "radeon_reg.h"
 
 /*
+ * GART
+ * The GART (Graphics Aperture Remapping Table) is an aperture
+ * in the GPU's address space.  System pages can be mapped into
+ * the aperture and look like contiguous pages from the GPU's
+ * perspective.  A page table maps the pages in the aperture
+ * to the actual backing pages in system memory.
+ *
+ * Radeon GPUs support both an internal GART, as described above,
+ * and AGP.  AGP works similarly, but the GART table is configured
+ * and maintained by the northbridge rather than the driver.
+ * Radeon hw has a separate AGP aperture that is programmed to
+ * point to the AGP aperture provided by the northbridge and the
+ * requests are passed through to the northbridge aperture.
+ * Both AGP and internal GART can be used at the same time, however
+ * that is not currently supported by the driver.
+ *
+ * This file handles the common internal GART management.
+ */
+
+/*
  * Common GART table functions.
  */
+/**
+ * radeon_gart_table_ram_alloc - allocate system ram for gart page table
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Allocate system memory for GART page table
+ * (r1xx-r3xx, non-pcie r4xx, rs400).  These asics require the
+ * gart table to be in system memory.
+ * Returns 0 for success, -ENOMEM for failure.
+ */
 int radeon_gart_table_ram_alloc(struct radeon_device *rdev)
 {
         void *ptr;
@@ -54,6 +84,15 @@ int radeon_gart_table_ram_alloc(struct radeon_device *rdev)
         return 0;
 }
 
+/**
+ * radeon_gart_table_ram_free - free system ram for gart page table
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Free system memory for GART page table
+ * (r1xx-r3xx, non-pcie r4xx, rs400).  These asics require the
+ * gart table to be in system memory.
+ */
 void radeon_gart_table_ram_free(struct radeon_device *rdev)
 {
         if (rdev->gart.ptr == NULL) {
@@ -73,6 +112,16 @@ void radeon_gart_table_ram_free(struct radeon_device *rdev)
         rdev->gart.table_addr = 0;
 }
 
+/**
+ * radeon_gart_table_vram_alloc - allocate vram for gart page table
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Allocate video memory for GART page table
+ * (pcie r4xx, r5xx+).  These asics require the
+ * gart table to be in video memory.
+ * Returns 0 for success, error for failure.
+ */
 int radeon_gart_table_vram_alloc(struct radeon_device *rdev)
 {
         int r;
@@ -88,6 +137,16 @@ int radeon_gart_table_vram_alloc(struct radeon_device *rdev)
         return 0;
 }
 
+/**
+ * radeon_gart_table_vram_pin - pin gart page table in vram
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Pin the GART page table in vram so it will not be moved
+ * by the memory manager (pcie r4xx, r5xx+).  These asics require the
+ * gart table to be in video memory.
+ * Returns 0 for success, error for failure.
+ */
 int radeon_gart_table_vram_pin(struct radeon_device *rdev)
 {
         uint64_t gpu_addr;
@@ -110,6 +169,14 @@ int radeon_gart_table_vram_pin(struct radeon_device *rdev)
         return r;
 }
 
+/**
+ * radeon_gart_table_vram_unpin - unpin gart page table in vram
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Unpin the GART page table in vram (pcie r4xx, r5xx+).
+ * These asics require the gart table to be in video memory.
+ */
 void radeon_gart_table_vram_unpin(struct radeon_device *rdev)
 {
         int r;
@@ -126,6 +193,15 @@ void radeon_gart_table_vram_unpin(struct radeon_device *rdev)
         }
 }
 
+/**
+ * radeon_gart_table_vram_free - free gart page table vram
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Free the video memory used for the GART page table
+ * (pcie r4xx, r5xx+).  These asics require the gart table to
+ * be in video memory.
+ */
 void radeon_gart_table_vram_free(struct radeon_device *rdev)
 {
         if (rdev->gart.robj == NULL) {
@@ -135,12 +211,19 @@ void radeon_gart_table_vram_free(struct radeon_device *rdev)
         radeon_bo_unref(&rdev->gart.robj);
 }
 
-
-
-
 /*
  * Common gart functions.
  */
+/**
+ * radeon_gart_unbind - unbind pages from the gart page table
+ *
+ * @rdev: radeon_device pointer
+ * @offset: offset into the GPU's gart aperture
+ * @pages: number of pages to unbind
+ *
+ * Unbinds the requested pages from the gart page table and
+ * replaces them with the dummy page (all asics).
+ */
 void radeon_gart_unbind(struct radeon_device *rdev, unsigned offset,
                         int pages)
 {
@@ -172,6 +255,19 @@ void radeon_gart_unbind(struct radeon_device *rdev, unsigned offset,
         radeon_gart_tlb_flush(rdev);
 }
 
+/**
+ * radeon_gart_bind - bind pages into the gart page table
+ *
+ * @rdev: radeon_device pointer
+ * @offset: offset into the GPU's gart aperture
+ * @pages: number of pages to bind
+ * @pagelist: pages to bind
+ * @dma_addr: DMA addresses of pages
+ *
+ * Binds the requested pages to the gart page table
+ * (all asics).
+ * Returns 0 for success, -EINVAL for failure.
+ */
 int radeon_gart_bind(struct radeon_device *rdev, unsigned offset,
                      int pages, struct page **pagelist, dma_addr_t *dma_addr)
 {
@@ -203,6 +299,14 @@ int radeon_gart_bind(struct radeon_device *rdev, unsigned offset,
         return 0;
 }
 
+/**
+ * radeon_gart_restore - bind all pages in the gart page table
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Binds all pages in the gart page table (all asics).
+ * Used to rebuild the gart table on device startup or resume.
+ */
 void radeon_gart_restore(struct radeon_device *rdev)
 {
         int i, j, t;
@@ -222,6 +326,14 @@ void radeon_gart_restore(struct radeon_device *rdev)
         radeon_gart_tlb_flush(rdev);
 }
 
+/**
+ * radeon_gart_init - init the driver info for managing the gart
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Allocate the dummy page and init the gart driver info (all asics).
+ * Returns 0 for success, error for failure.
+ */
 int radeon_gart_init(struct radeon_device *rdev)
 {
         int r, i;
@@ -262,6 +374,13 @@ int radeon_gart_init(struct radeon_device *rdev)
         return 0;
 }
 
+/**
+ * radeon_gart_fini - tear down the driver info for managing the gart
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Tear down the gart driver info and free the dummy page (all asics).
+ */
 void radeon_gart_fini(struct radeon_device *rdev)
 {
         if (rdev->gart.pages && rdev->gart.pages_addr && rdev->gart.ready) {
@@ -278,35 +397,104 @@ void radeon_gart_fini(struct radeon_device *rdev)
 }
 
 /*
+ * GPUVM
+ * GPUVM is similar to the legacy gart on older asics, however
+ * rather than there being a single global gart table
+ * for the entire GPU, there are multiple VM page tables active
+ * at any given time.  The VM page tables can contain a mix
+ * vram pages and system memory pages and system memory pages
+ * can be mapped as snooped (cached system pages) or unsnooped
+ * (uncached system pages).
+ * Each VM has an ID associated with it and there is a page table
+ * associated with each VMID.  When execting a command buffer,
+ * the kernel tells the the ring what VMID to use for that command
+ * buffer.  VMIDs are allocated dynamically as commands are submitted.
+ * The userspace drivers maintain their own address space and the kernel
+ * sets up their pages tables accordingly when they submit their
+ * command buffers and a VMID is assigned.
+ * Cayman/Trinity support up to 8 active VMs at any given time;
+ * SI supports 16.
+ */
+
+/*
  * vm helpers
  *
  * TODO bind a default page at vm initialization for default address
  */
+
+/**
+ * radeon_vm_manager_init - init the vm manager
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Init the vm manager (cayman+).
+ * Returns 0 for success, error for failure.
+ */
 int radeon_vm_manager_init(struct radeon_device *rdev)
 {
+        struct radeon_vm *vm;
+        struct radeon_bo_va *bo_va;
         int r;
 
-        rdev->vm_manager.enabled = false;
+        if (!rdev->vm_manager.enabled) {
+                /* mark first vm as always in use, it's the system one */
+                /* allocate enough for 2 full VM pts */
+                r = radeon_sa_bo_manager_init(rdev, &rdev->vm_manager.sa_manager,
+                                              rdev->vm_manager.max_pfn * 8 * 2,
+                                              RADEON_GEM_DOMAIN_VRAM);
+                if (r) {
+                        dev_err(rdev->dev, "failed to allocate vm bo (%dKB)\n",
+                                (rdev->vm_manager.max_pfn * 8) >> 10);
+                        return r;
+                }
 
-        /* mark first vm as always in use, it's the system one */
-        /* allocate enough for 2 full VM pts */
-        r = radeon_sa_bo_manager_init(rdev, &rdev->vm_manager.sa_manager,
-                                      rdev->vm_manager.max_pfn * 8 * 2,
-                                      RADEON_GEM_DOMAIN_VRAM);
-        if (r) {
-                dev_err(rdev->dev, "failed to allocate vm bo (%dKB)\n",
-                        (rdev->vm_manager.max_pfn * 8) >> 10);
-                return r;
+                r = rdev->vm_manager.funcs->init(rdev);
+                if (r)
+                        return r;
+        
+                rdev->vm_manager.enabled = true;
+
+                r = radeon_sa_bo_manager_start(rdev, &rdev->vm_manager.sa_manager);
+                if (r)
+                        return r;
         }
 
-        r = rdev->vm_manager.funcs->init(rdev);
-        if (r == 0)
-                rdev->vm_manager.enabled = true;
+        /* restore page table */
+        list_for_each_entry(vm, &rdev->vm_manager.lru_vm, list) {
+                if (vm->id == -1)
+                        continue;
 
-        return r;
+                list_for_each_entry(bo_va, &vm->va, vm_list) {
+                        struct ttm_mem_reg *mem = NULL;
+                        if (bo_va->valid)
+                                mem = &bo_va->bo->tbo.mem;
+
+                        bo_va->valid = false;
+                        r = radeon_vm_bo_update_pte(rdev, vm, bo_va->bo, mem);
+                        if (r) {
+                                DRM_ERROR("Failed to update pte for vm %d!\n", vm->id);
+                        }
+                }
+
+                r = rdev->vm_manager.funcs->bind(rdev, vm, vm->id);
+                if (r) {
+                        DRM_ERROR("Failed to bind vm %d!\n", vm->id);
+                }
+        }
+        return 0;
 }
 
-/* cs mutex must be lock */
+/* global mutex must be lock */
+/**
+ * radeon_vm_unbind_locked - unbind a specific vm
+ *
+ * @rdev: radeon_device pointer
+ * @vm: vm to unbind
+ *
+ * Unbind the requested vm (cayman+).
+ * Wait for use of the VM to finish, then unbind the page table,
+ * and free the page table memory.
+ */
 static void radeon_vm_unbind_locked(struct radeon_device *rdev,
                                     struct radeon_vm *vm)
 {
@@ -317,10 +505,21 @@ static void radeon_vm_unbind_locked(struct radeon_device *rdev,
         }
 
         /* wait for vm use to end */
-        if (vm->fence) {
-                radeon_fence_wait(vm->fence, false);
-                radeon_fence_unref(&vm->fence);
+        while (vm->fence) {
+                int r;
+                r = radeon_fence_wait(vm->fence, false);
+                if (r)
+                        DRM_ERROR("error while waiting for fence: %d\n", r);
+                if (r == -EDEADLK) {
+                        mutex_unlock(&rdev->vm_manager.lock);
+                        r = radeon_gpu_reset(rdev);
+                        mutex_lock(&rdev->vm_manager.lock);
+                        if (!r)
+                                continue;
+                }
+                break;
         }
+        radeon_fence_unref(&vm->fence);
 
         /* hw unbind */
         rdev->vm_manager.funcs->unbind(rdev, vm);
@@ -335,39 +534,42 @@ static void radeon_vm_unbind_locked(struct radeon_device *rdev,
         }
 }
 
+/**
+ * radeon_vm_manager_fini - tear down the vm manager
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Tear down the VM manager (cayman+).
+ */
 void radeon_vm_manager_fini(struct radeon_device *rdev)
 {
-        if (rdev->vm_manager.sa_manager.bo == NULL)
-                return;
-        radeon_vm_manager_suspend(rdev);
-        rdev->vm_manager.funcs->fini(rdev);
-        radeon_sa_bo_manager_fini(rdev, &rdev->vm_manager.sa_manager);
-        rdev->vm_manager.enabled = false;
-}
-
-int radeon_vm_manager_start(struct radeon_device *rdev)
-{
-        if (rdev->vm_manager.sa_manager.bo == NULL) {
-                return -EINVAL;
-        }
-        return radeon_sa_bo_manager_start(rdev, &rdev->vm_manager.sa_manager);
-}
-
-int radeon_vm_manager_suspend(struct radeon_device *rdev)
-{
         struct radeon_vm *vm, *tmp;
 
-        radeon_mutex_lock(&rdev->cs_mutex);
+        if (!rdev->vm_manager.enabled)
+                return;
+
+        mutex_lock(&rdev->vm_manager.lock);
         /* unbind all active vm */
         list_for_each_entry_safe(vm, tmp, &rdev->vm_manager.lru_vm, list) {
                 radeon_vm_unbind_locked(rdev, vm);
         }
         rdev->vm_manager.funcs->fini(rdev);
-        radeon_mutex_unlock(&rdev->cs_mutex);
-        return radeon_sa_bo_manager_suspend(rdev, &rdev->vm_manager.sa_manager);
+        mutex_unlock(&rdev->vm_manager.lock);
+
+        radeon_sa_bo_manager_suspend(rdev, &rdev->vm_manager.sa_manager);
+        radeon_sa_bo_manager_fini(rdev, &rdev->vm_manager.sa_manager);
+        rdev->vm_manager.enabled = false;
 }
 
-/* cs mutex must be lock */
+/* global mutex must be locked */
+/**
+ * radeon_vm_unbind - locked version of unbind
+ *
+ * @rdev: radeon_device pointer
+ * @vm: vm to unbind
+ *
+ * Locked version that wraps radeon_vm_unbind_locked (cayman+).
+ */
 void radeon_vm_unbind(struct radeon_device *rdev, struct radeon_vm *vm)
 {
         mutex_lock(&vm->mutex);
@@ -375,7 +577,19 @@ void radeon_vm_unbind(struct radeon_device *rdev, struct radeon_vm *vm)
         mutex_unlock(&vm->mutex);
 }
 
-/* cs mutex must be lock & vm mutex must be lock */
+/* global and local mutex must be locked */
+/**
+ * radeon_vm_bind - bind a page table to a VMID
+ *
+ * @rdev: radeon_device pointer
+ * @vm: vm to bind
+ *
+ * Bind the requested vm (cayman+).
+ * Suballocate memory for the page table, allocate a VMID
+ * and bind the page table to it, and finally start to populate
+ * the page table.
+ * Returns 0 for success, error for failure.
+ */
 int radeon_vm_bind(struct radeon_device *rdev, struct radeon_vm *vm)
 {
         struct radeon_vm *vm_evict;
@@ -438,6 +652,20 @@ retry_id:
 }
 
 /* object have to be reserved */
+/**
+ * radeon_vm_bo_add - add a bo to a specific vm
+ *
+ * @rdev: radeon_device pointer
+ * @vm: requested vm
+ * @bo: radeon buffer object
+ * @offset: requested offset of the buffer in the VM address space
+ * @flags: attributes of pages (read/write/valid/etc.)
+ *
+ * Add @bo into the requested vm (cayman+).
+ * Add @bo to the list of bos associated with the vm and validate
+ * the offset requested within the vm address space.
+ * Returns 0 for success, error for failure.
+ */
 int radeon_vm_bo_add(struct radeon_device *rdev,
                      struct radeon_vm *vm,
                      struct radeon_bo *bo,
@@ -479,7 +707,7 @@ int radeon_vm_bo_add(struct radeon_device *rdev,
         if (last_pfn > vm->last_pfn) {
                 /* release mutex and lock in right order */
                 mutex_unlock(&vm->mutex);
-                radeon_mutex_lock(&rdev->cs_mutex);
+                mutex_lock(&rdev->vm_manager.lock);
                 mutex_lock(&vm->mutex);
                 /* and check again */
                 if (last_pfn > vm->last_pfn) {
@@ -488,7 +716,7 @@ int radeon_vm_bo_add(struct radeon_device *rdev,
                         radeon_vm_unbind_locked(rdev, vm);
                         vm->last_pfn = (last_pfn + align) & ~align;
                 }
-                radeon_mutex_unlock(&rdev->cs_mutex);
+                mutex_unlock(&rdev->vm_manager.lock);
         }
         head = &vm->va;
         last_offset = 0;
@@ -515,6 +743,17 @@ int radeon_vm_bo_add(struct radeon_device *rdev,
         return 0;
 }
 
+/**
+ * radeon_vm_get_addr - get the physical address of the page
+ *
+ * @rdev: radeon_device pointer
+ * @mem: ttm mem
+ * @pfn: pfn
+ *
+ * Look up the physical address of the page that the pte resolves
+ * to (cayman+).
+ * Returns the physical address of the page.
+ */
 static u64 radeon_vm_get_addr(struct radeon_device *rdev,
                               struct ttm_mem_reg *mem,
                               unsigned pfn)
@@ -543,7 +782,18 @@ static u64 radeon_vm_get_addr(struct radeon_device *rdev,
         return addr;
 }
 
-/* object have to be reserved & cs mutex took & vm mutex took */
+/* object have to be reserved & global and local mutex must be locked */
+/**
+ * radeon_vm_bo_update_pte - map a bo into the vm page table
+ *
+ * @rdev: radeon_device pointer
+ * @vm: requested vm
+ * @bo: radeon buffer object
+ * @mem: ttm mem
+ *
+ * Fill in the page table entries for @bo (cayman+).
+ * Returns 0 for success, -EINVAL for failure.
+ */
 int radeon_vm_bo_update_pte(struct radeon_device *rdev,
                             struct radeon_vm *vm,
                             struct radeon_bo *bo,
@@ -592,6 +842,18 @@ int radeon_vm_bo_update_pte(struct radeon_device *rdev,
 }
 
 /* object have to be reserved */
+/**
+ * radeon_vm_bo_rmv - remove a bo to a specific vm
+ *
+ * @rdev: radeon_device pointer
+ * @vm: requested vm
+ * @bo: radeon buffer object
+ *
+ * Remove @bo from the requested vm (cayman+).
+ * Remove @bo from the list of bos associated with the vm and
+ * remove the ptes for @bo in the page table.
+ * Returns 0 for success.
+ */
 int radeon_vm_bo_rmv(struct radeon_device *rdev,
                      struct radeon_vm *vm,
                      struct radeon_bo *bo)
@@ -602,10 +864,10 @@ int radeon_vm_bo_rmv(struct radeon_device *rdev,
         if (bo_va == NULL)
                 return 0;
 
-        radeon_mutex_lock(&rdev->cs_mutex);
+        mutex_lock(&rdev->vm_manager.lock);
         mutex_lock(&vm->mutex);
         radeon_vm_bo_update_pte(rdev, vm, bo, NULL);
-        radeon_mutex_unlock(&rdev->cs_mutex);
+        mutex_unlock(&rdev->vm_manager.lock);
         list_del(&bo_va->vm_list);
         mutex_unlock(&vm->mutex);
         list_del(&bo_va->bo_list);
@@ -614,6 +876,15 @@ int radeon_vm_bo_rmv(struct radeon_device *rdev,
         return 0;
 }
 
+/**
+ * radeon_vm_bo_invalidate - mark the bo as invalid
+ *
+ * @rdev: radeon_device pointer
+ * @vm: requested vm
+ * @bo: radeon buffer object
+ *
+ * Mark @bo as invalid (cayman+).
+ */
 void radeon_vm_bo_invalidate(struct radeon_device *rdev,
                              struct radeon_bo *bo)
 {
@@ -625,6 +896,17 @@ void radeon_vm_bo_invalidate(struct radeon_device *rdev,
         }
 }
 
+/**
+ * radeon_vm_init - initialize a vm instance
+ *
+ * @rdev: radeon_device pointer
+ * @vm: requested vm
+ *
+ * Init @vm (cayman+).
+ * Map the IB pool and any other shared objects into the VM
+ * by default as it's used by all VMs.
+ * Returns 0 for success, error for failure.
+ */
 int radeon_vm_init(struct radeon_device *rdev, struct radeon_vm *vm)
 {
         int r;
@@ -651,15 +933,24 @@ int radeon_vm_init(struct radeon_device *rdev, struct radeon_vm *vm)
         return r;
 }
 
+/**
+ * radeon_vm_init - tear down a vm instance
+ *
+ * @rdev: radeon_device pointer
+ * @vm: requested vm
+ *
+ * Tear down @vm (cayman+).
+ * Unbind the VM and remove all bos from the vm bo list
+ */
 void radeon_vm_fini(struct radeon_device *rdev, struct radeon_vm *vm)
 {
         struct radeon_bo_va *bo_va, *tmp;
         int r;
 
-        radeon_mutex_lock(&rdev->cs_mutex);
+        mutex_lock(&rdev->vm_manager.lock);
         mutex_lock(&vm->mutex);
         radeon_vm_unbind_locked(rdev, vm);
-        radeon_mutex_unlock(&rdev->cs_mutex);
+        mutex_unlock(&rdev->vm_manager.lock);
 
         /* remove all bo */
         r = radeon_bo_reserve(rdev->ring_tmp_bo.bo, false);
diff --git a/drivers/gpu/drm/radeon/radeon_gem.c b/drivers/gpu/drm/radeon/radeon_gem.c
index 21ec9f5653ce..84d045245739 100644
--- a/drivers/gpu/drm/radeon/radeon_gem.c
+++ b/drivers/gpu/drm/radeon/radeon_gem.c
@@ -159,11 +159,9 @@ void radeon_gem_object_close(struct drm_gem_object *obj,
 static int radeon_gem_handle_lockup(struct radeon_device *rdev, int r)
 {
         if (r == -EDEADLK) {
-                radeon_mutex_lock(&rdev->cs_mutex);
                 r = radeon_gpu_reset(rdev);
                 if (!r)
                         r = -EAGAIN;
-                radeon_mutex_unlock(&rdev->cs_mutex);
         }
         return r;
 }
@@ -217,12 +215,14 @@ int radeon_gem_create_ioctl(struct drm_device *dev, void *data,
         uint32_t handle;
         int r;
 
+        down_read(&rdev->exclusive_lock);
         /* create a gem object to contain this object in */
         args->size = roundup(args->size, PAGE_SIZE);
         r = radeon_gem_object_create(rdev, args->size, args->alignment,
                                         args->initial_domain, false,
                                         false, &gobj);
         if (r) {
+                up_read(&rdev->exclusive_lock);
                 r = radeon_gem_handle_lockup(rdev, r);
                 return r;
         }
@@ -230,10 +230,12 @@ int radeon_gem_create_ioctl(struct drm_device *dev, void *data,
         /* drop reference from allocate - handle holds it now */
         drm_gem_object_unreference_unlocked(gobj);
         if (r) {
+                up_read(&rdev->exclusive_lock);
                 r = radeon_gem_handle_lockup(rdev, r);
                 return r;
         }
         args->handle = handle;
+        up_read(&rdev->exclusive_lock);
         return 0;
 }
 
@@ -242,6 +244,7 @@ int radeon_gem_set_domain_ioctl(struct drm_device *dev, void *data,
 {
         /* transition the BO to a domain -
          * just validate the BO into a certain domain */
+        struct radeon_device *rdev = dev->dev_private;
         struct drm_radeon_gem_set_domain *args = data;
         struct drm_gem_object *gobj;
         struct radeon_bo *robj;
@@ -249,10 +252,12 @@ int radeon_gem_set_domain_ioctl(struct drm_device *dev, void *data,
 
         /* for now if someone requests domain CPU -
          * just make sure the buffer is finished with */
+        down_read(&rdev->exclusive_lock);
 
         /* just do a BO wait for now */
         gobj = drm_gem_object_lookup(dev, filp, args->handle);
         if (gobj == NULL) {
+                up_read(&rdev->exclusive_lock);
                 return -ENOENT;
         }
         robj = gem_to_radeon_bo(gobj);
@@ -260,6 +265,7 @@ int radeon_gem_set_domain_ioctl(struct drm_device *dev, void *data,
         r = radeon_gem_set_domain(gobj, args->read_domains, args->write_domain);
 
         drm_gem_object_unreference_unlocked(gobj);
+        up_read(&rdev->exclusive_lock);
         r = radeon_gem_handle_lockup(robj->rdev, r);
         return r;
 }
diff --git a/drivers/gpu/drm/radeon/radeon_irq_kms.c b/drivers/gpu/drm/radeon/radeon_irq_kms.c
index 5df58d1aba06..afaa1727abd2 100644
--- a/drivers/gpu/drm/radeon/radeon_irq_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_irq_kms.c
@@ -32,6 +32,17 @@
 #include "radeon.h"
 #include "atom.h"
 
+#define RADEON_WAIT_IDLE_TIMEOUT 200
+
+/**
+ * radeon_driver_irq_handler_kms - irq handler for KMS
+ *
+ * @DRM_IRQ_ARGS: args
+ *
+ * This is the irq handler for the radeon KMS driver (all asics).
+ * radeon_irq_process is a macro that points to the per-asic
+ * irq handler callback.
+ */
 irqreturn_t radeon_driver_irq_handler_kms(DRM_IRQ_ARGS)
 {
         struct drm_device *dev = (struct drm_device *) arg;
@@ -43,6 +54,17 @@ irqreturn_t radeon_driver_irq_handler_kms(DRM_IRQ_ARGS)
 /*
  * Handle hotplug events outside the interrupt handler proper.
  */
+/**
+ * radeon_hotplug_work_func - display hotplug work handler
+ *
+ * @work: work struct
+ *
+ * This is the hot plug event work handler (all asics).
+ * The work gets scheduled from the irq handler if there
+ * was a hot plug interrupt.  It walks the connector table
+ * and calls the hotplug handler for each one, then sends
+ * a drm hotplug event to alert userspace.
+ */
 static void radeon_hotplug_work_func(struct work_struct *work)
 {
         struct radeon_device *rdev = container_of(work, struct radeon_device,
@@ -59,61 +81,94 @@ static void radeon_hotplug_work_func(struct work_struct *work)
         drm_helper_hpd_irq_event(dev);
 }
 
+/**
+ * radeon_driver_irq_preinstall_kms - drm irq preinstall callback
+ *
+ * @dev: drm dev pointer
+ *
+ * Gets the hw ready to enable irqs (all asics).
+ * This function disables all interrupt sources on the GPU.
+ */
 void radeon_driver_irq_preinstall_kms(struct drm_device *dev)
 {
         struct radeon_device *rdev = dev->dev_private;
+        unsigned long irqflags;
         unsigned i;
 
+        spin_lock_irqsave(&rdev->irq.lock, irqflags);
         /* Disable *all* interrupts */
         for (i = 0; i < RADEON_NUM_RINGS; i++)
-                rdev->irq.sw_int[i] = false;
+                atomic_set(&rdev->irq.ring_int[i], 0);
         rdev->irq.gui_idle = false;
         for (i = 0; i < RADEON_MAX_HPD_PINS; i++)
                 rdev->irq.hpd[i] = false;
         for (i = 0; i < RADEON_MAX_CRTCS; i++) {
                 rdev->irq.crtc_vblank_int[i] = false;
-                rdev->irq.pflip[i] = false;
+                atomic_set(&rdev->irq.pflip[i], 0);
                 rdev->irq.afmt[i] = false;
         }
         radeon_irq_set(rdev);
+        spin_unlock_irqrestore(&rdev->irq.lock, irqflags);
         /* Clear bits */
         radeon_irq_process(rdev);
 }
 
+/**
+ * radeon_driver_irq_postinstall_kms - drm irq preinstall callback
+ *
+ * @dev: drm dev pointer
+ *
+ * Handles stuff to be done after enabling irqs (all asics).
+ * Returns 0 on success.
+ */
 int radeon_driver_irq_postinstall_kms(struct drm_device *dev)
 {
-        struct radeon_device *rdev = dev->dev_private;
-        unsigned i;
-
         dev->max_vblank_count = 0x001fffff;
-        for (i = 0; i < RADEON_NUM_RINGS; i++)
-                rdev->irq.sw_int[i] = true;
-        radeon_irq_set(rdev);
         return 0;
 }
 
