Merge branch 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc

* 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc: (34 commits)
  powerpc: Wireup new syscalls
  Move update_mmu_cache() declaration from tlbflush.h to pgtable.h
  powerpc/pseries: Remove kmalloc call in handling writes to lparcfg
  powerpc/pseries: Update arch vector to indicate support for CMO
  ibmvfc: Add support for collaborative memory overcommit
  ibmvscsi: driver enablement for CMO
  ibmveth: enable driver for CMO
  ibmveth: Automatically enable larger rx buffer pools for larger mtu
  powerpc/pseries: Verify CMO memory entitlement updates with virtual I/O
  powerpc/pseries: vio bus support for CMO
  powerpc/pseries: iommu enablement for CMO
  powerpc/pseries: Add CMO paging statistics
  powerpc/pseries: Add collaborative memory manager
  powerpc/pseries: Utilities to set firmware page state
  powerpc/pseries: Enable CMO feature during platform setup
  powerpc/pseries: Split retrieval of processor entitlement data into a helper routine
  powerpc/pseries: Add memory entitlement capabilities to /proc/ppc64/lparcfg
  powerpc/pseries: Split processor entitlement retrieval and gathering to helper routines
  powerpc/pseries: Remove extraneous error reporting for hcall failures in lparcfg
  powerpc: Fix compile error with binutils 2.15
  ...

Fixed up conflict in arch/powerpc/platforms/52xx/Kconfig manually.

12 files changed, 648 insertions, 29 deletions

diff --git a/arch/powerpc/platforms/52xx/Kconfig b/arch/powerpc/platforms/52xx/Kconfig
index ccbd4958412e..696a5ee4962d 100644
--- a/arch/powerpc/platforms/52xx/Kconfig
+++ b/arch/powerpc/platforms/52xx/Kconfig
@@ -1,7 +1,6 @@
 config PPC_MPC52xx
         bool "52xx-based boards"
         depends on PPC_MULTIPLATFORM && PPC32
-        select FSL_SOC
         select PPC_CLOCK
         select PPC_PCI_CHOICE
 
@@ -49,5 +48,6 @@ config PPC_MPC5200_GPIO
         bool "MPC5200 GPIO support"
         depends on PPC_MPC52xx
         select ARCH_REQUIRE_GPIOLIB
+        select GENERIC_GPIO
         help
           Enable gpiolib support for mpc5200 based boards
diff --git a/arch/powerpc/platforms/cell/iommu.c b/arch/powerpc/platforms/cell/iommu.c
index 208005ca262c..e06420af5fe9 100644
--- a/arch/powerpc/platforms/cell/iommu.c
+++ b/arch/powerpc/platforms/cell/iommu.c
@@ -172,7 +172,7 @@ static void invalidate_tce_cache(struct cbe_iommu *iommu, unsigned long *pte,
         }
 }
 
-static void tce_build_cell(struct iommu_table *tbl, long index, long npages,
+static int tce_build_cell(struct iommu_table *tbl, long index, long npages,
                 unsigned long uaddr, enum dma_data_direction direction,
                 struct dma_attrs *attrs)
 {
@@ -213,6 +213,7 @@ static void tce_build_cell(struct iommu_table *tbl, long index, long npages,
 
         pr_debug("tce_build_cell(index=%lx,n=%lx,dir=%d,base_pte=%lx)\n",
                  index, npages, direction, base_pte);
+        return 0;
 }
 
 static void tce_free_cell(struct iommu_table *tbl, long index, long npages)
@@ -1150,12 +1151,23 @@ static int iommu_fixed_disabled;
 
 static int __init setup_iommu_fixed(char *str)
 {
+        struct device_node *pciep;
+
         if (strcmp(str, "off") == 0)
                 iommu_fixed_disabled = 1;
 
-        else if (strcmp(str, "weak") == 0)
+        /* If we can find a pcie-endpoint in the device tree assume that
+         * we're on a triblade or a CAB so by default the fixed mapping
+         * should be set to be weakly ordered; but only if the boot
+         * option WASN'T set for strong ordering
+         */
+        pciep = of_find_node_by_type(NULL, "pcie-endpoint");
+
+        if (strcmp(str, "weak") == 0 || (pciep && strcmp(str, "strong") != 0))
                 iommu_fixed_is_weak = 1;
 
