Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux into drm-next

This backmerges drm-fixes into drm-next mainly for the amdkfd
stuff, I'm not 100% confident, but it builds and the amdkfd
folks can fix anything up.

Signed-off-by: Dave Airlie <airlied@redhat.com>
Conflicts:
	drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
	drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h

14 files changed, 112 insertions, 54 deletions

diff --git a/drivers/gpu/drm/radeon/cik_sdma.c b/drivers/gpu/drm/radeon/cik_sdma.c
index 7d2ff31c35a5..f86eb54e7763 100644
--- a/drivers/gpu/drm/radeon/cik_sdma.c
+++ b/drivers/gpu/drm/radeon/cik_sdma.c
@@ -845,7 +845,6 @@ void cik_sdma_vm_write_pages(struct radeon_device *rdev,
                 for (; ndw > 0; ndw -= 2, --count, pe += 8) {
                         if (flags & R600_PTE_SYSTEM) {
                                 value = radeon_vm_map_gart(rdev, addr);
-                                value &= 0xFFFFFFFFFFFFF000ULL;
                         } else if (flags & R600_PTE_VALID) {
                                 value = addr;
                         } else {
diff --git a/drivers/gpu/drm/radeon/ni_dma.c b/drivers/gpu/drm/radeon/ni_dma.c
index 4be2bb7cbef3..ce787a9f12c0 100644
--- a/drivers/gpu/drm/radeon/ni_dma.c
+++ b/drivers/gpu/drm/radeon/ni_dma.c
@@ -372,7 +372,6 @@ void cayman_dma_vm_write_pages(struct radeon_device *rdev,
                 for (; ndw > 0; ndw -= 2, --count, pe += 8) {
                         if (flags & R600_PTE_SYSTEM) {
                                 value = radeon_vm_map_gart(rdev, addr);
-                                value &= 0xFFFFFFFFFFFFF000ULL;
                         } else if (flags & R600_PTE_VALID) {
                                 value = addr;
                         } else {
diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c
index 74f06d540591..279801ca5110 100644
--- a/drivers/gpu/drm/radeon/r100.c
+++ b/drivers/gpu/drm/radeon/r100.c
@@ -644,6 +644,7 @@ int r100_pci_gart_init(struct radeon_device *rdev)
                 return r;
         rdev->gart.table_size = rdev->gart.num_gpu_pages * 4;
         rdev->asic->gart.tlb_flush = &r100_pci_gart_tlb_flush;
+        rdev->asic->gart.get_page_entry = &r100_pci_gart_get_page_entry;
         rdev->asic->gart.set_page = &r100_pci_gart_set_page;
         return radeon_gart_table_ram_alloc(rdev);
 }
@@ -681,11 +682,16 @@ void r100_pci_gart_disable(struct radeon_device *rdev)
         WREG32(RADEON_AIC_HI_ADDR, 0);
 }
 
+uint64_t r100_pci_gart_get_page_entry(uint64_t addr, uint32_t flags)
+{
+        return addr;
+}
+
 void r100_pci_gart_set_page(struct radeon_device *rdev, unsigned i,
-                            uint64_t addr, uint32_t flags)
+                            uint64_t entry)
 {
         u32 *gtt = rdev->gart.ptr;
-        gtt[i] = cpu_to_le32(lower_32_bits(addr));
+        gtt[i] = cpu_to_le32(lower_32_bits(entry));
 }
 
 void r100_pci_gart_fini(struct radeon_device *rdev)
diff --git a/drivers/gpu/drm/radeon/r300.c b/drivers/gpu/drm/radeon/r300.c
index 064ad5569cca..08d68f3e13e9 100644
--- a/drivers/gpu/drm/radeon/r300.c
+++ b/drivers/gpu/drm/radeon/r300.c
@@ -73,11 +73,8 @@ void rv370_pcie_gart_tlb_flush(struct radeon_device *rdev)
 #define R300_PTE_WRITEABLE (1 << 2)
 #define R300_PTE_READABLE  (1 << 3)
 
