3 files changed, 716 insertions, 280 deletions

diff --git a/arch/sparc64/mm/init.c b/arch/sparc64/mm/init.c
index f37078d96407..177d8aaeec42 100644
--- a/arch/sparc64/mm/init.c
+++ b/arch/sparc64/mm/init.c
@@ -24,6 +24,8 @@
 #include <linux/cache.h>
 #include <linux/sort.h>
 #include <linux/percpu.h>
+#include <linux/lmb.h>
+#include <linux/mmzone.h>
 
 #include <asm/head.h>
 #include <asm/system.h>
@@ -72,9 +74,7 @@ extern struct tsb swapper_4m_tsb[KERNEL_TSB4M_NENTRIES];
 #define MAX_BANKS       32
 
 static struct linux_prom64_registers pavail[MAX_BANKS] __initdata;
-static struct linux_prom64_registers pavail_rescan[MAX_BANKS] __initdata;
 static int pavail_ents __initdata;
-static int pavail_rescan_ents __initdata;
 
 static int cmp_p64(const void *a, const void *b)
 {
@@ -715,285 +715,684 @@ out:
                 smp_new_mmu_context_version();
 }
 
-/* Find a free area for the bootmem map, avoiding the kernel image
- * and the initial ramdisk.
- */
-static unsigned long __init choose_bootmap_pfn(unsigned long start_pfn,
-                                               unsigned long end_pfn)
+static int numa_enabled = 1;
+static int numa_debug;
+
+static int __init early_numa(char *p)
 {
-        unsigned long avoid_start, avoid_end, bootmap_size;
-        int i;
+        if (!p)
+                return 0;
+
+        if (strstr(p, "off"))
+                numa_enabled = 0;
+
+        if (strstr(p, "debug"))
+                numa_debug = 1;
+
+        return 0;
+}
+early_param("numa", early_numa);
 
-        bootmap_size = bootmem_bootmap_pages(end_pfn - start_pfn);
-        bootmap_size <<= PAGE_SHIFT;
+#define numadbg(f, a...) \
+do {    if (numa_debug) \
+                printk(KERN_INFO f, ## a); \
+} while (0)
 
-        avoid_start = avoid_end = 0;
+static void __init find_ramdisk(unsigned long phys_base)
+{
 #ifdef CONFIG_BLK_DEV_INITRD
-        avoid_start = initrd_start;
-        avoid_end = PAGE_ALIGN(initrd_end);
+        if (sparc_ramdisk_image || sparc_ramdisk_image64) {
+                unsigned long ramdisk_image;
+
+                /* Older versions of the bootloader only supported a
+                 * 32-bit physical address for the ramdisk image
+                 * location, stored at sparc_ramdisk_image.  Newer
+                 * SILO versions set sparc_ramdisk_image to zero and
+                 * provide a full 64-bit physical address at
+                 * sparc_ramdisk_image64.
+                 */
+                ramdisk_image = sparc_ramdisk_image;
+                if (!ramdisk_image)
+                        ramdisk_image = sparc_ramdisk_image64;
+
+                /* Another bootloader quirk.  The bootloader normalizes
+                 * the physical address to KERNBASE, so we have to
+                 * factor that back out and add in the lowest valid
+                 * physical page address to get the true physical address.
+                 */
+                ramdisk_image -= KERNBASE;
+                ramdisk_image += phys_base;
+
+                numadbg("Found ramdisk at physical address 0x%lx, size %u\n",
+                        ramdisk_image, sparc_ramdisk_size);
+
+                initrd_start = ramdisk_image;
+                initrd_end = ramdisk_image + sparc_ramdisk_size;
+
+                lmb_reserve(initrd_start, initrd_end);
+        }
 #endif
+}
 
-        for (i = 0; i < pavail_ents; i++) {
-                unsigned long start, end;
+struct node_mem_mask {
+        unsigned long mask;
+        unsigned long val;
+        unsigned long bootmem_paddr;
+};
+static struct node_mem_mask node_masks[MAX_NUMNODES];
+static int num_node_masks;
 
-                start = pavail[i].phys_addr;
-                end = start + pavail[i].reg_size;
+int numa_cpu_lookup_table[NR_CPUS];
+cpumask_t numa_cpumask_lookup_table[MAX_NUMNODES];
 
-                while (start < end) {
-                        if (start >= kern_base &&
-                            start < PAGE_ALIGN(kern_base + kern_size)) {
-                                start = PAGE_ALIGN(kern_base + kern_size);
-                                continue;
-                        }
-                        if (start >= avoid_start && start < avoid_end) {
-                                start = avoid_end;
-                                continue;
-                        }
+#ifdef CONFIG_NEED_MULTIPLE_NODES
+static bootmem_data_t plat_node_bdata[MAX_NUMNODES];
 
-                        if ((end - start) < bootmap_size)
-                                break;
+struct mdesc_mblock {
+        u64     base;
+        u64     size;
+        u64     offset; /* RA-to-PA */
+};
+static struct mdesc_mblock *mblocks;
+static int num_mblocks;
 
-                        if (start < kern_base &&
-                            (start + bootmap_size) > kern_base) {
-                                start = PAGE_ALIGN(kern_base + kern_size);
-                                continue;
-                        }
+static unsigned long ra_to_pa(unsigned long addr)
+{
+        int i;
 