+        of_node_put(pciep);
+
         return 1;
 }
 __setup("iommu_fixed=", setup_iommu_fixed);
diff --git a/arch/powerpc/platforms/cell/spufs/sched.c b/arch/powerpc/platforms/cell/spufs/sched.c
index 34654743363d..2deeeba7eccf 100644
--- a/arch/powerpc/platforms/cell/spufs/sched.c
+++ b/arch/powerpc/platforms/cell/spufs/sched.c
@@ -312,11 +312,28 @@ static struct spu *aff_ref_location(struct spu_context *ctx, int mem_aff,
          */
         node = cpu_to_node(raw_smp_processor_id());
         for (n = 0; n < MAX_NUMNODES; n++, node++) {
+                int available_spus;
+
                 node = (node < MAX_NUMNODES) ? node : 0;
                 if (!node_allowed(ctx, node))
                         continue;
+
+                available_spus = 0;
                 mutex_lock(&cbe_spu_info[node].list_mutex);
                 list_for_each_entry(spu, &cbe_spu_info[node].spus, cbe_list) {
+                        if (spu->ctx && spu->ctx->gang
+                                        && spu->ctx->aff_offset == 0)
+                                available_spus -=
+                                        (spu->ctx->gang->contexts - 1);
+                        else
+                                available_spus++;
+                }
+                if (available_spus < ctx->gang->contexts) {
+                        mutex_unlock(&cbe_spu_info[node].list_mutex);
+                        continue;
+                }
+
+                list_for_each_entry(spu, &cbe_spu_info[node].spus, cbe_list) {
                         if ((!mem_aff || spu->has_mem_affinity) &&
                                                         sched_spu(spu)) {
                                 mutex_unlock(&cbe_spu_info[node].list_mutex);
@@ -389,6 +406,9 @@ static int has_affinity(struct spu_context *ctx)
         if (list_empty(&ctx->aff_list))
                 return 0;
 
+        if (atomic_read(&ctx->gang->aff_sched_count) == 0)
+                ctx->gang->aff_ref_spu = NULL;
+
         if (!gang->aff_ref_spu) {
                 if (!(gang->aff_flags & AFF_MERGED))
                         aff_merge_remaining_ctxs(gang);
@@ -416,14 +436,8 @@ static void spu_unbind_context(struct spu *spu, struct spu_context *ctx)
         if (spu->ctx->flags & SPU_CREATE_NOSCHED)
                 atomic_dec(&cbe_spu_info[spu->node].reserved_spus);
 
-        if (ctx->gang){
-                mutex_lock(&ctx->gang->aff_mutex);
-                if (has_affinity(ctx)) {
-                        if (atomic_dec_and_test(&ctx->gang->aff_sched_count))
-                                ctx->gang->aff_ref_spu = NULL;
-                }
-                mutex_unlock(&ctx->gang->aff_mutex);
-        }
+        if (ctx->gang)
+                atomic_dec_if_positive(&ctx->gang->aff_sched_count);
 
         spu_switch_notify(spu, NULL);
         spu_unmap_mappings(ctx);
@@ -562,10 +576,7 @@ static struct spu *spu_get_idle(struct spu_context *ctx)
                                 goto found;
                         mutex_unlock(&cbe_spu_info[node].list_mutex);
 
-                        mutex_lock(&ctx->gang->aff_mutex);
-                        if (atomic_dec_and_test(&ctx->gang->aff_sched_count))
-                                ctx->gang->aff_ref_spu = NULL;
-                        mutex_unlock(&ctx->gang->aff_mutex);
+                        atomic_dec(&ctx->gang->aff_sched_count);
                         goto not_found;
                 }
                 mutex_unlock(&ctx->gang->aff_mutex);
diff --git a/arch/powerpc/platforms/cell/spufs/sputrace.c b/arch/powerpc/platforms/cell/spufs/sputrace.c
index 8c0e95766a62..92d20e993ede 100644
--- a/arch/powerpc/platforms/cell/spufs/sputrace.c
+++ b/arch/powerpc/platforms/cell/spufs/sputrace.c
@@ -196,8 +196,7 @@ static int __init sputrace_init(void)
         struct proc_dir_entry *entry;
         int i, error = -ENOMEM;
 
-        sputrace_log = kcalloc(sizeof(struct sputrace),
-                                bufsize, GFP_KERNEL);
+        sputrace_log = kcalloc(bufsize, sizeof(struct sputrace), GFP_KERNEL);
         if (!sputrace_log)
                 goto out;
 
diff --git a/arch/powerpc/platforms/iseries/iommu.c b/arch/powerpc/platforms/iseries/iommu.c
index bc818e4e2033..bb464d1211b2 100644
--- a/arch/powerpc/platforms/iseries/iommu.c
+++ b/arch/powerpc/platforms/iseries/iommu.c
@@ -41,7 +41,7 @@
 #include <asm/iseries/hv_call_event.h>
 #include <asm/iseries/iommu.h>
 
-static void tce_build_iSeries(struct iommu_table *tbl, long index, long npages,
+static int tce_build_iSeries(struct iommu_table *tbl, long index, long npages,
                 unsigned long uaddr, enum dma_data_direction direction,
                 struct dma_attrs *attrs)
 {
@@ -71,6 +71,7 @@ static void tce_build_iSeries(struct iommu_table *tbl, long index, long npages,
                 index++;
                 uaddr += TCE_PAGE_SIZE;
         }
+        return 0;
 }
 
 static void tce_free_iSeries(struct iommu_table *tbl, long index, long npages)
diff --git a/arch/powerpc/platforms/pasemi/iommu.c b/arch/powerpc/platforms/pasemi/iommu.c
index 70541b7a5013..a0ff03a3d8da 100644
--- a/arch/powerpc/platforms/pasemi/iommu.c
+++ b/arch/powerpc/platforms/pasemi/iommu.c
@@ -83,7 +83,7 @@ static u32 *iob_l2_base;
 static struct iommu_table iommu_table_iobmap;
 static int iommu_table_iobmap_inited;
 
