-/* $Id: init.c,v 1.209 2002/02/09 19:49:31 davem Exp $
+/*
* arch/sparc64/mm/init.c
*
* Copyright (C) 1996-1999 David S. Miller (davem@caip.rutgers.edu)
* Copyright (C) 1997-1999 Jakub Jelinek (jj@sunsite.mff.cuni.cz)
*/
-#include <linux/config.h>
#include <linux/module.h>
#include <linux/kernel.h>
#include <linux/sched.h>
#include <linux/initrd.h>
#include <linux/swap.h>
#include <linux/pagemap.h>
+#include <linux/poison.h>
#include <linux/fs.h>
#include <linux/seq_file.h>
#include <linux/kprobes.h>
#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>
#include <asm/sections.h>
#include <asm/tsb.h>
#include <asm/hypervisor.h>
-
-extern void device_scan(void);
+#include <asm/prom.h>
+#include <asm/sstate.h>
+#include <asm/mdesc.h>
+#include <asm/cpudata.h>
#define MAX_PHYS_ADDRESS (1UL << 42UL)
#define KPTE_BITMAP_CHUNK_SZ (256UL * 1024UL * 1024UL)
*/
unsigned long kpte_linear_bitmap[KPTE_BITMAP_BYTES / sizeof(unsigned long)];
-/* A special kernel TSB for 4MB and 256MB linear mappings. */
-struct tsb swapper_4m_tsb[KERNEL_TSB4M_NENTRIES];
+#ifndef CONFIG_DEBUG_PAGEALLOC
+/* A special kernel TSB for 4MB and 256MB linear mappings.
+ * Space is allocated for this right after the trap table
+ * in arch/sparc64/kernel/head.S
+ */
+extern struct tsb swapper_4m_tsb[KERNEL_TSB4M_NENTRIES];
+#endif
#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)
{
prom_halt();
}
- *num_ents = ents;
-
/* Sanitize what we got from the firmware, by page aligning
* everything.
*/
size = 0UL;
base = new_base;
}
+ if (size == 0UL) {
+ /* If it is empty, simply get rid of it.
+ * This simplifies the logic of the other
+ * functions that process these arrays.
+ */
+ memmove(®s[i], ®s[i + 1],
+ (ents - i - 1) * sizeof(regs[0]));
+ i--;
+ ents--;
+ continue;
+ }
regs[i].phys_addr = base;
regs[i].reg_size = size;
}
+
+ *num_ents = ents;
+
sort(regs, ents, sizeof(struct linux_prom64_registers),
cmp_p64, NULL);
}
unsigned long kern_base __read_mostly;
unsigned long kern_size __read_mostly;
-/* get_new_mmu_context() uses "cache + 1". */
-DEFINE_SPINLOCK(ctx_alloc_lock);
-unsigned long tlb_context_cache = CTX_FIRST_VERSION - 1;
-#define CTX_BMAP_SLOTS (1UL << (CTX_NR_BITS - 6))
-unsigned long mmu_context_bmap[CTX_BMAP_SLOTS];
-
-/* References to special section boundaries */
-extern char _start[], _end[];
-
/* Initial ramdisk setup */
extern unsigned long sparc_ramdisk_image64;
extern unsigned int sparc_ramdisk_image;
extern unsigned int sparc_ramdisk_size;
struct page *mem_map_zero __read_mostly;
+EXPORT_SYMBOL(mem_map_zero);
unsigned int sparc64_highest_unlocked_tlb_ent __read_mostly;
unsigned long sparc64_kern_pri_nuc_bits __read_mostly;
unsigned long sparc64_kern_sec_context __read_mostly;