+/**
+ * radeon_driver_irq_uninstall_kms - drm irq uninstall callback
+ *
+ * @dev: drm dev pointer
+ *
+ * This function disables all interrupt sources on the GPU (all asics).
+ */
 void radeon_driver_irq_uninstall_kms(struct drm_device *dev)
 {
         struct radeon_device *rdev = dev->dev_private;
+        unsigned long irqflags;
         unsigned i;
 
         if (rdev == NULL) {
                 return;
         }
+        spin_lock_irqsave(&rdev->irq.lock, irqflags);
         /* Disable *all* interrupts */
         for (i = 0; i < RADEON_NUM_RINGS; i++)
-                rdev->irq.sw_int[i] = false;
+                atomic_set(&rdev->irq.ring_int[i], 0);
         rdev->irq.gui_idle = false;
         for (i = 0; i < RADEON_MAX_HPD_PINS; i++)
                 rdev->irq.hpd[i] = false;
         for (i = 0; i < RADEON_MAX_CRTCS; i++) {
                 rdev->irq.crtc_vblank_int[i] = false;
-                rdev->irq.pflip[i] = false;
+                atomic_set(&rdev->irq.pflip[i], 0);
                 rdev->irq.afmt[i] = false;
         }
         radeon_irq_set(rdev);
+        spin_unlock_irqrestore(&rdev->irq.lock, irqflags);
 }
 
+/**
+ * radeon_msi_ok - asic specific msi checks
+ *
+ * @rdev: radeon device pointer
+ *
+ * Handles asic specific MSI checks to determine if
+ * MSIs should be enabled on a particular chip (all asics).
+ * Returns true if MSIs should be enabled, false if MSIs
+ * should not be enabled.
+ */
 static bool radeon_msi_ok(struct radeon_device *rdev)
 {
         /* RV370/RV380 was first asic with MSI support */
@@ -166,17 +221,22 @@ static bool radeon_msi_ok(struct radeon_device *rdev)
         return true;
 }
 
+/**
+ * radeon_irq_kms_init - init driver interrupt info
+ *
+ * @rdev: radeon device pointer
+ *
+ * Sets up the work irq handlers, vblank init, MSIs, etc. (all asics).
+ * Returns 0 for success, error for failure.
+ */
 int radeon_irq_kms_init(struct radeon_device *rdev)
 {
-        int i;
         int r = 0;
 
         INIT_WORK(&rdev->hotplug_work, radeon_hotplug_work_func);
         INIT_WORK(&rdev->audio_work, r600_audio_update_hdmi);
 
-        spin_lock_init(&rdev->irq.sw_lock);
-        for (i = 0; i < rdev->num_crtc; i++)
-                spin_lock_init(&rdev->irq.pflip_lock[i]);
+        spin_lock_init(&rdev->irq.lock);
         r = drm_vblank_init(rdev->ddev, rdev->num_crtc);
         if (r) {
                 return r;
@@ -201,6 +261,13 @@ int radeon_irq_kms_init(struct radeon_device *rdev)
         return 0;
 }
 
+/**
+ * radeon_irq_kms_fini - tear down driver interrrupt info
+ *
+ * @rdev: radeon device pointer
+ *
+ * Tears down the work irq handlers, vblank handlers, MSIs, etc. (all asics).
+ */
 void radeon_irq_kms_fini(struct radeon_device *rdev)
 {
         drm_vblank_cleanup(rdev->ddev);
@@ -213,31 +280,63 @@ void radeon_irq_kms_fini(struct radeon_device *rdev)
         flush_work_sync(&rdev->hotplug_work);
 }
 
+/**
+ * radeon_irq_kms_sw_irq_get - enable software interrupt
+ *
+ * @rdev: radeon device pointer
+ * @ring: ring whose interrupt you want to enable
+ *
+ * Enables the software interrupt for a specific ring (all asics).
+ * The software interrupt is generally used to signal a fence on
+ * a particular ring.
+ */
 void radeon_irq_kms_sw_irq_get(struct radeon_device *rdev, int ring)
 {
         unsigned long irqflags;
 
-        spin_lock_irqsave(&rdev->irq.sw_lock, irqflags);
-        if (rdev->ddev->irq_enabled && (++rdev->irq.sw_refcount[ring] == 1)) {
-                rdev->irq.sw_int[ring] = true;
+        if (!rdev->ddev->irq_enabled)
+                return;
+
+        if (atomic_inc_return(&rdev->irq.ring_int[ring]) == 1) {
+                spin_lock_irqsave(&rdev->irq.lock, irqflags);
                 radeon_irq_set(rdev);
+                spin_unlock_irqrestore(&rdev->irq.lock, irqflags);
         }
-        spin_unlock_irqrestore(&rdev->irq.sw_lock, irqflags);
 }
 
+/**
+ * radeon_irq_kms_sw_irq_put - disable software interrupt
+ *
+ * @rdev: radeon device pointer
+ * @ring: ring whose interrupt you want to disable
+ *
+ * Disables the software interrupt for a specific ring (all asics).
+ * The software interrupt is generally used to signal a fence on
+ * a particular ring.
+ */
 void radeon_irq_kms_sw_irq_put(struct radeon_device *rdev, int ring)
 {
         unsigned long irqflags;
 
-        spin_lock_irqsave(&rdev->irq.sw_lock, irqflags);
-        BUG_ON(rdev->ddev->irq_enabled && rdev->irq.sw_refcount[ring] <= 0);
-        if (rdev->ddev->irq_enabled && (--rdev->irq.sw_refcount[ring] == 0)) {
-                rdev->irq.sw_int[ring] = false;
+        if (!rdev->ddev->irq_enabled)
+                return;
+
+        if (atomic_dec_and_test(&rdev->irq.ring_int[ring])) {
+                spin_lock_irqsave(&rdev->irq.lock, irqflags);
                 radeon_irq_set(rdev);
+                spin_unlock_irqrestore(&rdev->irq.lock, irqflags);
         }
-        spin_unlock_irqrestore(&rdev->irq.sw_lock, irqflags);
 }
 
+/**
+ * radeon_irq_kms_pflip_irq_get - enable pageflip interrupt
+ *
+ * @rdev: radeon device pointer
+ * @crtc: crtc whose interrupt you want to enable
+ *
+ * Enables the pageflip interrupt for a specific crtc (all asics).
+ * For pageflips we use the vblank interrupt source.
+ */
 void radeon_irq_kms_pflip_irq_get(struct radeon_device *rdev, int crtc)
 {
         unsigned long irqflags;
@@ -245,14 +344,25 @@ void radeon_irq_kms_pflip_irq_get(struct radeon_device *rdev, int crtc)
         if (crtc < 0 || crtc >= rdev->num_crtc)
                 return;
 
-        spin_lock_irqsave(&rdev->irq.pflip_lock[crtc], irqflags);
-        if (rdev->ddev->irq_enabled && (++rdev->irq.pflip_refcount[crtc] == 1)) {
-                rdev->irq.pflip[crtc] = true;
+        if (!rdev->ddev->irq_enabled)
+                return;
+
+        if (atomic_inc_return(&rdev->irq.pflip[crtc]) == 1) {
+                spin_lock_irqsave(&rdev->irq.lock, irqflags);
                 radeon_irq_set(rdev);
+                spin_unlock_irqrestore(&rdev->irq.lock, irqflags);
         }
-        spin_unlock_irqrestore(&rdev->irq.pflip_lock[crtc], irqflags);
 }
 
+/**
+ * radeon_irq_kms_pflip_irq_put - disable pageflip interrupt
+ *
+ * @rdev: radeon device pointer
+ * @crtc: crtc whose interrupt you want to disable
+ *
+ * Disables the pageflip interrupt for a specific crtc (all asics).
+ * For pageflips we use the vblank interrupt source.
+ */
 void radeon_irq_kms_pflip_irq_put(struct radeon_device *rdev, int crtc)
 {
         unsigned long irqflags;
@@ -260,12 +370,121 @@ void radeon_irq_kms_pflip_irq_put(struct radeon_device *rdev, int crtc)
         if (crtc < 0 || crtc >= rdev->num_crtc)
                 return;
 
-        spin_lock_irqsave(&rdev->irq.pflip_lock[crtc], irqflags);
-        BUG_ON(rdev->ddev->irq_enabled && rdev->irq.pflip_refcount[crtc] <= 0);
-        if (rdev->ddev->irq_enabled && (--rdev->irq.pflip_refcount[crtc] == 0)) {
-                rdev->irq.pflip[crtc] = false;
+        if (!rdev->ddev->irq_enabled)
+                return;
+
+        if (atomic_dec_and_test(&rdev->irq.pflip[crtc])) {
+                spin_lock_irqsave(&rdev->irq.lock, irqflags);
                 radeon_irq_set(rdev);
+                spin_unlock_irqrestore(&rdev->irq.lock, irqflags);
         }
-        spin_unlock_irqrestore(&rdev->irq.pflip_lock[crtc], irqflags);
 }
 
+/**
+ * radeon_irq_kms_enable_afmt - enable audio format change interrupt
+ *
+ * @rdev: radeon device pointer
+ * @block: afmt block whose interrupt you want to enable
+ *
+ * Enables the afmt change interrupt for a specific afmt block (all asics).
+ */
+void radeon_irq_kms_enable_afmt(struct radeon_device *rdev, int block)
+{
+        unsigned long irqflags;
+
+        spin_lock_irqsave(&rdev->irq.lock, irqflags);
+        rdev->irq.afmt[block] = true;
+        radeon_irq_set(rdev);
+        spin_unlock_irqrestore(&rdev->irq.lock, irqflags);
+
+}
+
+/**
+ * radeon_irq_kms_disable_afmt - disable audio format change interrupt
+ *
+ * @rdev: radeon device pointer
+ * @block: afmt block whose interrupt you want to disable
+ *
+ * Disables the afmt change interrupt for a specific afmt block (all asics).
+ */
+void radeon_irq_kms_disable_afmt(struct radeon_device *rdev, int block)
+{
+        unsigned long irqflags;
+
+        spin_lock_irqsave(&rdev->irq.lock, irqflags);
+        rdev->irq.afmt[block] = false;
+        radeon_irq_set(rdev);
+        spin_unlock_irqrestore(&rdev->irq.lock, irqflags);
+}
+
+/**
+ * radeon_irq_kms_enable_hpd - enable hotplug detect interrupt
+ *
+ * @rdev: radeon device pointer
+ * @hpd_mask: mask of hpd pins you want to enable.
+ *
+ * Enables the hotplug detect interrupt for a specific hpd pin (all asics).
+ */
+void radeon_irq_kms_enable_hpd(struct radeon_device *rdev, unsigned hpd_mask)
+{
+        unsigned long irqflags;
+        int i;
+
+        spin_lock_irqsave(&rdev->irq.lock, irqflags);
+        for (i = 0; i < RADEON_MAX_HPD_PINS; ++i)
+                rdev->irq.hpd[i] |= !!(hpd_mask & (1 << i));
+        radeon_irq_set(rdev);
+        spin_unlock_irqrestore(&rdev->irq.lock, irqflags);
+}
+
+/**
+ * radeon_irq_kms_disable_hpd - disable hotplug detect interrupt
+ *
+ * @rdev: radeon device pointer
+ * @hpd_mask: mask of hpd pins you want to disable.
+ *
+ * Disables the hotplug detect interrupt for a specific hpd pin (all asics).
+ */
+void radeon_irq_kms_disable_hpd(struct radeon_device *rdev, unsigned hpd_mask)
+{
+        unsigned long irqflags;
+        int i;
+
+        spin_lock_irqsave(&rdev->irq.lock, irqflags);
+        for (i = 0; i < RADEON_MAX_HPD_PINS; ++i)
+                rdev->irq.hpd[i] &= !(hpd_mask & (1 << i));
+        radeon_irq_set(rdev);
+        spin_unlock_irqrestore(&rdev->irq.lock, irqflags);
+}
+
+/**
+ * radeon_irq_kms_wait_gui_idle - waits for drawing engine to be idle
+ *
+ * @rdev: radeon device pointer
+ *
+ * Enabled the GUI idle interrupt and waits for it to fire (r6xx+).
+ * This is currently used to make sure the 3D engine is idle for power
+ * management, but should be replaces with proper fence waits.
+ * GUI idle interrupts don't work very well on pre-r6xx hw and it also
+ * does not take into account other aspects of the chip that may be busy.
+ * DO NOT USE GOING FORWARD.
+ */
+int radeon_irq_kms_wait_gui_idle(struct radeon_device *rdev)
+{
+        unsigned long irqflags;
+        int r;
+
+        spin_lock_irqsave(&rdev->irq.lock, irqflags);
+        rdev->irq.gui_idle = true;
+        radeon_irq_set(rdev);
+        spin_unlock_irqrestore(&rdev->irq.lock, irqflags);
+
+        r = wait_event_timeout(rdev->irq.idle_queue, radeon_gui_idle(rdev),
+                               msecs_to_jiffies(RADEON_WAIT_IDLE_TIMEOUT));
+
+        spin_lock_irqsave(&rdev->irq.lock, irqflags);
+        rdev->irq.gui_idle = false;
+        radeon_irq_set(rdev);
+        spin_unlock_irqrestore(&rdev->irq.lock, irqflags);
+        return r;
+}
diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c
index 5c58d7d90cb2..1d73f16b5d97 100644
--- a/drivers/gpu/drm/radeon/radeon_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_kms.c
@@ -33,6 +33,17 @@
 #include <linux/vga_switcheroo.h>
 #include <linux/slab.h>
 
+/**
+ * radeon_driver_unload_kms - Main unload function for KMS.
+ *
+ * @dev: drm dev pointer
+ *
+ * This is the main unload function for KMS (all asics).
+ * It calls radeon_modeset_fini() to tear down the
+ * displays, and radeon_device_fini() to tear down
+ * the rest of the device (CP, writeback, etc.).
+ * Returns 0 on success.
+ */
 int radeon_driver_unload_kms(struct drm_device *dev)
 {
         struct radeon_device *rdev = dev->dev_private;
@@ -46,6 +57,19 @@ int radeon_driver_unload_kms(struct drm_device *dev)
         return 0;
 }
 
+/**
+ * radeon_driver_load_kms - Main load function for KMS.
+ *
+ * @dev: drm dev pointer
+ * @flags: device flags
+ *
+ * This is the main load function for KMS (all asics).
+ * It calls radeon_device_init() to set up the non-display
+ * parts of the chip (asic init, CP, writeback, etc.), and
+ * radeon_modeset_init() to set up the display parts
+ * (crtcs, encoders, hotplug detect, etc.).
+ * Returns 0 on success, error on failure.
+ */
 int radeon_driver_load_kms(struct drm_device *dev, unsigned long flags)
 {
         struct radeon_device *rdev;
@@ -96,6 +120,16 @@ out:
         return r;
 }
 
+/**
+ * radeon_set_filp_rights - Set filp right.
+ *
+ * @dev: drm dev pointer
+ * @owner: drm file
+ * @applier: drm file
+ * @value: value
+ *
+ * Sets the filp rights for the device (all asics).
+ */
 static void radeon_set_filp_rights(struct drm_device *dev,
                                    struct drm_file **owner,
                                    struct drm_file *applier,
@@ -118,6 +152,18 @@ static void radeon_set_filp_rights(struct drm_device *dev,
 /*
  * Userspace get information ioctl
  */
+/**
+ * radeon_info_ioctl - answer a device specific request.
+ *
+ * @rdev: radeon device pointer
+ * @data: request object
+ * @filp: drm filp
+ *
+ * This function is used to pass device specific parameters to the userspace
+ * drivers.  Examples include: pci device id, pipeline parms, tiling params,
+ * etc. (all asics).
+ * Returns 0 on success, -EINVAL on failure.
+ */
 int radeon_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 {
         struct radeon_device *rdev = dev->dev_private;
@@ -301,16 +347,40 @@ int radeon_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 /*
  * Outdated mess for old drm with Xorg being in charge (void function now).
  */
+/**
+ * radeon_driver_firstopen_kms - drm callback for first open
+ *
+ * @dev: drm dev pointer
+ *
+ * Nothing to be done for KMS (all asics).
+ * Returns 0 on success.
+ */
 int radeon_driver_firstopen_kms(struct drm_device *dev)
 {
         return 0;
 }
 
+/**
+ * radeon_driver_firstopen_kms - drm callback for last close
+ *
+ * @dev: drm dev pointer
+ *
+ * Switch vga switcheroo state after last close (all asics).
+ */
 void radeon_driver_lastclose_kms(struct drm_device *dev)
 {
         vga_switcheroo_process_delayed_switch();
 }
 
+/**
+ * radeon_driver_open_kms - drm callback for open
+ *
+ * @dev: drm dev pointer
+ * @file_priv: drm file
+ *
+ * On device open, init vm on cayman+ (all asics).
+ * Returns 0 on success, error on failure.
+ */
 int radeon_driver_open_kms(struct drm_device *dev, struct drm_file *file_priv)
 {
         struct radeon_device *rdev = dev->dev_private;
@@ -339,6 +409,14 @@ int radeon_driver_open_kms(struct drm_device *dev, struct drm_file *file_priv)
         return 0;
 }
 
+/**
+ * radeon_driver_postclose_kms - drm callback for post close
+ *
+ * @dev: drm dev pointer
+ * @file_priv: drm file
+ *
+ * On device post close, tear down vm on cayman+ (all asics).
+ */
 void radeon_driver_postclose_kms(struct drm_device *dev,
                                  struct drm_file *file_priv)
 {
@@ -354,6 +432,15 @@ void radeon_driver_postclose_kms(struct drm_device *dev,
         }
 }
 
+/**
+ * radeon_driver_preclose_kms - drm callback for pre close
+ *
+ * @dev: drm dev pointer
+ * @file_priv: drm file
+ *
+ * On device pre close, tear down hyperz and cmask filps on r1xx-r5xx
+ * (all asics).
+ */
 void radeon_driver_preclose_kms(struct drm_device *dev,
                                 struct drm_file *file_priv)
 {
@@ -367,6 +454,15 @@ void radeon_driver_preclose_kms(struct drm_device *dev,
 /*
  * VBlank related functions.
  */
+/**
+ * radeon_get_vblank_counter_kms - get frame count
+ *
+ * @dev: drm dev pointer
+ * @crtc: crtc to get the frame count from
+ *
+ * Gets the frame count on the requested crtc (all asics).
+ * Returns frame count on success, -EINVAL on failure.
+ */
 u32 radeon_get_vblank_counter_kms(struct drm_device *dev, int crtc)
 {
         struct radeon_device *rdev = dev->dev_private;
@@ -379,34 +475,70 @@ u32 radeon_get_vblank_counter_kms(struct drm_device *dev, int crtc)
         return radeon_get_vblank_counter(rdev, crtc);
 }
 
+/**
+ * radeon_enable_vblank_kms - enable vblank interrupt
+ *
+ * @dev: drm dev pointer
+ * @crtc: crtc to enable vblank interrupt for
+ *
+ * Enable the interrupt on the requested crtc (all asics).
+ * Returns 0 on success, -EINVAL on failure.
+ */
 int radeon_enable_vblank_kms(struct drm_device *dev, int crtc)
 {
         struct radeon_device *rdev = dev->dev_private;
+        unsigned long irqflags;
+        int r;
 
         if (crtc < 0 || crtc >= rdev->num_crtc) {
                 DRM_ERROR("Invalid crtc %d\n", crtc);
                 return -EINVAL;
         }
 
+        spin_lock_irqsave(&rdev->irq.lock, irqflags);
         rdev->irq.crtc_vblank_int[crtc] = true;
-
-        return radeon_irq_set(rdev);
+        r = radeon_irq_set(rdev);
+        spin_unlock_irqrestore(&rdev->irq.lock, irqflags);
+        return r;
 }
 
+/**
+ * radeon_disable_vblank_kms - disable vblank interrupt
+ *
+ * @dev: drm dev pointer
+ * @crtc: crtc to disable vblank interrupt for
+ *
+ * Disable the interrupt on the requested crtc (all asics).
+ */
 void radeon_disable_vblank_kms(struct drm_device *dev, int crtc)
 {
         struct radeon_device *rdev = dev->dev_private;
+        unsigned long irqflags;
 
         if (crtc < 0 || crtc >= rdev->num_crtc) {
                 DRM_ERROR("Invalid crtc %d\n", crtc);
                 return;
         }
 
+        spin_lock_irqsave(&rdev->irq.lock, irqflags);
         rdev->irq.crtc_vblank_int[crtc] = false;
-
         radeon_irq_set(rdev);
+        spin_unlock_irqrestore(&rdev->irq.lock, irqflags);
 }
 
+/**
+ * radeon_get_vblank_timestamp_kms - get vblank timestamp
+ *
+ * @dev: drm dev pointer
+ * @crtc: crtc to get the timestamp for
+ * @max_error: max error
+ * @vblank_time: time value
+ * @flags: flags passed to the driver
+ *
+ * Gets the timestamp on the requested crtc based on the
+ * scanout position.  (all asics).
+ * Returns postive status flags on success, negative error on failure.
+ */
 int radeon_get_vblank_timestamp_kms(struct drm_device *dev, int crtc,
                                     int *max_error,
                                     struct timeval *vblank_time,
diff --git a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c
index 210317c7045e..d5fd615897ec 100644
--- a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c
+++ b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c
@@ -990,7 +990,7 @@ static void radeon_set_pll(struct drm_crtc *crtc, struct drm_display_mode *mode)
 }
 
 static bool radeon_crtc_mode_fixup(struct drm_crtc *crtc,
-                                   struct drm_display_mode *mode,
+                                   const struct drm_display_mode *mode,
                                    struct drm_display_mode *adjusted_mode)
 {
         if (!radeon_crtc_scaling_mode_fixup(crtc, mode, adjusted_mode))
diff --git a/drivers/gpu/drm/radeon/radeon_legacy_encoders.c b/drivers/gpu/drm/radeon/radeon_legacy_encoders.c
index a0c82229e8f0..670e9910f869 100644
--- a/drivers/gpu/drm/radeon/radeon_legacy_encoders.c
+++ b/drivers/gpu/drm/radeon/radeon_legacy_encoders.c
@@ -244,7 +244,7 @@ static void radeon_legacy_lvds_mode_set(struct drm_encoder *encoder,
 }
 
 static bool radeon_legacy_mode_fixup(struct drm_encoder *encoder,
-                                     struct drm_display_mode *mode,
+                                     const struct drm_display_mode *mode,
                                      struct drm_display_mode *adjusted_mode)
 {
         struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h
index 5b10ffd7bb2f..f380d59c5763 100644
--- a/drivers/gpu/drm/radeon/radeon_mode.h
+++ b/drivers/gpu/drm/radeon/radeon_mode.h
@@ -488,7 +488,7 @@ extern void radeon_connector_hotplug(struct drm_connector *connector);
 extern int radeon_dp_mode_valid_helper(struct drm_connector *connector,
                                        struct drm_display_mode *mode);
 extern void radeon_dp_set_link_config(struct drm_connector *connector,
-                                      struct drm_display_mode *mode);
+                                      const struct drm_display_mode *mode);
 extern void radeon_dp_link_train(struct drm_encoder *encoder,
                                  struct drm_connector *connector);
 extern bool radeon_dp_needs_link_train(struct radeon_connector *radeon_connector);
@@ -678,7 +678,7 @@ void radeon_enc_destroy(struct drm_encoder *encoder);
 void radeon_copy_fb(struct drm_device *dev, struct drm_gem_object *dst_obj);
 void radeon_combios_asic_init(struct drm_device *dev);
 bool radeon_crtc_scaling_mode_fixup(struct drm_crtc *crtc,
-                                        struct drm_display_mode *mode,
+                                        const struct drm_display_mode *mode,
                                         struct drm_display_mode *adjusted_mode);
 void radeon_panel_mode_fixup(struct drm_encoder *encoder,
                              struct drm_display_mode *adjusted_mode);
diff --git a/drivers/gpu/drm/radeon/radeon_object.c b/drivers/gpu/drm/radeon/radeon_object.c
index 830f1a7b486f..1f1a4c803c1d 100644
--- a/drivers/gpu/drm/radeon/radeon_object.c
+++ b/drivers/gpu/drm/radeon/radeon_object.c
@@ -115,9 +115,7 @@ int radeon_bo_create(struct radeon_device *rdev,
 
         size = ALIGN(size, PAGE_SIZE);
 
-        if (unlikely(rdev->mman.bdev.dev_mapping == NULL)) {
-                rdev->mman.bdev.dev_mapping = rdev->ddev->dev_mapping;
-        }
+        rdev->mman.bdev.dev_mapping = rdev->ddev->dev_mapping;
         if (kernel) {
                 type = ttm_bo_type_kernel;
         } else if (sg) {
@@ -138,7 +136,6 @@ int radeon_bo_create(struct radeon_device *rdev,
         acc_size = ttm_bo_dma_acc_size(&rdev->mman.bdev, size,
                                        sizeof(struct radeon_bo));
 
-retry:
         bo = kzalloc(sizeof(struct radeon_bo), GFP_KERNEL);
         if (bo == NULL)
                 return -ENOMEM;
@@ -152,13 +149,15 @@ retry:
         bo->surface_reg = -1;
         INIT_LIST_HEAD(&bo->list);
         INIT_LIST_HEAD(&bo->va);
+
+retry:
         radeon_ttm_placement_from_domain(bo, domain);
         /* Kernel allocation are uninterruptible */
-        mutex_lock(&rdev->vram_mutex);
+        down_read(&rdev->pm.mclk_lock);
         r = ttm_bo_init(&rdev->mman.bdev, &bo->tbo, size, type,
                         &bo->placement, page_align, 0, !kernel, NULL,
                         acc_size, sg, &radeon_ttm_bo_destroy);
-        mutex_unlock(&rdev->vram_mutex);
+        up_read(&rdev->pm.mclk_lock);
         if (unlikely(r != 0)) {
                 if (r != -ERESTARTSYS) {
                         if (domain == RADEON_GEM_DOMAIN_VRAM) {
@@ -219,9 +218,9 @@ void radeon_bo_unref(struct radeon_bo **bo)
                 return;
         rdev = (*bo)->rdev;
         tbo = &((*bo)->tbo);
-        mutex_lock(&rdev->vram_mutex);
+        down_read(&rdev->pm.mclk_lock);
         ttm_bo_unref(&tbo);
-        mutex_unlock(&rdev->vram_mutex);
+        up_read(&rdev->pm.mclk_lock);
         if (tbo == NULL)
                 *bo = NULL;
 }
diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
index 5b37e283ec38..7ae606600107 100644
--- a/drivers/gpu/drm/radeon/radeon_pm.c
+++ b/drivers/gpu/drm/radeon/radeon_pm.c
@@ -34,7 +34,6 @@
 #define RADEON_IDLE_LOOP_MS 100
 #define RADEON_RECLOCK_DELAY_MS 200
 #define RADEON_WAIT_VBLANK_TIMEOUT 200
-#define RADEON_WAIT_IDLE_TIMEOUT 200
 
 static const char *radeon_pm_state_type_name[5] = {
         "Default",
@@ -251,21 +250,14 @@ static void radeon_pm_set_clocks(struct radeon_device *rdev)
                 return;
 
         mutex_lock(&rdev->ddev->struct_mutex);
-        mutex_lock(&rdev->vram_mutex);
+        down_write(&rdev->pm.mclk_lock);
         mutex_lock(&rdev->ring_lock);
 
         /* gui idle int has issues on older chips it seems */
         if (rdev->family >= CHIP_R600) {
                 if (rdev->irq.installed) {
-                        /* wait for GPU idle */
-                        rdev->pm.gui_idle = false;
-                        rdev->irq.gui_idle = true;
-                        radeon_irq_set(rdev);
-                        wait_event_interruptible_timeout(
-                                rdev->irq.idle_queue, rdev->pm.gui_idle,
-                                msecs_to_jiffies(RADEON_WAIT_IDLE_TIMEOUT));
-                        rdev->irq.gui_idle = false;
-                        radeon_irq_set(rdev);
+                        /* wait for GPU to become idle */
+                        radeon_irq_kms_wait_gui_idle(rdev);
                 }
         } else {
                 struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
@@ -303,7 +295,7 @@ static void radeon_pm_set_clocks(struct radeon_device *rdev)
         rdev->pm.dynpm_planned_action = DYNPM_ACTION_NONE;
 
         mutex_unlock(&rdev->ring_lock);
-        mutex_unlock(&rdev->vram_mutex);
+        up_write(&rdev->pm.mclk_lock);
         mutex_unlock(&rdev->ddev->struct_mutex);
 }
 
diff --git a/drivers/gpu/drm/radeon/radeon_ring.c b/drivers/gpu/drm/radeon/radeon_ring.c
index 983658c91358..ec79b3750430 100644
--- a/drivers/gpu/drm/radeon/radeon_ring.c
+++ b/drivers/gpu/drm/radeon/radeon_ring.c
@@ -35,47 +35,97 @@
 #include "atom.h"
 