-static void iobmap_build(struct iommu_table *tbl, long index,
+static int iobmap_build(struct iommu_table *tbl, long index,
                          long npages, unsigned long uaddr,
                          enum dma_data_direction direction,
                          struct dma_attrs *attrs)
@@ -108,6 +108,7 @@ static void iobmap_build(struct iommu_table *tbl, long index,
                 uaddr += IOBMAP_PAGE_SIZE;
                 bus_addr += IOBMAP_PAGE_SIZE;
         }
+        return 0;
 }
 
 
diff --git a/arch/powerpc/platforms/pseries/Kconfig b/arch/powerpc/platforms/pseries/Kconfig
index 757c0296e0b8..97619fd51e39 100644
--- a/arch/powerpc/platforms/pseries/Kconfig
+++ b/arch/powerpc/platforms/pseries/Kconfig
@@ -40,3 +40,26 @@ config PPC_PSERIES_DEBUG
         depends on PPC_PSERIES && PPC_EARLY_DEBUG
         bool "Enable extra debug logging in platforms/pseries"
         default y
+
+config PPC_SMLPAR
+        bool "Support for shared-memory logical partitions"
+        depends on PPC_PSERIES
+        select LPARCFG
+        default n
+        help
+          Select this option to enable shared memory partition support.
+          With this option a system running in an LPAR can be given more
+          memory than physically available and will allow firmware to
+          balance memory across many LPARs.
+
+config CMM
+        tristate "Collaborative memory management"
+        depends on PPC_SMLPAR
+        default y
+        help
+          Select this option, if you want to enable the kernel interface
+          to reduce the memory size of the system. This is accomplished
+          by allocating pages of memory and put them "on hold". This only
+          makes sense for a system running in an LPAR where the unused pages
+          will be reused for other LPARs. The interface allows firmware to
+          balance memory across many LPARs.
diff --git a/arch/powerpc/platforms/pseries/Makefile b/arch/powerpc/platforms/pseries/Makefile
index 554c6e42ef2a..dfe574af2dc0 100644
--- a/arch/powerpc/platforms/pseries/Makefile
+++ b/arch/powerpc/platforms/pseries/Makefile
@@ -24,3 +24,4 @@ obj-$(CONFIG_HVC_CONSOLE)	+= hvconsole.o
 obj-$(CONFIG_HVCS)              += hvcserver.o
 obj-$(CONFIG_HCALL_STATS)       += hvCall_inst.o
 obj-$(CONFIG_PHYP_DUMP) += phyp_dump.o
+obj-$(CONFIG_CMM)               += cmm.o
diff --git a/arch/powerpc/platforms/pseries/cmm.c b/arch/powerpc/platforms/pseries/cmm.c
new file mode 100644
index 000000000000..c6b3be03168b
--- /dev/null
+++ b/arch/powerpc/platforms/pseries/cmm.c
@@ -0,0 +1,468 @@
+/*
+ * Collaborative memory management interface.
+ *
+ * Copyright (C) 2008 IBM Corporation
+ * Author(s): Brian King (brking@linux.vnet.ibm.com),
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ */
+
+#include <linux/ctype.h>
+#include <linux/delay.h>
+#include <linux/errno.h>
+#include <linux/fs.h>
+#include <linux/init.h>
+#include <linux/kthread.h>
+#include <linux/module.h>
+#include <linux/oom.h>
+#include <linux/sched.h>
+#include <linux/stringify.h>
+#include <linux/swap.h>
+#include <linux/sysdev.h>
+#include <asm/firmware.h>
+#include <asm/hvcall.h>
+#include <asm/mmu.h>
+#include <asm/pgalloc.h>
+#include <asm/uaccess.h>
+
+#include "plpar_wrappers.h"
+
+#define CMM_DRIVER_VERSION      "1.0.0"
+#define CMM_DEFAULT_DELAY       1
+#define CMM_DEBUG                       0
+#define CMM_DISABLE             0
+#define CMM_OOM_KB              1024
+#define CMM_MIN_MEM_MB          256
+#define KB2PAGES(_p)            ((_p)>>(PAGE_SHIFT-10))
+#define PAGES2KB(_p)            ((_p)<<(PAGE_SHIFT-10))
+
+static unsigned int delay = CMM_DEFAULT_DELAY;
+static unsigned int oom_kb = CMM_OOM_KB;
+static unsigned int cmm_debug = CMM_DEBUG;
+static unsigned int cmm_disabled = CMM_DISABLE;
+static unsigned long min_mem_mb = CMM_MIN_MEM_MB;
+static struct sys_device cmm_sysdev;
+
+MODULE_AUTHOR("Brian King <brking@linux.vnet.ibm.com>");
+MODULE_DESCRIPTION("IBM System p Collaborative Memory Manager");
+MODULE_LICENSE("GPL");
+MODULE_VERSION(CMM_DRIVER_VERSION);
+
+module_param_named(delay, delay, uint, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(delay, "Delay (in seconds) between polls to query hypervisor paging requests. "
+                 "[Default=" __stringify(CMM_DEFAULT_DELAY) "]");