-int bigkernel = 0;
-
-kmem_cache_t *pgtable_cache __read_mostly;
-
-static void zero_ctor(void *addr, kmem_cache_t *cache, unsigned long flags)
-{
- clear_page(addr);
-}
-
-void pgtable_cache_init(void)
-{
- pgtable_cache = kmem_cache_create("pgtable_cache",
- PAGE_SIZE, PAGE_SIZE,
- SLAB_HWCACHE_ALIGN |
- SLAB_MUST_HWCACHE_ALIGN,
- zero_ctor,
- NULL);
- if (!pgtable_cache) {
- prom_printf("pgtable_cache_init(): Could not create!\n");
- prom_halt();
- }
-}
+int num_kernel_image_mappings;
#ifdef CONFIG_DEBUG_DCFLUSH
atomic_t dcpage_flushes = ATOMIC_INIT(0);
}
#define PG_dcache_dirty PG_arch_1
-#define PG_dcache_cpu_shift 24UL
-#define PG_dcache_cpu_mask (256UL - 1UL)
-
-#if NR_CPUS > 256
-#error D-cache dirty tracking and thread_info->cpu need fixing for > 256 cpus
-#endif
+#define PG_dcache_cpu_shift 32UL
+#define PG_dcache_cpu_mask \
+ ((1UL<<ilog2(roundup_pow_of_two(NR_CPUS)))-1UL)
#define dcache_dirty_cpu(page) \
(((page)->flags >> PG_dcache_cpu_shift) & PG_dcache_cpu_mask)
-static __inline__ void set_dcache_dirty(struct page *page, int this_cpu)
+static inline void set_dcache_dirty(struct page *page, int this_cpu)
{
unsigned long mask = this_cpu;
unsigned long non_cpu_bits;
: "g1", "g7");
}
-static __inline__ void clear_dcache_dirty_cpu(struct page *page, unsigned long cpu)
+static inline void clear_dcache_dirty_cpu(struct page *page, unsigned long cpu)
{
unsigned long mask = (1UL << PG_dcache_dirty);
{
struct mm_struct *mm;
struct tsb *tsb;
- unsigned long tag;
+ unsigned long tag, flags;
+ unsigned long tsb_index, tsb_hash_shift;
if (tlb_type != hypervisor) {
unsigned long pfn = pte_pfn(pte);
}
mm = vma->vm_mm;
- tsb = &mm->context.tsb[(address >> PAGE_SHIFT) &
- (mm->context.tsb_nentries - 1UL)];
+
+ tsb_index = MM_TSB_BASE;
+ tsb_hash_shift = PAGE_SHIFT;
+
+ spin_lock_irqsave(&mm->context.lock, flags);
+
+#ifdef CONFIG_HUGETLB_PAGE
+ if (mm->context.tsb_block[MM_TSB_HUGE].tsb != NULL) {
+ if ((tlb_type == hypervisor &&
+ (pte_val(pte) & _PAGE_SZALL_4V) == _PAGE_SZHUGE_4V) ||
+ (tlb_type != hypervisor &&
+ (pte_val(pte) & _PAGE_SZALL_4U) == _PAGE_SZHUGE_4U)) {
+ tsb_index = MM_TSB_HUGE;
+ tsb_hash_shift = HPAGE_SHIFT;
+ }
+ }
+#endif
+
+ tsb = mm->context.tsb_block[tsb_index].tsb;
+ tsb += ((address >> tsb_hash_shift) &
+ (mm->context.tsb_block[tsb_index].tsb_nentries - 1UL));
tag = (address >> 22UL);
tsb_insert(tsb, tag, pte_val(pte));
+
+ spin_unlock_irqrestore(&mm->context.lock, flags);
}
void flush_dcache_page(struct page *page)
if (tlb_type == spitfire) {
unsigned long kaddr;
- for (kaddr = start; kaddr < end; kaddr += PAGE_SIZE)
- __flush_icache_page(__get_phys(kaddr));
+ /* This code only runs on Spitfire cpus so this is
+ * why we can assume _PAGE_PADDR_4U.