 /*
- * IB.
+ * IB
+ * IBs (Indirect Buffers) and areas of GPU accessible memory where
+ * commands are stored.  You can put a pointer to the IB in the
+ * command ring and the hw will fetch the commands from the IB
+ * and execute them.  Generally userspace acceleration drivers
+ * produce command buffers which are send to the kernel and
+ * put in IBs for execution by the requested ring.
  */
 int radeon_debugfs_sa_init(struct radeon_device *rdev);
 
+/**
+ * radeon_ib_get - request an IB (Indirect Buffer)
+ *
+ * @rdev: radeon_device pointer
+ * @ring: ring index the IB is associated with
+ * @ib: IB object returned
+ * @size: requested IB size
+ *
+ * Request an IB (all asics).  IBs are allocated using the
+ * suballocator.
+ * Returns 0 on success, error on failure.
+ */
 int radeon_ib_get(struct radeon_device *rdev, int ring,
                   struct radeon_ib *ib, unsigned size)
 {
-        int r;
+        int i, r;
 
         r = radeon_sa_bo_new(rdev, &rdev->ring_tmp_bo, &ib->sa_bo, size, 256, true);
         if (r) {
                 dev_err(rdev->dev, "failed to get a new IB (%d)\n", r);
                 return r;
         }
-        r = radeon_fence_create(rdev, &ib->fence, ring);
+
+        r = radeon_semaphore_create(rdev, &ib->semaphore);
         if (r) {
-                dev_err(rdev->dev, "failed to create fence for new IB (%d)\n", r);
-                radeon_sa_bo_free(rdev, &ib->sa_bo, NULL);
                 return r;
         }
 
+        ib->ring = ring;
+        ib->fence = NULL;
         ib->ptr = radeon_sa_bo_cpu_addr(ib->sa_bo);
         ib->gpu_addr = radeon_sa_bo_gpu_addr(ib->sa_bo);
         ib->vm_id = 0;
         ib->is_const_ib = false;
-        ib->semaphore = NULL;
+        for (i = 0; i < RADEON_NUM_RINGS; ++i)
+                ib->sync_to[i] = NULL;
 
         return 0;
 }
 
+/**
+ * radeon_ib_free - free an IB (Indirect Buffer)
+ *
+ * @rdev: radeon_device pointer
+ * @ib: IB object to free
+ *
+ * Free an IB (all asics).
+ */
 void radeon_ib_free(struct radeon_device *rdev, struct radeon_ib *ib)
 {
-        radeon_semaphore_free(rdev, ib->semaphore, ib->fence);
+        radeon_semaphore_free(rdev, &ib->semaphore, ib->fence);
         radeon_sa_bo_free(rdev, &ib->sa_bo, ib->fence);
         radeon_fence_unref(&ib->fence);
 }
 
-int radeon_ib_schedule(struct radeon_device *rdev, struct radeon_ib *ib)
+/**
+ * radeon_ib_schedule - schedule an IB (Indirect Buffer) on the ring
+ *
+ * @rdev: radeon_device pointer
+ * @ib: IB object to schedule
+ * @const_ib: Const IB to schedule (SI only)
+ *
+ * Schedule an IB on the associated ring (all asics).
+ * Returns 0 on success, error on failure.
+ *
+ * On SI, there are two parallel engines fed from the primary ring,
+ * the CE (Constant Engine) and the DE (Drawing Engine).  Since
+ * resource descriptors have moved to memory, the CE allows you to
+ * prime the caches while the DE is updating register state so that
+ * the resource descriptors will be already in cache when the draw is
+ * processed.  To accomplish this, the userspace driver submits two
+ * IBs, one for the CE and one for the DE.  If there is a CE IB (called
+ * a CONST_IB), it will be put on the ring prior to the DE IB.  Prior
+ * to SI there was just a DE IB.
+ */
+int radeon_ib_schedule(struct radeon_device *rdev, struct radeon_ib *ib,
+                       struct radeon_ib *const_ib)
 {
-        struct radeon_ring *ring = &rdev->ring[ib->fence->ring];
-        int r = 0;
+        struct radeon_ring *ring = &rdev->ring[ib->ring];
+        bool need_sync = false;
+        int i, r = 0;
 
         if (!ib->length_dw || !ring->ready) {
                 /* TODO: Nothings in the ib we should report. */
@@ -84,17 +134,51 @@ int radeon_ib_schedule(struct radeon_device *rdev, struct radeon_ib *ib)
         }
 
         /* 64 dwords should be enough for fence too */
-        r = radeon_ring_lock(rdev, ring, 64);
+        r = radeon_ring_lock(rdev, ring, 64 + RADEON_NUM_RINGS * 8);
         if (r) {
                 dev_err(rdev->dev, "scheduling IB failed (%d).\n", r);
                 return r;
         }
-        radeon_ring_ib_execute(rdev, ib->fence->ring, ib);
-        radeon_fence_emit(rdev, ib->fence);
+        for (i = 0; i < RADEON_NUM_RINGS; ++i) {
+                struct radeon_fence *fence = ib->sync_to[i];
+                if (radeon_fence_need_sync(fence, ib->ring)) {
+                        need_sync = true;
+                        radeon_semaphore_sync_rings(rdev, ib->semaphore,
+                                                    fence->ring, ib->ring);
+                        radeon_fence_note_sync(fence, ib->ring);
+                }
+        }
+        /* immediately free semaphore when we don't need to sync */
+        if (!need_sync) {
+                radeon_semaphore_free(rdev, &ib->semaphore, NULL);
+        }
+        if (const_ib) {
+                radeon_ring_ib_execute(rdev, const_ib->ring, const_ib);
+                radeon_semaphore_free(rdev, &const_ib->semaphore, NULL);
+        }
+        radeon_ring_ib_execute(rdev, ib->ring, ib);
+        r = radeon_fence_emit(rdev, &ib->fence, ib->ring);
+        if (r) {
+                dev_err(rdev->dev, "failed to emit fence for new IB (%d)\n", r);
+                radeon_ring_unlock_undo(rdev, ring);
+                return r;
+        }
+        if (const_ib) {
+                const_ib->fence = radeon_fence_ref(ib->fence);
+        }
         radeon_ring_unlock_commit(rdev, ring);
         return 0;
 }
 
+/**
+ * radeon_ib_pool_init - Init the IB (Indirect Buffer) pool
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Initialize the suballocator to manage a pool of memory
+ * for use as IBs (all asics).
+ * Returns 0 on success, error on failure.
+ */
 int radeon_ib_pool_init(struct radeon_device *rdev)
 {
         int r;
@@ -108,6 +192,12 @@ int radeon_ib_pool_init(struct radeon_device *rdev)
         if (r) {
                 return r;
         }
+
+        r = radeon_sa_bo_manager_start(rdev, &rdev->ring_tmp_bo);
+        if (r) {
+                return r;
+        }
+
         rdev->ib_pool_ready = true;
         if (radeon_debugfs_sa_init(rdev)) {
                 dev_err(rdev->dev, "failed to register debugfs file for SA\n");
@@ -115,24 +205,33 @@ int radeon_ib_pool_init(struct radeon_device *rdev)
         return 0;
 }
 
+/**
+ * radeon_ib_pool_fini - Free the IB (Indirect Buffer) pool
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Tear down the suballocator managing the pool of memory
+ * for use as IBs (all asics).
+ */
 void radeon_ib_pool_fini(struct radeon_device *rdev)
 {
         if (rdev->ib_pool_ready) {
+                radeon_sa_bo_manager_suspend(rdev, &rdev->ring_tmp_bo);
                 radeon_sa_bo_manager_fini(rdev, &rdev->ring_tmp_bo);
                 rdev->ib_pool_ready = false;
         }
 }
 
-int radeon_ib_pool_start(struct radeon_device *rdev)
-{
-        return radeon_sa_bo_manager_start(rdev, &rdev->ring_tmp_bo);
-}
-
-int radeon_ib_pool_suspend(struct radeon_device *rdev)
-{
-        return radeon_sa_bo_manager_suspend(rdev, &rdev->ring_tmp_bo);
-}
-
+/**
+ * radeon_ib_ring_tests - test IBs on the rings
+ *
+ * @rdev: radeon_device pointer
+ *
+ * Test an IB (Indirect Buffer) on each ring.
+ * If the test fails, disable the ring.
+ * Returns 0 on success, error if the primary GFX ring
+ * IB test fails.
+ */
 int radeon_ib_ring_tests(struct radeon_device *rdev)
 {
         unsigned i;
@@ -164,10 +263,28 @@ int radeon_ib_ring_tests(struct radeon_device *rdev)
 }
 
 /*
- * Ring.
+ * Rings
+ * Most engines on the GPU are fed via ring buffers.  Ring
+ * buffers are areas of GPU accessible memory that the host
+ * writes commands into and the GPU reads commands out of.
+ * There is a rptr (read pointer) that determines where the
+ * GPU is currently reading, and a wptr (write pointer)
+ * which determines where the host has written.  When the
+ * pointers are equal, the ring is idle.  When the host
+ * writes commands to the ring buffer, it increments the
+ * wptr.  The GPU then starts fetching commands and executes
+ * them until the pointers are equal again.
  */
 int radeon_debugfs_ring_init(struct radeon_device *rdev, struct radeon_ring *ring);
 
+/**
+ * radeon_ring_write - write a value to the ring
+ *
+ * @ring: radeon_ring structure holding ring information
+ * @v: dword (dw) value to write
+ *
+ * Write a value to the requested ring buffer (all asics).
+ */
 void radeon_ring_write(struct radeon_ring *ring, uint32_t v)
 {
 #if DRM_DEBUG_CODE
@@ -181,21 +298,37 @@ void radeon_ring_write(struct radeon_ring *ring, uint32_t v)
         ring->ring_free_dw--;
 }
 
-int radeon_ring_index(struct radeon_device *rdev, struct radeon_ring *ring)
+/**
+ * radeon_ring_supports_scratch_reg - check if the ring supports
+ * writing to scratch registers
+ *
+ * @rdev: radeon_device pointer
+ * @ring: radeon_ring structure holding ring information
+ *
+ * Check if a specific ring supports writing to scratch registers (all asics).
+ * Returns true if the ring supports writing to scratch regs, false if not.
+ */
+bool radeon_ring_supports_scratch_reg(struct radeon_device *rdev,
+                                      struct radeon_ring *ring)
 {
-        /* r1xx-r5xx only has CP ring */
-        if (rdev->family < CHIP_R600)
-                return RADEON_RING_TYPE_GFX_INDEX;
-
-        if (rdev->family >= CHIP_CAYMAN) {
-                if (ring == &rdev->ring[CAYMAN_RING_TYPE_CP1_INDEX])
-                        return CAYMAN_RING_TYPE_CP1_INDEX;
-                else if (ring == &rdev->ring[CAYMAN_RING_TYPE_CP2_INDEX])
-                        return CAYMAN_RING_TYPE_CP2_INDEX;
+        switch (ring->idx) {
+        case RADEON_RING_TYPE_GFX_INDEX:
+        case CAYMAN_RING_TYPE_CP1_INDEX:
+        case CAYMAN_RING_TYPE_CP2_INDEX:
+                return true;
+        default:
+                return false;
         }
-        return RADEON_RING_TYPE_GFX_INDEX;
 }
 
+/**
+ * radeon_ring_free_size - update the free size
+ *
+ * @rdev: radeon_device pointer
+ * @ring: radeon_ring structure holding ring information
+ *
+ * Update the free dw slots in the ring buffer (all asics).
+ */
 void radeon_ring_free_size(struct radeon_device *rdev, struct radeon_ring *ring)
 {
         u32 rptr;
@@ -214,7 +347,16 @@ void radeon_ring_free_size(struct radeon_device *rdev, struct radeon_ring *ring)
         }
 }
 
-
+/**
+ * radeon_ring_alloc - allocate space on the ring buffer
+ *
+ * @rdev: radeon_device pointer
+ * @ring: radeon_ring structure holding ring information
+ * @ndw: number of dwords to allocate in the ring buffer
+ *
+ * Allocate @ndw dwords in the ring buffer (all asics).
+ * Returns 0 on success, error on failure.
+ */
 int radeon_ring_alloc(struct radeon_device *rdev, struct radeon_ring *ring, unsigned ndw)
 {
         int r;
@@ -227,7 +369,7 @@ int radeon_ring_alloc(struct radeon_device *rdev, struct radeon_ring *ring, unsi
                 if (ndw < ring->ring_free_dw) {
                         break;
                 }
-                r = radeon_fence_wait_next_locked(rdev, radeon_ring_index(rdev, ring));
+                r = radeon_fence_wait_next_locked(rdev, ring->idx);
                 if (r)
                         return r;
         }
@@ -236,6 +378,17 @@ int radeon_ring_alloc(struct radeon_device *rdev, struct radeon_ring *ring, unsi
         return 0;
 }
 
+/**
+ * radeon_ring_lock - lock the ring and allocate space on it
+ *
+ * @rdev: radeon_device pointer
+ * @ring: radeon_ring structure holding ring information
+ * @ndw: number of dwords to allocate in the ring buffer
+ *
+ * Lock the ring and allocate @ndw dwords in the ring buffer
+ * (all asics).
+ * Returns 0 on success, error on failure.
+ */
 int radeon_ring_lock(struct radeon_device *rdev, struct radeon_ring *ring, unsigned ndw)
 {
         int r;
@@ -249,15 +402,20 @@ int radeon_ring_lock(struct radeon_device *rdev, struct radeon_ring *ring, unsig
         return 0;
 }
 
+/**
+ * radeon_ring_commit - tell the GPU to execute the new
+ * commands on the ring buffer
+ *
+ * @rdev: radeon_device pointer
+ * @ring: radeon_ring structure holding ring information
+ *
+ * Update the wptr (write pointer) to tell the GPU to
+ * execute new commands on the ring buffer (all asics).
+ */
 void radeon_ring_commit(struct radeon_device *rdev, struct radeon_ring *ring)
 {
-        unsigned count_dw_pad;
-        unsigned i;
-
         /* We pad to match fetch size */
-        count_dw_pad = (ring->align_mask + 1) -
-                       (ring->wptr & ring->align_mask);
-        for (i = 0; i < count_dw_pad; i++) {
+        while (ring->wptr & ring->align_mask) {
                 radeon_ring_write(ring, ring->nop);
         }
         DRM_MEMORYBARRIER();
@@ -265,23 +423,55 @@ void radeon_ring_commit(struct radeon_device *rdev, struct radeon_ring *ring)
         (void)RREG32(ring->wptr_reg);
 }
 
+/**
+ * radeon_ring_unlock_commit - tell the GPU to execute the new
+ * commands on the ring buffer and unlock it
+ *
+ * @rdev: radeon_device pointer
+ * @ring: radeon_ring structure holding ring information
+ *
+ * Call radeon_ring_commit() then unlock the ring (all asics).
+ */
 void radeon_ring_unlock_commit(struct radeon_device *rdev, struct radeon_ring *ring)
 {
         radeon_ring_commit(rdev, ring);
         mutex_unlock(&rdev->ring_lock);
 }
 
+/**
+ * radeon_ring_undo - reset the wptr
+ *
+ * @ring: radeon_ring structure holding ring information
+ *
+ * Reset the driver's copy of the wtpr (all asics).
+ */
 void radeon_ring_undo(struct radeon_ring *ring)
 {
         ring->wptr = ring->wptr_old;
 }
 
+/**
+ * radeon_ring_unlock_undo - reset the wptr and unlock the ring
+ *
+ * @ring: radeon_ring structure holding ring information
+ *
+ * Call radeon_ring_undo() then unlock the ring (all asics).
+ */
 void radeon_ring_unlock_undo(struct radeon_device *rdev, struct radeon_ring *ring)
 {
         radeon_ring_undo(ring);
         mutex_unlock(&rdev->ring_lock);
 }
 
+/**
+ * radeon_ring_force_activity - add some nop packets to the ring
+ *
+ * @rdev: radeon_device pointer
+ * @ring: radeon_ring structure holding ring information
+ *
+ * Add some nop packets to the ring to force activity (all asics).
+ * Used for lockup detection to see if the rptr is advancing.
+ */
 void radeon_ring_force_activity(struct radeon_device *rdev, struct radeon_ring *ring)
 {
         int r;
@@ -296,6 +486,13 @@ void radeon_ring_force_activity(struct radeon_device *rdev, struct radeon_ring *
         }
 }
 
+/**
+ * radeon_ring_force_activity - update lockup variables
+ *
+ * @ring: radeon_ring structure holding ring information
+ *
+ * Update the last rptr value and timestamp (all asics).
+ */
 void radeon_ring_lockup_update(struct radeon_ring *ring)
 {
         ring->last_rptr = ring->rptr;
@@ -349,6 +546,116 @@ bool radeon_ring_test_lockup(struct radeon_device *rdev, struct radeon_ring *rin
         return false;
 }
 
+/**
+ * radeon_ring_backup - Back up the content of a ring
+ *
+ * @rdev: radeon_device pointer
+ * @ring: the ring we want to back up
+ *
+ * Saves all unprocessed commits from a ring, returns the number of dwords saved.
+ */
+unsigned radeon_ring_backup(struct radeon_device *rdev, struct radeon_ring *ring,
+                            uint32_t **data)
+{
+        unsigned size, ptr, i;
+
+        /* just in case lock the ring */
+        mutex_lock(&rdev->ring_lock);
+        *data = NULL;
+
+        if (ring->ring_obj == NULL) {
+                mutex_unlock(&rdev->ring_lock);
+                return 0;
+        }
+
+        /* it doesn't make sense to save anything if all fences are signaled */
+        if (!radeon_fence_count_emitted(rdev, ring->idx)) {
+                mutex_unlock(&rdev->ring_lock);
+                return 0;
+        }
+
+        /* calculate the number of dw on the ring */
+        if (ring->rptr_save_reg)
+                ptr = RREG32(ring->rptr_save_reg);
+        else if (rdev->wb.enabled)
+                ptr = le32_to_cpu(*ring->next_rptr_cpu_addr);
+        else {
+                /* no way to read back the next rptr */
+                mutex_unlock(&rdev->ring_lock);
+                return 0;
+        }
+
+        size = ring->wptr + (ring->ring_size / 4);
+        size -= ptr;
+        size &= ring->ptr_mask;
+        if (size == 0) {
+                mutex_unlock(&rdev->ring_lock);
+                return 0;
+        }
+
+        /* and then save the content of the ring */
+        *data = kmalloc_array(size, sizeof(uint32_t), GFP_KERNEL);
+        if (!*data) {
+                mutex_unlock(&rdev->ring_lock);
+                return 0;
+        }
+        for (i = 0; i < size; ++i) {
+                (*data)[i] = ring->ring[ptr++];
+                ptr &= ring->ptr_mask;
+        }
+
+        mutex_unlock(&rdev->ring_lock);
+        return size;
+}
+
+/**
+ * radeon_ring_restore - append saved commands to the ring again
+ *
+ * @rdev: radeon_device pointer
+ * @ring: ring to append commands to
+ * @size: number of dwords we want to write
+ * @data: saved commands
+ *
+ * Allocates space on the ring and restore the previously saved commands.
+ */
+int radeon_ring_restore(struct radeon_device *rdev, struct radeon_ring *ring,
+                        unsigned size, uint32_t *data)
+{
+        int i, r;
+
+        if (!size || !data)
+                return 0;
+
+        /* restore the saved ring content */
+        r = radeon_ring_lock(rdev, ring, size);
+        if (r)
+                return r;
+
+        for (i = 0; i < size; ++i) {
+                radeon_ring_write(ring, data[i]);
+        }
+
+        radeon_ring_unlock_commit(rdev, ring);
+        kfree(data);
+        return 0;
+}
+
+/**
+ * radeon_ring_init - init driver ring struct.
+ *
+ * @rdev: radeon_device pointer
+ * @ring: radeon_ring structure holding ring information
+ * @ring_size: size of the ring
+ * @rptr_offs: offset of the rptr writeback location in the WB buffer
+ * @rptr_reg: MMIO offset of the rptr register
+ * @wptr_reg: MMIO offset of the wptr register
+ * @ptr_reg_shift: bit offset of the rptr/wptr values
+ * @ptr_reg_mask: bit mask of the rptr/wptr values
+ * @nop: nop packet for this ring
+ *
+ * Initialize the driver information for the selected ring (all asics).
+ * Returns 0 on success, error on failure.
+ */
 int radeon_ring_init(struct radeon_device *rdev, struct radeon_ring *ring, unsigned ring_size,
                      unsigned rptr_offs, unsigned rptr_reg, unsigned wptr_reg,
                      u32 ptr_reg_shift, u32 ptr_reg_mask, u32 nop)
@@ -391,12 +698,25 @@ int radeon_ring_init(struct radeon_device *rdev, struct radeon_ring *ring, unsig
         }
         ring->ptr_mask = (ring->ring_size / 4) - 1;
         ring->ring_free_dw = ring->ring_size / 4;
+        if (rdev->wb.enabled) {
+                u32 index = RADEON_WB_RING0_NEXT_RPTR + (ring->idx * 4);
+                ring->next_rptr_gpu_addr = rdev->wb.gpu_addr + index;
+                ring->next_rptr_cpu_addr = &rdev->wb.wb[index/4];
+        }
         if (radeon_debugfs_ring_init(rdev, ring)) {
                 DRM_ERROR("Failed to register debugfs file for rings !\n");
         }
         return 0;
 }
 
+/**
+ * radeon_ring_fini - tear down the driver ring struct.
+ *
+ * @rdev: radeon_device pointer
+ * @ring: radeon_ring structure holding ring information
+ *
+ * Tear down the driver information for the selected ring (all asics).
+ */
 void radeon_ring_fini(struct radeon_device *rdev, struct radeon_ring *ring)
 {
         int r;
@@ -438,6 +758,10 @@ static int radeon_debugfs_ring_info(struct seq_file *m, void *data)
         count = (ring->ring_size / 4) - ring->ring_free_dw;
         seq_printf(m, "wptr(0x%04x): 0x%08x\n", ring->wptr_reg, RREG32(ring->wptr_reg));
         seq_printf(m, "rptr(0x%04x): 0x%08x\n", ring->rptr_reg, RREG32(ring->rptr_reg));
+        if (ring->rptr_save_reg) {
+                seq_printf(m, "rptr next(0x%04x): 0x%08x\n", ring->rptr_save_reg,
+                           RREG32(ring->rptr_save_reg));
+        }
         seq_printf(m, "driver's copy of the wptr: 0x%08x\n", ring->wptr);
         seq_printf(m, "driver's copy of the rptr: 0x%08x\n", ring->rptr);
         seq_printf(m, "%u free dwords in ring\n", ring->ring_free_dw);
diff --git a/drivers/gpu/drm/radeon/radeon_sa.c b/drivers/gpu/drm/radeon/radeon_sa.c
index 32059b745728..4e771240fdd0 100644
--- a/drivers/gpu/drm/radeon/radeon_sa.c
+++ b/drivers/gpu/drm/radeon/radeon_sa.c
@@ -54,7 +54,7 @@ int radeon_sa_bo_manager_init(struct radeon_device *rdev,
 {
         int i, r;
 
-        spin_lock_init(&sa_manager->lock);
+        init_waitqueue_head(&sa_manager->wq);
         sa_manager->bo = NULL;
         sa_manager->size = size;
         sa_manager->domain = domain;
@@ -211,6 +211,39 @@ static bool radeon_sa_bo_try_alloc(struct radeon_sa_manager *sa_manager,
         return false;
 }
 
+/**
+ * radeon_sa_event - Check if we can stop waiting
+ *
+ * @sa_manager: pointer to the sa_manager
+ * @size: number of bytes we want to allocate
+ * @align: alignment we need to match
+ *
+ * Check if either there is a fence we can wait for or
+ * enough free memory to satisfy the allocation directly
+ */
+static bool radeon_sa_event(struct radeon_sa_manager *sa_manager,
+                            unsigned size, unsigned align)
+{
+        unsigned soffset, eoffset, wasted;
+        int i;
+
+        for (i = 0; i < RADEON_NUM_RINGS; ++i) {
+                if (!list_empty(&sa_manager->flist[i])) {
+                        return true;
+                }
+        }
+
+        soffset = radeon_sa_bo_hole_soffset(sa_manager);
+        eoffset = radeon_sa_bo_hole_eoffset(sa_manager);
+        wasted = (align - (soffset % align)) % align;
+
+        if ((eoffset - soffset) >= (size + wasted)) {
+                return true;
+        }
+
+        return false;
+}
+
 static bool radeon_sa_bo_next_hole(struct radeon_sa_manager *sa_manager,
                                    struct radeon_fence **fences,
                                    unsigned *tries)
@@ -297,8 +330,8 @@ int radeon_sa_bo_new(struct radeon_device *rdev,
         INIT_LIST_HEAD(&(*sa_bo)->olist);
         INIT_LIST_HEAD(&(*sa_bo)->flist);
 
-        spin_lock(&sa_manager->lock);
-        do {
+        spin_lock(&sa_manager->wq.lock);
+        while(1) {
                 for (i = 0; i < RADEON_NUM_RINGS; ++i) {
                         fences[i] = NULL;
                         tries[i] = 0;
@@ -309,30 +342,34 @@ int radeon_sa_bo_new(struct radeon_device *rdev,
 
                         if (radeon_sa_bo_try_alloc(sa_manager, *sa_bo,
                                                    size, align)) {
-                                spin_unlock(&sa_manager->lock);
+                                spin_unlock(&sa_manager->wq.lock);
                                 return 0;
                         }
 
                         /* see if we can skip over some allocations */
                 } while (radeon_sa_bo_next_hole(sa_manager, fences, tries));
 
-                if (block) {
-                        spin_unlock(&sa_manager->lock);
-                        r = radeon_fence_wait_any(rdev, fences, false);
-                        spin_lock(&sa_manager->lock);
-                        if (r) {
-                                /* if we have nothing to wait for we
-                                   are practically out of memory */
-                                if (r == -ENOENT) {
-                                        r = -ENOMEM;
-                                }
-                                goto out_err;
-                        }
+                if (!block) {
+                        break;
+                }
+
+                spin_unlock(&sa_manager->wq.lock);
+                r = radeon_fence_wait_any(rdev, fences, false);
+                spin_lock(&sa_manager->wq.lock);
+                /* if we have nothing to wait for block */
+                if (r == -ENOENT) {
+                        r = wait_event_interruptible_locked(
+                                sa_manager->wq, 
+                                radeon_sa_event(sa_manager, size, align)
+                        );
+                }
+                if (r) {
+                        goto out_err;
                 }
-        } while (block);
+        };
 
 out_err:
-        spin_unlock(&sa_manager->lock);
+        spin_unlock(&sa_manager->wq.lock);
         kfree(*sa_bo);
         *sa_bo = NULL;
         return r;
@@ -348,15 +385,16 @@ void radeon_sa_bo_free(struct radeon_device *rdev, struct radeon_sa_bo **sa_bo,
         }
 
         sa_manager = (*sa_bo)->manager;
-        spin_lock(&sa_manager->lock);
-        if (fence && fence->seq && fence->seq < RADEON_FENCE_NOTEMITED_SEQ) {
+        spin_lock(&sa_manager->wq.lock);
+        if (fence && !radeon_fence_signaled(fence)) {
                 (*sa_bo)->fence = radeon_fence_ref(fence);
                 list_add_tail(&(*sa_bo)->flist,
                               &sa_manager->flist[fence->ring]);
         } else {
                 radeon_sa_bo_remove_locked(*sa_bo);
         }
-        spin_unlock(&sa_manager->lock);
+        wake_up_all_locked(&sa_manager->wq);
+        spin_unlock(&sa_manager->wq.lock);
         *sa_bo = NULL;
 }
 
@@ -366,7 +404,7 @@ void radeon_sa_bo_dump_debug_info(struct radeon_sa_manager *sa_manager,
 {
         struct radeon_sa_bo *i;
 