+module_param_named(oom_kb, oom_kb, uint, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(oom_kb, "Amount of memory in kb to free on OOM. "
+                 "[Default=" __stringify(CMM_OOM_KB) "]");
+module_param_named(min_mem_mb, min_mem_mb, ulong, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(min_mem_mb, "Minimum amount of memory (in MB) to not balloon. "
+                 "[Default=" __stringify(CMM_MIN_MEM_MB) "]");
+module_param_named(debug, cmm_debug, uint, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(debug, "Enable module debugging logging. Set to 1 to enable. "
+                 "[Default=" __stringify(CMM_DEBUG) "]");
+
+#define CMM_NR_PAGES ((PAGE_SIZE - sizeof(void *) - sizeof(unsigned long)) / sizeof(unsigned long))
+
+#define cmm_dbg(...) if (cmm_debug) { printk(KERN_INFO "cmm: "__VA_ARGS__); }
+
+struct cmm_page_array {
+        struct cmm_page_array *next;
+        unsigned long index;
+        unsigned long page[CMM_NR_PAGES];
+};
+
+static unsigned long loaned_pages;
+static unsigned long loaned_pages_target;
+static unsigned long oom_freed_pages;
+
+static struct cmm_page_array *cmm_page_list;
+static DEFINE_SPINLOCK(cmm_lock);
+
+static struct task_struct *cmm_thread_ptr;
+
+/**
+ * cmm_alloc_pages - Allocate pages and mark them as loaned
+ * @nr: number of pages to allocate
+ *
+ * Return value:
+ *      number of pages requested to be allocated which were not
+ **/
+static long cmm_alloc_pages(long nr)
+{
+        struct cmm_page_array *pa, *npa;
+        unsigned long addr;
+        long rc;
+
+        cmm_dbg("Begin request for %ld pages\n", nr);
+
+        while (nr) {
+                addr = __get_free_page(GFP_NOIO | __GFP_NOWARN |
+                                       __GFP_NORETRY | __GFP_NOMEMALLOC);
+                if (!addr)
+                        break;
+                spin_lock(&cmm_lock);
+                pa = cmm_page_list;
+                if (!pa || pa->index >= CMM_NR_PAGES) {
+                        /* Need a new page for the page list. */
+                        spin_unlock(&cmm_lock);
+                        npa = (struct cmm_page_array *)__get_free_page(GFP_NOIO | __GFP_NOWARN |
+                                                                       __GFP_NORETRY | __GFP_NOMEMALLOC);
+                        if (!npa) {
+                                pr_info("%s: Can not allocate new page list\n", __FUNCTION__);
+                                free_page(addr);
+                                break;
+                        }
+                        spin_lock(&cmm_lock);
+                        pa = cmm_page_list;
+
+                        if (!pa || pa->index >= CMM_NR_PAGES) {
+                                npa->next = pa;
+                                npa->index = 0;
+                                pa = npa;
+                                cmm_page_list = pa;
+                        } else
+                                free_page((unsigned long) npa);
+                }
+
+                if ((rc = plpar_page_set_loaned(__pa(addr)))) {
+                        pr_err("%s: Can not set page to loaned. rc=%ld\n", __FUNCTION__, rc);
+                        spin_unlock(&cmm_lock);
+                        free_page(addr);
+                        break;
+                }
+
+                pa->page[pa->index++] = addr;
+                loaned_pages++;
+                totalram_pages--;
+                spin_unlock(&cmm_lock);
+                nr--;
+        }
+
+        cmm_dbg("End request with %ld pages unfulfilled\n", nr);
+        return nr;
+}
+
+/**
+ * cmm_free_pages - Free pages and mark them as active
+ * @nr: number of pages to free
+ *
+ * Return value:
+ *      number of pages requested to be freed which were not
+ **/
+static long cmm_free_pages(long nr)
+{
+        struct cmm_page_array *pa;
+        unsigned long addr;
+
+        cmm_dbg("Begin free of %ld pages.\n", nr);
+        spin_lock(&cmm_lock);
+        pa = cmm_page_list;
+        while (nr) {
+                if (!pa || pa->index <= 0)
+                        break;
+                addr = pa->page[--pa->index];
+
+                if (pa->index == 0) {
+                        pa = pa->next;