+ */
+ for (kaddr = start; kaddr < end; kaddr += PAGE_SIZE) {
+ unsigned long paddr, mask = _PAGE_PADDR_4U;
+
+ if (kaddr >= PAGE_OFFSET)
+ paddr = kaddr & mask;
+ else {
+ pgd_t *pgdp = pgd_offset_k(kaddr);
+ pud_t *pudp = pud_offset(pgdp, kaddr);
+ pmd_t *pmdp = pmd_offset(pudp, kaddr);
+ pte_t *ptep = pte_offset_kernel(pmdp, kaddr);
+
+ paddr = pte_val(*ptep) & mask;
+ }
+ __flush_icache_page(paddr);
+ }
}
}
void show_mem(void)
{
- printk("Mem-info:\n");
+ unsigned long total = 0, reserved = 0;
+ unsigned long shared = 0, cached = 0;
+ pg_data_t *pgdat;
+
+ printk(KERN_INFO "Mem-info:\n");
show_free_areas();
- printk("Free swap: %6ldkB\n",
+ printk(KERN_INFO "Free swap: %6ldkB\n",
nr_swap_pages << (PAGE_SHIFT-10));
- printk("%ld pages of RAM\n", num_physpages);
- printk("%d free pages\n", nr_free_pages());
+ for_each_online_pgdat(pgdat) {
+ unsigned long i, flags;
+
+ pgdat_resize_lock(pgdat, &flags);
+ for (i = 0; i < pgdat->node_spanned_pages; i++) {
+ struct page *page = pgdat_page_nr(pgdat, i);
+ total++;
+ if (PageReserved(page))
+ reserved++;
+ else if (PageSwapCache(page))
+ cached++;
+ else if (page_count(page))
+ shared += page_count(page) - 1;
+ }
+ pgdat_resize_unlock(pgdat, &flags);
+ }
+
+ printk(KERN_INFO "%lu pages of RAM\n", total);
+ printk(KERN_INFO "%lu reserved pages\n", reserved);
+ printk(KERN_INFO "%lu pages shared\n", shared);
+ printk(KERN_INFO "%lu pages swap cached\n", cached);
+
+ printk(KERN_INFO "%lu pages dirty\n",
+ global_page_state(NR_FILE_DIRTY));
+ printk(KERN_INFO "%lu pages writeback\n",
+ global_page_state(NR_WRITEBACK));
+ printk(KERN_INFO "%lu pages mapped\n",
+ global_page_state(NR_FILE_MAPPED));
+ printk(KERN_INFO "%lu pages slab\n",
+ global_page_state(NR_SLAB_RECLAIMABLE) +
+ global_page_state(NR_SLAB_UNRECLAIMABLE));
+ printk(KERN_INFO "%lu pages pagetables\n",
+ global_page_state(NR_PAGETABLE));
}
void mmu_info(struct seq_file *m)
unsigned long pte,
unsigned long mmu)
{
- register unsigned long func asm("%o5");
- register unsigned long arg0 asm("%o0");
- register unsigned long arg1 asm("%o1");
- register unsigned long arg2 asm("%o2");
- register unsigned long arg3 asm("%o3");
-
- func = HV_FAST_MMU_MAP_PERM_ADDR;
- arg0 = vaddr;
- arg1 = 0;
- arg2 = pte;
- arg3 = mmu;
- __asm__ __volatile__("ta 0x80"
- : "=&r" (func), "=&r" (arg0),
- "=&r" (arg1), "=&r" (arg2),
- "=&r" (arg3)
- : "0" (func), "1" (arg0), "2" (arg1),
- "3" (arg2), "4" (arg3));
- if (arg0 != 0) {
+ unsigned long ret = sun4v_mmu_map_perm_addr(vaddr, 0, pte, mmu);
+
+ if (ret != 0) {