-void rv370_pcie_gart_set_page(struct radeon_device *rdev, unsigned i,
-                              uint64_t addr, uint32_t flags)
+uint64_t rv370_pcie_gart_get_page_entry(uint64_t addr, uint32_t flags)
 {
-        void __iomem *ptr = rdev->gart.ptr;
-
         addr = (lower_32_bits(addr) >> 8) |
                 ((upper_32_bits(addr) & 0xff) << 24);
         if (flags & RADEON_GART_PAGE_READ)
@@ -86,10 +83,18 @@ void rv370_pcie_gart_set_page(struct radeon_device *rdev, unsigned i,
                 addr |= R300_PTE_WRITEABLE;
         if (!(flags & RADEON_GART_PAGE_SNOOP))
                 addr |= R300_PTE_UNSNOOPED;
+        return addr;
+}
+
+void rv370_pcie_gart_set_page(struct radeon_device *rdev, unsigned i,
+                              uint64_t entry)
+{
+        void __iomem *ptr = rdev->gart.ptr;
+
         /* on x86 we want this to be CPU endian, on powerpc
          * on powerpc without HW swappers, it'll get swapped on way
          * into VRAM - so no need for cpu_to_le32 on VRAM tables */
-        writel(addr, ((void __iomem *)ptr) + (i * 4));
+        writel(entry, ((void __iomem *)ptr) + (i * 4));
 }
 
 int rv370_pcie_gart_init(struct radeon_device *rdev)
@@ -109,6 +114,7 @@ int rv370_pcie_gart_init(struct radeon_device *rdev)
                 DRM_ERROR("Failed to register debugfs file for PCIE gart !\n");
         rdev->gart.table_size = rdev->gart.num_gpu_pages * 4;
         rdev->asic->gart.tlb_flush = &rv370_pcie_gart_tlb_flush;
+        rdev->asic->gart.get_page_entry = &rv370_pcie_gart_get_page_entry;
         rdev->asic->gart.set_page = &rv370_pcie_gart_set_page;
         return radeon_gart_table_vram_alloc(rdev);
 }
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index 93e407b7e7a7..5587603b4a89 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -242,6 +242,7 @@ bool radeon_get_bios(struct radeon_device *rdev);
  * Dummy page
  */
 struct radeon_dummy_page {
+        uint64_t        entry;
         struct page     *page;
         dma_addr_t      addr;
 };
@@ -645,7 +646,7 @@ struct radeon_gart {
         unsigned                        num_cpu_pages;
         unsigned                        table_size;
         struct page                     **pages;
-        dma_addr_t                      *pages_addr;
+        uint64_t                        *pages_entry;
         bool                            ready;
 };
 