+                        free_page((unsigned long) cmm_page_list);
+                        cmm_page_list = pa;
+                }
+
+                plpar_page_set_active(__pa(addr));
+                free_page(addr);
+                loaned_pages--;
+                nr--;
+                totalram_pages++;
+        }
+        spin_unlock(&cmm_lock);
+        cmm_dbg("End request with %ld pages unfulfilled\n", nr);
+        return nr;
+}
+
+/**
+ * cmm_oom_notify - OOM notifier
+ * @self:       notifier block struct
+ * @dummy:      not used
+ * @parm:       returned - number of pages freed
+ *
+ * Return value:
+ *      NOTIFY_OK
+ **/
+static int cmm_oom_notify(struct notifier_block *self,
+                          unsigned long dummy, void *parm)
+{
+        unsigned long *freed = parm;
+        long nr = KB2PAGES(oom_kb);
+
+        cmm_dbg("OOM processing started\n");
+        nr = cmm_free_pages(nr);
+        loaned_pages_target = loaned_pages;
+        *freed += KB2PAGES(oom_kb) - nr;
+        oom_freed_pages += KB2PAGES(oom_kb) - nr;
+        cmm_dbg("OOM processing complete\n");
+        return NOTIFY_OK;
+}
+
+/**
+ * cmm_get_mpp - Read memory performance parameters
+ *
+ * Makes hcall to query the current page loan request from the hypervisor.
+ *
+ * Return value:
+ *      nothing
+ **/
+static void cmm_get_mpp(void)
+{
+        int rc;
+        struct hvcall_mpp_data mpp_data;
+        unsigned long active_pages_target;
+        signed long page_loan_request;
+
+        rc = h_get_mpp(&mpp_data);
+
+        if (rc != H_SUCCESS)
+                return;
+
+        page_loan_request = div_s64((s64)mpp_data.loan_request, PAGE_SIZE);
+        loaned_pages_target = page_loan_request + loaned_pages;
+        if (loaned_pages_target > oom_freed_pages)
+                loaned_pages_target -= oom_freed_pages;
+        else
+                loaned_pages_target = 0;
+
+        active_pages_target = totalram_pages + loaned_pages - loaned_pages_target;
+
+        if ((min_mem_mb * 1024 * 1024) > (active_pages_target * PAGE_SIZE))
+                loaned_pages_target = totalram_pages + loaned_pages -
+                        ((min_mem_mb * 1024 * 1024) / PAGE_SIZE);
+
+        cmm_dbg("delta = %ld, loaned = %lu, target = %lu, oom = %lu, totalram = %lu\n",
+                page_loan_request, loaned_pages, loaned_pages_target,
+                oom_freed_pages, totalram_pages);
+}
+
+static struct notifier_block cmm_oom_nb = {
+        .notifier_call = cmm_oom_notify
+};
+
+/**
+ * cmm_thread - CMM task thread
+ * @dummy:      not used
+ *
+ * Return value:
+ *      0
+ **/
+static int cmm_thread(void *dummy)
+{
+        unsigned long timeleft;
+
+        while (1) {
+                timeleft = msleep_interruptible(delay * 1000);
+
+                if (kthread_should_stop() || timeleft) {
+                        loaned_pages_target = loaned_pages;
+                        break;
+                }
+
+                cmm_get_mpp();
+
+                if (loaned_pages_target > loaned_pages) {
+                        if (cmm_alloc_pages(loaned_pages_target - loaned_pages))
+                                loaned_pages_target = loaned_pages;
+                } else if (loaned_pages_target < loaned_pages)
+                        cmm_free_pages(loaned_pages - loaned_pages_target);
+        }
+        return 0;
+}
+
+#define CMM_SHOW(name, format, args...)                 \
+        static ssize_t show_##name(struct sys_device *dev, char *buf)   \
+        {                                                       \
+                return sprintf(buf, format, ##args);            \
+        }                                                       \
+        static SYSDEV_ATTR(name, S_IRUGO, show_##name, NULL)
+
+CMM_SHOW(loaned_kb, "%lu\n", PAGES2KB(loaned_pages));
+CMM_SHOW(loaned_target_kb, "%lu\n", PAGES2KB(loaned_pages_target));
+
+static ssize_t show_oom_pages(struct sys_device *dev, char *buf)
+{
+        return sprintf(buf, "%lu\n", PAGES2KB(oom_freed_pages));
+}
+
+static ssize_t store_oom_pages(struct sys_device *dev,
+                               const char *buf, size_t count)
+{
+        unsigned long val = simple_strtoul (buf, NULL, 10);
+
+        if (!capable(CAP_SYS_ADMIN))
+                return -EPERM;
+        if (val != 0)