prom_printf("hypervisor_tlb_lock[%lx:%lx:%lx:%lx]: "
- "errors with %lx\n", vaddr, 0, pte, mmu, arg0);
+ "errors with %lx\n", vaddr, 0, pte, mmu, ret);
prom_halt();
}
}
static void __init remap_kernel(void)
{
unsigned long phys_page, tte_vaddr, tte_data;
- int tlb_ent = sparc64_highest_locked_tlbent();
+ int i, tlb_ent = sparc64_highest_locked_tlbent();
tte_vaddr = (unsigned long) KERNBASE;
phys_page = (prom_boot_mapping_phys_low >> 22UL) << 22UL;
/* Now lock us into the TLBs via Hypervisor or OBP. */
if (tlb_type == hypervisor) {
- hypervisor_tlb_lock(tte_vaddr, tte_data, HV_MMU_DMMU);
- hypervisor_tlb_lock(tte_vaddr, tte_data, HV_MMU_IMMU);
- if (bigkernel) {
- tte_vaddr += 0x400000;
- tte_data += 0x400000;
+ for (i = 0; i < num_kernel_image_mappings; i++) {
hypervisor_tlb_lock(tte_vaddr, tte_data, HV_MMU_DMMU);
hypervisor_tlb_lock(tte_vaddr, tte_data, HV_MMU_IMMU);
+ tte_vaddr += 0x400000;
+ tte_data += 0x400000;
}
} else {
- prom_dtlb_load(tlb_ent, tte_data, tte_vaddr);
- prom_itlb_load(tlb_ent, tte_data, tte_vaddr);
- if (bigkernel) {
- tlb_ent -= 1;
- prom_dtlb_load(tlb_ent,
- tte_data + 0x400000,
- tte_vaddr + 0x400000);
- prom_itlb_load(tlb_ent,
- tte_data + 0x400000,
- tte_vaddr + 0x400000);
+ for (i = 0; i < num_kernel_image_mappings; i++) {
+ prom_dtlb_load(tlb_ent - i, tte_data, tte_vaddr);
+ prom_itlb_load(tlb_ent - i, tte_data, tte_vaddr);
+ tte_vaddr += 0x400000;
+ tte_data += 0x400000;
}
- sparc64_highest_unlocked_tlb_ent = tlb_ent - 1;
+ sparc64_highest_unlocked_tlb_ent = tlb_ent - i;
}
if (tlb_type == cheetah_plus) {
sparc64_kern_pri_context = (CTX_CHEETAH_PLUS_CTX0 |
static void __init inherit_prom_mappings(void)
{
- read_obp_translations();
-
/* Now fixup OBP's idea about where we really are mapped. */
- prom_printf("Remapping the kernel... ");
+ printk("Remapping the kernel... ");
remap_kernel();
- prom_printf("done.\n");
+ printk("done.\n");
}
void prom_world(int enter)
__asm__ __volatile__("flushw");
}
-#ifdef DCACHE_ALIASING_POSSIBLE
void __flush_dcache_range(unsigned long start, unsigned long end)
{
unsigned long va;
"i" (ASI_DCACHE_INVALIDATE));
}
}
-#endif /* DCACHE_ALIASING_POSSIBLE */
+
+/* get_new_mmu_context() uses "cache + 1". */
+DEFINE_SPINLOCK(ctx_alloc_lock);
+unsigned long tlb_context_cache = CTX_FIRST_VERSION - 1;
+#define MAX_CTX_NR (1UL << CTX_NR_BITS)
+#define CTX_BMAP_SLOTS BITS_TO_LONGS(MAX_CTX_NR)
+DECLARE_BITMAP(mmu_context_bmap, MAX_CTX_NR);
/* Caller does TLB context flushing on local CPU if necessary.
* The caller also ensures that CTX_VALID(mm->context) is false.