@@ -1858,8 +1859,9 @@ struct radeon_asic {
         /* gart */
         struct {
                 void (*tlb_flush)(struct radeon_device *rdev);
+                uint64_t (*get_page_entry)(uint64_t addr, uint32_t flags);
                 void (*set_page)(struct radeon_device *rdev, unsigned i,
-                                 uint64_t addr, uint32_t flags);
+                                 uint64_t entry);
         } gart;
         struct {
                 int (*init)(struct radeon_device *rdev);
@@ -2867,7 +2869,8 @@ static inline void radeon_ring_write(struct radeon_ring *ring, uint32_t v)
 #define radeon_vga_set_state(rdev, state) (rdev)->asic->vga_set_state((rdev), (state))
 #define radeon_asic_reset(rdev) (rdev)->asic->asic_reset((rdev))
 #define radeon_gart_tlb_flush(rdev) (rdev)->asic->gart.tlb_flush((rdev))
-#define radeon_gart_set_page(rdev, i, p, f) (rdev)->asic->gart.set_page((rdev), (i), (p), (f))
+#define radeon_gart_get_page_entry(a, f) (rdev)->asic->gart.get_page_entry((a), (f))
+#define radeon_gart_set_page(rdev, i, e) (rdev)->asic->gart.set_page((rdev), (i), (e))
 #define radeon_asic_vm_init(rdev) (rdev)->asic->vm.init((rdev))
 #define radeon_asic_vm_fini(rdev) (rdev)->asic->vm.fini((rdev))
 #define radeon_asic_vm_copy_pages(rdev, ib, pe, src, count) ((rdev)->asic->vm.copy_pages((rdev), (ib), (pe), (src), (count)))
diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c
index f811ee14a237..c0ecd128b14b 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.c
+++ b/drivers/gpu/drm/radeon/radeon_asic.c
@@ -159,11 +159,13 @@ void radeon_agp_disable(struct radeon_device *rdev)
                 DRM_INFO("Forcing AGP to PCIE mode\n");
                 rdev->flags |= RADEON_IS_PCIE;
                 rdev->asic->gart.tlb_flush = &rv370_pcie_gart_tlb_flush;
+                rdev->asic->gart.get_page_entry = &rv370_pcie_gart_get_page_entry;
                 rdev->asic->gart.set_page = &rv370_pcie_gart_set_page;
         } else {
                 DRM_INFO("Forcing AGP to PCI mode\n");
                 rdev->flags |= RADEON_IS_PCI;
                 rdev->asic->gart.tlb_flush = &r100_pci_gart_tlb_flush;
+                rdev->asic->gart.get_page_entry = &r100_pci_gart_get_page_entry;
                 rdev->asic->gart.set_page = &r100_pci_gart_set_page;
         }
         rdev->mc.gtt_size = radeon_gart_size * 1024 * 1024;
@@ -199,6 +201,7 @@ static struct radeon_asic r100_asic = {
         .mc_wait_for_idle = &r100_mc_wait_for_idle,
         .gart = {
                 .tlb_flush = &r100_pci_gart_tlb_flush,
+                .get_page_entry = &r100_pci_gart_get_page_entry,
                 .set_page = &r100_pci_gart_set_page,
         },
         .ring = {
@@ -265,6 +268,7 @@ static struct radeon_asic r200_asic = {
         .mc_wait_for_idle = &r100_mc_wait_for_idle,
         .gart = {
                 .tlb_flush = &r100_pci_gart_tlb_flush,
+                .get_page_entry = &r100_pci_gart_get_page_entry,
                 .set_page = &r100_pci_gart_set_page,
         },
         .ring = {
@@ -359,6 +363,7 @@ static struct radeon_asic r300_asic = {
         .mc_wait_for_idle = &r300_mc_wait_for_idle,
         .gart = {
                 .tlb_flush = &r100_pci_gart_tlb_flush,
+                .get_page_entry = &r100_pci_gart_get_page_entry,
                 .set_page = &r100_pci_gart_set_page,
         },
         .ring = {
@@ -425,6 +430,7 @@ static struct radeon_asic r300_asic_pcie = {
         .mc_wait_for_idle = &r300_mc_wait_for_idle,
         .gart = {
                 .tlb_flush = &rv370_pcie_gart_tlb_flush,
+                .get_page_entry = &rv370_pcie_gart_get_page_entry,
                 .set_page = &rv370_pcie_gart_set_page,
         },
         .ring = {
@@ -491,6 +497,7 @@ static struct radeon_asic r420_asic = {
         .mc_wait_for_idle = &r300_mc_wait_for_idle,
         .gart = {
                 .tlb_flush = &rv370_pcie_gart_tlb_flush,
+                .get_page_entry = &rv370_pcie_gart_get_page_entry,
                 .set_page = &rv370_pcie_gart_set_page,
         },
         .ring = {
@@ -557,6 +564,7 @@ static struct radeon_asic rs400_asic = {
         .mc_wait_for_idle = &rs400_mc_wait_for_idle,
         .gart = {
                 .tlb_flush = &rs400_gart_tlb_flush,
+                .get_page_entry = &rs400_gart_get_page_entry,
                 .set_page = &rs400_gart_set_page,
         },
         .ring = {
@@ -623,6 +631,7 @@ static struct radeon_asic rs600_asic = {