-        spin_lock(&sa_manager->lock);
+        spin_lock(&sa_manager->wq.lock);
         list_for_each_entry(i, &sa_manager->olist, olist) {
                 if (&i->olist == sa_manager->hole) {
                         seq_printf(m, ">");
@@ -381,6 +419,6 @@ void radeon_sa_bo_dump_debug_info(struct radeon_sa_manager *sa_manager,
                 }
                 seq_printf(m, "\n");
         }
-        spin_unlock(&sa_manager->lock);
+        spin_unlock(&sa_manager->wq.lock);
 }
 #endif
diff --git a/drivers/gpu/drm/radeon/radeon_semaphore.c b/drivers/gpu/drm/radeon/radeon_semaphore.c
index e2ace5dce117..7cc78de6ddc3 100644
--- a/drivers/gpu/drm/radeon/radeon_semaphore.c
+++ b/drivers/gpu/drm/radeon/radeon_semaphore.c
@@ -68,70 +68,49 @@ void radeon_semaphore_emit_wait(struct radeon_device *rdev, int ring,
         radeon_semaphore_ring_emit(rdev, ring, &rdev->ring[ring], semaphore, true);
 }
 
+/* caller must hold ring lock */
 int radeon_semaphore_sync_rings(struct radeon_device *rdev,
                                 struct radeon_semaphore *semaphore,
-                                bool sync_to[RADEON_NUM_RINGS],
-                                int dst_ring)
+                                int signaler, int waiter)
 {
-        int i = 0, r;
+        int r;
 
-        mutex_lock(&rdev->ring_lock);
-        r = radeon_ring_alloc(rdev, &rdev->ring[dst_ring], RADEON_NUM_RINGS * 8);
-        if (r) {
-                goto error;
+        /* no need to signal and wait on the same ring */
+        if (signaler == waiter) {
+                return 0;
         }
 
-        for (i = 0; i < RADEON_NUM_RINGS; ++i) {
-                /* no need to sync to our own or unused rings */
-                if (!sync_to[i] || i == dst_ring)
-                        continue;
-
-                /* prevent GPU deadlocks */
-                if (!rdev->ring[i].ready) {
-                        dev_err(rdev->dev, "Trying to sync to a disabled ring!");
-                        r = -EINVAL;
-                        goto error;
-                }
-
-                r = radeon_ring_alloc(rdev, &rdev->ring[i], 8);
-                if (r) {
-                        goto error;
-                }
-
-                radeon_semaphore_emit_signal(rdev, i, semaphore);
-                radeon_semaphore_emit_wait(rdev, dst_ring, semaphore);
+        /* prevent GPU deadlocks */
+        if (!rdev->ring[signaler].ready) {
+                dev_err(rdev->dev, "Trying to sync to a disabled ring!");
+                return -EINVAL;
+        }
 
-                radeon_ring_commit(rdev, &rdev->ring[i]);
+        r = radeon_ring_alloc(rdev, &rdev->ring[signaler], 8);
+        if (r) {
+                return r;
         }
+        radeon_semaphore_emit_signal(rdev, signaler, semaphore);
+        radeon_ring_commit(rdev, &rdev->ring[signaler]);
 
-        radeon_ring_commit(rdev, &rdev->ring[dst_ring]);
-        mutex_unlock(&rdev->ring_lock);
+        /* we assume caller has already allocated space on waiters ring */
+        radeon_semaphore_emit_wait(rdev, waiter, semaphore);
 
         return 0;
-
-error:
-        /* unlock all locks taken so far */
-        for (--i; i >= 0; --i) {
-                if (sync_to[i] || i == dst_ring) {
-                        radeon_ring_undo(&rdev->ring[i]);
-                }
-        }
-        radeon_ring_undo(&rdev->ring[dst_ring]);
-        mutex_unlock(&rdev->ring_lock);
-        return r;
 }
 
 void radeon_semaphore_free(struct radeon_device *rdev,
-                           struct radeon_semaphore *semaphore,
+                           struct radeon_semaphore **semaphore,
                            struct radeon_fence *fence)
 {
-        if (semaphore == NULL) {
+        if (semaphore == NULL || *semaphore == NULL) {
                 return;
         }
-        if (semaphore->waiters > 0) {
+        if ((*semaphore)->waiters > 0) {
                 dev_err(rdev->dev, "semaphore %p has more waiters than signalers,"
-                        " hardware lockup imminent!\n", semaphore);
+                        " hardware lockup imminent!\n", *semaphore);
         }
-        radeon_sa_bo_free(rdev, &semaphore->sa_bo, fence);
-        kfree(semaphore);
+        radeon_sa_bo_free(rdev, &(*semaphore)->sa_bo, fence);
+        kfree(*semaphore);
+        *semaphore = NULL;
 }
diff --git a/drivers/gpu/drm/radeon/radeon_test.c b/drivers/gpu/drm/radeon/radeon_test.c
index efff929ea49d..7c16540c10ff 100644
--- a/drivers/gpu/drm/radeon/radeon_test.c
+++ b/drivers/gpu/drm/radeon/radeon_test.c
@@ -106,13 +106,7 @@ void radeon_test_moves(struct radeon_device *rdev)
 
                 radeon_bo_kunmap(gtt_obj[i]);
 
-                r = radeon_fence_create(rdev, &fence, RADEON_RING_TYPE_GFX_INDEX);
-                if (r) {
-                        DRM_ERROR("Failed to create GTT->VRAM fence %d\n", i);
-                        goto out_cleanup;
-                }
-
-                r = radeon_copy(rdev, gtt_addr, vram_addr, size / RADEON_GPU_PAGE_SIZE, fence);
+                r = radeon_copy(rdev, gtt_addr, vram_addr, size / RADEON_GPU_PAGE_SIZE, &fence);
                 if (r) {
                         DRM_ERROR("Failed GTT->VRAM copy %d\n", i);
                         goto out_cleanup;
@@ -155,13 +149,7 @@ void radeon_test_moves(struct radeon_device *rdev)
 
                 radeon_bo_kunmap(vram_obj);
 
-                r = radeon_fence_create(rdev, &fence, RADEON_RING_TYPE_GFX_INDEX);
-                if (r) {
-                        DRM_ERROR("Failed to create VRAM->GTT fence %d\n", i);
-                        goto out_cleanup;
-                }
-
-                r = radeon_copy(rdev, vram_addr, gtt_addr, size / RADEON_GPU_PAGE_SIZE, fence);
+                r = radeon_copy(rdev, vram_addr, gtt_addr, size / RADEON_GPU_PAGE_SIZE, &fence);
                 if (r) {
                         DRM_ERROR("Failed VRAM->GTT copy %d\n", i);
                         goto out_cleanup;
@@ -241,36 +229,33 @@ void radeon_test_ring_sync(struct radeon_device *rdev,
 {
         struct radeon_fence *fence1 = NULL, *fence2 = NULL;
         struct radeon_semaphore *semaphore = NULL;
-        int ridxA = radeon_ring_index(rdev, ringA);
-        int ridxB = radeon_ring_index(rdev, ringB);
         int r;
 
-        r = radeon_fence_create(rdev, &fence1, ridxA);
+        r = radeon_semaphore_create(rdev, &semaphore);
         if (r) {
-                DRM_ERROR("Failed to create sync fence 1\n");
+                DRM_ERROR("Failed to create semaphore\n");
                 goto out_cleanup;
         }
-        r = radeon_fence_create(rdev, &fence2, ridxA);
+
+        r = radeon_ring_lock(rdev, ringA, 64);
         if (r) {
-                DRM_ERROR("Failed to create sync fence 2\n");
+                DRM_ERROR("Failed to lock ring A %d\n", ringA->idx);
                 goto out_cleanup;
         }
-
-        r = radeon_semaphore_create(rdev, &semaphore);
+        radeon_semaphore_emit_wait(rdev, ringA->idx, semaphore);
+        r = radeon_fence_emit(rdev, &fence1, ringA->idx);
         if (r) {
-                DRM_ERROR("Failed to create semaphore\n");
+                DRM_ERROR("Failed to emit fence 1\n");
+                radeon_ring_unlock_undo(rdev, ringA);
                 goto out_cleanup;
         }
-
-        r = radeon_ring_lock(rdev, ringA, 64);
+        radeon_semaphore_emit_wait(rdev, ringA->idx, semaphore);
+        r = radeon_fence_emit(rdev, &fence2, ringA->idx);
         if (r) {
-                DRM_ERROR("Failed to lock ring A %d\n", ridxA);
+                DRM_ERROR("Failed to emit fence 2\n");
+                radeon_ring_unlock_undo(rdev, ringA);
                 goto out_cleanup;
         }
-        radeon_semaphore_emit_wait(rdev, ridxA, semaphore);
-        radeon_fence_emit(rdev, fence1);
-        radeon_semaphore_emit_wait(rdev, ridxA, semaphore);
-        radeon_fence_emit(rdev, fence2);
         radeon_ring_unlock_commit(rdev, ringA);
 
         mdelay(1000);
@@ -285,7 +270,7 @@ void radeon_test_ring_sync(struct radeon_device *rdev,
                 DRM_ERROR("Failed to lock ring B %p\n", ringB);
                 goto out_cleanup;
         }
-        radeon_semaphore_emit_signal(rdev, ridxB, semaphore);
+        radeon_semaphore_emit_signal(rdev, ringB->idx, semaphore);
         radeon_ring_unlock_commit(rdev, ringB);
 
         r = radeon_fence_wait(fence1, false);
@@ -306,7 +291,7 @@ void radeon_test_ring_sync(struct radeon_device *rdev,
                 DRM_ERROR("Failed to lock ring B %p\n", ringB);
                 goto out_cleanup;
         }
-        radeon_semaphore_emit_signal(rdev, ridxB, semaphore);
+        radeon_semaphore_emit_signal(rdev, ringB->idx, semaphore);
         radeon_ring_unlock_commit(rdev, ringB);
 
         r = radeon_fence_wait(fence2, false);
@@ -316,8 +301,7 @@ void radeon_test_ring_sync(struct radeon_device *rdev,
         }
 
 out_cleanup:
-        if (semaphore)
-                radeon_semaphore_free(rdev, semaphore, NULL);
+        radeon_semaphore_free(rdev, &semaphore, NULL);
 
         if (fence1)
                 radeon_fence_unref(&fence1);
@@ -336,23 +320,9 @@ void radeon_test_ring_sync2(struct radeon_device *rdev,
 {
         struct radeon_fence *fenceA = NULL, *fenceB = NULL;
         struct radeon_semaphore *semaphore = NULL;
-        int ridxA = radeon_ring_index(rdev, ringA);
-        int ridxB = radeon_ring_index(rdev, ringB);
-        int ridxC = radeon_ring_index(rdev, ringC);
         bool sigA, sigB;
         int i, r;
 
-        r = radeon_fence_create(rdev, &fenceA, ridxA);
-        if (r) {
-                DRM_ERROR("Failed to create sync fence 1\n");
-                goto out_cleanup;
-        }
-        r = radeon_fence_create(rdev, &fenceB, ridxB);
-        if (r) {
-                DRM_ERROR("Failed to create sync fence 2\n");
-                goto out_cleanup;
-        }
-
         r = radeon_semaphore_create(rdev, &semaphore);
         if (r) {
                 DRM_ERROR("Failed to create semaphore\n");
@@ -361,20 +331,30 @@ void radeon_test_ring_sync2(struct radeon_device *rdev,
 
         r = radeon_ring_lock(rdev, ringA, 64);
         if (r) {
-                DRM_ERROR("Failed to lock ring A %d\n", ridxA);
+                DRM_ERROR("Failed to lock ring A %d\n", ringA->idx);
+                goto out_cleanup;
+        }
+        radeon_semaphore_emit_wait(rdev, ringA->idx, semaphore);
+        r = radeon_fence_emit(rdev, &fenceA, ringA->idx);
+        if (r) {
+                DRM_ERROR("Failed to emit sync fence 1\n");
+                radeon_ring_unlock_undo(rdev, ringA);
                 goto out_cleanup;
         }
-        radeon_semaphore_emit_wait(rdev, ridxA, semaphore);
-        radeon_fence_emit(rdev, fenceA);
         radeon_ring_unlock_commit(rdev, ringA);
 
         r = radeon_ring_lock(rdev, ringB, 64);
         if (r) {
-                DRM_ERROR("Failed to lock ring B %d\n", ridxB);
+                DRM_ERROR("Failed to lock ring B %d\n", ringB->idx);
+                goto out_cleanup;
+        }
+        radeon_semaphore_emit_wait(rdev, ringB->idx, semaphore);
+        r = radeon_fence_emit(rdev, &fenceB, ringB->idx);
+        if (r) {
+                DRM_ERROR("Failed to create sync fence 2\n");
+                radeon_ring_unlock_undo(rdev, ringB);
                 goto out_cleanup;
         }
-        radeon_semaphore_emit_wait(rdev, ridxB, semaphore);
-        radeon_fence_emit(rdev, fenceB);
         radeon_ring_unlock_commit(rdev, ringB);
 
         mdelay(1000);
@@ -393,7 +373,7 @@ void radeon_test_ring_sync2(struct radeon_device *rdev,
                 DRM_ERROR("Failed to lock ring B %p\n", ringC);
                 goto out_cleanup;
         }
-        radeon_semaphore_emit_signal(rdev, ridxC, semaphore);
+        radeon_semaphore_emit_signal(rdev, ringC->idx, semaphore);
         radeon_ring_unlock_commit(rdev, ringC);
 
         for (i = 0; i < 30; ++i) {
@@ -419,7 +399,7 @@ void radeon_test_ring_sync2(struct radeon_device *rdev,
                 DRM_ERROR("Failed to lock ring B %p\n", ringC);
                 goto out_cleanup;
         }
-        radeon_semaphore_emit_signal(rdev, ridxC, semaphore);
+        radeon_semaphore_emit_signal(rdev, ringC->idx, semaphore);
         radeon_ring_unlock_commit(rdev, ringC);
 
         mdelay(1000);
@@ -436,8 +416,7 @@ void radeon_test_ring_sync2(struct radeon_device *rdev,
         }
 
 out_cleanup:
-        if (semaphore)
-                radeon_semaphore_free(rdev, semaphore, NULL);
+        radeon_semaphore_free(rdev, &semaphore, NULL);
 
         if (fenceA)
                 radeon_fence_unref(&fenceA);
diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c
index c94a2257761f..5b71c716d83f 100644
--- a/drivers/gpu/drm/radeon/radeon_ttm.c
+++ b/drivers/gpu/drm/radeon/radeon_ttm.c
@@ -222,15 +222,11 @@ static int radeon_move_blit(struct ttm_buffer_object *bo,
 {
         struct radeon_device *rdev;
         uint64_t old_start, new_start;
-        struct radeon_fence *fence, *old_fence;
-        struct radeon_semaphore *sem = NULL;
-        int r;
+        struct radeon_fence *fence;
+        int r, ridx;
 
         rdev = radeon_get_rdev(bo->bdev);
-        r = radeon_fence_create(rdev, &fence, radeon_copy_ring_index(rdev));
-        if (unlikely(r)) {
-                return r;
-        }
+        ridx = radeon_copy_ring_index(rdev);
         old_start = old_mem->start << PAGE_SHIFT;
         new_start = new_mem->start << PAGE_SHIFT;
 
@@ -243,7 +239,6 @@ static int radeon_move_blit(struct ttm_buffer_object *bo,
                 break;
         default:
                 DRM_ERROR("Unknown placement %d\n", old_mem->mem_type);
-                radeon_fence_unref(&fence);
                 return -EINVAL;
         }
         switch (new_mem->mem_type) {
@@ -255,46 +250,23 @@ static int radeon_move_blit(struct ttm_buffer_object *bo,
                 break;
         default:
                 DRM_ERROR("Unknown placement %d\n", old_mem->mem_type);
-                radeon_fence_unref(&fence);
                 return -EINVAL;
         }
-        if (!rdev->ring[radeon_copy_ring_index(rdev)].ready) {
+        if (!rdev->ring[ridx].ready) {
                 DRM_ERROR("Trying to move memory with ring turned off.\n");
-                radeon_fence_unref(&fence);
                 return -EINVAL;
         }
 
         BUILD_BUG_ON((PAGE_SIZE % RADEON_GPU_PAGE_SIZE) != 0);
 
         /* sync other rings */
-        old_fence = bo->sync_obj;
-        if (old_fence && old_fence->ring != fence->ring
-            && !radeon_fence_signaled(old_fence)) {
-                bool sync_to_ring[RADEON_NUM_RINGS] = { };
-                sync_to_ring[old_fence->ring] = true;
-
-                r = radeon_semaphore_create(rdev, &sem);
-                if (r) {
-                        radeon_fence_unref(&fence);
-                        return r;
-                }
-
-                r = radeon_semaphore_sync_rings(rdev, sem,
-                                                sync_to_ring, fence->ring);
-                if (r) {
-                        radeon_semaphore_free(rdev, sem, NULL);
-                        radeon_fence_unref(&fence);
-                        return r;
-                }
-        }
-
+        fence = bo->sync_obj;
         r = radeon_copy(rdev, old_start, new_start,
                         new_mem->num_pages * (PAGE_SIZE / RADEON_GPU_PAGE_SIZE), /* GPU pages */
-                        fence);
+                        &fence);
         /* FIXME: handle copy error */
         r = ttm_bo_move_accel_cleanup(bo, (void *)fence, NULL,
                                       evict, no_wait_reserve, no_wait_gpu, new_mem);
-        radeon_semaphore_free(rdev, sem, fence);
         radeon_fence_unref(&fence);
         return r;
 }
@@ -762,9 +734,7 @@ int radeon_ttm_init(struct radeon_device *rdev)
         }
         DRM_INFO("radeon: %uM of GTT memory ready.\n",
                  (unsigned)(rdev->mc.gtt_size / (1024 * 1024)));
-        if (unlikely(rdev->mman.bdev.dev_mapping == NULL)) {
-                rdev->mman.bdev.dev_mapping = rdev->ddev->dev_mapping;
-        }
+        rdev->mman.bdev.dev_mapping = rdev->ddev->dev_mapping;
 
         r = radeon_ttm_debugfs_init(rdev);
         if (r) {
@@ -825,9 +795,9 @@ static int radeon_ttm_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
                 return VM_FAULT_NOPAGE;
         }
         rdev = radeon_get_rdev(bo->bdev);
-        mutex_lock(&rdev->vram_mutex);
+        down_read(&rdev->pm.mclk_lock);
         r = ttm_vm_ops->fault(vma, vmf);
-        mutex_unlock(&rdev->vram_mutex);
+        up_read(&rdev->pm.mclk_lock);
         return r;
 }
 
diff --git a/drivers/gpu/drm/radeon/rs400.c b/drivers/gpu/drm/radeon/rs400.c
index a464eb5e2df2..2752f7f78237 100644
--- a/drivers/gpu/drm/radeon/rs400.c
+++ b/drivers/gpu/drm/radeon/rs400.c
@@ -426,13 +426,11 @@ static int rs400_startup(struct radeon_device *rdev)
                 return r;
         }
 
-        r = radeon_ib_pool_start(rdev);
-        if (r)
-                return r;
-
-        r = radeon_ib_ring_tests(rdev);
-        if (r)
+        r = radeon_ib_pool_init(rdev);
+        if (r) {
+                dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
                 return r;
+        }
 
         return 0;
 }
@@ -470,7 +468,6 @@ int rs400_resume(struct radeon_device *rdev)
 
 int rs400_suspend(struct radeon_device *rdev)
 {
-        radeon_ib_pool_suspend(rdev);
         r100_cp_disable(rdev);
         radeon_wb_disable(rdev);
         r100_irq_disable(rdev);
@@ -482,7 +479,7 @@ void rs400_fini(struct radeon_device *rdev)
 {
         r100_cp_fini(rdev);
         radeon_wb_fini(rdev);
-        r100_ib_fini(rdev);
+        radeon_ib_pool_fini(rdev);
         radeon_gem_fini(rdev);
         rs400_gart_fini(rdev);
         radeon_irq_kms_fini(rdev);
@@ -550,20 +547,14 @@ int rs400_init(struct radeon_device *rdev)
                 return r;
         r300_set_reg_safe(rdev);
 
-        r = radeon_ib_pool_init(rdev);
         rdev->accel_working = true;
-        if (r) {
-                dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
-                rdev->accel_working = false;
-        }
-
         r = rs400_startup(rdev);
         if (r) {
                 /* Somethings want wront with the accel init stop accel */
                 dev_err(rdev->dev, "Disabling GPU acceleration\n");
                 r100_cp_fini(rdev);
                 radeon_wb_fini(rdev);
-                r100_ib_fini(rdev);
+                radeon_ib_pool_fini(rdev);
                 rs400_gart_fini(rdev);
                 radeon_irq_kms_fini(rdev);
                 rdev->accel_working = false;
diff --git a/drivers/gpu/drm/radeon/rs600.c b/drivers/gpu/drm/radeon/rs600.c
index e95c5e61d4e2..5301b3df8466 100644
--- a/drivers/gpu/drm/radeon/rs600.c
+++ b/drivers/gpu/drm/radeon/rs600.c
@@ -294,6 +294,7 @@ void rs600_hpd_init(struct radeon_device *rdev)
 {
         struct drm_device *dev = rdev->ddev;
         struct drm_connector *connector;
+        unsigned enable = 0;
 
         list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
                 struct radeon_connector *radeon_connector = to_radeon_connector(connector);
@@ -301,26 +302,25 @@ void rs600_hpd_init(struct radeon_device *rdev)
                 case RADEON_HPD_1:
                         WREG32(R_007D00_DC_HOT_PLUG_DETECT1_CONTROL,
                                S_007D00_DC_HOT_PLUG_DETECT1_EN(1));
-                        rdev->irq.hpd[0] = true;
                         break;
                 case RADEON_HPD_2:
                         WREG32(R_007D10_DC_HOT_PLUG_DETECT2_CONTROL,
                                S_007D10_DC_HOT_PLUG_DETECT2_EN(1));
-                        rdev->irq.hpd[1] = true;
                         break;
                 default:
                         break;
                 }
+                enable |= 1 << radeon_connector->hpd.hpd;
                 radeon_hpd_set_polarity(rdev, radeon_connector->hpd.hpd);
         }
-        if (rdev->irq.installed)
-                rs600_irq_set(rdev);
+        radeon_irq_kms_enable_hpd(rdev, enable);
 }
 
 void rs600_hpd_fini(struct radeon_device *rdev)
 {
         struct drm_device *dev = rdev->ddev;
         struct drm_connector *connector;
+        unsigned disable = 0;
 
         list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
                 struct radeon_connector *radeon_connector = to_radeon_connector(connector);
@@ -328,17 +328,17 @@ void rs600_hpd_fini(struct radeon_device *rdev)
                 case RADEON_HPD_1:
                         WREG32(R_007D00_DC_HOT_PLUG_DETECT1_CONTROL,
                                S_007D00_DC_HOT_PLUG_DETECT1_EN(0));
-                        rdev->irq.hpd[0] = false;
                         break;
                 case RADEON_HPD_2:
                         WREG32(R_007D10_DC_HOT_PLUG_DETECT2_CONTROL,
                                S_007D10_DC_HOT_PLUG_DETECT2_EN(0));
-                        rdev->irq.hpd[1] = false;
                         break;
                 default:
                         break;
                 }
+                disable |= 1 << radeon_connector->hpd.hpd;
         }
+        radeon_irq_kms_disable_hpd(rdev, disable);
 }
 
 int rs600_asic_reset(struct radeon_device *rdev)
@@ -564,18 +564,18 @@ int rs600_irq_set(struct radeon_device *rdev)
                 WREG32(R_000040_GEN_INT_CNTL, 0);
                 return -EINVAL;
         }
-        if (rdev->irq.sw_int[RADEON_RING_TYPE_GFX_INDEX]) {
+        if (atomic_read(&rdev->irq.ring_int[RADEON_RING_TYPE_GFX_INDEX])) {
                 tmp |= S_000040_SW_INT_EN(1);
         }
         if (rdev->irq.gui_idle) {
                 tmp |= S_000040_GUI_IDLE(1);
         }
         if (rdev->irq.crtc_vblank_int[0] ||
-            rdev->irq.pflip[0]) {
+            atomic_read(&rdev->irq.pflip[0])) {
                 mode_int |= S_006540_D1MODE_VBLANK_INT_MASK(1);
         }
         if (rdev->irq.crtc_vblank_int[1] ||
-            rdev->irq.pflip[1]) {
+            atomic_read(&rdev->irq.pflip[1])) {
                 mode_int |= S_006540_D2MODE_VBLANK_INT_MASK(1);
         }
         if (rdev->irq.hpd[0]) {
@@ -686,7 +686,6 @@ int rs600_irq_process(struct radeon_device *rdev)
                 /* GUI idle */
                 if (G_000040_GUI_IDLE(status)) {
                         rdev->irq.gui_idle_acked = true;
-                        rdev->pm.gui_idle = true;
                         wake_up(&rdev->irq.idle_queue);
                 }
                 /* Vertical blank interrupts */
@@ -696,7 +695,7 @@ int rs600_irq_process(struct radeon_device *rdev)
                                 rdev->pm.vblank_sync = true;
                                 wake_up(&rdev->irq.vblank_queue);
                         }
-                        if (rdev->irq.pflip[0])
+                        if (atomic_read(&rdev->irq.pflip[0]))
                                 radeon_crtc_handle_flip(rdev, 0);
                 }
                 if (G_007EDC_LB_D2_VBLANK_INTERRUPT(rdev->irq.stat_regs.r500.disp_int)) {
@@ -705,7 +704,7 @@ int rs600_irq_process(struct radeon_device *rdev)
                                 rdev->pm.vblank_sync = true;
                                 wake_up(&rdev->irq.vblank_queue);
                         }
-                        if (rdev->irq.pflip[1])
+                        if (atomic_read(&rdev->irq.pflip[1]))
                                 radeon_crtc_handle_flip(rdev, 1);
                 }
                 if (G_007EDC_DC_HOT_PLUG_DETECT1_INTERRUPT(rdev->irq.stat_regs.r500.disp_int)) {
@@ -908,13 +907,11 @@ static int rs600_startup(struct radeon_device *rdev)
                 return r;
         }
 
-        r = radeon_ib_pool_start(rdev);
-        if (r)
-                return r;
-
-        r = radeon_ib_ring_tests(rdev);
-        if (r)
+        r = radeon_ib_pool_init(rdev);
+        if (r) {
+                dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
                 return r;
+        }
 
         r = r600_audio_init(rdev);
         if (r) {
@@ -956,7 +953,6 @@ int rs600_resume(struct radeon_device *rdev)
 
 int rs600_suspend(struct radeon_device *rdev)
 {
-        radeon_ib_pool_suspend(rdev);
         r600_audio_fini(rdev);
         r100_cp_disable(rdev);
         radeon_wb_disable(rdev);
@@ -970,7 +966,7 @@ void rs600_fini(struct radeon_device *rdev)
         r600_audio_fini(rdev);
         r100_cp_fini(rdev);
         radeon_wb_fini(rdev);
-        r100_ib_fini(rdev);
+        radeon_ib_pool_fini(rdev);
         radeon_gem_fini(rdev);
         rs600_gart_fini(rdev);
         radeon_irq_kms_fini(rdev);
@@ -1038,20 +1034,14 @@ int rs600_init(struct radeon_device *rdev)
                 return r;
         rs600_set_safe_registers(rdev);
 
-        r = radeon_ib_pool_init(rdev);
         rdev->accel_working = true;
-        if (r) {
-                dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
-                rdev->accel_working = false;
-        }
-
         r = rs600_startup(rdev);
         if (r) {
                 /* Somethings want wront with the accel init stop accel */
                 dev_err(rdev->dev, "Disabling GPU acceleration\n");
                 r100_cp_fini(rdev);
                 radeon_wb_fini(rdev);
-                r100_ib_fini(rdev);
+                radeon_ib_pool_fini(rdev);
                 rs600_gart_fini(rdev);
                 radeon_irq_kms_fini(rdev);
                 rdev->accel_working = false;
diff --git a/drivers/gpu/drm/radeon/rs690.c b/drivers/gpu/drm/radeon/rs690.c
index 159b6a43fda0..3b663fcfe061 100644
--- a/drivers/gpu/drm/radeon/rs690.c
+++ b/drivers/gpu/drm/radeon/rs690.c
@@ -637,13 +637,11 @@ static int rs690_startup(struct radeon_device *rdev)
                 return r;
         }
 