smp_new_mmu_context_version();
}
-void sparc_ultra_dump_itlb(void)
+static int numa_enabled = 1;
+static int numa_debug;
+
+static int __init early_numa(char *p)
{
- int slot;
+ if (!p)
+ return 0;
- if (tlb_type == spitfire) {
- printk ("Contents of itlb: ");
- for (slot = 0; slot < 14; slot++) printk (" ");
- printk ("%2x:%016lx,%016lx\n",
- 0,
- spitfire_get_itlb_tag(0), spitfire_get_itlb_data(0));
- for (slot = 1; slot < 64; slot+=3) {
- printk ("%2x:%016lx,%016lx %2x:%016lx,%016lx %2x:%016lx,%016lx\n",
- slot,
- spitfire_get_itlb_tag(slot), spitfire_get_itlb_data(slot),
- slot+1,
- spitfire_get_itlb_tag(slot+1), spitfire_get_itlb_data(slot+1),
- slot+2,
- spitfire_get_itlb_tag(slot+2), spitfire_get_itlb_data(slot+2));
- }
- } else if (tlb_type == cheetah || tlb_type == cheetah_plus) {
- printk ("Contents of itlb0:\n");
- for (slot = 0; slot < 16; slot+=2) {
- printk ("%2x:%016lx,%016lx %2x:%016lx,%016lx\n",
- slot,
- cheetah_get_litlb_tag(slot), cheetah_get_litlb_data(slot),
- slot+1,
- cheetah_get_litlb_tag(slot+1), cheetah_get_litlb_data(slot+1));
- }
- printk ("Contents of itlb2:\n");
- for (slot = 0; slot < 128; slot+=2) {
- printk ("%2x:%016lx,%016lx %2x:%016lx,%016lx\n",
- slot,
- cheetah_get_itlb_tag(slot), cheetah_get_itlb_data(slot),
- slot+1,
- cheetah_get_itlb_tag(slot+1), cheetah_get_itlb_data(slot+1));
- }
+ if (strstr(p, "off"))
+ numa_enabled = 0;
+
+ if (strstr(p, "debug"))
+ numa_debug = 1;
+
+ return 0;
+}
+early_param("numa", early_numa);
+
+#define numadbg(f, a...) \
+do { if (numa_debug) \
+ printk(KERN_INFO f, ## a); \
+} while (0)
+
+static void __init find_ramdisk(unsigned long phys_base)
+{
+#ifdef CONFIG_BLK_DEV_INITRD
+ 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, sparc_ramdisk_size);
+
+ initrd_start += PAGE_OFFSET;
+ initrd_end += PAGE_OFFSET;
}
+#endif
}
-void sparc_ultra_dump_dtlb(void)
+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;
+
+int numa_cpu_lookup_table[NR_CPUS];
+cpumask_t numa_cpumask_lookup_table[MAX_NUMNODES];
+
+#ifdef CONFIG_NEED_MULTIPLE_NODES
+
+struct mdesc_mblock {
+ u64 base;
+ u64 size;
+ u64 offset; /* RA-to-PA */
+};
+static struct mdesc_mblock *mblocks;
+static int num_mblocks;
+
+static unsigned long ra_to_pa(unsigned long addr)
{
- int slot;
+ int i;
- if (tlb_type == spitfire) {
- printk ("Contents of dtlb: ");
- for (slot = 0; slot < 14; slot++) printk (" ");
- printk ("%2x:%016lx,%016lx\n", 0,
- spitfire_get_dtlb_tag(0), spitfire_get_dtlb_data(0));
- for (slot = 1; slot < 64; slot+=3) {
- printk ("%2x:%016lx,%016lx %2x:%016lx,%016lx %2x:%016lx,%016lx\n",
- slot,
- spitfire_get_dtlb_tag(slot), spitfire_get_dtlb_data(slot),
- slot+1,
- spitfire_get_dtlb_tag(slot+1), spitfire_get_dtlb_data(slot+1),
- slot+2,
- spitfire_get_dtlb_tag(slot+2), spitfire_get_dtlb_data(slot+2));
- }
- } else if (tlb_type == cheetah || tlb_type == cheetah_plus) {
- printk ("Contents of dtlb0:\n");
- for (slot = 0; slot < 16; slot+=2) {
- printk ("%2x:%016lx,%016lx %2x:%016lx,%016lx\n",
- slot,
- cheetah_get_ldtlb_tag(slot), cheetah_get_ldtlb_data(slot),
- slot+1,
- cheetah_get_ldtlb_tag(slot+1), cheetah_get_ldtlb_data(slot+1));
- }
- printk ("Contents of dtlb2:\n");
- for (slot = 0; slot < 512; slot+=2) {
- printk ("%2x:%016lx,%016lx %2x:%016lx,%016lx\n",
- slot,
- cheetah_get_dtlb_tag(slot, 2), cheetah_get_dtlb_data(slot, 2),
- slot+1,
- cheetah_get_dtlb_tag(slot+1, 2), cheetah_get_dtlb_data(slot+1, 2));
- }
- if (tlb_type == cheetah_plus) {
- printk ("Contents of dtlb3:\n");
- for (slot = 0; slot < 512; slot+=2) {
- printk ("%2x:%016lx,%016lx %2x:%016lx,%016lx\n",
- slot,
- cheetah_get_dtlb_tag(slot, 3), cheetah_get_dtlb_data(slot, 3),
- slot+1,
- cheetah_get_dtlb_tag(slot+1, 3), cheetah_get_dtlb_data(slot+1, 3));
- }
+ for (i = 0; i < num_mblocks; i++) {
+ struct mdesc_mblock *m = &mblocks[i];
+
+ if (addr >= m->base &&
+ addr < (m->base + m->size)) {
+ addr += m->offset;
+ break;
}
}