         .mc_wait_for_idle = &rs600_mc_wait_for_idle,
         .gart = {
                 .tlb_flush = &rs600_gart_tlb_flush,
+                .get_page_entry = &rs600_gart_get_page_entry,
                 .set_page = &rs600_gart_set_page,
         },
         .ring = {
@@ -689,6 +698,7 @@ static struct radeon_asic rs690_asic = {
         .mc_wait_for_idle = &rs690_mc_wait_for_idle,
         .gart = {
                 .tlb_flush = &rs400_gart_tlb_flush,
+                .get_page_entry = &rs400_gart_get_page_entry,
                 .set_page = &rs400_gart_set_page,
         },
         .ring = {
@@ -755,6 +765,7 @@ static struct radeon_asic rv515_asic = {
         .mc_wait_for_idle = &rv515_mc_wait_for_idle,
         .gart = {
                 .tlb_flush = &rv370_pcie_gart_tlb_flush,
+                .get_page_entry = &rv370_pcie_gart_get_page_entry,
                 .set_page = &rv370_pcie_gart_set_page,
         },
         .ring = {
@@ -821,6 +832,7 @@ static struct radeon_asic r520_asic = {
         .mc_wait_for_idle = &r520_mc_wait_for_idle,
         .gart = {
                 .tlb_flush = &rv370_pcie_gart_tlb_flush,
+                .get_page_entry = &rv370_pcie_gart_get_page_entry,
                 .set_page = &rv370_pcie_gart_set_page,
         },
         .ring = {
@@ -915,6 +927,7 @@ static struct radeon_asic r600_asic = {
         .get_gpu_clock_counter = &r600_get_gpu_clock_counter,
         .gart = {
                 .tlb_flush = &r600_pcie_gart_tlb_flush,
+                .get_page_entry = &rs600_gart_get_page_entry,
                 .set_page = &rs600_gart_set_page,
         },
         .ring = {
@@ -998,6 +1011,7 @@ static struct radeon_asic rv6xx_asic = {
         .get_gpu_clock_counter = &r600_get_gpu_clock_counter,
         .gart = {
                 .tlb_flush = &r600_pcie_gart_tlb_flush,
+                .get_page_entry = &rs600_gart_get_page_entry,
                 .set_page = &rs600_gart_set_page,
         },
         .ring = {
@@ -1087,6 +1101,7 @@ static struct radeon_asic rs780_asic = {
         .get_gpu_clock_counter = &r600_get_gpu_clock_counter,
         .gart = {
                 .tlb_flush = &r600_pcie_gart_tlb_flush,
+                .get_page_entry = &rs600_gart_get_page_entry,
                 .set_page = &rs600_gart_set_page,
         },
         .ring = {
@@ -1189,6 +1204,7 @@ static struct radeon_asic rv770_asic = {
         .get_gpu_clock_counter = &r600_get_gpu_clock_counter,
         .gart = {
                 .tlb_flush = &r600_pcie_gart_tlb_flush,
+                .get_page_entry = &rs600_gart_get_page_entry,
                 .set_page = &rs600_gart_set_page,
         },
         .ring = {
@@ -1305,6 +1321,7 @@ static struct radeon_asic evergreen_asic = {
         .get_gpu_clock_counter = &r600_get_gpu_clock_counter,
         .gart = {
                 .tlb_flush = &evergreen_pcie_gart_tlb_flush,
+                .get_page_entry = &rs600_gart_get_page_entry,
                 .set_page = &rs600_gart_set_page,
         },
         .ring = {
@@ -1395,6 +1412,7 @@ static struct radeon_asic sumo_asic = {
         .get_gpu_clock_counter = &r600_get_gpu_clock_counter,
         .gart = {
                 .tlb_flush = &evergreen_pcie_gart_tlb_flush,
+                .get_page_entry = &rs600_gart_get_page_entry,
                 .set_page = &rs600_gart_set_page,
         },
         .ring = {
@@ -1484,6 +1502,7 @@ static struct radeon_asic btc_asic = {
         .get_gpu_clock_counter = &r600_get_gpu_clock_counter,
         .gart = {
                 .tlb_flush = &evergreen_pcie_gart_tlb_flush,
+                .get_page_entry = &rs600_gart_get_page_entry,
                 .set_page = &rs600_gart_set_page,
         },
         .ring = {
@@ -1617,6 +1636,7 @@ static struct radeon_asic cayman_asic = {
         .get_gpu_clock_counter = &r600_get_gpu_clock_counter,
         .gart = {
                 .tlb_flush = &cayman_pcie_gart_tlb_flush,
+                .get_page_entry = &rs600_gart_get_page_entry,
                 .set_page = &rs600_gart_set_page,
         },
         .vm = {
@@ -1718,6 +1738,7 @@ static struct radeon_asic trinity_asic = {
         .get_gpu_clock_counter = &r600_get_gpu_clock_counter,
         .gart = {
                 .tlb_flush = &cayman_pcie_gart_tlb_flush,
+                .get_page_entry = &rs600_gart_get_page_entry,
                 .set_page = &rs600_gart_set_page,
         },
         .vm = {