-        r = radeon_ib_pool_start(rdev);
-        if (r)
-                return r;
-
-        r = radeon_ib_ring_tests(rdev);
-        if (r)
+        r = radeon_ib_pool_init(rdev);
+        if (r) {
+                dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
                 return r;
+        }
 
         r = r600_audio_init(rdev);
         if (r) {
@@ -685,7 +683,6 @@ int rs690_resume(struct radeon_device *rdev)
 
 int rs690_suspend(struct radeon_device *rdev)
 {
-        radeon_ib_pool_suspend(rdev);
         r600_audio_fini(rdev);
         r100_cp_disable(rdev);
         radeon_wb_disable(rdev);
@@ -699,7 +696,7 @@ void rs690_fini(struct radeon_device *rdev)
         r600_audio_fini(rdev);
         r100_cp_fini(rdev);
         radeon_wb_fini(rdev);
-        r100_ib_fini(rdev);
+        radeon_ib_pool_fini(rdev);
         radeon_gem_fini(rdev);
         rs400_gart_fini(rdev);
         radeon_irq_kms_fini(rdev);
@@ -768,20 +765,14 @@ int rs690_init(struct radeon_device *rdev)
                 return r;
         rs600_set_safe_registers(rdev);
 
-        r = radeon_ib_pool_init(rdev);
         rdev->accel_working = true;
-        if (r) {
-                dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
-                rdev->accel_working = false;
-        }
-
         r = rs690_startup(rdev);
         if (r) {
                 /* Somethings want wront with the accel init stop accel */
                 dev_err(rdev->dev, "Disabling GPU acceleration\n");
                 r100_cp_fini(rdev);
                 radeon_wb_fini(rdev);
-                r100_ib_fini(rdev);
+                radeon_ib_pool_fini(rdev);
                 rs400_gart_fini(rdev);
                 radeon_irq_kms_fini(rdev);
                 rdev->accel_working = false;
diff --git a/drivers/gpu/drm/radeon/rv515.c b/drivers/gpu/drm/radeon/rv515.c
index 7f08cedb5333..a12fbcc8ccb6 100644
--- a/drivers/gpu/drm/radeon/rv515.c
+++ b/drivers/gpu/drm/radeon/rv515.c
@@ -408,13 +408,11 @@ static int rv515_startup(struct radeon_device *rdev)
                 return r;
         }
 
-        r = radeon_ib_pool_start(rdev);
-        if (r)
-                return r;
-
-        r = radeon_ib_ring_tests(rdev);
-        if (r)
+        r = radeon_ib_pool_init(rdev);
+        if (r) {
+                dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
                 return r;
+        }
 
         return 0;
 }
@@ -469,7 +467,7 @@ void rv515_fini(struct radeon_device *rdev)
 {
         r100_cp_fini(rdev);
         radeon_wb_fini(rdev);
-        r100_ib_fini(rdev);
+        radeon_ib_pool_fini(rdev);
         radeon_gem_fini(rdev);
         rv370_pcie_gart_fini(rdev);
         radeon_agp_fini(rdev);
@@ -543,20 +541,14 @@ int rv515_init(struct radeon_device *rdev)
                 return r;
         rv515_set_safe_registers(rdev);
 
-        r = radeon_ib_pool_init(rdev);
         rdev->accel_working = true;
-        if (r) {
-                dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
-                rdev->accel_working = false;
-        }
-
         r = rv515_startup(rdev);
         if (r) {
                 /* Somethings want wront with the accel init stop accel */
                 dev_err(rdev->dev, "Disabling GPU acceleration\n");
                 r100_cp_fini(rdev);
                 radeon_wb_fini(rdev);
-                r100_ib_fini(rdev);
+                radeon_ib_pool_fini(rdev);
                 radeon_irq_kms_fini(rdev);
                 rv370_pcie_gart_fini(rdev);
                 radeon_agp_fini(rdev);
diff --git a/drivers/gpu/drm/radeon/rv770.c b/drivers/gpu/drm/radeon/rv770.c
index b4f51c569c36..ca8ffec10ff6 100644
--- a/drivers/gpu/drm/radeon/rv770.c
+++ b/drivers/gpu/drm/radeon/rv770.c
@@ -358,8 +358,10 @@ static int rv770_cp_load_microcode(struct radeon_device *rdev)
 
 void r700_cp_fini(struct radeon_device *rdev)
 {
+        struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
         r700_cp_stop(rdev);
-        radeon_ring_fini(rdev, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]);
+        radeon_ring_fini(rdev, ring);
+        radeon_scratch_free(rdev, ring->rptr_save_reg);
 }
 
 /*
@@ -951,13 +953,11 @@ static int rv770_startup(struct radeon_device *rdev)
         if (r)
                 return r;
 
-        r = radeon_ib_pool_start(rdev);
-        if (r)
-                return r;
-
-        r = radeon_ib_ring_tests(rdev);
-        if (r)
+        r = radeon_ib_pool_init(rdev);
+        if (r) {
+                dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
                 return r;
+        }
 
         r = r600_audio_init(rdev);
         if (r) {
@@ -994,9 +994,6 @@ int rv770_resume(struct radeon_device *rdev)
 int rv770_suspend(struct radeon_device *rdev)
 {
         r600_audio_fini(rdev);
-        radeon_ib_pool_suspend(rdev);
-        r600_blit_suspend(rdev);
-        /* FIXME: we should wait for ring to be empty */
         r700_cp_stop(rdev);
         rdev->ring[RADEON_RING_TYPE_GFX_INDEX].ready = false;
         r600_irq_suspend(rdev);
@@ -1076,20 +1073,14 @@ int rv770_init(struct radeon_device *rdev)
         if (r)
                 return r;
 
-        r = radeon_ib_pool_init(rdev);
         rdev->accel_working = true;
-        if (r) {
-                dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
-                rdev->accel_working = false;
-        }
-
         r = rv770_startup(rdev);
         if (r) {
                 dev_err(rdev->dev, "disabling GPU acceleration\n");
                 r700_cp_fini(rdev);
                 r600_irq_fini(rdev);
                 radeon_wb_fini(rdev);
-                r100_ib_fini(rdev);
+                radeon_ib_pool_fini(rdev);
                 radeon_irq_kms_fini(rdev);
                 rv770_pcie_gart_fini(rdev);
                 rdev->accel_working = false;
@@ -1104,7 +1095,7 @@ void rv770_fini(struct radeon_device *rdev)
         r700_cp_fini(rdev);
         r600_irq_fini(rdev);
         radeon_wb_fini(rdev);
-        r100_ib_fini(rdev);
+        radeon_ib_pool_fini(rdev);
         radeon_irq_kms_fini(rdev);
         rv770_pcie_gart_fini(rdev);
         r600_vram_scratch_fini(rdev);
@@ -1121,6 +1112,8 @@ static void rv770_pcie_gen2_enable(struct radeon_device *rdev)
 {
         u32 link_width_cntl, lanes, speed_cntl, tmp;
         u16 link_cntl2;
+        u32 mask;
+        int ret;
 
         if (radeon_pcie_gen2 == 0)
                 return;
@@ -1135,6 +1128,15 @@ static void rv770_pcie_gen2_enable(struct radeon_device *rdev)
         if (ASIC_IS_X2(rdev))
                 return;
 
+        ret = drm_pcie_get_speed_cap_mask(rdev->ddev, &mask);
+        if (ret != 0)
+                return;
+
+        if (!(mask & DRM_PCIE_SPEED_50))
+                return;
+
+        DRM_INFO("enabling PCIE gen 2 link speeds, disable with radeon.pcie_gen2=0\n");
+
         /* advertise upconfig capability */
         link_width_cntl = RREG32_PCIE_P(PCIE_LC_LINK_WIDTH_CNTL);
         link_width_cntl &= ~LC_UPCONFIGURE_DIS;
diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c
index 0b0279291a73..c053f8193771 100644
--- a/drivers/gpu/drm/radeon/si.c
+++ b/drivers/gpu/drm/radeon/si.c
@@ -1762,13 +1762,34 @@ void si_fence_ring_emit(struct radeon_device *rdev,
  */
 void si_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib)
 {
-        struct radeon_ring *ring = &rdev->ring[ib->fence->ring];
+        struct radeon_ring *ring = &rdev->ring[ib->ring];
         u32 header;
 
-        if (ib->is_const_ib)
+        if (ib->is_const_ib) {
+                /* set switch buffer packet before const IB */
+                radeon_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0));
+                radeon_ring_write(ring, 0);
+
                 header = PACKET3(PACKET3_INDIRECT_BUFFER_CONST, 2);
-        else
+        } else {
+                u32 next_rptr;
+                if (ring->rptr_save_reg) {
+                        next_rptr = ring->wptr + 3 + 4 + 8;
+                        radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1));
+                        radeon_ring_write(ring, ((ring->rptr_save_reg -
+                                                  PACKET3_SET_CONFIG_REG_START) >> 2));
+                        radeon_ring_write(ring, next_rptr);
+                } else if (rdev->wb.enabled) {
+                        next_rptr = ring->wptr + 5 + 4 + 8;
+                        radeon_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
+                        radeon_ring_write(ring, (1 << 8));
+                        radeon_ring_write(ring, ring->next_rptr_gpu_addr & 0xfffffffc);
+                        radeon_ring_write(ring, upper_32_bits(ring->next_rptr_gpu_addr) & 0xffffffff);
+                        radeon_ring_write(ring, next_rptr);
+                }
+
                 header = PACKET3(PACKET3_INDIRECT_BUFFER, 2);
+        }
 
         radeon_ring_write(ring, header);
         radeon_ring_write(ring,
@@ -1779,18 +1800,20 @@ void si_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib)
         radeon_ring_write(ring, upper_32_bits(ib->gpu_addr) & 0xFFFF);
         radeon_ring_write(ring, ib->length_dw | (ib->vm_id << 24));
 
-        /* flush read cache over gart for this vmid */
-        radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1));
-        radeon_ring_write(ring, (CP_COHER_CNTL2 - PACKET3_SET_CONFIG_REG_START) >> 2);
-        radeon_ring_write(ring, ib->vm_id);
-        radeon_ring_write(ring, PACKET3(PACKET3_SURFACE_SYNC, 3));
-        radeon_ring_write(ring, PACKET3_TCL1_ACTION_ENA |
-                          PACKET3_TC_ACTION_ENA |
-                          PACKET3_SH_KCACHE_ACTION_ENA |
-                          PACKET3_SH_ICACHE_ACTION_ENA);
-        radeon_ring_write(ring, 0xFFFFFFFF);
-        radeon_ring_write(ring, 0);
-        radeon_ring_write(ring, 10); /* poll interval */
+        if (!ib->is_const_ib) {
+                /* flush read cache over gart for this vmid */
+                radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1));
+                radeon_ring_write(ring, (CP_COHER_CNTL2 - PACKET3_SET_CONFIG_REG_START) >> 2);
+                radeon_ring_write(ring, ib->vm_id);
+                radeon_ring_write(ring, PACKET3(PACKET3_SURFACE_SYNC, 3));
+                radeon_ring_write(ring, PACKET3_TCL1_ACTION_ENA |
+                                  PACKET3_TC_ACTION_ENA |
+                                  PACKET3_SH_KCACHE_ACTION_ENA |
+                                  PACKET3_SH_ICACHE_ACTION_ENA);
+                radeon_ring_write(ring, 0xFFFFFFFF);
+                radeon_ring_write(ring, 0);
+                radeon_ring_write(ring, 10); /* poll interval */
+        }
 }
 
 /*
@@ -1917,10 +1940,20 @@ static int si_cp_start(struct radeon_device *rdev)
 
 static void si_cp_fini(struct radeon_device *rdev)
 {
+        struct radeon_ring *ring;
         si_cp_enable(rdev, false);
-        radeon_ring_fini(rdev, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]);
-        radeon_ring_fini(rdev, &rdev->ring[CAYMAN_RING_TYPE_CP1_INDEX]);
-        radeon_ring_fini(rdev, &rdev->ring[CAYMAN_RING_TYPE_CP2_INDEX]);
+
+        ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX];
+        radeon_ring_fini(rdev, ring);
+        radeon_scratch_free(rdev, ring->rptr_save_reg);
+
+        ring = &rdev->ring[CAYMAN_RING_TYPE_CP1_INDEX];
+        radeon_ring_fini(rdev, ring);
+        radeon_scratch_free(rdev, ring->rptr_save_reg);
+
+        ring = &rdev->ring[CAYMAN_RING_TYPE_CP2_INDEX];
+        radeon_ring_fini(rdev, ring);
+        radeon_scratch_free(rdev, ring->rptr_save_reg);
 }
 
 static int si_cp_resume(struct radeon_device *rdev)
@@ -2702,7 +2735,7 @@ int si_ib_parse(struct radeon_device *rdev, struct radeon_ib *ib)
                         if (ib->is_const_ib)
                                 ret = si_vm_packet3_ce_check(rdev, ib->ptr, &pkt);
                         else {
-                                switch (ib->fence->ring) {
+                                switch (ib->ring) {
                                 case RADEON_RING_TYPE_GFX_INDEX:
                                         ret = si_vm_packet3_gfx_check(rdev, ib->ptr, &pkt);
                                         break;
@@ -2711,7 +2744,7 @@ int si_ib_parse(struct radeon_device *rdev, struct radeon_ib *ib)
                                         ret = si_vm_packet3_compute_check(rdev, ib->ptr, &pkt);
                                         break;
                                 default:
-                                        dev_err(rdev->dev, "Non-PM4 ring %d !\n", ib->fence->ring);
+                                        dev_err(rdev->dev, "Non-PM4 ring %d !\n", ib->ring);
                                         ret = -EINVAL;
                                         break;
                                 }
@@ -2942,7 +2975,6 @@ static void si_disable_interrupts(struct radeon_device *rdev)
         WREG32(IH_RB_RPTR, 0);
         WREG32(IH_RB_WPTR, 0);
         rdev->ih.enabled = false;
-        rdev->ih.wptr = 0;
         rdev->ih.rptr = 0;
 }
 
@@ -3093,45 +3125,45 @@ int si_irq_set(struct radeon_device *rdev)
         hpd6 = RREG32(DC_HPD6_INT_CONTROL) & ~DC_HPDx_INT_EN;
 
         /* enable CP interrupts on all rings */
-        if (rdev->irq.sw_int[RADEON_RING_TYPE_GFX_INDEX]) {
+        if (atomic_read(&rdev->irq.ring_int[RADEON_RING_TYPE_GFX_INDEX])) {
                 DRM_DEBUG("si_irq_set: sw int gfx\n");
                 cp_int_cntl |= TIME_STAMP_INT_ENABLE;
         }
-        if (rdev->irq.sw_int[CAYMAN_RING_TYPE_CP1_INDEX]) {
+        if (atomic_read(&rdev->irq.ring_int[CAYMAN_RING_TYPE_CP1_INDEX])) {
                 DRM_DEBUG("si_irq_set: sw int cp1\n");
                 cp_int_cntl1 |= TIME_STAMP_INT_ENABLE;
         }
-        if (rdev->irq.sw_int[CAYMAN_RING_TYPE_CP2_INDEX]) {
+        if (atomic_read(&rdev->irq.ring_int[CAYMAN_RING_TYPE_CP2_INDEX])) {
                 DRM_DEBUG("si_irq_set: sw int cp2\n");
                 cp_int_cntl2 |= TIME_STAMP_INT_ENABLE;
         }
         if (rdev->irq.crtc_vblank_int[0] ||
-            rdev->irq.pflip[0]) {
+            atomic_read(&rdev->irq.pflip[0])) {
                 DRM_DEBUG("si_irq_set: vblank 0\n");
                 crtc1 |= VBLANK_INT_MASK;
         }
         if (rdev->irq.crtc_vblank_int[1] ||
-            rdev->irq.pflip[1]) {
+            atomic_read(&rdev->irq.pflip[1])) {
                 DRM_DEBUG("si_irq_set: vblank 1\n");
                 crtc2 |= VBLANK_INT_MASK;
         }
         if (rdev->irq.crtc_vblank_int[2] ||
-            rdev->irq.pflip[2]) {
+            atomic_read(&rdev->irq.pflip[2])) {
                 DRM_DEBUG("si_irq_set: vblank 2\n");
                 crtc3 |= VBLANK_INT_MASK;
         }
         if (rdev->irq.crtc_vblank_int[3] ||
-            rdev->irq.pflip[3]) {
+            atomic_read(&rdev->irq.pflip[3])) {
                 DRM_DEBUG("si_irq_set: vblank 3\n");
                 crtc4 |= VBLANK_INT_MASK;
         }
         if (rdev->irq.crtc_vblank_int[4] ||
-            rdev->irq.pflip[4]) {
+            atomic_read(&rdev->irq.pflip[4])) {
                 DRM_DEBUG("si_irq_set: vblank 4\n");
                 crtc5 |= VBLANK_INT_MASK;
         }
         if (rdev->irq.crtc_vblank_int[5] ||
-            rdev->irq.pflip[5]) {
+            atomic_read(&rdev->irq.pflip[5])) {
                 DRM_DEBUG("si_irq_set: vblank 5\n");
                 crtc6 |= VBLANK_INT_MASK;
         }
@@ -3359,29 +3391,27 @@ int si_irq_process(struct radeon_device *rdev)
         u32 rptr;
         u32 src_id, src_data, ring_id;
         u32 ring_index;
-        unsigned long flags;
         bool queue_hotplug = false;
 
         if (!rdev->ih.enabled || rdev->shutdown)
                 return IRQ_NONE;
 
         wptr = si_get_ih_wptr(rdev);
+
+restart_ih:
+        /* is somebody else already processing irqs? */
+        if (atomic_xchg(&rdev->ih.lock, 1))
+                return IRQ_NONE;
+
         rptr = rdev->ih.rptr;
         DRM_DEBUG("si_irq_process start: rptr %d, wptr %d\n", rptr, wptr);
 
-        spin_lock_irqsave(&rdev->ih.lock, flags);
-        if (rptr == wptr) {
-                spin_unlock_irqrestore(&rdev->ih.lock, flags);
-                return IRQ_NONE;
-        }
-restart_ih:
         /* Order reading of wptr vs. reading of IH ring data */
         rmb();
 
         /* display interrupts */
         si_irq_ack(rdev);
 
-        rdev->ih.wptr = wptr;
         while (rptr != wptr) {
                 /* wptr/rptr are in bytes! */
                 ring_index = rptr / 4;
@@ -3399,7 +3429,7 @@ restart_ih:
                                                 rdev->pm.vblank_sync = true;
                                                 wake_up(&rdev->irq.vblank_queue);
                                         }
-                                        if (rdev->irq.pflip[0])
+                                        if (atomic_read(&rdev->irq.pflip[0]))
                                                 radeon_crtc_handle_flip(rdev, 0);
                                         rdev->irq.stat_regs.evergreen.disp_int &= ~LB_D1_VBLANK_INTERRUPT;
                                         DRM_DEBUG("IH: D1 vblank\n");
@@ -3425,7 +3455,7 @@ restart_ih:
                                                 rdev->pm.vblank_sync = true;
                                                 wake_up(&rdev->irq.vblank_queue);
                                         }
-                                        if (rdev->irq.pflip[1])
+                                        if (atomic_read(&rdev->irq.pflip[1]))
                                                 radeon_crtc_handle_flip(rdev, 1);
                                         rdev->irq.stat_regs.evergreen.disp_int_cont &= ~LB_D2_VBLANK_INTERRUPT;
                                         DRM_DEBUG("IH: D2 vblank\n");
@@ -3451,7 +3481,7 @@ restart_ih:
                                                 rdev->pm.vblank_sync = true;
                                                 wake_up(&rdev->irq.vblank_queue);
                                         }
-                                        if (rdev->irq.pflip[2])
+                                        if (atomic_read(&rdev->irq.pflip[2]))
                                                 radeon_crtc_handle_flip(rdev, 2);
                                         rdev->irq.stat_regs.evergreen.disp_int_cont2 &= ~LB_D3_VBLANK_INTERRUPT;
                                         DRM_DEBUG("IH: D3 vblank\n");
@@ -3477,7 +3507,7 @@ restart_ih:
                                                 rdev->pm.vblank_sync = true;
                                                 wake_up(&rdev->irq.vblank_queue);
                                         }
-                                        if (rdev->irq.pflip[3])
+                                        if (atomic_read(&rdev->irq.pflip[3]))
                                                 radeon_crtc_handle_flip(rdev, 3);
                                         rdev->irq.stat_regs.evergreen.disp_int_cont3 &= ~LB_D4_VBLANK_INTERRUPT;
                                         DRM_DEBUG("IH: D4 vblank\n");
@@ -3503,7 +3533,7 @@ restart_ih:
                                                 rdev->pm.vblank_sync = true;
                                                 wake_up(&rdev->irq.vblank_queue);
                                         }
-                                        if (rdev->irq.pflip[4])
+                                        if (atomic_read(&rdev->irq.pflip[4]))
                                                 radeon_crtc_handle_flip(rdev, 4);
                                         rdev->irq.stat_regs.evergreen.disp_int_cont4 &= ~LB_D5_VBLANK_INTERRUPT;
                                         DRM_DEBUG("IH: D5 vblank\n");
@@ -3529,7 +3559,7 @@ restart_ih:
                                                 rdev->pm.vblank_sync = true;
                                                 wake_up(&rdev->irq.vblank_queue);
                                         }
-                                        if (rdev->irq.pflip[5])
+                                        if (atomic_read(&rdev->irq.pflip[5]))
                                                 radeon_crtc_handle_flip(rdev, 5);
                                         rdev->irq.stat_regs.evergreen.disp_int_cont5 &= ~LB_D6_VBLANK_INTERRUPT;
                                         DRM_DEBUG("IH: D6 vblank\n");
@@ -3620,7 +3650,6 @@ restart_ih:
                         break;
                 case 233: /* GUI IDLE */
                         DRM_DEBUG("IH: GUI idle\n");
-                        rdev->pm.gui_idle = true;
                         wake_up(&rdev->irq.idle_queue);
                         break;
                 default:
@@ -3632,15 +3661,17 @@ restart_ih:
                 rptr += 16;
                 rptr &= rdev->ih.ptr_mask;
         }
-        /* make sure wptr hasn't changed while processing */
-        wptr = si_get_ih_wptr(rdev);
-        if (wptr != rdev->ih.wptr)
-                goto restart_ih;
         if (queue_hotplug)
                 schedule_work(&rdev->hotplug_work);
         rdev->ih.rptr = rptr;
         WREG32(IH_RB_RPTR, rdev->ih.rptr);
-        spin_unlock_irqrestore(&rdev->ih.lock, flags);
+        atomic_set(&rdev->ih.lock, 0);
+
+        /* make sure wptr hasn't changed while processing */
+        wptr = si_get_ih_wptr(rdev);
+        if (wptr != rptr)
+                goto restart_ih;
+
         return IRQ_HANDLED;
 }
 
@@ -3752,35 +3783,18 @@ static int si_startup(struct radeon_device *rdev)
         if (r)
                 return r;
 
-        r = radeon_ib_pool_start(rdev);
-        if (r)
-                return r;
-
-        r = radeon_ib_test(rdev, RADEON_RING_TYPE_GFX_INDEX, &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]);
-        if (r) {
-                DRM_ERROR("radeon: failed testing IB (%d) on CP ring 0\n", r);
-                rdev->accel_working = false;
-                return r;
-        }
-
-        r = radeon_ib_test(rdev, CAYMAN_RING_TYPE_CP1_INDEX, &rdev->ring[CAYMAN_RING_TYPE_CP1_INDEX]);
+        r = radeon_ib_pool_init(rdev);
         if (r) {
-                DRM_ERROR("radeon: failed testing IB (%d) on CP ring 1\n", r);
-                rdev->accel_working = false;
+                dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
                 return r;
         }
 
-        r = radeon_ib_test(rdev, CAYMAN_RING_TYPE_CP2_INDEX, &rdev->ring[CAYMAN_RING_TYPE_CP2_INDEX]);
+        r = radeon_vm_manager_init(rdev);
         if (r) {
-                DRM_ERROR("radeon: failed testing IB (%d) on CP ring 2\n", r);
-                rdev->accel_working = false;
+                dev_err(rdev->dev, "vm manager initialization failed (%d).\n", r);
                 return r;
         }
 
-        r = radeon_vm_manager_start(rdev);
-        if (r)
-                return r;
-
         return 0;
 }
 
@@ -3809,12 +3823,6 @@ int si_resume(struct radeon_device *rdev)
 
 int si_suspend(struct radeon_device *rdev)
 {
-        /* FIXME: we should wait for ring to be empty */
-        radeon_ib_pool_suspend(rdev);
-        radeon_vm_manager_suspend(rdev);
-#if 0
-        r600_blit_suspend(rdev);
-#endif
         si_cp_enable(rdev, false);
         rdev->ring[RADEON_RING_TYPE_GFX_INDEX].ready = false;
         rdev->ring[CAYMAN_RING_TYPE_CP1_INDEX].ready = false;
@@ -3903,17 +3911,7 @@ int si_init(struct radeon_device *rdev)
         if (r)
                 return r;
 
-        r = radeon_ib_pool_init(rdev);
         rdev->accel_working = true;
-        if (r) {
-                dev_err(rdev->dev, "IB initialization failed (%d).\n", r);
-                rdev->accel_working = false;
-        }
-        r = radeon_vm_manager_init(rdev);
-        if (r) {
-                dev_err(rdev->dev, "vm manager initialization failed (%d).\n", r);
-        }
-
         r = si_startup(rdev);
         if (r) {
                 dev_err(rdev->dev, "disabling GPU acceleration\n");
@@ -3921,7 +3919,7 @@ int si_init(struct radeon_device *rdev)
                 si_irq_fini(rdev);
                 si_rlc_fini(rdev);
                 radeon_wb_fini(rdev);
-                r100_ib_fini(rdev);
+                radeon_ib_pool_fini(rdev);
                 radeon_vm_manager_fini(rdev);
                 radeon_irq_kms_fini(rdev);
                 si_pcie_gart_fini(rdev);
@@ -3950,7 +3948,7 @@ void si_fini(struct radeon_device *rdev)
         si_rlc_fini(rdev);
         radeon_wb_fini(rdev);
         radeon_vm_manager_fini(rdev);
-        r100_ib_fini(rdev);
+        radeon_ib_pool_fini(rdev);
         radeon_irq_kms_fini(rdev);
         si_pcie_gart_fini(rdev);
         r600_vram_scratch_fini(rdev);
diff --git a/drivers/gpu/drm/radeon/sid.h b/drivers/gpu/drm/radeon/sid.h
index db4067962868..7869089e8761 100644
--- a/drivers/gpu/drm/radeon/sid.h
+++ b/drivers/gpu/drm/radeon/sid.h
@@ -901,5 +901,6 @@
 #define PACKET3_WAIT_ON_DE_COUNTER_DIFF                 0x88
 #define PACKET3_SET_CE_DE_COUNTERS                      0x89
 #define PACKET3_WAIT_ON_AVAIL_BUFFER                    0x8A
+#define PACKET3_SWITCH_BUFFER                           0x8B
 
 #endif
diff --git a/drivers/gpu/drm/savage/savage_bci.c b/drivers/gpu/drm/savage/savage_bci.c
index 6eb507a5d130..1efbb9075837 100644
--- a/drivers/gpu/drm/savage/savage_bci.c
+++ b/drivers/gpu/drm/savage/savage_bci.c
@@ -1050,6 +1050,7 @@ void savage_reclaim_buffers(struct drm_device *dev, struct drm_file *file_priv)
 {
         struct drm_device_dma *dma = dev->dma;
         drm_savage_private_t *dev_priv = dev->dev_private;
+        int release_idlelock = 0;
         int i;
 
         if (!dma)
@@ -1059,7 +1060,10 @@ void savage_reclaim_buffers(struct drm_device *dev, struct drm_file *file_priv)
         if (!dma->buflist)
                 return;
 
-        /*i830_flush_queue(dev); */
+        if (file_priv->master && file_priv->master->lock.hw_lock) {
+                drm_idlelock_take(&file_priv->master->lock);
+                release_idlelock = 1;
+        }
 
         for (i = 0; i < dma->buf_count; i++) {
                 struct drm_buf *buf = dma->buflist[i];
@@ -1075,7 +1079,8 @@ void savage_reclaim_buffers(struct drm_device *dev, struct drm_file *file_priv)
                 }
         }
 