-                        if (start < avoid_start &&
-                            (start + bootmap_size) > avoid_start) {
-                                start = avoid_end;
-                                continue;
-                        }
+        for (i = 0; i < num_mblocks; i++) {
+                struct mdesc_mblock *m = &mblocks[i];
 
-                        /* OK, it doesn't overlap anything, use it.  */
-                        return start >> PAGE_SHIFT;
+                if (addr >= m->base &&
+                    addr < (m->base + m->size)) {
+                        addr += m->offset;
+                        break;
                 }
         }
-
-        prom_printf("Cannot find free area for bootmap, aborting.\n");
-        prom_halt();
+        return addr;
 }
 
-static void __init trim_pavail(unsigned long *cur_size_p,
-                               unsigned long *end_of_phys_p)
+static int find_node(unsigned long addr)
 {
-        unsigned long to_trim = *cur_size_p - cmdline_memory_size;
-        unsigned long avoid_start, avoid_end;
         int i;
 
-        to_trim = PAGE_ALIGN(to_trim);
+        addr = ra_to_pa(addr);
+        for (i = 0; i < num_node_masks; i++) {
+                struct node_mem_mask *p = &node_masks[i];
 
-        avoid_start = avoid_end = 0;
-#ifdef CONFIG_BLK_DEV_INITRD
-        avoid_start = initrd_start;
-        avoid_end = PAGE_ALIGN(initrd_end);
+                if ((addr & p->mask) == p->val)
+                        return i;
+        }
+        return -1;
+}
+
+static unsigned long nid_range(unsigned long start, unsigned long end,
+                               int *nid)
+{
+        *nid = find_node(start);
+        start += PAGE_SIZE;
+        while (start < end) {
+                int n = find_node(start);
+
+                if (n != *nid)
+                        break;
+                start += PAGE_SIZE;
+        }
+
+        return start;
+}
+#else
+static unsigned long nid_range(unsigned long start, unsigned long end,
+                               int *nid)
+{
+        *nid = 0;
+        return end;
+}
 #endif
 
-        /* Trim some pavail[] entries in order to satisfy the
-         * requested "mem=xxx" kernel command line specification.
-         *
-         * We must not trim off the kernel image area nor the
-         * initial ramdisk range (if any).  Also, we must not trim
-         * any pavail[] entry down to zero in order to preserve
-         * the invariant that all pavail[] entries have a non-zero
-         * size which is assumed by all of the code in here.
-         */
-        for (i = 0; i < pavail_ents; i++) {
-                unsigned long start, end, kern_end;
-                unsigned long trim_low, trim_high, n;
+/* This must be invoked after performing all of the necessary
+ * add_active_range() calls for 'nid'.  We need to be able to get
+ * correct data from get_pfn_range_for_nid().
+ */
+static void __init allocate_node_data(int nid)
+{
+        unsigned long paddr, num_pages, start_pfn, end_pfn;
+        struct pglist_data *p;
+
+#ifdef CONFIG_NEED_MULTIPLE_NODES
+        paddr = lmb_alloc_nid(sizeof(struct pglist_data),
+                              SMP_CACHE_BYTES, nid, nid_range);
+        if (!paddr) {
+                prom_printf("Cannot allocate pglist_data for nid[%d]\n", nid);
+                prom_halt();
+        }
+        NODE_DATA(nid) = __va(paddr);
+        memset(NODE_DATA(nid), 0, sizeof(struct pglist_data));
 
-                kern_end = PAGE_ALIGN(kern_base + kern_size);
+        NODE_DATA(nid)->bdata = &plat_node_bdata[nid];
+#endif
 
-                trim_low = start = pavail[i].phys_addr;
-                trim_high = end = start + pavail[i].reg_size;
+        p = NODE_DATA(nid);
 
-                if (kern_base >= start &&
-                    kern_base < end) {
-                        trim_low = kern_base;
-                        if (kern_end >= end)
-                                continue;
-                }
-                if (kern_end >= start &&
-                    kern_end < end) {
-                        trim_high = kern_end;
-                }
-                if (avoid_start &&
-                    avoid_start >= start &&
-                    avoid_start < end) {
-                        if (trim_low > avoid_start)
-                                trim_low = avoid_start;
-                        if (avoid_end >= end)
-                                continue;
-                }
-                if (avoid_end &&
-                    avoid_end >= start &&
-                    avoid_end < end) {
-                        if (trim_high < avoid_end)
-                                trim_high = avoid_end;
+        get_pfn_range_for_nid(nid, &start_pfn, &end_pfn);
+        p->node_start_pfn = start_pfn;
+        p->node_spanned_pages = end_pfn - start_pfn;
+
+        if (p->node_spanned_pages) {
+                num_pages = bootmem_bootmap_pages(p->node_spanned_pages);
+
+                paddr = lmb_alloc_nid(num_pages << PAGE_SHIFT, PAGE_SIZE, nid,
+                                      nid_range);
+                if (!paddr) {
+                        prom_printf("Cannot allocate bootmap for nid[%d]\n",
+                                  nid);
+                        prom_halt();
                 }
+                node_masks[nid].bootmem_paddr = paddr;
+        }
+}
+
+static void init_node_masks_nonnuma(void)
+{
+        int i;
+
+        numadbg("Initializing tables for non-numa.\n");
+
+        node_masks[0].mask = node_masks[0].val = 0;
+        num_node_masks = 1;
+
+        for (i = 0; i < NR_CPUS; i++)
+                numa_cpu_lookup_table[i] = 0;
+
+        numa_cpumask_lookup_table[0] = CPU_MASK_ALL;
+}
+
+#ifdef CONFIG_NEED_MULTIPLE_NODES
+struct pglist_data *node_data[MAX_NUMNODES];
+
+EXPORT_SYMBOL(numa_cpu_lookup_table);
+EXPORT_SYMBOL(numa_cpumask_lookup_table);
+EXPORT_SYMBOL(node_data);
+
+struct mdesc_mlgroup {
+        u64     node;
+        u64     latency;
+        u64     match;
+        u64     mask;
+};
+static struct mdesc_mlgroup *mlgroups;
+static int num_mlgroups;
+
+static int scan_pio_for_cfg_handle(struct mdesc_handle *md, u64 pio,
+                                   u32 cfg_handle)
+{
+        u64 arc;
 