+ return addr;
}
-extern unsigned long cmdline_memory_size;
-
-/* 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 find_node(unsigned long addr)
{
- unsigned long avoid_start, avoid_end, bootmap_size;
int i;
- bootmap_size = ((end_pfn - start_pfn) + 7) / 8;
- bootmap_size = ALIGN(bootmap_size, sizeof(long));
+ 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
-#ifdef CONFIG_DEBUG_BOOTMEM
- prom_printf("choose_bootmap_pfn: kern[%lx:%lx] avoid[%lx:%lx]\n",
- kern_base, PAGE_ALIGN(kern_base + kern_size),
- avoid_start, avoid_end);
+/* 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));
+
+ NODE_DATA(nid)->bdata = &bootmem_node_data[nid];
#endif
- for (i = 0; i < pavail_ents; i++) {
- unsigned long start, end;
- start = pavail[i].phys_addr;
- end = start + pavail[i].reg_size;
+ p = NODE_DATA(nid);
- 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;
- }
+ 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 ((end - start) < bootmap_size)
- break;
+ if (p->node_spanned_pages) {
+ num_pages = bootmem_bootmap_pages(p->node_spanned_pages);
- if (start < kern_base &&
- (start + bootmap_size) > kern_base) {
- start = PAGE_ALIGN(kern_base + kern_size);
- continue;
- }
+ 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;
+ }
+}
- if (start < avoid_start &&
- (start + bootmap_size) > avoid_start) {
- start = avoid_end;
- continue;
- }
+static void init_node_masks_nonnuma(void)
+{
+ int i;
- /* OK, it doesn't overlap anything, use it. */
-#ifdef CONFIG_DEBUG_BOOTMEM
- prom_printf("choose_bootmap_pfn: Using %lx [%lx]\n",
- start >> PAGE_SHIFT, start);
-#endif
- return start >> PAGE_SHIFT;
+ 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;
+
+ 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;
+
+ 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 (!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++;
}
- prom_printf("Cannot find free area for bootmap, aborting.\n");
- prom_halt();
+ mdesc_release(md);
+
+ return nid;
}
-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;
-#ifdef CONFIG_DEBUG_BOOTMEM
- prom_printf("bootmem_init: Scan pavail, ");
-#endif
+ for (i = 0; i < lmb.memory.cnt; i++) {
+ unsigned long size = lmb_size_bytes(&lmb.memory, i);
+ unsigned long start, end;
- 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;
- if (cmdline_memory_size) {
- if (bytes_avail > cmdline_memory_size) {
- unsigned long slack = bytes_avail - cmdline_memory_size;
-
- bytes_avail -= slack;
- end_of_phys_memory -= slack;
-
- pavail[i].reg_size -= slack;
- if ((long)pavail[i].reg_size <= 0L) {
- pavail[i].phys_addr = 0xdeadbeefUL;
- pavail[i].reg_size = 0UL;
- pavail_ents = i;
- } else {
- pavail[i+1].reg_size = 0Ul;
- pavail[i+1].phys_addr = 0xdeadbeefUL;
- pavail_ents = i + 1;
- }
- break;
- }
+ 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;
}
}
+}
+
+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);
+ }
- *pages_avail = bytes_avail >> PAGE_SHIFT;
+ return 0;
+}
- end_pfn = end_of_phys_memory >> PAGE_SHIFT;
+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);
+ }
-#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;
- if (ramdisk_image >= (unsigned long)_end - 2 * PAGE_SIZE)
- 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 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
- /* Initialize the boot-time allocator. */
- max_pfn = max_low_pfn = end_pfn;
- min_low_pfn = (phys_base >> PAGE_SHIFT);
+ 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;
+ }
- bootmap_pfn = choose_bootmap_pfn(min_low_pfn, end_pfn);
+ n = &node_masks[num_node_masks++];
-#ifdef CONFIG_DEBUG_BOOTMEM
- prom_printf("init_bootmem(min[%lx], bootmap[%lx], max[%lx])\n",
- min_low_pfn, bootmap_pfn, max_low_pfn);
-#endif
- bootmap_size = init_bootmem_node(NODE_DATA(0), bootmap_pfn,
- min_low_pfn, end_pfn);
+ n->mask = candidate->mask;
+ n->val = candidate->match;
- /* Now register the available physical memory with the
- * allocator.