-        drm_core_reclaim_buffers(dev, file_priv);
+        if (release_idlelock)
+                drm_idlelock_release(&file_priv->master->lock);
 }
 
 struct drm_ioctl_desc savage_ioctls[] = {
diff --git a/drivers/gpu/drm/savage/savage_drv.c b/drivers/gpu/drm/savage/savage_drv.c
index 89afe0b83643..d31d4cca9a4c 100644
--- a/drivers/gpu/drm/savage/savage_drv.c
+++ b/drivers/gpu/drm/savage/savage_drv.c
@@ -52,9 +52,9 @@ static struct drm_driver driver = {
         .dev_priv_size = sizeof(drm_savage_buf_priv_t),
         .load = savage_driver_load,
         .firstopen = savage_driver_firstopen,
+        .preclose = savage_reclaim_buffers,
         .lastclose = savage_driver_lastclose,
         .unload = savage_driver_unload,
-        .reclaim_buffers = savage_reclaim_buffers,
         .ioctls = savage_ioctls,
         .dma_ioctl = savage_bci_buffers,
         .fops = &savage_driver_fops,
diff --git a/drivers/gpu/drm/sis/sis_drv.c b/drivers/gpu/drm/sis/sis_drv.c
index dd14cd1a0033..7f119870147c 100644
--- a/drivers/gpu/drm/sis/sis_drv.c
+++ b/drivers/gpu/drm/sis/sis_drv.c
@@ -105,10 +105,9 @@ static struct drm_driver driver = {
         .load = sis_driver_load,
         .unload = sis_driver_unload,
         .open = sis_driver_open,
+        .preclose = sis_reclaim_buffers_locked,
         .postclose = sis_driver_postclose,
         .dma_quiescent = sis_idle,
-        .reclaim_buffers = NULL,
-        .reclaim_buffers_idlelocked = sis_reclaim_buffers_locked,
         .lastclose = sis_lastclose,
         .ioctls = sis_ioctls,
         .fops = &sis_driver_fops,
diff --git a/drivers/gpu/drm/sis/sis_mm.c b/drivers/gpu/drm/sis/sis_mm.c
index dd4a316c3d74..2c231070d250 100644
--- a/drivers/gpu/drm/sis/sis_mm.c
+++ b/drivers/gpu/drm/sis/sis_mm.c
@@ -74,7 +74,7 @@ static int sis_fb_init(struct drm_device *dev, void *data, struct drm_file *file
         dev_priv->vram_offset = fb->offset;
 
         mutex_unlock(&dev->struct_mutex);
-        DRM_DEBUG("offset = %u, size = %u\n", fb->offset, fb->size);
+        DRM_DEBUG("offset = %lu, size = %lu\n", fb->offset, fb->size);
 
         return 0;
 }
@@ -161,7 +161,7 @@ fail_alloc:
         mem->size = 0;
         mem->free = 0;
 
-        DRM_DEBUG("alloc %d, size = %d, offset = %d\n", pool, mem->size,
+        DRM_DEBUG("alloc %d, size = %ld, offset = %ld\n", pool, mem->size,
                   mem->offset);
 
         return retval;
@@ -215,7 +215,7 @@ static int sis_ioctl_agp_init(struct drm_device *dev, void *data,
         dev_priv->agp_offset = agp->offset;
         mutex_unlock(&dev->struct_mutex);
 
-        DRM_DEBUG("offset = %u, size = %u\n", agp->offset, agp->size);
+        DRM_DEBUG("offset = %lu, size = %lu\n", agp->offset, agp->size);
         return 0;
 }
 
@@ -321,14 +321,20 @@ void sis_reclaim_buffers_locked(struct drm_device *dev,
         struct sis_file_private *file_priv = file->driver_priv;
         struct sis_memblock *entry, *next;
 
+        if (!(file->minor->master && file->master->lock.hw_lock))
+                return;
+
+        drm_idlelock_take(&file->master->lock);
+
         mutex_lock(&dev->struct_mutex);
         if (list_empty(&file_priv->obj_list)) {
                 mutex_unlock(&dev->struct_mutex);
+                drm_idlelock_release(&file->master->lock);
+
                 return;
         }
 
-        if (dev->driver->dma_quiescent)
-                dev->driver->dma_quiescent(dev);
+        sis_idle(dev);
 
 
         list_for_each_entry_safe(entry, next, &file_priv->obj_list,
@@ -343,6 +349,9 @@ void sis_reclaim_buffers_locked(struct drm_device *dev,
                 kfree(entry);
         }
         mutex_unlock(&dev->struct_mutex);
+
+        drm_idlelock_release(&file->master->lock);
+
         return;
 }
 
diff --git a/drivers/gpu/drm/tdfx/tdfx_drv.c b/drivers/gpu/drm/tdfx/tdfx_drv.c
index 1613c78544c0..90f6b13acfac 100644
--- a/drivers/gpu/drm/tdfx/tdfx_drv.c
+++ b/drivers/gpu/drm/tdfx/tdfx_drv.c
@@ -54,7 +54,6 @@ static const struct file_operations tdfx_driver_fops = {
 
 static struct drm_driver driver = {
         .driver_features = DRIVER_USE_MTRR,
-        .reclaim_buffers = drm_core_reclaim_buffers,
         .fops = &tdfx_driver_fops,
         .name = DRIVER_NAME,
         .desc = DRIVER_DESC,
diff --git a/drivers/gpu/drm/udl/udl_encoder.c b/drivers/gpu/drm/udl/udl_encoder.c
index 56e75f0f1df5..0731ab2e6c06 100644
--- a/drivers/gpu/drm/udl/udl_encoder.c
+++ b/drivers/gpu/drm/udl/udl_encoder.c
@@ -27,7 +27,7 @@ static void udl_encoder_disable(struct drm_encoder *encoder)
 }
 
 static bool udl_mode_fixup(struct drm_encoder *encoder,
-                           struct drm_display_mode *mode,
+                           const struct drm_display_mode *mode,
                            struct drm_display_mode *adjusted_mode)
 {
         return true;
diff --git a/drivers/gpu/drm/udl/udl_modeset.c b/drivers/gpu/drm/udl/udl_modeset.c
index 0d7816789da1..f5dd89e891de 100644
--- a/drivers/gpu/drm/udl/udl_modeset.c
+++ b/drivers/gpu/drm/udl/udl_modeset.c
@@ -45,12 +45,25 @@ static char *udl_vidreg_unlock(char *buf)
  *  0x01 H and V sync off (screen blank but powered)
  *  0x07 DPMS powerdown (requires modeset to come back)
  */
-static char *udl_enable_hvsync(char *buf, bool enable)
+static char *udl_set_blank(char *buf, int dpms_mode)
 {
-        if (enable)
-                return udl_set_register(buf, 0x1F, 0x00);
-        else
-                return udl_set_register(buf, 0x1F, 0x07);
+        u8 reg;
+        switch (dpms_mode) {
+        case DRM_MODE_DPMS_OFF:
+                reg = 0x07;
+                break;
+        case DRM_MODE_DPMS_STANDBY:
+                reg = 0x05;
+                break;
+        case DRM_MODE_DPMS_SUSPEND:
+                reg = 0x01;
+                break;
+        case DRM_MODE_DPMS_ON:
+                reg = 0x00;
+                break;
+        }
+
+        return udl_set_register(buf, 0x1f, reg);
 }
 
 static char *udl_set_color_depth(char *buf, u8 selection)
@@ -199,6 +212,20 @@ static char *udl_set_vid_cmds(char *wrptr, struct drm_display_mode *mode)
         return wrptr;
 }
 
+static char *udl_dummy_render(char *wrptr)
+{
+        *wrptr++ = 0xAF;
+        *wrptr++ = 0x6A; /* copy */
+        *wrptr++ = 0x00; /* from addr */
+        *wrptr++ = 0x00;
+        *wrptr++ = 0x00;
+        *wrptr++ = 0x01; /* one pixel */
+        *wrptr++ = 0x00; /* to address */
+        *wrptr++ = 0x00;
+        *wrptr++ = 0x00;
+        return wrptr;
+}
+
 static int udl_crtc_write_mode_to_hw(struct drm_crtc *crtc)
 {
         struct drm_device *dev = crtc->dev;
@@ -235,9 +262,10 @@ static void udl_crtc_dpms(struct drm_crtc *crtc, int mode)
 
                 buf = (char *)urb->transfer_buffer;
                 buf = udl_vidreg_lock(buf);
-                buf = udl_enable_hvsync(buf, false);
+                buf = udl_set_blank(buf, mode);
                 buf = udl_vidreg_unlock(buf);
 
+                buf = udl_dummy_render(buf);
                 retval = udl_submit_urb(dev, urb, buf - (char *)
                                         urb->transfer_buffer);
         } else {
@@ -251,7 +279,7 @@ static void udl_crtc_dpms(struct drm_crtc *crtc, int mode)
 }
 
 static bool udl_crtc_mode_fixup(struct drm_crtc *crtc,
-                                  struct drm_display_mode *mode,
+                                  const struct drm_display_mode *mode,
                                   struct drm_display_mode *adjusted_mode)
 
 {
@@ -306,9 +334,11 @@ static int udl_crtc_mode_set(struct drm_crtc *crtc,
         wrptr = udl_set_base8bpp(wrptr, 2 * mode->vdisplay * mode->hdisplay);
 
         wrptr = udl_set_vid_cmds(wrptr, adjusted_mode);
-        wrptr = udl_enable_hvsync(wrptr, true);
+        wrptr = udl_set_blank(wrptr, DRM_MODE_DPMS_ON);
         wrptr = udl_vidreg_unlock(wrptr);
 
+        wrptr = udl_dummy_render(wrptr);
+
         ufb->active_16 = true;
         if (old_fb) {
                 struct udl_framebuffer *uold_fb = to_udl_fb(old_fb);
diff --git a/drivers/gpu/drm/via/via_drv.c b/drivers/gpu/drm/via/via_drv.c
index 02661f35f7a0..e927b4c052f5 100644
--- a/drivers/gpu/drm/via/via_drv.c
+++ b/drivers/gpu/drm/via/via_drv.c
@@ -75,6 +75,7 @@ static struct drm_driver driver = {
         .load = via_driver_load,
         .unload = via_driver_unload,
         .open = via_driver_open,
+        .preclose = via_reclaim_buffers_locked,
         .postclose = via_driver_postclose,
         .context_dtor = via_final_context,
         .get_vblank_counter = via_get_vblank_counter,
@@ -85,9 +86,6 @@ static struct drm_driver driver = {
         .irq_uninstall = via_driver_irq_uninstall,
         .irq_handler = via_driver_irq_handler,
         .dma_quiescent = via_driver_dma_quiescent,
-        .reclaim_buffers = drm_core_reclaim_buffers,
-        .reclaim_buffers_locked = NULL,
-        .reclaim_buffers_idlelocked = via_reclaim_buffers_locked,
         .lastclose = via_lastclose,
         .ioctls = via_ioctls,
         .fops = &via_driver_fops,
diff --git a/drivers/gpu/drm/via/via_mm.c b/drivers/gpu/drm/via/via_mm.c
index a3574d09a07d..acfcb358e7b7 100644
--- a/drivers/gpu/drm/via/via_mm.c
+++ b/drivers/gpu/drm/via/via_mm.c
@@ -215,14 +215,20 @@ void via_reclaim_buffers_locked(struct drm_device *dev,
         struct via_file_private *file_priv = file->driver_priv;
         struct via_memblock *entry, *next;
 
+        if (!(file->minor->master && file->master->lock.hw_lock))
+                return;
+
+        drm_idlelock_take(&file->master->lock);
+
         mutex_lock(&dev->struct_mutex);
         if (list_empty(&file_priv->obj_list)) {
                 mutex_unlock(&dev->struct_mutex);
+                drm_idlelock_release(&file->master->lock);
+
                 return;
         }
 
-        if (dev->driver->dma_quiescent)
-                dev->driver->dma_quiescent(dev);
+        via_driver_dma_quiescent(dev);
 
         list_for_each_entry_safe(entry, next, &file_priv->obj_list,
                                  owner_list) {
@@ -231,5 +237,8 @@ void via_reclaim_buffers_locked(struct drm_device *dev,
                 kfree(entry);
         }
         mutex_unlock(&dev->struct_mutex);
+
+        drm_idlelock_release(&file->master->lock);
+
         return;
 }
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
index ee24d216aa85..4d9edead01ac 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
@@ -769,10 +769,7 @@ static int vmw_driver_open(struct drm_device *dev, struct drm_file *file_priv)
                 goto out_no_tfile;
 
         file_priv->driver_priv = vmw_fp;
-
-        if (unlikely(dev_priv->bdev.dev_mapping == NULL))
-                dev_priv->bdev.dev_mapping =
-                        file_priv->filp->f_path.dentry->d_inode->i_mapping;
+        dev_priv->bdev.dev_mapping = dev->dev_mapping;
 
         return 0;
 
@@ -1147,7 +1144,6 @@ static struct drm_driver driver = {
         .get_vblank_counter = vmw_get_vblank_counter,
         .enable_vblank = vmw_enable_vblank,
         .disable_vblank = vmw_disable_vblank,
-        .reclaim_buffers_locked = NULL,
         .ioctls = vmw_ioctls,
         .num_ioctls = DRM_ARRAY_SIZE(vmw_ioctls),
         .dma_quiescent = NULL,  /*vmw_dma_quiescent, */
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index 8c2754f835ef..240ff3125040 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -2485,9 +2485,10 @@ static sector_t sync_request(struct mddev *mddev, sector_t sector_nr, int *skipp
          */
         if (test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery)) {
                 atomic_set(&r1_bio->remaining, read_targets);
-                for (i = 0; i < conf->raid_disks * 2; i++) {
+                for (i = 0; i < conf->raid_disks * 2 && read_targets; i++) {
                         bio = r1_bio->bios[i];
                         if (bio->bi_end_io == end_sync_read) {
+                                read_targets--;
                                 md_sync_acct(bio->bi_bdev, nr_sectors);
                                 generic_make_request(bio);
                         }
diff --git a/drivers/staging/omapdrm/omap_crtc.c b/drivers/staging/omapdrm/omap_crtc.c
index 8b864afb40b6..62e0022561bc 100644
--- a/drivers/staging/omapdrm/omap_crtc.c
+++ b/drivers/staging/omapdrm/omap_crtc.c
@@ -60,7 +60,7 @@ static void omap_crtc_dpms(struct drm_crtc *crtc, int mode)
 }
 
 static bool omap_crtc_mode_fixup(struct drm_crtc *crtc,
-                struct drm_display_mode *mode,
+                const struct drm_display_mode *mode,
                 struct drm_display_mode *adjusted_mode)
 {
         return true;
diff --git a/drivers/staging/omapdrm/omap_encoder.c b/drivers/staging/omapdrm/omap_encoder.c
index 06c52cb62d2f..31c735d39217 100644
--- a/drivers/staging/omapdrm/omap_encoder.c
+++ b/drivers/staging/omapdrm/omap_encoder.c
@@ -48,7 +48,7 @@ static void omap_encoder_dpms(struct drm_encoder *encoder, int mode)
 }
 
 static bool omap_encoder_mode_fixup(struct drm_encoder *encoder,
-                                  struct drm_display_mode *mode,
+                                  const struct drm_display_mode *mode,
                                   struct drm_display_mode *adjusted_mode)
 {
         struct omap_encoder *omap_encoder = to_omap_encoder(encoder);
diff --git a/include/drm/drmP.h b/include/drm/drmP.h
index 31ad880ca2ef..d6b67bb9075f 100644
--- a/include/drm/drmP.h
+++ b/include/drm/drmP.h
@@ -348,7 +348,6 @@ struct drm_buf {
         struct drm_buf *next;          /**< Kernel-only: used for free list */
         __volatile__ int waiting;      /**< On kernel DMA queue */
         __volatile__ int pending;      /**< On hardware DMA queue */
-        wait_queue_head_t dma_wait;    /**< Processes waiting */
         struct drm_file *file_priv;    /**< Private of holding file descr */
         int context;                   /**< Kernel queue for this buffer */
         int while_locked;              /**< Dispatch this buffer while locked */
@@ -876,12 +875,6 @@ struct drm_driver {
         void (*irq_preinstall) (struct drm_device *dev);
         int (*irq_postinstall) (struct drm_device *dev);
         void (*irq_uninstall) (struct drm_device *dev);
-        void (*reclaim_buffers) (struct drm_device *dev,
-                                 struct drm_file * file_priv);
-        void (*reclaim_buffers_locked) (struct drm_device *dev,
-                                        struct drm_file *file_priv);
-        void (*reclaim_buffers_idlelocked) (struct drm_device *dev,
-                                            struct drm_file *file_priv);
         void (*set_version) (struct drm_device *dev,
                              struct drm_set_version *sv);
 
@@ -1108,12 +1101,8 @@ struct drm_device {
 
         /*@} */
 
-        /** \name DMA queues (contexts) */
+        /** \name DMA support */
         /*@{ */
-        int queue_count;                /**< Number of active DMA queues */
-        int queue_reserved;               /**< Number of reserved DMA queues */
-        int queue_slots;                /**< Actual length of queuelist */
-        struct drm_queue **queuelist;   /**< Vector of pointers to DMA queues */
         struct drm_device_dma *dma;             /**< Optional pointer for DMA support */
         /*@} */
 
@@ -1540,7 +1529,6 @@ extern int drm_debugfs_cleanup(struct drm_minor *minor);
                                 /* Info file support */
 extern int drm_name_info(struct seq_file *m, void *data);
 extern int drm_vm_info(struct seq_file *m, void *data);
-extern int drm_queues_info(struct seq_file *m, void *data);
 extern int drm_bufs_info(struct seq_file *m, void *data);
 extern int drm_vblank_info(struct seq_file *m, void *data);
 extern int drm_clients_info(struct seq_file *m, void* data);
@@ -1761,6 +1749,11 @@ extern int drm_get_pci_dev(struct pci_dev *pdev,
                            const struct pci_device_id *ent,
                            struct drm_driver *driver);
 
+#define DRM_PCIE_SPEED_25 1
+#define DRM_PCIE_SPEED_50 2
+#define DRM_PCIE_SPEED_80 4
+
+extern int drm_pcie_get_speed_cap_mask(struct drm_device *dev, u32 *speed_mask);
 
 /* platform section */
 extern int drm_platform_init(struct drm_driver *driver, struct platform_device *platform_device);
diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h
index bac55c215113..a1a0386e0160 100644
--- a/include/drm/drm_crtc.h
+++ b/include/drm/drm_crtc.h
@@ -676,8 +676,6 @@ struct drm_plane {
  * This is used to set modes.
  */
 struct drm_mode_set {
-        struct list_head head;
-
         struct drm_framebuffer *fb;
         struct drm_crtc *crtc;
         struct drm_display_mode *mode;
diff --git a/include/drm/drm_crtc_helper.h b/include/drm/drm_crtc_helper.h
index 7988e55c98d0..e01cc80c9c30 100644
--- a/include/drm/drm_crtc_helper.h
+++ b/include/drm/drm_crtc_helper.h
@@ -62,7 +62,7 @@ struct drm_crtc_helper_funcs {
 
         /* Provider can fixup or change mode timings before modeset occurs */
         bool (*mode_fixup)(struct drm_crtc *crtc,
-                           struct drm_display_mode *mode,
+                           const struct drm_display_mode *mode,
                            struct drm_display_mode *adjusted_mode);
         /* Actually set the mode */
         int (*mode_set)(struct drm_crtc *crtc, struct drm_display_mode *mode,
@@ -96,7 +96,7 @@ struct drm_encoder_helper_funcs {
         void (*restore)(struct drm_encoder *encoder);
 
         bool (*mode_fixup)(struct drm_encoder *encoder,
-                           struct drm_display_mode *mode,
+                           const struct drm_display_mode *mode,
                            struct drm_display_mode *adjusted_mode);
         void (*prepare)(struct drm_encoder *encoder);
         void (*commit)(struct drm_encoder *encoder);
diff --git a/include/drm/drm_encoder_slave.h b/include/drm/drm_encoder_slave.h
index 2f65633d28a7..7dc385233805 100644
--- a/include/drm/drm_encoder_slave.h
+++ b/include/drm/drm_encoder_slave.h
@@ -54,7 +54,7 @@ struct drm_encoder_slave_funcs {
         void (*save)(struct drm_encoder *encoder);
         void (*restore)(struct drm_encoder *encoder);
         bool (*mode_fixup)(struct drm_encoder *encoder,
-                           struct drm_display_mode *mode,
+                           const struct drm_display_mode *mode,
                            struct drm_display_mode *adjusted_mode);
         int (*mode_valid)(struct drm_encoder *encoder,
                           struct drm_display_mode *mode);
diff --git a/include/drm/drm_mm.h b/include/drm/drm_mm.h
index 564b14aa7e16..06d7f798a08c 100644
--- a/include/drm/drm_mm.h
+++ b/include/drm/drm_mm.h
@@ -50,6 +50,7 @@ struct drm_mm_node {
         unsigned scanned_next_free : 1;
         unsigned scanned_preceeds_hole : 1;
         unsigned allocated : 1;
+        unsigned long color;
         unsigned long start;
         unsigned long size;
         struct drm_mm *mm;
@@ -66,6 +67,7 @@ struct drm_mm {
         spinlock_t unused_lock;
         unsigned int scan_check_range : 1;
         unsigned scan_alignment;
+        unsigned long scan_color;
         unsigned long scan_size;
         unsigned long scan_hit_start;
         unsigned scan_hit_size;
@@ -73,6 +75,9 @@ struct drm_mm {
         unsigned long scan_start;
         unsigned long scan_end;
         struct drm_mm_node *prev_scanned_node;
+
+        void (*color_adjust)(struct drm_mm_node *node, unsigned long color,
+                             unsigned long *start, unsigned long *end);
 };
 
 static inline bool drm_mm_node_allocated(struct drm_mm_node *node)
@@ -100,11 +105,13 @@ static inline bool drm_mm_initialized(struct drm_mm *mm)
 extern struct drm_mm_node *drm_mm_get_block_generic(struct drm_mm_node *node,
                                                     unsigned long size,
                                                     unsigned alignment,
+                                                    unsigned long color,
                                                     int atomic);
 extern struct drm_mm_node *drm_mm_get_block_range_generic(
                                                 struct drm_mm_node *node,
                                                 unsigned long size,
                                                 unsigned alignment,
+                                                unsigned long color,
                                                 unsigned long start,
                                                 unsigned long end,
                                                 int atomic);
@@ -112,13 +119,13 @@ static inline struct drm_mm_node *drm_mm_get_block(struct drm_mm_node *parent,
                                                    unsigned long size,
                                                    unsigned alignment)
 {
-        return drm_mm_get_block_generic(parent, size, alignment, 0);
+        return drm_mm_get_block_generic(parent, size, alignment, 0, 0);
 }
 static inline struct drm_mm_node *drm_mm_get_block_atomic(struct drm_mm_node *parent,
                                                           unsigned long size,
                                                           unsigned alignment)
 {
-        return drm_mm_get_block_generic(parent, size, alignment, 1);
+        return drm_mm_get_block_generic(parent, size, alignment, 0, 1);
 }
 static inline struct drm_mm_node *drm_mm_get_block_range(
                                                 struct drm_mm_node *parent,
@@ -127,8 +134,19 @@ static inline struct drm_mm_node *drm_mm_get_block_range(
                                                 unsigned long start,
                                                 unsigned long end)
 {
-        return drm_mm_get_block_range_generic(parent, size, alignment,
-                                                start, end, 0);
+        return drm_mm_get_block_range_generic(parent, size, alignment, 0,
+                                              start, end, 0);
+}
+static inline struct drm_mm_node *drm_mm_get_color_block_range(
+                                                struct drm_mm_node *parent,
+                                                unsigned long size,
+                                                unsigned alignment,
+                                                unsigned long color,
+                                                unsigned long start,
+                                                unsigned long end)
+{
+        return drm_mm_get_block_range_generic(parent, size, alignment, color,
+                                              start, end, 0);
 }
 static inline struct drm_mm_node *drm_mm_get_block_atomic_range(
                                                 struct drm_mm_node *parent,
@@ -137,7 +155,7 @@ static inline struct drm_mm_node *drm_mm_get_block_atomic_range(
                                                 unsigned long start,
                                                 unsigned long end)
 {
-        return drm_mm_get_block_range_generic(parent, size, alignment,
+        return drm_mm_get_block_range_generic(parent, size, alignment, 0,
                                                 start, end, 1);
 }
 extern int drm_mm_insert_node(struct drm_mm *mm, struct drm_mm_node *node,
@@ -149,18 +167,59 @@ extern int drm_mm_insert_node_in_range(struct drm_mm *mm,
 extern void drm_mm_put_block(struct drm_mm_node *cur);
 extern void drm_mm_remove_node(struct drm_mm_node *node);
 extern void drm_mm_replace_node(struct drm_mm_node *old, struct drm_mm_node *new);
-extern struct drm_mm_node *drm_mm_search_free(const struct drm_mm *mm,
-                                              unsigned long size,
-                                              unsigned alignment,
-                                              int best_match);
-extern struct drm_mm_node *drm_mm_search_free_in_range(
+extern struct drm_mm_node *drm_mm_search_free_generic(const struct drm_mm *mm,
+                                                      unsigned long size,
+                                                      unsigned alignment,
+                                                      unsigned long color,
+                                                      bool best_match);
+extern struct drm_mm_node *drm_mm_search_free_in_range_generic(
+                                                const struct drm_mm *mm,
+                                                unsigned long size,
+                                                unsigned alignment,
+                                                unsigned long color,
+                                                unsigned long start,
+                                                unsigned long end,
+                                                bool best_match);
+static inline struct drm_mm_node *drm_mm_search_free(const struct drm_mm *mm,
+                                                     unsigned long size,
+                                                     unsigned alignment,
+                                                     bool best_match)
+{
+        return drm_mm_search_free_generic(mm,size, alignment, 0, best_match);
+}
+static inline  struct drm_mm_node *drm_mm_search_free_in_range(
                                                 const struct drm_mm *mm,
                                                 unsigned long size,
                                                 unsigned alignment,
                                                 unsigned long start,
                                                 unsigned long end,
-                                                int best_match);
-extern int drm_mm_init(struct drm_mm *mm, unsigned long start,
+                                                bool best_match)
+{
+        return drm_mm_search_free_in_range_generic(mm, size, alignment, 0,
+                                                   start, end, best_match);
+}
+static inline struct drm_mm_node *drm_mm_search_free_color(const struct drm_mm *mm,
+                                                           unsigned long size,
+                                                           unsigned alignment,
+                                                           unsigned long color,
+                                                           bool best_match)
+{
+        return drm_mm_search_free_generic(mm,size, alignment, color, best_match);
+}
+static inline  struct drm_mm_node *drm_mm_search_free_in_range_color(
+                                                const struct drm_mm *mm,
+                                                unsigned long size,
+                                                unsigned alignment,
+                                                unsigned long color,
+                                                unsigned long start,
+                                                unsigned long end,
+                                                bool best_match)
+{
+        return drm_mm_search_free_in_range_generic(mm, size, alignment, color,
+                                                   start, end, best_match);
+}
+extern int drm_mm_init(struct drm_mm *mm,
+                       unsigned long start,
                        unsigned long size);
 extern void drm_mm_takedown(struct drm_mm *mm);
 extern int drm_mm_clean(struct drm_mm *mm);
@@ -171,10 +230,14 @@ static inline struct drm_mm *drm_get_mm(struct drm_mm_node *block)
         return block->mm;
 }
 
-void drm_mm_init_scan(struct drm_mm *mm, unsigned long size,
-                      unsigned alignment);
-void drm_mm_init_scan_with_range(struct drm_mm *mm, unsigned long size,
+void drm_mm_init_scan(struct drm_mm *mm,
+                      unsigned long size,
+                      unsigned alignment,
+                      unsigned long color);
+void drm_mm_init_scan_with_range(struct drm_mm *mm,
+                                 unsigned long size,
                                  unsigned alignment,
+                                 unsigned long color,
                                  unsigned long start,
                                  unsigned long end);
 int drm_mm_scan_add_block(struct drm_mm_node *node);
diff --git a/include/drm/drm_pciids.h b/include/drm/drm_pciids.h
index a7aec391b7b7..7ff5c99b1638 100644
--- a/include/drm/drm_pciids.h
+++ b/include/drm/drm_pciids.h
@@ -686,14 +686,6 @@
         {0x8086, 0x1132, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \
         {0, 0, 0}
 