-                if (trim_high <= trim_low)
+        mdesc_for_each_arc(arc, md, pio, MDESC_ARC_TYPE_FWD) {
+                u64 target = mdesc_arc_target(md, arc);
+                const u64 *val;
+
+                val = mdesc_get_property(md, target,
+                                         "cfg-handle", NULL);
+                if (val && *val == cfg_handle)
+                        return 0;
+        }
+        return -ENODEV;
+}
+
+static int scan_arcs_for_cfg_handle(struct mdesc_handle *md, u64 grp,
+                                    u32 cfg_handle)
+{
+        u64 arc, candidate, best_latency = ~(u64)0;
+
+        candidate = MDESC_NODE_NULL;
+        mdesc_for_each_arc(arc, md, grp, MDESC_ARC_TYPE_FWD) {
+                u64 target = mdesc_arc_target(md, arc);
+                const char *name = mdesc_node_name(md, target);
+                const u64 *val;
+
+                if (strcmp(name, "pio-latency-group"))
                         continue;
 
-                if (trim_low == start && trim_high == end) {
-                        /* Whole chunk is available for trimming.
-                         * Trim all except one page, in order to keep
-                         * entry non-empty.
-                         */
-                        n = (end - start) - PAGE_SIZE;
-                        if (n > to_trim)
-                                n = to_trim;
-
-                        if (n) {
-                                pavail[i].phys_addr += n;
-                                pavail[i].reg_size -= n;
-                                to_trim -= n;
-                        }
-                } else {
-                        n = (trim_low - start);
-                        if (n > to_trim)
-                                n = to_trim;
-
-                        if (n) {
-                                pavail[i].phys_addr += n;
-                                pavail[i].reg_size -= n;
-                                to_trim -= n;
-                        }
-                        if (to_trim) {
-                                n = end - trim_high;
-                                if (n > to_trim)
-                                        n = to_trim;
-                                if (n) {
-                                        pavail[i].reg_size -= n;
-                                        to_trim -= n;
-                                }
-                        }
+                val = mdesc_get_property(md, target, "latency", NULL);
+                if (!val)
+                        continue;
+
+                if (*val < best_latency) {
+                        candidate = target;
+                        best_latency = *val;
                 }
+        }
+
+        if (candidate == MDESC_NODE_NULL)
+                return -ENODEV;
+
+        return scan_pio_for_cfg_handle(md, candidate, cfg_handle);
+}
+
+int of_node_to_nid(struct device_node *dp)
+{
+        const struct linux_prom64_registers *regs;
+        struct mdesc_handle *md;
+        u32 cfg_handle;
+        int count, nid;
+        u64 grp;
 
-                if (!to_trim)
+        if (!mlgroups)
+                return -1;
+
+        regs = of_get_property(dp, "reg", NULL);
+        if (!regs)
+                return -1;
+
+        cfg_handle = (regs->phys_addr >> 32UL) & 0x0fffffff;
+
+        md = mdesc_grab();
+
+        count = 0;
+        nid = -1;
+        mdesc_for_each_node_by_name(md, grp, "group") {
+                if (!scan_arcs_for_cfg_handle(md, grp, cfg_handle)) {
+                        nid = count;
                         break;
+                }
+                count++;
         }
 
-        /* Recalculate.  */
-        *cur_size_p = 0UL;
-        for (i = 0; i < pavail_ents; i++) {
-                *end_of_phys_p = pavail[i].phys_addr +
-                        pavail[i].reg_size;
-                *cur_size_p += pavail[i].reg_size;
-        }
+        mdesc_release(md);
+
+        return nid;
 }
 
-/* About pages_avail, this is the value we will use to calculate
- * the zholes_size[] argument given to free_area_init_node().  The
- * page allocator uses this to calculate nr_kernel_pages,
- * nr_all_pages and zone->present_pages.  On NUMA it is used
- * to calculate zone->min_unmapped_pages and zone->min_slab_pages.
- *
- * So this number should really be set to what the page allocator
- * actually ends up with.  This means:
- * 1) It should include bootmem map pages, we'll release those.
- * 2) It should not include the kernel image, except for the
- *    __init sections which we will also release.
- * 3) It should include the initrd image, since we'll release
- *    that too.
- */
-static unsigned long __init bootmem_init(unsigned long *pages_avail,
-                                         unsigned long phys_base)
+static void add_node_ranges(void)
 {
-        unsigned long bootmap_size, end_pfn;
-        unsigned long end_of_phys_memory = 0UL;
-        unsigned long bootmap_pfn, bytes_avail, size;
         int i;
 