+                return -EBADMSG;
+
+        oom_freed_pages = 0;
+        return count;
+}
+
+static SYSDEV_ATTR(oom_freed_kb, S_IWUSR| S_IRUGO,
+                   show_oom_pages, store_oom_pages);
+
+static struct sysdev_attribute *cmm_attrs[] = {
+        &attr_loaned_kb,
+        &attr_loaned_target_kb,
+        &attr_oom_freed_kb,
+};
+
+static struct sysdev_class cmm_sysdev_class = {
+        .name = "cmm",
+};
+
+/**
+ * cmm_sysfs_register - Register with sysfs
+ *
+ * Return value:
+ *      0 on success / other on failure
+ **/
+static int cmm_sysfs_register(struct sys_device *sysdev)
+{
+        int i, rc;
+
+        if ((rc = sysdev_class_register(&cmm_sysdev_class)))
+                return rc;
+
+        sysdev->id = 0;
+        sysdev->cls = &cmm_sysdev_class;
+
+        if ((rc = sysdev_register(sysdev)))
+                goto class_unregister;
+
+        for (i = 0; i < ARRAY_SIZE(cmm_attrs); i++) {
+                if ((rc = sysdev_create_file(sysdev, cmm_attrs[i])))
+                        goto fail;
+        }
+
+        return 0;
+
+fail:
+        while (--i >= 0)
+                sysdev_remove_file(sysdev, cmm_attrs[i]);
+        sysdev_unregister(sysdev);
+class_unregister:
+        sysdev_class_unregister(&cmm_sysdev_class);
+        return rc;
+}
+
+/**
+ * cmm_unregister_sysfs - Unregister from sysfs
+ *
+ **/
+static void cmm_unregister_sysfs(struct sys_device *sysdev)
+{
+        int i;
+
+        for (i = 0; i < ARRAY_SIZE(cmm_attrs); i++)
+                sysdev_remove_file(sysdev, cmm_attrs[i]);
+        sysdev_unregister(sysdev);
+        sysdev_class_unregister(&cmm_sysdev_class);
+}
+
+/**
+ * cmm_init - Module initialization
+ *
+ * Return value:
+ *      0 on success / other on failure
+ **/
+static int cmm_init(void)
+{
+        int rc = -ENOMEM;
+
+        if (!firmware_has_feature(FW_FEATURE_CMO))
+                return -EOPNOTSUPP;
+
+        if ((rc = register_oom_notifier(&cmm_oom_nb)) < 0)
+                return rc;
+
+        if ((rc = cmm_sysfs_register(&cmm_sysdev)))
+                goto out_oom_notifier;
+
+        if (cmm_disabled)
+                return rc;
+
+        cmm_thread_ptr = kthread_run(cmm_thread, NULL, "cmmthread");
+        if (IS_ERR(cmm_thread_ptr)) {
+                rc = PTR_ERR(cmm_thread_ptr);
+                goto out_unregister_sysfs;
+        }
+
+        return rc;
+
+out_unregister_sysfs:
+        cmm_unregister_sysfs(&cmm_sysdev);
+out_oom_notifier:
+        unregister_oom_notifier(&cmm_oom_nb);
+        return rc;
+}
+
+/**
+ * cmm_exit - Module exit
+ *
+ * Return value:
+ *      nothing
+ **/
+static void cmm_exit(void)
+{
+        if (cmm_thread_ptr)
+                kthread_stop(cmm_thread_ptr);
+        unregister_oom_notifier(&cmm_oom_nb);
+        cmm_free_pages(loaned_pages);
+        cmm_unregister_sysfs(&cmm_sysdev);
+}
+
+/**
+ * cmm_set_disable - Disable/Enable CMM
+ *
+ * Return value:
+ *      0 on success / other on failure
+ **/
+static int cmm_set_disable(const char *val, struct kernel_param *kp)
+{
+        int disable = simple_strtoul(val, NULL, 10);
+
+        if (disable != 0 && disable != 1)
+                return -EINVAL;
+
+        if (disable && !cmm_disabled) {
+                if (cmm_thread_ptr)
+                        kthread_stop(cmm_thread_ptr);
+                cmm_thread_ptr = NULL;
+                cmm_free_pages(loaned_pages);
+        } else if (!disable && cmm_disabled) {
+                cmm_thread_ptr = kthread_run(cmm_thread, NULL, "cmmthread");
+                if (IS_ERR(cmm_thread_ptr))
+                        return PTR_ERR(cmm_thread_ptr);
+        }
+
+        cmm_disabled = disable;
+        return 0;
+}
+
+module_param_call(disable, cmm_set_disable, param_get_uint,
+                  &cmm_disabled, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(disable, "Disable CMM. Set to 1 to disable. "
+                 "[Default=" __stringify(CMM_DISABLE) "]");
+
+module_init(cmm_init);
+module_exit(cmm_exit);
diff --git a/arch/powerpc/platforms/pseries/iommu.c b/arch/powerpc/platforms/pseries/iommu.c
index 5377dd4b849a..a8c446697f9e 100644
--- a/arch/powerpc/platforms/pseries/iommu.c
+++ b/arch/powerpc/platforms/pseries/iommu.c
@@ -48,7 +48,7 @@
 #include "plpar_wrappers.h"
 