@@ -1849,6 +1870,7 @@ static struct radeon_asic si_asic = {
         .get_gpu_clock_counter = &si_get_gpu_clock_counter,
         .gart = {
                 .tlb_flush = &si_pcie_gart_tlb_flush,
+                .get_page_entry = &rs600_gart_get_page_entry,
                 .set_page = &rs600_gart_set_page,
         },
         .vm = {
@@ -2012,6 +2034,7 @@ static struct radeon_asic ci_asic = {
         .get_gpu_clock_counter = &cik_get_gpu_clock_counter,
         .gart = {
                 .tlb_flush = &cik_pcie_gart_tlb_flush,
+                .get_page_entry = &rs600_gart_get_page_entry,
                 .set_page = &rs600_gart_set_page,
         },
         .vm = {
@@ -2121,6 +2144,7 @@ static struct radeon_asic kv_asic = {
         .get_gpu_clock_counter = &cik_get_gpu_clock_counter,
         .gart = {
                 .tlb_flush = &cik_pcie_gart_tlb_flush,
+                .get_page_entry = &rs600_gart_get_page_entry,
                 .set_page = &rs600_gart_set_page,
         },
         .vm = {
diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h
index 4045a320a424..72bdd3bf0d8e 100644
--- a/drivers/gpu/drm/radeon/radeon_asic.h
+++ b/drivers/gpu/drm/radeon/radeon_asic.h
@@ -67,8 +67,9 @@ bool r100_gpu_is_lockup(struct radeon_device *rdev, struct radeon_ring *cp);
 int r100_asic_reset(struct radeon_device *rdev);
 u32 r100_get_vblank_counter(struct radeon_device *rdev, int crtc);
 void r100_pci_gart_tlb_flush(struct radeon_device *rdev);
+uint64_t r100_pci_gart_get_page_entry(uint64_t addr, uint32_t flags);
 void r100_pci_gart_set_page(struct radeon_device *rdev, unsigned i,
-                            uint64_t addr, uint32_t flags);
+                            uint64_t entry);
 void r100_ring_start(struct radeon_device *rdev, struct radeon_ring *ring);
 int r100_irq_set(struct radeon_device *rdev);
 int r100_irq_process(struct radeon_device *rdev);
@@ -172,8 +173,9 @@ extern void r300_fence_ring_emit(struct radeon_device *rdev,
                                 struct radeon_fence *fence);
 extern int r300_cs_parse(struct radeon_cs_parser *p);
 extern void rv370_pcie_gart_tlb_flush(struct radeon_device *rdev);
+extern uint64_t rv370_pcie_gart_get_page_entry(uint64_t addr, uint32_t flags);
 extern void rv370_pcie_gart_set_page(struct radeon_device *rdev, unsigned i,
-                                     uint64_t addr, uint32_t flags);
+                                     uint64_t entry);
 extern void rv370_set_pcie_lanes(struct radeon_device *rdev, int lanes);
 extern int rv370_get_pcie_lanes(struct radeon_device *rdev);
 extern void r300_set_reg_safe(struct radeon_device *rdev);
@@ -208,8 +210,9 @@ extern void rs400_fini(struct radeon_device *rdev);
 extern int rs400_suspend(struct radeon_device *rdev);
 extern int rs400_resume(struct radeon_device *rdev);
 void rs400_gart_tlb_flush(struct radeon_device *rdev);
+uint64_t rs400_gart_get_page_entry(uint64_t addr, uint32_t flags);
 void rs400_gart_set_page(struct radeon_device *rdev, unsigned i,
-                         uint64_t addr, uint32_t flags);
+                         uint64_t entry);
 uint32_t rs400_mc_rreg(struct radeon_device *rdev, uint32_t reg);
 void rs400_mc_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v);
 int rs400_gart_init(struct radeon_device *rdev);
@@ -232,8 +235,9 @@ int rs600_irq_process(struct radeon_device *rdev);
 void rs600_irq_disable(struct radeon_device *rdev);
 u32 rs600_get_vblank_counter(struct radeon_device *rdev, int crtc);
 void rs600_gart_tlb_flush(struct radeon_device *rdev);
+uint64_t rs600_gart_get_page_entry(uint64_t addr, uint32_t flags);
 void rs600_gart_set_page(struct radeon_device *rdev, unsigned i,
-                         uint64_t addr, uint32_t flags);
+                         uint64_t entry);
 uint32_t rs600_mc_rreg(struct radeon_device *rdev, uint32_t reg);
 void rs600_mc_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v);
 void rs600_bandwidth_update(struct radeon_device *rdev);
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index 0ec65168f331..bd7519fdd3f4 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -774,6 +774,8 @@ int radeon_dummy_page_init(struct radeon_device *rdev)
                 rdev->dummy_page.page = NULL;
                 return -ENOMEM;
         }
+        rdev->dummy_page.entry = radeon_gart_get_page_entry(rdev->dummy_page.addr,
+                                                            RADEON_GART_PAGE_DUMMY);
         return 0;
 }
 