-        bytes_avail = 0UL;
-        for (i = 0; i < pavail_ents; i++) {
-                end_of_phys_memory = pavail[i].phys_addr +
-                        pavail[i].reg_size;
-                bytes_avail += pavail[i].reg_size;
+        for (i = 0; i < lmb.memory.cnt; i++) {
+                unsigned long size = lmb_size_bytes(&lmb.memory, i);
+                unsigned long start, end;
+
+                start = lmb.memory.region[i].base;
+                end = start + size;
+                while (start < end) {
+                        unsigned long this_end;
+                        int nid;
+
+                        this_end = nid_range(start, end, &nid);
+
+                        numadbg("Adding active range nid[%d] "
+                                "start[%lx] end[%lx]\n",
+                                nid, start, this_end);
+
+                        add_active_range(nid,
+                                         start >> PAGE_SHIFT,
+                                         this_end >> PAGE_SHIFT);
+
+                        start = this_end;
+                }
         }
+}
 
-        /* Determine the location of the initial ramdisk before trying
-         * to honor the "mem=xxx" command line argument.  We must know
-         * where the kernel image and the ramdisk image are so that we
-         * do not trim those two areas from the physical memory map.
-         */
+static int __init grab_mlgroups(struct mdesc_handle *md)
+{
+        unsigned long paddr;
+        int count = 0;
+        u64 node;
+
+        mdesc_for_each_node_by_name(md, node, "memory-latency-group")
+                count++;
+        if (!count)
+                return -ENOENT;
+
+        paddr = lmb_alloc(count * sizeof(struct mdesc_mlgroup),
+                          SMP_CACHE_BYTES);
+        if (!paddr)
+                return -ENOMEM;
+
+        mlgroups = __va(paddr);
+        num_mlgroups = count;
+
+        count = 0;
+        mdesc_for_each_node_by_name(md, node, "memory-latency-group") {
+                struct mdesc_mlgroup *m = &mlgroups[count++];
+                const u64 *val;
+
+                m->node = node;
+
+                val = mdesc_get_property(md, node, "latency", NULL);
+                m->latency = *val;
+                val = mdesc_get_property(md, node, "address-match", NULL);
+                m->match = *val;
+                val = mdesc_get_property(md, node, "address-mask", NULL);
+                m->mask = *val;
+
+                numadbg("MLGROUP[%d]: node[%lx] latency[%lx] "
+                        "match[%lx] mask[%lx]\n",
+                        count - 1, m->node, m->latency, m->match, m->mask);
+        }
 
-#ifdef CONFIG_BLK_DEV_INITRD
-        /* Now have to check initial ramdisk, so that bootmap does not overwrite it */
-        if (sparc_ramdisk_image || sparc_ramdisk_image64) {
-                unsigned long ramdisk_image = sparc_ramdisk_image ?
-                        sparc_ramdisk_image : sparc_ramdisk_image64;
-                ramdisk_image -= KERNBASE;
-                initrd_start = ramdisk_image + phys_base;
-                initrd_end = initrd_start + sparc_ramdisk_size;
-                if (initrd_end > end_of_phys_memory) {
-                        printk(KERN_CRIT "initrd extends beyond end of memory "
-                                         "(0x%016lx > 0x%016lx)\ndisabling initrd\n",
-                               initrd_end, end_of_phys_memory);
-                        initrd_start = 0;
-                        initrd_end = 0;
+        return 0;
+}
+
+static int __init grab_mblocks(struct mdesc_handle *md)
+{
+        unsigned long paddr;
+        int count = 0;
+        u64 node;
+
+        mdesc_for_each_node_by_name(md, node, "mblock")
+                count++;
+        if (!count)
+                return -ENOENT;
+
+        paddr = lmb_alloc(count * sizeof(struct mdesc_mblock),
+                          SMP_CACHE_BYTES);
+        if (!paddr)
+                return -ENOMEM;
+
+        mblocks = __va(paddr);
+        num_mblocks = count;
+
+        count = 0;
+        mdesc_for_each_node_by_name(md, node, "mblock") {
+                struct mdesc_mblock *m = &mblocks[count++];
+                const u64 *val;
+
+                val = mdesc_get_property(md, node, "base", NULL);
+                m->base = *val;
+                val = mdesc_get_property(md, node, "size", NULL);
+                m->size = *val;
+                val = mdesc_get_property(md, node,
+                                         "address-congruence-offset", NULL);
+                m->offset = *val;
+
+                numadbg("MBLOCK[%d]: base[%lx] size[%lx] offset[%lx]\n",
+                        count - 1, m->base, m->size, m->offset);
+        }
+
+        return 0;
+}
+
+static void __init numa_parse_mdesc_group_cpus(struct mdesc_handle *md,
+                                               u64 grp, cpumask_t *mask)
+{
+        u64 arc;
+
+        cpus_clear(*mask);
+
+        mdesc_for_each_arc(arc, md, grp, MDESC_ARC_TYPE_BACK) {
+                u64 target = mdesc_arc_target(md, arc);
+                const char *name = mdesc_node_name(md, target);
+                const u64 *id;
+
+                if (strcmp(name, "cpu"))
+                        continue;
+                id = mdesc_get_property(md, target, "id", NULL);
+                if (*id < NR_CPUS)
+                        cpu_set(*id, *mask);
+        }
+}
+
+static struct mdesc_mlgroup * __init find_mlgroup(u64 node)
+{
+        int i;
+
+        for (i = 0; i < num_mlgroups; i++) {
+                struct mdesc_mlgroup *m = &mlgroups[i];
+                if (m->node == node)
+                        return m;
+        }
+        return NULL;
+}
+
+static int __init numa_attach_mlgroup(struct mdesc_handle *md, u64 grp,
+                                      int index)
+{
+        struct mdesc_mlgroup *candidate = NULL;
+        u64 arc, best_latency = ~(u64)0;
+        struct node_mem_mask *n;
+
+        mdesc_for_each_arc(arc, md, grp, MDESC_ARC_TYPE_FWD) {
+                u64 target = mdesc_arc_target(md, arc);
+                struct mdesc_mlgroup *m = find_mlgroup(target);
+                if (!m)
+                        continue;
+                if (m->latency < best_latency) {
+                        candidate = m;
+                        best_latency = m->latency;
                 }
         }
-#endif  
+        if (!candidate)
+                return -ENOENT;
+
+        if (num_node_masks != index) {
+                printk(KERN_ERR "Inconsistent NUMA state, "
+                       "index[%d] != num_node_masks[%d]\n",
+                       index, num_node_masks);
+                return -EINVAL;
+        }
 