 
-static void tce_build_pSeries(struct iommu_table *tbl, long index,
+static int tce_build_pSeries(struct iommu_table *tbl, long index,
                               long npages, unsigned long uaddr,
                               enum dma_data_direction direction,
                               struct dma_attrs *attrs)
@@ -72,6 +72,7 @@ static void tce_build_pSeries(struct iommu_table *tbl, long index,
                 uaddr += TCE_PAGE_SIZE;
                 tcep++;
         }
+        return 0;
 }
 
 
@@ -94,14 +95,19 @@ static unsigned long tce_get_pseries(struct iommu_table *tbl, long index)
         return *tcep;
 }
 
-static void tce_build_pSeriesLP(struct iommu_table *tbl, long tcenum,
+static void tce_free_pSeriesLP(struct iommu_table*, long, long);
+static void tce_freemulti_pSeriesLP(struct iommu_table*, long, long);
+
+static int tce_build_pSeriesLP(struct iommu_table *tbl, long tcenum,
                                 long npages, unsigned long uaddr,
                                 enum dma_data_direction direction,
                                 struct dma_attrs *attrs)
 {
-        u64 rc;
+        u64 rc = 0;
         u64 proto_tce, tce;
         u64 rpn;
+        int ret = 0;
+        long tcenum_start = tcenum, npages_start = npages;
 
         rpn = (virt_to_abs(uaddr)) >> TCE_SHIFT;
         proto_tce = TCE_PCI_READ;
@@ -112,6 +118,13 @@ static void tce_build_pSeriesLP(struct iommu_table *tbl, long tcenum,
                 tce = proto_tce | (rpn & TCE_RPN_MASK) << TCE_RPN_SHIFT;
                 rc = plpar_tce_put((u64)tbl->it_index, (u64)tcenum << 12, tce);
 
+                if (unlikely(rc == H_NOT_ENOUGH_RESOURCES)) {
+                        ret = (int)rc;
+                        tce_free_pSeriesLP(tbl, tcenum_start,
+                                           (npages_start - (npages + 1)));
+                        break;
+                }
+
                 if (rc && printk_ratelimit()) {
                         printk("tce_build_pSeriesLP: plpar_tce_put failed. rc=%ld\n", rc);
                         printk("\tindex   = 0x%lx\n", (u64)tbl->it_index);
@@ -123,25 +136,27 @@ static void tce_build_pSeriesLP(struct iommu_table *tbl, long tcenum,
                 tcenum++;
                 rpn++;
         }
+        return ret;
 }
 
 static DEFINE_PER_CPU(u64 *, tce_page) = NULL;
 
-static void tce_buildmulti_pSeriesLP(struct iommu_table *tbl, long tcenum,
+static int tce_buildmulti_pSeriesLP(struct iommu_table *tbl, long tcenum,
                                      long npages, unsigned long uaddr,
                                      enum dma_data_direction direction,
                                      struct dma_attrs *attrs)
 {
-        u64 rc;
+        u64 rc = 0;
         u64 proto_tce;
         u64 *tcep;
         u64 rpn;
         long l, limit;
+        long tcenum_start = tcenum, npages_start = npages;
+        int ret = 0;
 
         if (npages == 1) {
-                tce_build_pSeriesLP(tbl, tcenum, npages, uaddr,
-                                    direction, attrs);
-                return;
+                return tce_build_pSeriesLP(tbl, tcenum, npages, uaddr,
+                                           direction, attrs);
         }
 
         tcep = __get_cpu_var(tce_page);
@@ -153,9 +168,8 @@ static void tce_buildmulti_pSeriesLP(struct iommu_table *tbl, long tcenum,
                 tcep = (u64 *)__get_free_page(GFP_ATOMIC);
                 /* If allocation fails, fall back to the loop implementation */
                 if (!tcep) {
-                        tce_build_pSeriesLP(tbl, tcenum, npages, uaddr,
+                        return tce_build_pSeriesLP(tbl, tcenum, npages, uaddr,
                                             direction, attrs);
-                        return;
                 }
                 __get_cpu_var(tce_page) = tcep;
         }
@@ -187,6 +201,13 @@ static void tce_buildmulti_pSeriesLP(struct iommu_table *tbl, long tcenum,
                 tcenum += limit;
         } while (npages > 0 && !rc);
 
+        if (unlikely(rc == H_NOT_ENOUGH_RESOURCES)) {
+                ret = (int)rc;
+                tce_freemulti_pSeriesLP(tbl, tcenum_start,
+                                        (npages_start - (npages + limit)));
+                return ret;
+        }
+
         if (rc && printk_ratelimit()) {
                 printk("tce_buildmulti_pSeriesLP: plpar_tce_put failed. rc=%ld\n", rc);
                 printk("\tindex   = 0x%lx\n", (u64)tbl->it_index);
@@ -194,6 +215,7 @@ static void tce_buildmulti_pSeriesLP(struct iommu_table *tbl, long tcenum,
                 printk("\ttce[0] val = 0x%lx\n", tcep[0]);
                 show_stack(current, (unsigned long *)__get_SP());
         }
+        return ret;
 }
 
 static void tce_free_pSeriesLP(struct iommu_table *tbl, long tcenum, long npages)
diff --git a/arch/powerpc/platforms/pseries/plpar_wrappers.h b/arch/powerpc/platforms/pseries/plpar_wrappers.h
index d8680b589dc9..a437267c6bf8 100644
--- a/arch/powerpc/platforms/pseries/plpar_wrappers.h
+++ b/arch/powerpc/platforms/pseries/plpar_wrappers.h
@@ -42,6 +42,16 @@ static inline long register_slb_shadow(unsigned long cpu, unsigned long vpa)
         return vpa_call(0x3, cpu, vpa);
 }
 