diff --git a/drivers/gpu/drm/radeon/radeon_gart.c b/drivers/gpu/drm/radeon/radeon_gart.c
index 84146d5901aa..5450fa95a47e 100644
--- a/drivers/gpu/drm/radeon/radeon_gart.c
+++ b/drivers/gpu/drm/radeon/radeon_gart.c
@@ -165,6 +165,19 @@ int radeon_gart_table_vram_pin(struct radeon_device *rdev)
                 radeon_bo_unpin(rdev->gart.robj);
         radeon_bo_unreserve(rdev->gart.robj);
         rdev->gart.table_addr = gpu_addr;
+
+        if (!r) {
+                int i;
+
+                /* We might have dropped some GART table updates while it wasn't
+                 * mapped, restore all entries
+                 */
+                for (i = 0; i < rdev->gart.num_gpu_pages; i++)
+                        radeon_gart_set_page(rdev, i, rdev->gart.pages_entry[i]);
+                mb();
+                radeon_gart_tlb_flush(rdev);
+        }
+
         return r;
 }
 
@@ -228,7 +241,6 @@ void radeon_gart_unbind(struct radeon_device *rdev, unsigned offset,
         unsigned t;
         unsigned p;
         int i, j;
-        u64 page_base;
 
         if (!rdev->gart.ready) {
                 WARN(1, "trying to unbind memory from uninitialized GART !\n");
@@ -239,14 +251,12 @@ void radeon_gart_unbind(struct radeon_device *rdev, unsigned offset,
         for (i = 0; i < pages; i++, p++) {
                 if (rdev->gart.pages[p]) {
                         rdev->gart.pages[p] = NULL;
-                        rdev->gart.pages_addr[p] = rdev->dummy_page.addr;
-                        page_base = rdev->gart.pages_addr[p];
                         for (j = 0; j < (PAGE_SIZE / RADEON_GPU_PAGE_SIZE); j++, t++) {
+                                rdev->gart.pages_entry[t] = rdev->dummy_page.entry;
                                 if (rdev->gart.ptr) {
-                                        radeon_gart_set_page(rdev, t, page_base,
-                                                             RADEON_GART_PAGE_DUMMY);
+                                        radeon_gart_set_page(rdev, t,
+                                                             rdev->dummy_page.entry);
                                 }
-                                page_base += RADEON_GPU_PAGE_SIZE;
                         }
                 }
         }
@@ -274,7 +284,7 @@ int radeon_gart_bind(struct radeon_device *rdev, unsigned offset,
 {
         unsigned t;
         unsigned p;
-        uint64_t page_base;
+        uint64_t page_base, page_entry;
         int i, j;
 
         if (!rdev->gart.ready) {
@@ -285,14 +295,15 @@ int radeon_gart_bind(struct radeon_device *rdev, unsigned offset,
         p = t / (PAGE_SIZE / RADEON_GPU_PAGE_SIZE);
 