-        if (cmdline_memory_size &&
-            bytes_avail > cmdline_memory_size)
-                trim_pavail(&bytes_avail,
-                            &end_of_phys_memory);
+        n = &node_masks[num_node_masks++];
 
-        *pages_avail = bytes_avail >> PAGE_SHIFT;
+        n->mask = candidate->mask;
+        n->val = candidate->match;
 
-        end_pfn = end_of_phys_memory >> PAGE_SHIFT;
+        numadbg("NUMA NODE[%d]: mask[%lx] val[%lx] (latency[%lx])\n",
+                index, n->mask, n->val, candidate->latency);
 
-        /* Initialize the boot-time allocator. */
-        max_pfn = max_low_pfn = end_pfn;
-        min_low_pfn = (phys_base >> PAGE_SHIFT);
+        return 0;
+}
+
+static int __init numa_parse_mdesc_group(struct mdesc_handle *md, u64 grp,
+                                         int index)
+{
+        cpumask_t mask;
+        int cpu;
 
-        bootmap_pfn = choose_bootmap_pfn(min_low_pfn, end_pfn);
+        numa_parse_mdesc_group_cpus(md, grp, &mask);
 
-        bootmap_size = init_bootmem_node(NODE_DATA(0), bootmap_pfn,
-                                         min_low_pfn, end_pfn);
+        for_each_cpu_mask(cpu, mask)
+                numa_cpu_lookup_table[cpu] = index;
+        numa_cpumask_lookup_table[index] = mask;
 
-        /* Now register the available physical memory with the
-         * allocator.
-         */
-        for (i = 0; i < pavail_ents; i++)
-                free_bootmem(pavail[i].phys_addr, pavail[i].reg_size);
+        if (numa_debug) {
+                printk(KERN_INFO "NUMA GROUP[%d]: cpus [ ", index);
+                for_each_cpu_mask(cpu, mask)
+                        printk("%d ", cpu);
+                printk("]\n");
+        }
 
-#ifdef CONFIG_BLK_DEV_INITRD
-        if (initrd_start) {
-                size = initrd_end - initrd_start;
+        return numa_attach_mlgroup(md, grp, index);
+}
+
+static int __init numa_parse_mdesc(void)
+{
+        struct mdesc_handle *md = mdesc_grab();
+        int i, err, count;
+        u64 node;
+
+        node = mdesc_node_by_name(md, MDESC_NODE_NULL, "latency-groups");
+        if (node == MDESC_NODE_NULL) {
+                mdesc_release(md);
+                return -ENOENT;
+        }
+
+        err = grab_mblocks(md);
+        if (err < 0)
+                goto out;
+
+        err = grab_mlgroups(md);
+        if (err < 0)
+                goto out;
+
+        count = 0;
+        mdesc_for_each_node_by_name(md, node, "group") {
+                err = numa_parse_mdesc_group(md, node, count);
+                if (err < 0)
+                        break;
+                count++;
+        }
+
+        add_node_ranges();
+
+        for (i = 0; i < num_node_masks; i++) {
+                allocate_node_data(i);
+                node_set_online(i);
+        }
+
+        err = 0;
+out:
+        mdesc_release(md);
+        return err;
+}
+
+static int __init numa_parse_sun4u(void)
+{
+        return -1;
+}
 
-                /* Reserve the initrd image area. */
-                reserve_bootmem(initrd_start, size, BOOTMEM_DEFAULT);
+static int __init bootmem_init_numa(void)
+{
+        int err = -1;
 