-#define i830_PCI_IDS \
-        {0x8086, 0x3577, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \
-        {0x8086, 0x2562, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \
-        {0x8086, 0x3582, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \
-        {0x8086, 0x2572, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \
-        {0x8086, 0x358e, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \
-        {0, 0, 0}
-
 #define gamma_PCI_IDS \
         {0x3d3d, 0x0008, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0}, \
         {0, 0, 0}
@@ -726,37 +718,3 @@
 
 #define ffb_PCI_IDS \
         {0, 0, 0}
-
-#define i915_PCI_IDS \
-        {0x8086, 0x3577, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0x2562, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0x3582, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0x2572, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0x2582, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0x258a, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0x2592, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0x2772, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0x27a2, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0x27ae, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0x2972, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0x2982, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0x2992, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0x29a2, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0x29b2, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0x29c2, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0x29d2, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0x2a02, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0x2a12, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0x2a42, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0x2e02, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0x2e12, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0x2e22, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0x2e32, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0x2e42, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0xa001, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0xa011, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0x35e8, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0x0042, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0x0046, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0x8086, 0x0102, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8, 0xffff00, 0}, \
-        {0, 0, 0}
diff --git a/include/drm/i915_drm.h b/include/drm/i915_drm.h
index f3f82242bf1d..8cc70837f929 100644
--- a/include/drm/i915_drm.h
+++ b/include/drm/i915_drm.h
@@ -200,6 +200,9 @@ typedef struct _drm_i915_sarea {
 #define DRM_I915_GEM_EXECBUFFER2        0x29
 #define DRM_I915_GET_SPRITE_COLORKEY    0x2a
 #define DRM_I915_SET_SPRITE_COLORKEY    0x2b
+#define DRM_I915_GEM_WAIT       0x2c
+#define DRM_I915_GEM_CONTEXT_CREATE     0x2d
+#define DRM_I915_GEM_CONTEXT_DESTROY    0x2e
 
 #define DRM_IOCTL_I915_INIT             DRM_IOW( DRM_COMMAND_BASE + DRM_I915_INIT, drm_i915_init_t)
 #define DRM_IOCTL_I915_FLUSH            DRM_IO ( DRM_COMMAND_BASE + DRM_I915_FLUSH)
@@ -243,6 +246,9 @@ typedef struct _drm_i915_sarea {
 #define DRM_IOCTL_I915_OVERLAY_ATTRS    DRM_IOWR(DRM_COMMAND_BASE + DRM_I915_OVERLAY_ATTRS, struct drm_intel_overlay_attrs)
 #define DRM_IOCTL_I915_SET_SPRITE_COLORKEY DRM_IOWR(DRM_COMMAND_BASE + DRM_I915_SET_SPRITE_COLORKEY, struct drm_intel_sprite_colorkey)
 #define DRM_IOCTL_I915_GET_SPRITE_COLORKEY DRM_IOWR(DRM_COMMAND_BASE + DRM_I915_SET_SPRITE_COLORKEY, struct drm_intel_sprite_colorkey)
+#define DRM_IOCTL_I915_GEM_WAIT         DRM_IOWR(DRM_COMMAND_BASE + DRM_I915_GEM_WAIT, struct drm_i915_gem_wait)
+#define DRM_IOCTL_I915_GEM_CONTEXT_CREATE       DRM_IOWR (DRM_COMMAND_BASE + DRM_I915_GEM_CONTEXT_CREATE, struct drm_i915_gem_context_create)
+#define DRM_IOCTL_I915_GEM_CONTEXT_DESTROY      DRM_IOW (DRM_COMMAND_BASE + DRM_I915_GEM_CONTEXT_DESTROY, struct drm_i915_gem_context_destroy)
 
 /* Allow drivers to submit batchbuffers directly to hardware, relying
  * on the security mechanisms provided by hardware.
@@ -298,6 +304,7 @@ typedef struct drm_i915_irq_wait {
 #define I915_PARAM_HAS_GEN7_SOL_RESET    16
 #define I915_PARAM_HAS_LLC               17
 #define I915_PARAM_HAS_ALIASING_PPGTT    18
+#define I915_PARAM_HAS_WAIT_TIMEOUT      19
 
 typedef struct drm_i915_getparam {
         int param;
@@ -656,13 +663,19 @@ struct drm_i915_gem_execbuffer2 {
 #define I915_EXEC_CONSTANTS_ABSOLUTE    (1<<6)
 #define I915_EXEC_CONSTANTS_REL_SURFACE (2<<6) /* gen4/5 only */
         __u64 flags;
-        __u64 rsvd1;
+        __u64 rsvd1; /* now used for context info */
         __u64 rsvd2;
 };
 
 /** Resets the SO write offset registers for transform feedback on gen7. */
 #define I915_EXEC_GEN7_SOL_RESET        (1<<8)
 
+#define I915_EXEC_CONTEXT_ID_MASK       (0xffffffff)
+#define i915_execbuffer2_set_context_id(eb2, context) \
+        (eb2).rsvd1 = context & I915_EXEC_CONTEXT_ID_MASK
+#define i915_execbuffer2_get_context_id(eb2) \
+        ((eb2).rsvd1 & I915_EXEC_CONTEXT_ID_MASK)
+
 struct drm_i915_gem_pin {
         /** Handle of the buffer to be pinned. */
         __u32 handle;
@@ -886,4 +899,23 @@ struct drm_intel_sprite_colorkey {
         __u32 flags;
 };
 
+struct drm_i915_gem_wait {
+        /** Handle of BO we shall wait on */
+        __u32 bo_handle;
+        __u32 flags;
+        /** Number of nanoseconds to wait, Returns time remaining. */
+        __s64 timeout_ns;
+};
+
+struct drm_i915_gem_context_create {
+        /*  output: id of new context*/
+        __u32 ctx_id;
+        __u32 pad;
+};
+
+struct drm_i915_gem_context_destroy {
+        __u32 ctx_id;
+        __u32 pad;
+};
+
 #endif                          /* _I915_DRM_H_ */
diff --git a/include/drm/intel-gtt.h b/include/drm/intel-gtt.h
index 923afb5dcf0c..8e29d551bb3c 100644
--- a/include/drm/intel-gtt.h
+++ b/include/drm/intel-gtt.h
@@ -19,8 +19,16 @@ const struct intel_gtt {
         dma_addr_t scratch_page_dma;
         /* for ppgtt PDE access */
         u32 __iomem *gtt;
+        /* needed for ioremap in drm/i915 */
+        phys_addr_t gma_bus_addr;
 } *intel_gtt_get(void);
 
+int intel_gmch_probe(struct pci_dev *bridge_pdev, struct pci_dev *gpu_pdev,
+                     struct agp_bridge_data *bridge);
+void intel_gmch_remove(void);
+
+bool intel_enable_gtt(void);
+
 void intel_gtt_chipset_flush(void);
 void intel_gtt_unmap_memory(struct scatterlist *sg_list, int num_sg);
 void intel_gtt_clear_range(unsigned int first_entry, unsigned int num_entries);
diff --git a/include/drm/sis_drm.h b/include/drm/sis_drm.h
index 035b804dda6d..df3763222d73 100644
--- a/include/drm/sis_drm.h
+++ b/include/drm/sis_drm.h
@@ -51,17 +51,17 @@
 
 typedef struct {
         int context;
-        unsigned int offset;
-        unsigned int size;
+        unsigned long offset;
+        unsigned long size;
         unsigned long free;
 } drm_sis_mem_t;
 
 typedef struct {
-        unsigned int offset, size;
+        unsigned long offset, size;
 } drm_sis_agp_t;
 
 typedef struct {
-        unsigned int offset, size;
+        unsigned long offset, size;
 } drm_sis_fb_t;
 
 struct sis_file_private {
diff --git a/include/drm/ttm/ttm_bo_driver.h b/include/drm/ttm/ttm_bo_driver.h
index a05f1b55714d..084e8989a6e1 100644
--- a/include/drm/ttm/ttm_bo_driver.h
+++ b/include/drm/ttm/ttm_bo_driver.h
@@ -39,8 +39,6 @@
 #include "linux/fs.h"
 #include "linux/spinlock.h"
 
-struct ttm_backend;
-
 struct ttm_backend_func {
         /**
          * struct ttm_backend_func member bind
@@ -119,7 +117,6 @@ struct ttm_tt {
         unsigned long num_pages;
         struct sg_table *sg; /* for SG objects via dma-buf */
         struct ttm_bo_global *glob;
-        struct ttm_backend *be;
         struct file *swap_storage;
         enum ttm_caching_state caching_state;
         enum {
diff --git a/include/linux/hrtimer.h b/include/linux/hrtimer.h
index fd0dc30c9f15..cc07d2777bbe 100644
--- a/include/linux/hrtimer.h
+++ b/include/linux/hrtimer.h
@@ -165,6 +165,7 @@ enum  hrtimer_base_type {
  * @lock:               lock protecting the base and associated clock bases
  *                      and timers
  * @active_bases:       Bitfield to mark bases with active timers
+ * @clock_was_set:      Indicates that clock was set from irq context.
  * @expires_next:       absolute time of the next event which was scheduled
  *                      via clock_set_next_event()
  * @hres_active:        State of high resolution mode
@@ -177,7 +178,8 @@ enum  hrtimer_base_type {
  */
 struct hrtimer_cpu_base {
         raw_spinlock_t                  lock;
-        unsigned long                   active_bases;
+        unsigned int                    active_bases;
+        unsigned int                    clock_was_set;
 #ifdef CONFIG_HIGH_RES_TIMERS
         ktime_t                         expires_next;
         int                             hres_active;
@@ -286,6 +288,8 @@ extern void hrtimer_peek_ahead_timers(void);
 # define MONOTONIC_RES_NSEC     HIGH_RES_NSEC
 # define KTIME_MONOTONIC_RES    KTIME_HIGH_RES
 
+extern void clock_was_set_delayed(void);
+
 #else
 
 # define MONOTONIC_RES_NSEC     LOW_RES_NSEC
@@ -306,6 +310,9 @@ static inline int hrtimer_is_hres_active(struct hrtimer *timer)
 {
         return 0;
 }
+
+static inline void clock_was_set_delayed(void) { }
+
 #endif
 
 extern void clock_was_set(void);
@@ -320,6 +327,7 @@ extern ktime_t ktime_get(void);
 extern ktime_t ktime_get_real(void);
 extern ktime_t ktime_get_boottime(void);
 extern ktime_t ktime_get_monotonic_offset(void);
+extern ktime_t ktime_get_update_offsets(ktime_t *offs_real, ktime_t *offs_boot);
 
 DECLARE_PER_CPU(struct tick_device, tick_cpu_device);
 
diff --git a/include/linux/pci_regs.h b/include/linux/pci_regs.h
index 4b608f543412..7f04132eb02d 100644
--- a/include/linux/pci_regs.h
+++ b/include/linux/pci_regs.h
@@ -521,6 +521,11 @@
 #define  PCI_EXP_OBFF_MSGA_EN   0x2000  /* OBFF enable with Message type A */
 #define  PCI_EXP_OBFF_MSGB_EN   0x4000  /* OBFF enable with Message type B */
 #define  PCI_EXP_OBFF_WAKE_EN   0x6000  /* OBFF using WAKE# signaling */
+#define PCI_EXP_LNKCAP2         44      /* Link Capability 2 */
+#define  PCI_EXP_LNKCAP2_SLS_2_5GB 0x01 /* Current Link Speed 2.5GT/s */
+#define  PCI_EXP_LNKCAP2_SLS_5_0GB 0x02 /* Current Link Speed 5.0GT/s */
+#define  PCI_EXP_LNKCAP2_SLS_8_0GB 0x04 /* Current Link Speed 8.0GT/s */
+#define  PCI_EXP_LNKCAP2_CROSSLINK 0x100 /* Crosslink supported */
 #define PCI_EXP_LNKCTL2         48      /* Link Control 2 */
 #define PCI_EXP_SLTCTL2         56      /* Slot Control 2 */
 
diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index 26d1a47591f1..9cac722b169c 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -184,7 +184,6 @@ static inline int rcu_preempt_depth(void)
 /* Internal to kernel */
 extern void rcu_sched_qs(int cpu);
 extern void rcu_bh_qs(int cpu);
-extern void rcu_preempt_note_context_switch(void);
 extern void rcu_check_callbacks(int cpu, int user);
 struct notifier_block;
 extern void rcu_idle_enter(void);
diff --git a/include/linux/rcutiny.h b/include/linux/rcutiny.h
index 854dc4c5c271..4e56a9c69a35 100644
--- a/include/linux/rcutiny.h
+++ b/include/linux/rcutiny.h
@@ -87,6 +87,10 @@ static inline void kfree_call_rcu(struct rcu_head *head,
 
 #ifdef CONFIG_TINY_RCU
 
+static inline void rcu_preempt_note_context_switch(void)
+{
+}
+
 static inline int rcu_needs_cpu(int cpu, unsigned long *delta_jiffies)
 {
         *delta_jiffies = ULONG_MAX;
@@ -95,6 +99,7 @@ static inline int rcu_needs_cpu(int cpu, unsigned long *delta_jiffies)
 
 #else /* #ifdef CONFIG_TINY_RCU */
 
+void rcu_preempt_note_context_switch(void);
 int rcu_preempt_needs_cpu(void);
 
 static inline int rcu_needs_cpu(int cpu, unsigned long *delta_jiffies)
@@ -108,6 +113,7 @@ static inline int rcu_needs_cpu(int cpu, unsigned long *delta_jiffies)
 static inline void rcu_note_context_switch(int cpu)
 {
         rcu_sched_qs(cpu);
+        rcu_preempt_note_context_switch();
 }
 
 /*
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 4059c0f33f07..4a1f493e0fef 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1871,22 +1871,12 @@ static inline void rcu_copy_process(struct task_struct *p)
         INIT_LIST_HEAD(&p->rcu_node_entry);
 }
 
-static inline void rcu_switch_from(struct task_struct *prev)
-{
-        if (prev->rcu_read_lock_nesting != 0)
-                rcu_preempt_note_context_switch();
-}
-
 #else
 
 static inline void rcu_copy_process(struct task_struct *p)
 {
 }
 
-static inline void rcu_switch_from(struct task_struct *prev)
-{
-}
-
 #endif
 
 #ifdef CONFIG_SMP
@@ -1909,6 +1899,14 @@ static inline int set_cpus_allowed_ptr(struct task_struct *p,
 }
 #endif
 
+#ifdef CONFIG_NO_HZ
+void calc_load_enter_idle(void);
+void calc_load_exit_idle(void);
+#else
+static inline void calc_load_enter_idle(void) { }
+static inline void calc_load_exit_idle(void) { }
+#endif /* CONFIG_NO_HZ */
+
 #ifndef CONFIG_CPUMASK_OFFSTACK
 static inline int set_cpus_allowed(struct task_struct *p, cpumask_t new_mask)
 {
diff --git a/kernel/fork.c b/kernel/fork.c
index ab5211b9e622..f00e319d8376 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -304,12 +304,17 @@ static struct task_struct *dup_task_struct(struct task_struct *orig)
         }
 
         err = arch_dup_task_struct(tsk, orig);
-        if (err)
-                goto out;
 
+        /*
+         * We defer looking at err, because we will need this setup
+         * for the clean up path to work correctly.
+         */
         tsk->stack = ti;
-
         setup_thread_stack(tsk, orig);
+
+        if (err)
+                goto out;
+
         clear_user_return_notifier(tsk);
         clear_tsk_need_resched(tsk);
         stackend = end_of_stack(tsk);
diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
index ae34bf51682b..6db7a5ed52b5 100644
--- a/kernel/hrtimer.c
+++ b/kernel/hrtimer.c
@@ -657,6 +657,14 @@ static inline int hrtimer_enqueue_reprogram(struct hrtimer *timer,
         return 0;
 }
 
+static inline ktime_t hrtimer_update_base(struct hrtimer_cpu_base *base)
+{
+        ktime_t *offs_real = &base->clock_base[HRTIMER_BASE_REALTIME].offset;
+        ktime_t *offs_boot = &base->clock_base[HRTIMER_BASE_BOOTTIME].offset;
+
+        return ktime_get_update_offsets(offs_real, offs_boot);
+}
+
 /*
  * Retrigger next event is called after clock was set
  *
@@ -665,22 +673,12 @@ static inline int hrtimer_enqueue_reprogram(struct hrtimer *timer,
 static void retrigger_next_event(void *arg)
 {
         struct hrtimer_cpu_base *base = &__get_cpu_var(hrtimer_bases);
-        struct timespec realtime_offset, xtim, wtm, sleep;
 
         if (!hrtimer_hres_active())
                 return;
 
-        /* Optimized out for !HIGH_RES */
-        get_xtime_and_monotonic_and_sleep_offset(&xtim, &wtm, &sleep);
-        set_normalized_timespec(&realtime_offset, -wtm.tv_sec, -wtm.tv_nsec);
-
-        /* Adjust CLOCK_REALTIME offset */
         raw_spin_lock(&base->lock);
-        base->clock_base[HRTIMER_BASE_REALTIME].offset =
-                timespec_to_ktime(realtime_offset);
-        base->clock_base[HRTIMER_BASE_BOOTTIME].offset =
-                timespec_to_ktime(sleep);
-
+        hrtimer_update_base(base);
         hrtimer_force_reprogram(base, 0);
         raw_spin_unlock(&base->lock);
 }
@@ -710,13 +708,25 @@ static int hrtimer_switch_to_hres(void)
                 base->clock_base[i].resolution = KTIME_HIGH_RES;
 
         tick_setup_sched_timer();
-
         /* "Retrigger" the interrupt to get things going */
         retrigger_next_event(NULL);
         local_irq_restore(flags);
         return 1;
 }
 
+/*
+ * Called from timekeeping code to reprogramm the hrtimer interrupt
+ * device. If called from the timer interrupt context we defer it to
+ * softirq context.
+ */
+void clock_was_set_delayed(void)
+{
+        struct hrtimer_cpu_base *cpu_base = &__get_cpu_var(hrtimer_bases);
+
+        cpu_base->clock_was_set = 1;
+        __raise_softirq_irqoff(HRTIMER_SOFTIRQ);
+}
+
 #else
 
 static inline int hrtimer_hres_active(void) { return 0; }
@@ -1250,11 +1260,10 @@ void hrtimer_interrupt(struct clock_event_device *dev)
         cpu_base->nr_events++;
         dev->next_event.tv64 = KTIME_MAX;
 
-        entry_time = now = ktime_get();
+        raw_spin_lock(&cpu_base->lock);
+        entry_time = now = hrtimer_update_base(cpu_base);
 retry:
         expires_next.tv64 = KTIME_MAX;
-
-        raw_spin_lock(&cpu_base->lock);
         /*
          * We set expires_next to KTIME_MAX here with cpu_base->lock
          * held to prevent that a timer is enqueued in our queue via
@@ -1330,8 +1339,12 @@ retry:
          * We need to prevent that we loop forever in the hrtimer
          * interrupt routine. We give it 3 attempts to avoid
          * overreacting on some spurious event.
+         *
+         * Acquire base lock for updating the offsets and retrieving
+         * the current time.
          */
-        now = ktime_get();
+        raw_spin_lock(&cpu_base->lock);
+        now = hrtimer_update_base(cpu_base);
         cpu_base->nr_retries++;
         if (++retries < 3)
                 goto retry;
@@ -1343,6 +1356,7 @@ retry:
          */
         cpu_base->nr_hangs++;
         cpu_base->hang_detected = 1;
+        raw_spin_unlock(&cpu_base->lock);
         delta = ktime_sub(now, entry_time);
         if (delta.tv64 > cpu_base->max_hang_time.tv64)
                 cpu_base->max_hang_time = delta;
@@ -1395,6 +1409,13 @@ void hrtimer_peek_ahead_timers(void)
 
 static void run_hrtimer_softirq(struct softirq_action *h)
 {
+        struct hrtimer_cpu_base *cpu_base = &__get_cpu_var(hrtimer_bases);
+
+        if (cpu_base->clock_was_set) {
+                cpu_base->clock_was_set = 0;
+                clock_was_set();
+        }
+
         hrtimer_peek_ahead_timers();
 }
 
diff --git a/kernel/rcutree.c b/kernel/rcutree.c
index 38ecdda3f55f..4b97bba7396e 100644
--- a/kernel/rcutree.c
+++ b/kernel/rcutree.c
@@ -201,6 +201,7 @@ void rcu_note_context_switch(int cpu)
 {
         trace_rcu_utilization("Start context switch");
         rcu_sched_qs(cpu);
+        rcu_preempt_note_context_switch(cpu);
         trace_rcu_utilization("End context switch");
 }
 EXPORT_SYMBOL_GPL(rcu_note_context_switch);
diff --git a/kernel/rcutree.h b/kernel/rcutree.h
index ea056495783e..19b61ac1079f 100644
--- a/kernel/rcutree.h
+++ b/kernel/rcutree.h
@@ -444,6 +444,7 @@ DECLARE_PER_CPU(char, rcu_cpu_has_work);
 /* Forward declarations for rcutree_plugin.h */
 static void rcu_bootup_announce(void);
 long rcu_batches_completed(void);
+static void rcu_preempt_note_context_switch(int cpu);
 static int rcu_preempt_blocked_readers_cgp(struct rcu_node *rnp);
 #ifdef CONFIG_HOTPLUG_CPU
 static void rcu_report_unblock_qs_rnp(struct rcu_node *rnp,
diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h
index 5271a020887e..3e4899459f3d 100644
--- a/kernel/rcutree_plugin.h
+++ b/kernel/rcutree_plugin.h
@@ -153,7 +153,7 @@ static void rcu_preempt_qs(int cpu)
  *
  * Caller must disable preemption.
  */
-void rcu_preempt_note_context_switch(void)
+static void rcu_preempt_note_context_switch(int cpu)
 {
         struct task_struct *t = current;
         unsigned long flags;
@@ -164,7 +164,7 @@ void rcu_preempt_note_context_switch(void)
             (t->rcu_read_unlock_special & RCU_READ_UNLOCK_BLOCKED) == 0) {
 
                 /* Possibly blocking in an RCU read-side critical section. */
-                rdp = __this_cpu_ptr(rcu_preempt_state.rda);
+                rdp = per_cpu_ptr(rcu_preempt_state.rda, cpu);
                 rnp = rdp->mynode;
                 raw_spin_lock_irqsave(&rnp->lock, flags);
                 t->rcu_read_unlock_special |= RCU_READ_UNLOCK_BLOCKED;
@@ -228,7 +228,7 @@ void rcu_preempt_note_context_switch(void)
          * means that we continue to block the current grace period.
          */
         local_irq_save(flags);
-        rcu_preempt_qs(smp_processor_id());
+        rcu_preempt_qs(cpu);
         local_irq_restore(flags);
 }
 
@@ -1002,6 +1002,14 @@ void rcu_force_quiescent_state(void)
 EXPORT_SYMBOL_GPL(rcu_force_quiescent_state);
 
 /*
+ * Because preemptible RCU does not exist, we never have to check for
+ * CPUs being in quiescent states.
+ */
+static void rcu_preempt_note_context_switch(int cpu)
+{
+}
+
+/*
  * Because preemptible RCU does not exist, there are never any preempted
  * RCU readers.
  */
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index d5594a4268d4..468bdd44c1ba 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -2081,7 +2081,6 @@ context_switch(struct rq *rq, struct task_struct *prev,
 #endif
 
         /* Here we just switch the register state and the stack. */
-        rcu_switch_from(prev);
         switch_to(prev, next, prev);
 
         barrier();
@@ -2161,11 +2160,73 @@ unsigned long this_cpu_load(void)
 }
 
 
+/*
+ * Global load-average calculations
+ *
+ * We take a distributed and async approach to calculating the global load-avg
+ * in order to minimize overhead.
+ *
+ * The global load average is an exponentially decaying average of nr_running +
+ * nr_uninterruptible.
+ *
+ * Once every LOAD_FREQ:
+ *
+ *   nr_active = 0;
+ *   for_each_possible_cpu(cpu)
+ *      nr_active += cpu_of(cpu)->nr_running + cpu_of(cpu)->nr_uninterruptible;
+ *
+ *   avenrun[n] = avenrun[0] * exp_n + nr_active * (1 - exp_n)
+ *
+ * Due to a number of reasons the above turns in the mess below:
+ *
+ *  - for_each_possible_cpu() is prohibitively expensive on machines with
+ *    serious number of cpus, therefore we need to take a distributed approach
+ *    to calculating nr_active.
+ *
+ *        \Sum_i x_i(t) = \Sum_i x_i(t) - x_i(t_0) | x_i(t_0) := 0
+ *                      = \Sum_i { \Sum_j=1 x_i(t_j) - x_i(t_j-1) }
+ *
+ *    So assuming nr_active := 0 when we start out -- true per definition, we
+ *    can simply take per-cpu deltas and fold those into a global accumulate
+ *    to obtain the same result. See calc_load_fold_active().
+ *
+ *    Furthermore, in order to avoid synchronizing all per-cpu delta folding
+ *    across the machine, we assume 10 ticks is sufficient time for every
+ *    cpu to have completed this task.
+ *
+ *    This places an upper-bound on the IRQ-off latency of the machine. Then
+ *    again, being late doesn't loose the delta, just wrecks the sample.
+ *
+ *  - cpu_rq()->nr_uninterruptible isn't accurately tracked per-cpu because
+ *    this would add another cross-cpu cacheline miss and atomic operation
+ *    to the wakeup path. Instead we increment on whatever cpu the task ran
+ *    when it went into uninterruptible state and decrement on whatever cpu
+ *    did the wakeup. This means that only the sum of nr_uninterruptible over
+ *    all cpus yields the correct result.
+ *
+ *  This covers the NO_HZ=n code, for extra head-aches, see the comment below.
+ */
+
 /* Variables and functions for calc_load */
 static atomic_long_t calc_load_tasks;
 static unsigned long calc_load_update;
 unsigned long avenrun[3];
-EXPORT_SYMBOL(avenrun);
+EXPORT_SYMBOL(avenrun); /* should be removed */
+
+/**
+ * get_avenrun - get the load average array
+ * @loads:      pointer to dest load array
+ * @offset:     offset to add
+ * @shift:      shift count to shift the result left
+ *
+ * These values are estimates at best, so no need for locking.
+ */
+void get_avenrun(unsigned long *loads, unsigned long offset, int shift)
+{
+        loads[0] = (avenrun[0] + offset) << shift;
+        loads[1] = (avenrun[1] + offset) << shift;
+        loads[2] = (avenrun[2] + offset) << shift;
+}
 
 static long calc_load_fold_active(struct rq *this_rq)
 {
@@ -2182,6 +2243,9 @@ static long calc_load_fold_active(struct rq *this_rq)
         return delta;
 }
 
+/*
+ * a1 = a0 * e + a * (1 - e)
+ */
 static unsigned long
 calc_load(unsigned long load, unsigned long exp, unsigned long active)
 {
@@ -2193,30 +2257,118 @@ calc_load(unsigned long load, unsigned long exp, unsigned long active)
 
 #ifdef CONFIG_NO_HZ
 /*
- * For NO_HZ we delay the active fold to the next LOAD_FREQ update.
+ * Handle NO_HZ for the global load-average.
+ *
+ * Since the above described distributed algorithm to compute the global
+ * load-average relies on per-cpu sampling from the tick, it is affected by
+ * NO_HZ.
+ *
+ * The basic idea is to fold the nr_active delta into a global idle-delta upon
+ * entering NO_HZ state such that we can include this as an 'extra' cpu delta
+ * when we read the global state.
+ *
+ * Obviously reality has to ruin such a delightfully simple scheme:
+ *
+ *  - When we go NO_HZ idle during the window, we can negate our sample
+ *    contribution, causing under-accounting.
+ *
+ *    We avoid this by keeping two idle-delta counters and flipping them
+ *    when the window starts, thus separating old and new NO_HZ load.
+ *
+ *    The only trick is the slight shift in index flip for read vs write.
+ *
+ *        0s            5s            10s           15s
+ *          +10           +10           +10           +10
+ *        |-|-----------|-|-----------|-|-----------|-|
+ *    r:0 0 1           1 0           0 1           1 0
+ *    w:0 1 1           0 0           1 1           0 0
+ *
+ *    This ensures we'll fold the old idle contribution in this window while
+ *    accumlating the new one.
+ *
+ *  - When we wake up from NO_HZ idle during the window, we push up our
+ *    contribution, since we effectively move our sample point to a known
+ *    busy state.
+ *
+ *    This is solved by pushing the window forward, and thus skipping the
+ *    sample, for this cpu (effectively using the idle-delta for this cpu which
+ *    was in effect at the time the window opened). This also solves the issue
+ *    of having to deal with a cpu having been in NOHZ idle for multiple
+ *    LOAD_FREQ intervals.
  *
  * When making the ILB scale, we should try to pull this in as well.
  */
-static atomic_long_t calc_load_tasks_idle;
+static atomic_long_t calc_load_idle[2];
+static int calc_load_idx;
 