- */
- for (i = 0; i < pavail_ents; i++) {
-#ifdef CONFIG_DEBUG_BOOTMEM
- prom_printf("free_bootmem(pavail:%d): base[%lx] size[%lx]\n",
- i, pavail[i].phys_addr, pavail[i].reg_size);
-#endif
- free_bootmem(pavail[i].phys_addr, pavail[i].reg_size);
+ numadbg("NUMA NODE[%d]: mask[%lx] val[%lx] (latency[%lx])\n",
+ index, n->mask, n->val, candidate->latency);
+
+ return 0;
+}
+
+static int __init numa_parse_mdesc_group(struct mdesc_handle *md, u64 grp,
+ int index)
+{
+ cpumask_t mask;
+ int cpu;
+
+ numa_parse_mdesc_group_cpus(md, grp, &mask);
+
+ for_each_cpu_mask(cpu, mask)
+ numa_cpu_lookup_table[cpu] = index;
+ numa_cpumask_lookup_table[index] = mask;
+
+ 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);
+}
- /* Resert the initrd image area. */
-#ifdef CONFIG_DEBUG_BOOTMEM
- prom_printf("reserve_bootmem(initrd): base[%llx] size[%lx]\n",
- initrd_start, initrd_end);
-#endif
- reserve_bootmem(initrd_start, size);
- *pages_avail -= PAGE_ALIGN(size) >> PAGE_SHIFT;
+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;
+ }
- initrd_start += PAGE_OFFSET;
- initrd_end += PAGE_OFFSET;
+ 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++;
}
-#endif
- /* Reserve the kernel text/data/bss. */
-#ifdef CONFIG_DEBUG_BOOTMEM
- prom_printf("reserve_bootmem(kernel): base[%lx] size[%lx]\n", kern_base, kern_size);
-#endif
- reserve_bootmem(kern_base, kern_size);
- *pages_avail -= PAGE_ALIGN(kern_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;
-#ifdef CONFIG_DEBUG_BOOTMEM
- prom_printf("reserve_bootmem(bootmap): base[%lx] size[%lx]\n",
- (bootmap_pfn << PAGE_SHIFT), size);
+ 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;
+}
+
+static int __init bootmem_init_numa(void)
+{
+ int err = -1;
+
+ 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_bootmem((bootmap_pfn << PAGE_SHIFT), size);
- *pages_avail -= PAGE_ALIGN(size) >> PAGE_SHIFT;
- 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));
-#ifdef CONFIG_DEBUG_BOOTMEM
- prom_printf("memory_present(0, %lx, %lx)\n",
- start_pfn, end_pfn);
-#endif
- 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();
static int pall_ents __initdata;
#ifdef CONFIG_DEBUG_PAGEALLOC
-static unsigned long kernel_map_range(unsigned long pstart, unsigned long pend, pgprot_t prot)
+static unsigned long __ref kernel_map_range(unsigned long pstart,
+ unsigned long pend, pgprot_t prot)
{
unsigned long vstart = PAGE_OFFSET + pstart;
unsigned long vend = PAGE_OFFSET + pend;
}
}
-static void __init kernel_physical_mapping_init(void)
+static void __init init_kpte_bitmap(void)
{
unsigned long i;
-#ifdef CONFIG_DEBUG_PAGEALLOC
- unsigned long mem_alloced = 0UL;
-#endif
-