-                initrd_start += PAGE_OFFSET;
-                initrd_end += PAGE_OFFSET;
+        numadbg("bootmem_init_numa()\n");
+
+        if (numa_enabled) {
+                if (tlb_type == hypervisor)
+                        err = numa_parse_mdesc();
+                else
+                        err = numa_parse_sun4u();
         }
+        return err;
+}
+
+#else
+
+static int bootmem_init_numa(void)
+{
+        return -1;
+}
+
 #endif
-        /* Reserve the kernel text/data/bss. */
-        reserve_bootmem(kern_base, kern_size, BOOTMEM_DEFAULT);
-        *pages_avail -= PAGE_ALIGN(kern_size) >> PAGE_SHIFT;
-
-        /* Add back in the initmem pages. */
-        size = ((unsigned long)(__init_end) & PAGE_MASK) -
-                PAGE_ALIGN((unsigned long)__init_begin);
-        *pages_avail += size >> PAGE_SHIFT;
-
-        /* Reserve the bootmem map.   We do not account for it
-         * in pages_avail because we will release that memory
-         * in free_all_bootmem.
-         */
-        size = bootmap_size;
-        reserve_bootmem((bootmap_pfn << PAGE_SHIFT), size, BOOTMEM_DEFAULT);
 
-        for (i = 0; i < pavail_ents; i++) {
+static void __init bootmem_init_nonnuma(void)
+{
+        unsigned long top_of_ram = lmb_end_of_DRAM();
+        unsigned long total_ram = lmb_phys_mem_size();
+        unsigned int i;
+
+        numadbg("bootmem_init_nonnuma()\n");
+
+        printk(KERN_INFO "Top of RAM: 0x%lx, Total RAM: 0x%lx\n",
+               top_of_ram, total_ram);
+        printk(KERN_INFO "Memory hole size: %ldMB\n",
+               (top_of_ram - total_ram) >> 20);
+
+        init_node_masks_nonnuma();
+
+        for (i = 0; i < lmb.memory.cnt; i++) {
+                unsigned long size = lmb_size_bytes(&lmb.memory, i);
                 unsigned long start_pfn, end_pfn;
 
-                start_pfn = pavail[i].phys_addr >> PAGE_SHIFT;
-                end_pfn = (start_pfn + (pavail[i].reg_size >> PAGE_SHIFT));
-                memory_present(0, start_pfn, end_pfn);
+                if (!size)
+                        continue;
+
+                start_pfn = lmb.memory.region[i].base >> PAGE_SHIFT;
+                end_pfn = start_pfn + lmb_size_pages(&lmb.memory, i);
+                add_active_range(0, start_pfn, end_pfn);
+        }
+
+        allocate_node_data(0);
+
+        node_set_online(0);
+}
+
+static void __init reserve_range_in_node(int nid, unsigned long start,
+                                         unsigned long end)
+{
+        numadbg("    reserve_range_in_node(nid[%d],start[%lx],end[%lx]\n",
+                nid, start, end);
+        while (start < end) {
+                unsigned long this_end;
+                int n;
+
+                this_end = nid_range(start, end, &n);
+                if (n == nid) {
+                        numadbg("      MATCH reserving range [%lx:%lx]\n",
+                                start, this_end);
+                        reserve_bootmem_node(NODE_DATA(nid), start,
+                                             (this_end - start), BOOTMEM_DEFAULT);
+                } else
+                        numadbg("      NO MATCH, advancing start to %lx\n",
+                                this_end);
+
+                start = this_end;
+        }
+}
+
+static void __init trim_reserved_in_node(int nid)
+{
+        int i;
+
+        numadbg("  trim_reserved_in_node(%d)\n", nid);
+
+        for (i = 0; i < lmb.reserved.cnt; i++) {
+                unsigned long start = lmb.reserved.region[i].base;
+                unsigned long size = lmb_size_bytes(&lmb.reserved, i);
+                unsigned long end = start + size;
+
+                reserve_range_in_node(nid, start, end);
+        }
+}
+
+static void __init bootmem_init_one_node(int nid)
+{
+        struct pglist_data *p;
+
+        numadbg("bootmem_init_one_node(%d)\n", nid);
+
+        p = NODE_DATA(nid);
+
+        if (p->node_spanned_pages) {
+                unsigned long paddr = node_masks[nid].bootmem_paddr;
+                unsigned long end_pfn;
+
+                end_pfn = p->node_start_pfn + p->node_spanned_pages;
+
+                numadbg("  init_bootmem_node(%d, %lx, %lx, %lx)\n",
+                        nid, paddr >> PAGE_SHIFT, p->node_start_pfn, end_pfn);
+
+                init_bootmem_node(p, paddr >> PAGE_SHIFT,
+                                  p->node_start_pfn, end_pfn);
+
+                numadbg("  free_bootmem_with_active_regions(%d, %lx)\n",
+                        nid, end_pfn);
+                free_bootmem_with_active_regions(nid, end_pfn);
+
+                trim_reserved_in_node(nid);
+
+                numadbg("  sparse_memory_present_with_active_regions(%d)\n",
+                        nid);
+                sparse_memory_present_with_active_regions(nid);
         }
+}
+
+static unsigned long __init bootmem_init(unsigned long phys_base)
+{
+        unsigned long end_pfn;
+        int nid;
+
+        end_pfn = lmb_end_of_DRAM() >> PAGE_SHIFT;
+        max_pfn = max_low_pfn = end_pfn;
+        min_low_pfn = (phys_base >> PAGE_SHIFT);
+
+        if (bootmem_init_numa() < 0)
+                bootmem_init_nonnuma();
+
+        /* XXX cpu notifier XXX */
+
+        for_each_online_node(nid)
+                bootmem_init_one_node(nid);
 
         sparse_init();
 