         for (i = 0; i < pages; i++, p++) {
-                rdev->gart.pages_addr[p] = dma_addr[i];
                 rdev->gart.pages[p] = pagelist[i];
-                if (rdev->gart.ptr) {
-                        page_base = rdev->gart.pages_addr[p];
-                        for (j = 0; j < (PAGE_SIZE / RADEON_GPU_PAGE_SIZE); j++, t++) {
-                                radeon_gart_set_page(rdev, t, page_base, flags);
-                                page_base += RADEON_GPU_PAGE_SIZE;
+                page_base = dma_addr[i];
+                for (j = 0; j < (PAGE_SIZE / RADEON_GPU_PAGE_SIZE); j++, t++) {
+                        page_entry = radeon_gart_get_page_entry(page_base, flags);
+                        rdev->gart.pages_entry[t] = page_entry;
+                        if (rdev->gart.ptr) {
+                                radeon_gart_set_page(rdev, t, page_entry);
                         }
+                        page_base += RADEON_GPU_PAGE_SIZE;
                 }
         }
         mb();
@@ -334,16 +345,15 @@ int radeon_gart_init(struct radeon_device *rdev)
                 radeon_gart_fini(rdev);
                 return -ENOMEM;
         }
-        rdev->gart.pages_addr = vzalloc(sizeof(dma_addr_t) *
-                                        rdev->gart.num_cpu_pages);
-        if (rdev->gart.pages_addr == NULL) {
+        rdev->gart.pages_entry = vmalloc(sizeof(uint64_t) *
+                                         rdev->gart.num_gpu_pages);
+        if (rdev->gart.pages_entry == NULL) {
                 radeon_gart_fini(rdev);
                 return -ENOMEM;
         }
         /* set GART entry to point to the dummy page by default */
-        for (i = 0; i < rdev->gart.num_cpu_pages; i++) {
-                rdev->gart.pages_addr[i] = rdev->dummy_page.addr;
-        }
+        for (i = 0; i < rdev->gart.num_gpu_pages; i++)
+                rdev->gart.pages_entry[i] = rdev->dummy_page.entry;
         return 0;
 }
 
@@ -356,15 +366,15 @@ int radeon_gart_init(struct radeon_device *rdev)
  */
 void radeon_gart_fini(struct radeon_device *rdev)
 {
-        if (rdev->gart.pages && rdev->gart.pages_addr && rdev->gart.ready) {
+        if (rdev->gart.ready) {
                 /* unbind pages */
                 radeon_gart_unbind(rdev, 0, rdev->gart.num_cpu_pages);
         }
         rdev->gart.ready = false;
         vfree(rdev->gart.pages);
-        vfree(rdev->gart.pages_addr);
+        vfree(rdev->gart.pages_entry);
         rdev->gart.pages = NULL;
-        rdev->gart.pages_addr = NULL;
+        rdev->gart.pages_entry = NULL;
 
         radeon_dummy_page_fini(rdev);
 }
diff --git a/drivers/gpu/drm/radeon/radeon_kfd.c b/drivers/gpu/drm/radeon/radeon_kfd.c
index 7b274205eeaf..061eaa9c19c7 100644
--- a/drivers/gpu/drm/radeon/radeon_kfd.c
+++ b/drivers/gpu/drm/radeon/radeon_kfd.c
@@ -392,7 +392,7 @@ static int kgd_set_pasid_vmid_mapping(struct kgd_dev *kgd, unsigned int pasid,
 static int kgd_init_pipeline(struct kgd_dev *kgd, uint32_t pipe_id,
                                 uint32_t hpd_size, uint64_t hpd_gpu_addr)
 {
-        uint32_t mec = (++pipe_id / CIK_PIPE_PER_MEC) + 1;
+        uint32_t mec = (pipe_id / CIK_PIPE_PER_MEC) + 1;
         uint32_t pipe = (pipe_id % CIK_PIPE_PER_MEC);
 
         lock_srbm(kgd, mec, pipe, 0, 0);
diff --git a/drivers/gpu/drm/radeon/radeon_vm.c b/drivers/gpu/drm/radeon/radeon_vm.c
index cde48c42b30a..06d2246d07f1 100644
--- a/drivers/gpu/drm/radeon/radeon_vm.c
+++ b/drivers/gpu/drm/radeon/radeon_vm.c
@@ -587,10 +587,8 @@ uint64_t radeon_vm_map_gart(struct radeon_device *rdev, uint64_t addr)
         uint64_t result;
 