+static inline long plpar_page_set_loaned(unsigned long vpa)
+{
+        return plpar_hcall_norets(H_PAGE_INIT, H_PAGE_SET_LOANED, vpa, 0);
+}
+
+static inline long plpar_page_set_active(unsigned long vpa)
+{
+        return plpar_hcall_norets(H_PAGE_INIT, H_PAGE_SET_ACTIVE, vpa, 0);
+}
+
 extern void vpa_init(int cpu);
 
 static inline long plpar_pte_enter(unsigned long flags,
diff --git a/arch/powerpc/platforms/pseries/setup.c b/arch/powerpc/platforms/pseries/setup.c
index 90beb444e1dd..063a0d2fba30 100644
--- a/arch/powerpc/platforms/pseries/setup.c
+++ b/arch/powerpc/platforms/pseries/setup.c
@@ -314,6 +314,76 @@ static int pseries_set_xdabr(unsigned long dabr)
                         H_DABRX_KERNEL | H_DABRX_USER);
 }
 
+#define CMO_CHARACTERISTICS_TOKEN 44
+#define CMO_MAXLENGTH 1026
+
+/**
+ * fw_cmo_feature_init - FW_FEATURE_CMO is not stored in ibm,hypertas-functions,
+ * handle that here. (Stolen from parse_system_parameter_string)
+ */
+void pSeries_cmo_feature_init(void)
+{
+        char *ptr, *key, *value, *end;
+        int call_status;
+        int PrPSP = -1;
+        int SecPSP = -1;
+
+        pr_debug(" -> fw_cmo_feature_init()\n");
+        spin_lock(&rtas_data_buf_lock);
+        memset(rtas_data_buf, 0, RTAS_DATA_BUF_SIZE);
+        call_status = rtas_call(rtas_token("ibm,get-system-parameter"), 3, 1,
+                                NULL,
+                                CMO_CHARACTERISTICS_TOKEN,
+                                __pa(rtas_data_buf),
+                                RTAS_DATA_BUF_SIZE);
+
+        if (call_status != 0) {
+                spin_unlock(&rtas_data_buf_lock);
+                pr_debug("CMO not available\n");
+                pr_debug(" <- fw_cmo_feature_init()\n");
+                return;
+        }
+
+        end = rtas_data_buf + CMO_MAXLENGTH - 2;
+        ptr = rtas_data_buf + 2;        /* step over strlen value */
+        key = value = ptr;
+
+        while (*ptr && (ptr <= end)) {
+                /* Separate the key and value by replacing '=' with '\0' and
+                 * point the value at the string after the '='
+                 */
+                if (ptr[0] == '=') {
+                        ptr[0] = '\0';
+                        value = ptr + 1;
+                } else if (ptr[0] == '\0' || ptr[0] == ',') {
+                        /* Terminate the string containing the key/value pair */
+                        ptr[0] = '\0';
+
+                        if (key == value) {
+                                pr_debug("Malformed key/value pair\n");
+                                /* Never found a '=', end processing */
+                                break;
+                        }
+
+                        if (0 == strcmp(key, "PrPSP"))
+                                PrPSP = simple_strtol(value, NULL, 10);
+                        else if (0 == strcmp(key, "SecPSP"))
+                                SecPSP = simple_strtol(value, NULL, 10);
+                        value = key = ptr + 1;
+                }
+                ptr++;
+        }
+
+        if (PrPSP != -1 || SecPSP != -1) {
+                pr_info("CMO enabled\n");
+                pr_debug("CMO enabled, PrPSP=%d, SecPSP=%d\n", PrPSP, SecPSP);
+                powerpc_firmware_features |= FW_FEATURE_CMO;
+        } else
+                pr_debug("CMO not enabled, PrPSP=%d, SecPSP=%d\n", PrPSP, SecPSP);
+        spin_unlock(&rtas_data_buf_lock);
+        pr_debug(" <- fw_cmo_feature_init()\n");
+}
+
 /*
  * Early initialization.  Relocation is on but do not reference unbolted pages
  */
@@ -329,6 +399,7 @@ static void __init pSeries_init_early(void)
         else if (firmware_has_feature(FW_FEATURE_XDABR))
                 ppc_md.set_dabr = pseries_set_xdabr;
 
+        pSeries_cmo_feature_init();
         iommu_init_early_pSeries();
 
         pr_debug(" <- pSeries_init_early()\n");