@@ -1289,7 +1688,7 @@ void __init setup_per_cpu_areas(void)
 
 void __init paging_init(void)
 {
-        unsigned long end_pfn, pages_avail, shift, phys_base;
+        unsigned long end_pfn, shift, phys_base;
         unsigned long real_end, i;
 
         /* These build time checkes make sure that the dcache_dirty_cpu()
@@ -1330,12 +1729,26 @@ void __init paging_init(void)
                 sun4v_ktsb_init();
         }
 
+        lmb_init();
+
         /* Find available physical memory... */
         read_obp_memory("available", &pavail[0], &pavail_ents);
 
         phys_base = 0xffffffffffffffffUL;
-        for (i = 0; i < pavail_ents; i++)
+        for (i = 0; i < pavail_ents; i++) {
                 phys_base = min(phys_base, pavail[i].phys_addr);
+                lmb_add(pavail[i].phys_addr, pavail[i].reg_size);
+        }
+
+        lmb_reserve(kern_base, kern_size);
+
+        find_ramdisk(phys_base);
+
+        if (cmdline_memory_size)
+                lmb_enforce_memory_limit(phys_base + cmdline_memory_size);
+
+        lmb_analyze();
+        lmb_dump_all();
 
         set_bit(0, mmu_context_bmap);
 
@@ -1371,14 +1784,10 @@ void __init paging_init(void)
         if (tlb_type == hypervisor)
                 sun4v_ktsb_register();
 
-        /* Setup bootmem... */
-        pages_avail = 0;
-        last_valid_pfn = end_pfn = bootmem_init(&pages_avail, phys_base);
-
-        max_mapnr = last_valid_pfn;
-
-        kernel_physical_mapping_init();
-
+        /* We must setup the per-cpu areas before we pull in the
+         * PROM and the MDESC.  The code there fills in cpu and
+         * other information into per-cpu data structures.
+         */
         real_setup_per_cpu_areas();
 
         prom_build_devicetree();
@@ -1386,20 +1795,22 @@ void __init paging_init(void)
         if (tlb_type == hypervisor)
                 sun4v_mdesc_init();
 
+        /* Setup bootmem... */
+        last_valid_pfn = end_pfn = bootmem_init(phys_base);
+
+#ifndef CONFIG_NEED_MULTIPLE_NODES
+        max_mapnr = last_valid_pfn;
+#endif
+        kernel_physical_mapping_init();
+
         {
-                unsigned long zones_size[MAX_NR_ZONES];
-                unsigned long zholes_size[MAX_NR_ZONES];
-                int znum;
+                unsigned long max_zone_pfns[MAX_NR_ZONES];
 
-                for (znum = 0; znum < MAX_NR_ZONES; znum++)
-                        zones_size[znum] = zholes_size[znum] = 0;
+                memset(max_zone_pfns, 0, sizeof(max_zone_pfns));
 
-                zones_size[ZONE_NORMAL] = end_pfn;
-                zholes_size[ZONE_NORMAL] = end_pfn - pages_avail;
+                max_zone_pfns[ZONE_NORMAL] = end_pfn;
 
-                free_area_init_node(0, &contig_page_data, zones_size,
-                                    __pa(PAGE_OFFSET) >> PAGE_SHIFT,
-                                    zholes_size);
+                free_area_init_nodes(max_zone_pfns);
         }
 
         printk("Booting Linux...\n");
@@ -1408,21 +1819,52 @@ void __init paging_init(void)
         cpu_probe();
 }
 
-static void __init taint_real_pages(void)
+int __init page_in_phys_avail(unsigned long paddr)
+{
+        int i;
+
+        paddr &= PAGE_MASK;
+
+        for (i = 0; i < pavail_ents; i++) {
+                unsigned long start, end;
+
+                start = pavail[i].phys_addr;
+                end = start + pavail[i].reg_size;
+
+                if (paddr >= start && paddr < end)
+                        return 1;
+        }
+        if (paddr >= kern_base && paddr < (kern_base + kern_size))
+                return 1;
+#ifdef CONFIG_BLK_DEV_INITRD
+        if (paddr >= __pa(initrd_start) &&
+            paddr < __pa(PAGE_ALIGN(initrd_end)))
+                return 1;
+#endif
+
+        return 0;
+}
+
+static struct linux_prom64_registers pavail_rescan[MAX_BANKS] __initdata;
+static int pavail_rescan_ents __initdata;
+
+/* Certain OBP calls, such as fetching "available" properties, can
+ * claim physical memory.  So, along with initializing the valid
+ * address bitmap, what we do here is refetch the physical available
+ * memory list again, and make sure it provides at least as much
+ * memory as 'pavail' does.
+ */
+static void setup_valid_addr_bitmap_from_pavail(void)
 {
         int i;
 
         read_obp_memory("available", &pavail_rescan[0], &pavail_rescan_ents);
 