         /* page table offset */
-        result = rdev->gart.pages_addr[addr >> PAGE_SHIFT];
-
-        /* in case cpu page size != gpu page size*/
-        result |= addr & (~PAGE_MASK);
+        result = rdev->gart.pages_entry[addr >> RADEON_GPU_PAGE_SHIFT];
+        result &= ~RADEON_GPU_PAGE_MASK;
 
         return result;
 }
diff --git a/drivers/gpu/drm/radeon/rs400.c b/drivers/gpu/drm/radeon/rs400.c
index c5799f16aa4b..34e3235f41d2 100644
--- a/drivers/gpu/drm/radeon/rs400.c
+++ b/drivers/gpu/drm/radeon/rs400.c
@@ -212,11 +212,9 @@ void rs400_gart_fini(struct radeon_device *rdev)
 #define RS400_PTE_WRITEABLE (1 << 2)
 #define RS400_PTE_READABLE  (1 << 3)
 
-void rs400_gart_set_page(struct radeon_device *rdev, unsigned i,
-                         uint64_t addr, uint32_t flags)
+uint64_t rs400_gart_get_page_entry(uint64_t addr, uint32_t flags)
 {
         uint32_t entry;
-        u32 *gtt = rdev->gart.ptr;
 
         entry = (lower_32_bits(addr) & PAGE_MASK) |
                 ((upper_32_bits(addr) & 0xff) << 4);
@@ -226,8 +224,14 @@ void rs400_gart_set_page(struct radeon_device *rdev, unsigned i,
                 entry |= RS400_PTE_WRITEABLE;
         if (!(flags & RADEON_GART_PAGE_SNOOP))
                 entry |= RS400_PTE_UNSNOOPED;
-        entry = cpu_to_le32(entry);
-        gtt[i] = entry;
+        return entry;
+}
+
+void rs400_gart_set_page(struct radeon_device *rdev, unsigned i,
+                         uint64_t entry)
+{
+        u32 *gtt = rdev->gart.ptr;
+        gtt[i] = cpu_to_le32(lower_32_bits(entry));
 }
 
 int rs400_mc_wait_for_idle(struct radeon_device *rdev)
diff --git a/drivers/gpu/drm/radeon/rs600.c b/drivers/gpu/drm/radeon/rs600.c
index 68f154a451c0..d81182ad53ec 100644
--- a/drivers/gpu/drm/radeon/rs600.c
+++ b/drivers/gpu/drm/radeon/rs600.c
@@ -626,11 +626,8 @@ static void rs600_gart_fini(struct radeon_device *rdev)
         radeon_gart_table_vram_free(rdev);
 }
 
-void rs600_gart_set_page(struct radeon_device *rdev, unsigned i,
-                         uint64_t addr, uint32_t flags)
+uint64_t rs600_gart_get_page_entry(uint64_t addr, uint32_t flags)
 {
-        void __iomem *ptr = (void *)rdev->gart.ptr;
-
         addr = addr & 0xFFFFFFFFFFFFF000ULL;
         addr |= R600_PTE_SYSTEM;
         if (flags & RADEON_GART_PAGE_VALID)
@@ -641,7 +638,14 @@ void rs600_gart_set_page(struct radeon_device *rdev, unsigned i,
                 addr |= R600_PTE_WRITEABLE;
         if (flags & RADEON_GART_PAGE_SNOOP)
                 addr |= R600_PTE_SNOOPED;
-        writeq(addr, ptr + (i * 8));
+        return addr;
+}
+
+void rs600_gart_set_page(struct radeon_device *rdev, unsigned i,
+                         uint64_t entry)
+{
+        void __iomem *ptr = (void *)rdev->gart.ptr;
+        writeq(entry, ptr + (i * 8));
 }
 
 int rs600_irq_set(struct radeon_device *rdev)
diff --git a/drivers/gpu/drm/radeon/si_dma.c b/drivers/gpu/drm/radeon/si_dma.c
index aa7b872b2c43..83207929fc62 100644
--- a/drivers/gpu/drm/radeon/si_dma.c
+++ b/drivers/gpu/drm/radeon/si_dma.c
@@ -123,7 +123,6 @@ void si_dma_vm_write_pages(struct radeon_device *rdev,
                 for (; ndw > 0; ndw -= 2, --count, pe += 8) {
                         if (flags & R600_PTE_SYSTEM) {
                                 value = radeon_vm_map_gart(rdev, addr);
-                                value &= 0xFFFFFFFFFFFFF000ULL;
                         } else if (flags & R600_PTE_VALID) {
                                 value = addr;
                         } else {