-void calc_load_account_idle(struct rq *this_rq)
+static inline int calc_load_write_idx(void)
 {
+        int idx = calc_load_idx;
+
+        /*
+         * See calc_global_nohz(), if we observe the new index, we also
+         * need to observe the new update time.
+         */
+        smp_rmb();
+
+        /*
+         * If the folding window started, make sure we start writing in the
+         * next idle-delta.
+         */
+        if (!time_before(jiffies, calc_load_update))
+                idx++;
+
+        return idx & 1;
+}
+
+static inline int calc_load_read_idx(void)
+{
+        return calc_load_idx & 1;
+}
+
+void calc_load_enter_idle(void)
+{
+        struct rq *this_rq = this_rq();
         long delta;
 
+        /*
+         * We're going into NOHZ mode, if there's any pending delta, fold it
+         * into the pending idle delta.
+         */
         delta = calc_load_fold_active(this_rq);
-        if (delta)
-                atomic_long_add(delta, &calc_load_tasks_idle);
+        if (delta) {
+                int idx = calc_load_write_idx();
+                atomic_long_add(delta, &calc_load_idle[idx]);
+        }
 }
 
-static long calc_load_fold_idle(void)
+void calc_load_exit_idle(void)
 {
-        long delta = 0;
+        struct rq *this_rq = this_rq();
+
+        /*
+         * If we're still before the sample window, we're done.
+         */
+        if (time_before(jiffies, this_rq->calc_load_update))
+                return;
 
         /*
-         * Its got a race, we don't care...
+         * We woke inside or after the sample window, this means we're already
+         * accounted through the nohz accounting, so skip the entire deal and
+         * sync up for the next window.
          */
-        if (atomic_long_read(&calc_load_tasks_idle))
-                delta = atomic_long_xchg(&calc_load_tasks_idle, 0);
+        this_rq->calc_load_update = calc_load_update;
+        if (time_before(jiffies, this_rq->calc_load_update + 10))
+                this_rq->calc_load_update += LOAD_FREQ;
+}
+
+static long calc_load_fold_idle(void)
+{
+        int idx = calc_load_read_idx();
+        long delta = 0;
+
+        if (atomic_long_read(&calc_load_idle[idx]))
+                delta = atomic_long_xchg(&calc_load_idle[idx], 0);
 
         return delta;
 }
@@ -2302,66 +2454,39 @@ static void calc_global_nohz(void)
 {
         long delta, active, n;
 
-        /*
-         * If we crossed a calc_load_update boundary, make sure to fold
-         * any pending idle changes, the respective CPUs might have
-         * missed the tick driven calc_load_account_active() update
-         * due to NO_HZ.
-         */
-        delta = calc_load_fold_idle();
-        if (delta)
-                atomic_long_add(delta, &calc_load_tasks);
-
-        /*
-         * It could be the one fold was all it took, we done!
-         */
-        if (time_before(jiffies, calc_load_update + 10))
-                return;
-
-        /*
-         * Catch-up, fold however many we are behind still
-         */
-        delta = jiffies - calc_load_update - 10;
-        n = 1 + (delta / LOAD_FREQ);
+        if (!time_before(jiffies, calc_load_update + 10)) {
+                /*
+                 * Catch-up, fold however many we are behind still
+                 */
+                delta = jiffies - calc_load_update - 10;
+                n = 1 + (delta / LOAD_FREQ);
 
-        active = atomic_long_read(&calc_load_tasks);
-        active = active > 0 ? active * FIXED_1 : 0;
+                active = atomic_long_read(&calc_load_tasks);
+                active = active > 0 ? active * FIXED_1 : 0;
 
-        avenrun[0] = calc_load_n(avenrun[0], EXP_1, active, n);
-        avenrun[1] = calc_load_n(avenrun[1], EXP_5, active, n);
-        avenrun[2] = calc_load_n(avenrun[2], EXP_15, active, n);
+                avenrun[0] = calc_load_n(avenrun[0], EXP_1, active, n);
+                avenrun[1] = calc_load_n(avenrun[1], EXP_5, active, n);
+                avenrun[2] = calc_load_n(avenrun[2], EXP_15, active, n);
 
-        calc_load_update += n * LOAD_FREQ;
-}
-#else
-void calc_load_account_idle(struct rq *this_rq)
-{
-}
+                calc_load_update += n * LOAD_FREQ;
+        }
 
-static inline long calc_load_fold_idle(void)
-{
-        return 0;
+        /*
+         * Flip the idle index...
+         *
+         * Make sure we first write the new time then flip the index, so that
+         * calc_load_write_idx() will see the new time when it reads the new
+         * index, this avoids a double flip messing things up.
+         */
+        smp_wmb();
+        calc_load_idx++;
 }
+#else /* !CONFIG_NO_HZ */
 
-static void calc_global_nohz(void)
-{
-}
-#endif
+static inline long calc_load_fold_idle(void) { return 0; }
+static inline void calc_global_nohz(void) { }
 
-/**
- * get_avenrun - get the load average array
- * @loads:      pointer to dest load array
- * @offset:     offset to add
- * @shift:      shift count to shift the result left
- *
- * These values are estimates at best, so no need for locking.
- */
-void get_avenrun(unsigned long *loads, unsigned long offset, int shift)
-{
-        loads[0] = (avenrun[0] + offset) << shift;
-        loads[1] = (avenrun[1] + offset) << shift;
-        loads[2] = (avenrun[2] + offset) << shift;
-}
+#endif /* CONFIG_NO_HZ */
 
 /*
  * calc_load - update the avenrun load estimates 10 ticks after the
@@ -2369,11 +2494,18 @@ void get_avenrun(unsigned long *loads, unsigned long offset, int shift)
  */
 void calc_global_load(unsigned long ticks)
 {
-        long active;
+        long active, delta;
 
         if (time_before(jiffies, calc_load_update + 10))
                 return;
 
+        /*
+         * Fold the 'old' idle-delta to include all NO_HZ cpus.
+         */
+        delta = calc_load_fold_idle();
+        if (delta)
+                atomic_long_add(delta, &calc_load_tasks);
+
         active = atomic_long_read(&calc_load_tasks);
         active = active > 0 ? active * FIXED_1 : 0;
 
@@ -2384,12 +2516,7 @@ void calc_global_load(unsigned long ticks)
         calc_load_update += LOAD_FREQ;
 
         /*
-         * Account one period with whatever state we found before
-         * folding in the nohz state and ageing the entire idle period.
-         *
-         * This avoids loosing a sample when we go idle between 
-         * calc_load_account_active() (10 ticks ago) and now and thus
-         * under-accounting.
+         * In case we idled for multiple LOAD_FREQ intervals, catch up in bulk.
          */
         calc_global_nohz();
 }
@@ -2406,7 +2533,6 @@ static void calc_load_account_active(struct rq *this_rq)
                 return;
 
         delta  = calc_load_fold_active(this_rq);
-        delta += calc_load_fold_idle();
         if (delta)
                 atomic_long_add(delta, &calc_load_tasks);
 
@@ -2414,6 +2540,10 @@ static void calc_load_account_active(struct rq *this_rq)
 }
 
 /*
+ * End of global load-average stuff
+ */
+
+/*
  * The exact cpuload at various idx values, calculated at every tick would be
  * load = (2^idx - 1) / 2^idx * load + 1 / 2^idx * cur_load
  *
diff --git a/kernel/sched/idle_task.c b/kernel/sched/idle_task.c
index b44d604b35d1..b6baf370cae9 100644
--- a/kernel/sched/idle_task.c
+++ b/kernel/sched/idle_task.c
@@ -25,7 +25,6 @@ static void check_preempt_curr_idle(struct rq *rq, struct task_struct *p, int fl
 static struct task_struct *pick_next_task_idle(struct rq *rq)
 {
         schedstat_inc(rq, sched_goidle);
-        calc_load_account_idle(rq);
         return rq->idle;
 }
 
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index 6d52cea7f33d..55844f24435a 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -942,8 +942,6 @@ static inline u64 sched_avg_period(void)
         return (u64)sysctl_sched_time_avg * NSEC_PER_MSEC / 2;
 }
 
-void calc_load_account_idle(struct rq *this_rq);
-
 #ifdef CONFIG_SCHED_HRTICK
 
 /*
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index 869997833928..4a08472c3ca7 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -406,6 +406,7 @@ static void tick_nohz_stop_sched_tick(struct tick_sched *ts)
                  */
                 if (!ts->tick_stopped) {
                         select_nohz_load_balancer(1);
+                        calc_load_enter_idle();
 
                         ts->idle_tick = hrtimer_get_expires(&ts->sched_timer);
                         ts->tick_stopped = 1;
@@ -597,6 +598,7 @@ void tick_nohz_idle_exit(void)
                 account_idle_ticks(ticks);
 #endif
 
+        calc_load_exit_idle();
         touch_softlockup_watchdog();
         /*
          * Cancel the scheduled timer and restore the tick
diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c
index 6f46a00a1e8a..269b1fe5f2ae 100644
--- a/kernel/time/timekeeping.c
+++ b/kernel/time/timekeeping.c
@@ -70,6 +70,12 @@ struct timekeeper {
         /* The raw monotonic time for the CLOCK_MONOTONIC_RAW posix clock. */
         struct timespec raw_time;
 
+        /* Offset clock monotonic -> clock realtime */
+        ktime_t offs_real;
+
+        /* Offset clock monotonic -> clock boottime */
+        ktime_t offs_boot;
+
         /* Seqlock for all timekeeper values */
         seqlock_t lock;
 };
@@ -172,6 +178,14 @@ static inline s64 timekeeping_get_ns_raw(void)
         return clocksource_cyc2ns(cycle_delta, clock->mult, clock->shift);
 }
 
+static void update_rt_offset(void)
+{
+        struct timespec tmp, *wtm = &timekeeper.wall_to_monotonic;
+
+        set_normalized_timespec(&tmp, -wtm->tv_sec, -wtm->tv_nsec);
+        timekeeper.offs_real = timespec_to_ktime(tmp);
+}
+
 /* must hold write on timekeeper.lock */
 static void timekeeping_update(bool clearntp)
 {
@@ -179,6 +193,7 @@ static void timekeeping_update(bool clearntp)
                 timekeeper.ntp_error = 0;
                 ntp_clear();
         }
+        update_rt_offset();
         update_vsyscall(&timekeeper.xtime, &timekeeper.wall_to_monotonic,
                          timekeeper.clock, timekeeper.mult);
 }
@@ -604,6 +619,7 @@ void __init timekeeping_init(void)
         }
         set_normalized_timespec(&timekeeper.wall_to_monotonic,
                                 -boot.tv_sec, -boot.tv_nsec);
+        update_rt_offset();
         timekeeper.total_sleep_time.tv_sec = 0;
         timekeeper.total_sleep_time.tv_nsec = 0;
         write_sequnlock_irqrestore(&timekeeper.lock, flags);
@@ -612,6 +628,12 @@ void __init timekeeping_init(void)
 /* time in seconds when suspend began */
 static struct timespec timekeeping_suspend_time;
 
+static void update_sleep_time(struct timespec t)
+{
+        timekeeper.total_sleep_time = t;
+        timekeeper.offs_boot = timespec_to_ktime(t);
+}
+
 /**
  * __timekeeping_inject_sleeptime - Internal function to add sleep interval
  * @delta: pointer to a timespec delta value
@@ -630,8 +652,7 @@ static void __timekeeping_inject_sleeptime(struct timespec *delta)
         timekeeper.xtime = timespec_add(timekeeper.xtime, *delta);
         timekeeper.wall_to_monotonic =
                         timespec_sub(timekeeper.wall_to_monotonic, *delta);
-        timekeeper.total_sleep_time = timespec_add(
-                                        timekeeper.total_sleep_time, *delta);
+        update_sleep_time(timespec_add(timekeeper.total_sleep_time, *delta));
 }
 
 
@@ -963,6 +984,8 @@ static cycle_t logarithmic_accumulation(cycle_t offset, int shift)
                 leap = second_overflow(timekeeper.xtime.tv_sec);
                 timekeeper.xtime.tv_sec += leap;
                 timekeeper.wall_to_monotonic.tv_sec -= leap;
+                if (leap)
+                        clock_was_set_delayed();
         }
 
         /* Accumulate raw time */
@@ -1079,6 +1102,8 @@ static void update_wall_time(void)
                 leap = second_overflow(timekeeper.xtime.tv_sec);
                 timekeeper.xtime.tv_sec += leap;
                 timekeeper.wall_to_monotonic.tv_sec -= leap;
+                if (leap)
+                        clock_was_set_delayed();
         }
 
         timekeeping_update(false);
@@ -1246,6 +1271,40 @@ void get_xtime_and_monotonic_and_sleep_offset(struct timespec *xtim,
         } while (read_seqretry(&timekeeper.lock, seq));
 }
 
+#ifdef CONFIG_HIGH_RES_TIMERS
+/**
+ * ktime_get_update_offsets - hrtimer helper
+ * @offs_real:  pointer to storage for monotonic -> realtime offset
+ * @offs_boot:  pointer to storage for monotonic -> boottime offset
+ *
+ * Returns current monotonic time and updates the offsets
+ * Called from hrtimer_interupt() or retrigger_next_event()
+ */
+ktime_t ktime_get_update_offsets(ktime_t *offs_real, ktime_t *offs_boot)
+{
+        ktime_t now;
+        unsigned int seq;
+        u64 secs, nsecs;
+
+        do {
+                seq = read_seqbegin(&timekeeper.lock);
+
+                secs = timekeeper.xtime.tv_sec;
+                nsecs = timekeeper.xtime.tv_nsec;
+                nsecs += timekeeping_get_ns();
+                /* If arch requires, add in gettimeoffset() */
+                nsecs += arch_gettimeoffset();
+
+                *offs_real = timekeeper.offs_real;
+                *offs_boot = timekeeper.offs_boot;
+        } while (read_seqretry(&timekeeper.lock, seq));
+
+        now = ktime_add_ns(ktime_set(secs, 0), nsecs);
+        now = ktime_sub(now, *offs_real);
+        return now;
+}
+#endif
+
 /**
  * ktime_get_monotonic_offset() - get wall_to_monotonic in ktime_t format
  */
diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c
index 1d0f6a8a0e5e..f765465bffe4 100644
--- a/kernel/trace/ring_buffer.c
+++ b/kernel/trace/ring_buffer.c
@@ -1075,6 +1075,7 @@ rb_allocate_cpu_buffer(struct ring_buffer *buffer, int nr_pages, int cpu)
         rb_init_page(bpage->page);
 
         INIT_LIST_HEAD(&cpu_buffer->reader_page->list);
+        INIT_LIST_HEAD(&cpu_buffer->new_pages);
 
         ret = rb_allocate_pages(cpu_buffer, nr_pages);
         if (ret < 0)
@@ -1346,10 +1347,9 @@ rb_remove_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned int nr_pages)
                          * If something was added to this page, it was full
                          * since it is not the tail page. So we deduct the
                          * bytes consumed in ring buffer from here.
-                         * No need to update overruns, since this page is
-                         * deleted from ring buffer and its entries are
-                         * already accounted for.
+                         * Increment overrun to account for the lost events.
                          */
+                        local_add(page_entries, &cpu_buffer->overrun);
                         local_sub(BUF_PAGE_SIZE, &cpu_buffer->entries_bytes);
                 }
 
diff --git a/sound/usb/endpoint.c b/sound/usb/endpoint.c
index e6906901debb..0f647d22cb4a 100644
--- a/sound/usb/endpoint.c
+++ b/sound/usb/endpoint.c
@@ -414,7 +414,7 @@ struct snd_usb_endpoint *snd_usb_add_endpoint(struct snd_usb_audio *chip,
 {
         struct list_head *p;
         struct snd_usb_endpoint *ep;
-        int ret, is_playback = direction == SNDRV_PCM_STREAM_PLAYBACK;
+        int is_playback = direction == SNDRV_PCM_STREAM_PLAYBACK;
 
         mutex_lock(&chip->mutex);
 
@@ -434,16 +434,6 @@ struct snd_usb_endpoint *snd_usb_add_endpoint(struct snd_usb_audio *chip,
                     type == SND_USB_ENDPOINT_TYPE_DATA ? "data" : "sync",
                     ep_num);
 
-        /* select the alt setting once so the endpoints become valid */
-        ret = usb_set_interface(chip->dev, alts->desc.bInterfaceNumber,
-                                alts->desc.bAlternateSetting);
-        if (ret < 0) {
-                snd_printk(KERN_ERR "%s(): usb_set_interface() failed, ret = %d\n",
-                                        __func__, ret);
-                ep = NULL;
-                goto __exit_unlock;
-        }
-
         ep = kzalloc(sizeof(*ep), GFP_KERNEL);
         if (!ep)
                 goto __exit_unlock;
@@ -831,9 +821,6 @@ int snd_usb_endpoint_start(struct snd_usb_endpoint *ep)
         if (++ep->use_count != 1)
                 return 0;
 
-        if (snd_BUG_ON(!test_bit(EP_FLAG_ACTIVATED, &ep->flags)))
-                return -EINVAL;
-
         /* just to be sure */
         deactivate_urbs(ep, 0, 1);
         wait_clear_urbs(ep);
@@ -911,9 +898,6 @@ void snd_usb_endpoint_stop(struct snd_usb_endpoint *ep,
         if (snd_BUG_ON(ep->use_count == 0))
                 return;
 
-        if (snd_BUG_ON(!test_bit(EP_FLAG_ACTIVATED, &ep->flags)))
-                return;
-
         if (--ep->use_count == 0) {
                 deactivate_urbs(ep, force, can_sleep);
                 ep->data_subs = NULL;
@@ -927,42 +911,6 @@ void snd_usb_endpoint_stop(struct snd_usb_endpoint *ep,
 }
 
 /**
- * snd_usb_endpoint_activate: activate an snd_usb_endpoint
- *
- * @ep: the endpoint to activate
- *
- * If the endpoint is not currently in use, this functions will select the
- * correct alternate interface setting for the interface of this endpoint.
- *
- * In case of any active users, this functions does nothing.
- *
- * Returns an error if usb_set_interface() failed, 0 in all other
- * cases.
- */
-int snd_usb_endpoint_activate(struct snd_usb_endpoint *ep)
-{
-        if (ep->use_count != 0)
-                return 0;
-
-        if (!ep->chip->shutdown &&
-            !test_and_set_bit(EP_FLAG_ACTIVATED, &ep->flags)) {
-                int ret;
-
-                ret = usb_set_interface(ep->chip->dev, ep->iface, ep->alt_idx);
-                if (ret < 0) {
-                        snd_printk(KERN_ERR "%s() usb_set_interface() failed, ret = %d\n",
-                                                __func__, ret);
-                        clear_bit(EP_FLAG_ACTIVATED, &ep->flags);
-                        return ret;
-                }
-
-                return 0;
-        }
-
-        return -EBUSY;
-}
-
-/**
  * snd_usb_endpoint_deactivate: deactivate an snd_usb_endpoint
  *
  * @ep: the endpoint to deactivate
@@ -980,24 +928,15 @@ int snd_usb_endpoint_deactivate(struct snd_usb_endpoint *ep)
         if (!ep)
                 return -EINVAL;
 
+        deactivate_urbs(ep, 1, 1);
+        wait_clear_urbs(ep);
+
         if (ep->use_count != 0)
                 return 0;
 
-        if (!ep->chip->shutdown &&
-            test_and_clear_bit(EP_FLAG_ACTIVATED, &ep->flags)) {
-                int ret;
-
-                ret = usb_set_interface(ep->chip->dev, ep->iface, 0);
-                if (ret < 0) {
-                        snd_printk(KERN_ERR "%s(): usb_set_interface() failed, ret = %d\n",
-                                                __func__, ret);
-                        return ret;
-                }
+        clear_bit(EP_FLAG_ACTIVATED, &ep->flags);
 
-                return 0;
-        }
-
-        return -EBUSY;
+        return 0;
 }
 
 /**
diff --git a/sound/usb/pcm.c b/sound/usb/pcm.c
index 54607f8c4f66..a1298f379428 100644
--- a/sound/usb/pcm.c
+++ b/sound/usb/pcm.c
@@ -261,19 +261,6 @@ static void stop_endpoints(struct snd_usb_substream *subs,
                                       force, can_sleep, wait);
 }
 
-static int activate_endpoints(struct snd_usb_substream *subs)
-{
-        if (subs->sync_endpoint) {
-                int ret;
-
-                ret = snd_usb_endpoint_activate(subs->sync_endpoint);
-                if (ret < 0)
-                        return ret;
-        }
-
-        return snd_usb_endpoint_activate(subs->data_endpoint);
-}
-
 static int deactivate_endpoints(struct snd_usb_substream *subs)
 {
         int reta, retb;
@@ -314,6 +301,33 @@ static int set_format(struct snd_usb_substream *subs, struct audioformat *fmt)
         if (fmt == subs->cur_audiofmt)
                 return 0;
 
+        /* close the old interface */
+        if (subs->interface >= 0 && subs->interface != fmt->iface) {
+                err = usb_set_interface(subs->dev, subs->interface, 0);
+                if (err < 0) {
+                        snd_printk(KERN_ERR "%d:%d:%d: return to setting 0 failed (%d)\n",
+                                dev->devnum, fmt->iface, fmt->altsetting, err);
+                        return -EIO;
+                }
+                subs->interface = -1;
+                subs->altset_idx = 0;
+        }
+
+        /* set interface */
+        if (subs->interface != fmt->iface ||
+            subs->altset_idx != fmt->altset_idx) {
+                err = usb_set_interface(dev, fmt->iface, fmt->altsetting);
+                if (err < 0) {
+                        snd_printk(KERN_ERR "%d:%d:%d: usb_set_interface failed (%d)\n",
+                                   dev->devnum, fmt->iface, fmt->altsetting, err);
+                        return -EIO;
+                }
+                snd_printdd(KERN_INFO "setting usb interface %d:%d\n",
+                                fmt->iface, fmt->altsetting);
+                subs->interface = fmt->iface;
+                subs->altset_idx = fmt->altset_idx;
+        }
+
         subs->data_endpoint = snd_usb_add_endpoint(subs->stream->chip,
                                                    alts, fmt->endpoint, subs->direction,
                                                    SND_USB_ENDPOINT_TYPE_DATA);
@@ -387,7 +401,7 @@ add_sync_ep:
                 subs->data_endpoint->sync_master = subs->sync_endpoint;
         }
 
-        if ((err = snd_usb_init_pitch(subs->stream->chip, subs->interface, alts, fmt)) < 0)
+        if ((err = snd_usb_init_pitch(subs->stream->chip, fmt->iface, alts, fmt)) < 0)
                 return err;
 
         subs->cur_audiofmt = fmt;
@@ -450,7 +464,7 @@ static int snd_usb_hw_params(struct snd_pcm_substream *substream,
                 struct usb_interface *iface;
                 iface = usb_ifnum_to_if(subs->dev, fmt->iface);
                 alts = &iface->altsetting[fmt->altset_idx];
-                ret = snd_usb_init_sample_rate(subs->stream->chip, subs->interface, alts, fmt, rate);
+                ret = snd_usb_init_sample_rate(subs->stream->chip, fmt->iface, alts, fmt, rate);
                 if (ret < 0)
                         return ret;
                 subs->cur_rate = rate;
@@ -460,12 +474,6 @@ static int snd_usb_hw_params(struct snd_pcm_substream *substream,
                 mutex_lock(&subs->stream->chip->shutdown_mutex);
                 /* format changed */
                 stop_endpoints(subs, 0, 0, 0);
-                deactivate_endpoints(subs);
-
-                ret = activate_endpoints(subs);
-                if (ret < 0)
-                        goto unlock;
-
                 ret = snd_usb_endpoint_set_params(subs->data_endpoint, hw_params, fmt,
                                                   subs->sync_endpoint);
                 if (ret < 0)
@@ -500,6 +508,7 @@ static int snd_usb_hw_free(struct snd_pcm_substream *substream)
         subs->period_bytes = 0;
         mutex_lock(&subs->stream->chip->shutdown_mutex);
         stop_endpoints(subs, 0, 1, 1);
+        deactivate_endpoints(subs);
         mutex_unlock(&subs->stream->chip->shutdown_mutex);
         return snd_pcm_lib_free_vmalloc_buffer(substream);
 }
@@ -938,16 +947,20 @@ static int snd_usb_pcm_open(struct snd_pcm_substream *substream, int direction)
 
 static int snd_usb_pcm_close(struct snd_pcm_substream *substream, int direction)
 {
-        int ret;
         struct snd_usb_stream *as = snd_pcm_substream_chip(substream);
         struct snd_usb_substream *subs = &as->substream[direction];
 
         stop_endpoints(subs, 0, 0, 0);
-        ret = deactivate_endpoints(subs);
+
+        if (!as->chip->shutdown && subs->interface >= 0) {
+                usb_set_interface(subs->dev, subs->interface, 0);
+                subs->interface = -1;
+        }
+
         subs->pcm_substream = NULL;
         snd_usb_autosuspend(subs->stream->chip);
 
-        return ret;
+        return 0;
 }
 
 /* Since a URB can handle only a single linear buffer, we must use double
diff --git a/tools/perf/util/map.c b/tools/perf/util/map.c
index 35ae56864e4f..a1f4e3669142 100644
--- a/tools/perf/util/map.c
+++ b/tools/perf/util/map.c
@@ -669,25 +669,26 @@ struct machine *machines__find(struct rb_root *self, pid_t pid)
 struct machine *machines__findnew(struct rb_root *self, pid_t pid)
 {
         char path[PATH_MAX];
-        const char *root_dir;
+        const char *root_dir = "";
         struct machine *machine = machines__find(self, pid);
 
-        if (!machine || machine->pid != pid) {
-                if (pid == HOST_KERNEL_ID || pid == DEFAULT_GUEST_KERNEL_ID)
-                        root_dir = "";
-                else {
-                        if (!symbol_conf.guestmount)
-                                goto out;
-                        sprintf(path, "%s/%d", symbol_conf.guestmount, pid);
-                        if (access(path, R_OK)) {
-                                pr_err("Can't access file %s\n", path);
-                                goto out;
-                        }
-                        root_dir = path;
+        if (machine && (machine->pid == pid))
+                goto out;
+
+        if ((pid != HOST_KERNEL_ID) &&
+            (pid != DEFAULT_GUEST_KERNEL_ID) &&
+            (symbol_conf.guestmount)) {
+                sprintf(path, "%s/%d", symbol_conf.guestmount, pid);
+                if (access(path, R_OK)) {
+                        pr_err("Can't access file %s\n", path);
+                        machine = NULL;
+                        goto out;
                 }
-                machine = machines__add(self, pid, root_dir);
+                root_dir = path;
         }
 
+        machine = machines__add(self, pid, root_dir);
+
 out:
         return machine;
 }
diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
index c3e399bcf18d..56142d0fb8d7 100644
--- a/tools/perf/util/session.c
+++ b/tools/perf/util/session.c
@@ -926,7 +926,7 @@ static struct machine *
                 else
                         pid = event->ip.pid;
 
-                return perf_session__find_machine(session, pid);
+                return perf_session__findnew_machine(session, pid);
         }
 
         return perf_session__find_host_machine(session);
diff --git a/tools/perf/util/trace-event-parse.c b/tools/perf/util/trace-event-parse.c
index df2fddbf0cd2..5dd3b5ec8411 100644
--- a/tools/perf/util/trace-event-parse.c
+++ b/tools/perf/util/trace-event-parse.c
@@ -198,9 +198,8 @@ void print_trace_event(int cpu, void *data, int size)
         record.data = data;
 
         trace_seq_init(&s);
-        pevent_print_event(pevent, &s, &record);
+        pevent_event_info(&s, event, &record);
         trace_seq_do_printf(&s);
-        printf("\n");
 }
 
 void print_event(int cpu, void *data, int size, unsigned long long nsecs,