-        /* Find changes discovered in the physmem available rescan and
-         * reserve the lost portions in the bootmem maps.
-         */
         for (i = 0; i < pavail_ents; i++) {
                 unsigned long old_start, old_end;
 
                 old_start = pavail[i].phys_addr;
-                old_end = old_start +
-                        pavail[i].reg_size;
+                old_end = old_start + pavail[i].reg_size;
                 while (old_start < old_end) {
                         int n;
 
@@ -1440,7 +1882,16 @@ static void __init taint_real_pages(void)
                                         goto do_next_page;
                                 }
                         }
-                        reserve_bootmem(old_start, PAGE_SIZE, BOOTMEM_DEFAULT);
+
+                        prom_printf("mem_init: Lost memory in pavail\n");
+                        prom_printf("mem_init: OLD start[%lx] size[%lx]\n",
+                                    pavail[i].phys_addr,
+                                    pavail[i].reg_size);
+                        prom_printf("mem_init: NEW start[%lx] size[%lx]\n",
+                                    pavail_rescan[i].phys_addr,
+                                    pavail_rescan[i].reg_size);
+                        prom_printf("mem_init: Cannot continue, aborting.\n");
+                        prom_halt();
 
                 do_next_page:
                         old_start += PAGE_SIZE;
@@ -1448,32 +1899,6 @@ static void __init taint_real_pages(void)
         }
 }
 
-int __init page_in_phys_avail(unsigned long paddr)
-{
-        int i;
-
-        paddr &= PAGE_MASK;
-
-        for (i = 0; i < pavail_rescan_ents; i++) {
-                unsigned long start, end;
-
-                start = pavail_rescan[i].phys_addr;
-                end = start + pavail_rescan[i].reg_size;
-
-                if (paddr >= start && paddr < end)
-                        return 1;
-        }
-        if (paddr >= kern_base && paddr < (kern_base + kern_size))
-                return 1;
-#ifdef CONFIG_BLK_DEV_INITRD
-        if (paddr >= __pa(initrd_start) &&
-            paddr < __pa(PAGE_ALIGN(initrd_end)))
-                return 1;
-#endif
-
-        return 0;
-}
-
 void __init mem_init(void)
 {
         unsigned long codepages, datapages, initpages;
@@ -1496,14 +1921,26 @@ void __init mem_init(void)
                 addr += PAGE_SIZE;
         }
 
-        taint_real_pages();
+        setup_valid_addr_bitmap_from_pavail();
 
         high_memory = __va(last_valid_pfn << PAGE_SHIFT);
 
+#ifdef CONFIG_NEED_MULTIPLE_NODES
+        for_each_online_node(i) {
+                if (NODE_DATA(i)->node_spanned_pages != 0) {
+                        totalram_pages +=
+                                free_all_bootmem_node(NODE_DATA(i));
+                }
+        }
+#else
+        totalram_pages = free_all_bootmem();
+#endif
+
         /* We subtract one to account for the mem_map_zero page
          * allocated below.
          */
-        totalram_pages = num_physpages = free_all_bootmem() - 1;
+        totalram_pages -= 1;
+        num_physpages = totalram_pages;
 
         /*
          * Set up the zero page, mark it reserved, so that page count
diff --git a/arch/sparc64/mm/tsb.c b/arch/sparc64/mm/tsb.c
index a3e6e4b635b3..fe70c8a557b5 100644
--- a/arch/sparc64/mm/tsb.c
+++ b/arch/sparc64/mm/tsb.c
@@ -321,7 +321,8 @@ retry_tsb_alloc:
         if (new_size > (PAGE_SIZE * 2))
                 gfp_flags = __GFP_NOWARN | __GFP_NORETRY;
 
-        new_tsb = kmem_cache_alloc(tsb_caches[new_cache_index], gfp_flags);
+        new_tsb = kmem_cache_alloc_node(tsb_caches[new_cache_index],
+                                        gfp_flags, numa_node_id());
         if (unlikely(!new_tsb)) {
                 /* Not being able to fork due to a high-order TSB
                  * allocation failure is very bad behavior.  Just back
diff --git a/arch/sparc64/mm/ultra.S b/arch/sparc64/mm/ultra.S
index 2865c105b6a4..e686a67561af 100644
--- a/arch/sparc64/mm/ultra.S
+++ b/arch/sparc64/mm/ultra.S
@@ -476,7 +476,6 @@ xcall_sync_tick:
 #endif
         call            smp_synchronize_tick_client
          nop
-        clr             %l6
         b               rtrap_xcall
          ldx            [%sp + PTREGS_OFF + PT_V9_TSTATE], %l1
 
@@ -511,7 +510,6 @@ xcall_report_regs:
 #endif
         call            __show_regs
          add            %sp, PTREGS_OFF, %o0
-        clr             %l6
         /* Has to be a non-v9 branch due to the large distance. */
         b               rtrap_xcall
          ldx            [%sp + PTREGS_OFF + PT_V9_TSTATE], %l1
@@ -576,7 +574,7 @@ __hypervisor_tlb_xcall_error:
         mov     %l4, %o0
         call    hypervisor_tlbop_error_xcall
          mov    %l5, %o1
-        ba,a,pt %xcc, rtrap_clr_l6
+        ba,a,pt %xcc, rtrap
 
         .globl          __hypervisor_xcall_flush_tlb_mm
 __hypervisor_xcall_flush_tlb_mm: /* 21 insns */