- read_obp_memory("reg", &pall[0], &pall_ents);
for (i = 0; i < pall_ents; i++) {
unsigned long phys_start, phys_end;
phys_end = phys_start + pall[i].reg_size;
mark_kpte_bitmap(phys_start, phys_end);
+ }
+}
+static void __init kernel_physical_mapping_init(void)
+{
#ifdef CONFIG_DEBUG_PAGEALLOC
+ unsigned long i, mem_alloced = 0UL;
+
+ for (i = 0; i < pall_ents; i++) {
+ unsigned long phys_start, phys_end;
+
+ phys_start = pall[i].phys_addr;
+ phys_end = phys_start + pall[i].reg_size;
+
mem_alloced += kernel_map_range(phys_start, phys_end,
PAGE_KERNEL);
-#endif
}
-#ifdef CONFIG_DEBUG_PAGEALLOC
printk("Allocated %ld bytes for kernel page tables.\n",
mem_alloced);
}
/* Don't mark as init, we give this to the Hypervisor. */
-static struct hv_tsb_descr ktsb_descr[2];
+#ifndef CONFIG_DEBUG_PAGEALLOC
+#define NUM_KTSB_DESCR 2
+#else
+#define NUM_KTSB_DESCR 1
+#endif
+static struct hv_tsb_descr ktsb_descr[NUM_KTSB_DESCR];
extern struct tsb swapper_tsb[KERNEL_TSB_NENTRIES];
static void __init sun4v_ktsb_init(void)
ktsb_descr[0].tsb_base = ktsb_pa;
ktsb_descr[0].resv = 0;
+#ifndef CONFIG_DEBUG_PAGEALLOC
/* Second KTSB for 4MB/256MB mappings. */
ktsb_pa = (kern_base +
((unsigned long)&swapper_4m_tsb[0] - KERNBASE));
ktsb_descr[1].ctx_idx = 0;
ktsb_descr[1].tsb_base = ktsb_pa;
ktsb_descr[1].resv = 0;
+#endif
}
void __cpuinit sun4v_ktsb_register(void)
{
- register unsigned long func asm("%o5");
- register unsigned long arg0 asm("%o0");
- register unsigned long arg1 asm("%o1");
- unsigned long pa;
+ unsigned long pa, ret;
pa = kern_base + ((unsigned long)&ktsb_descr[0] - KERNBASE);
- func = HV_FAST_MMU_TSB_CTX0;
- arg0 = 2;
- arg1 = pa;
- __asm__ __volatile__("ta %6"
- : "=&r" (func), "=&r" (arg0), "=&r" (arg1)
- : "0" (func), "1" (arg0), "2" (arg1),
- "i" (HV_FAST_TRAP));
+ ret = sun4v_mmu_tsb_ctx0(NUM_KTSB_DESCR, pa);
+ if (ret != 0) {
+ prom_printf("hypervisor_mmu_tsb_ctx0[%lx]: "
+ "errors with %lx\n", pa, ret);
+ prom_halt();
+ }
}
/* paging_init() sets up the page tables */
-extern void cheetah_ecache_flush_init(void);
-extern void sun4v_patch_tlb_handlers(void);
+extern void central_probe(void);
static unsigned long last_valid_pfn;
pgd_t swapper_pg_dir[2048];
static void sun4u_pgprot_init(void);
static void sun4v_pgprot_init(void);
+/* Dummy function */
+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()
+ * page->flags usage will work.
+ *
+ * When a page gets marked as dcache-dirty, we store the
+ * cpu number starting at bit 32 in the page->flags. Also,
+ * functions like clear_dcache_dirty_cpu use the cpu mask
+ * in 13-bit signed-immediate instruction fields.
+ */
+
+ /*
+ * Page flags must not reach into upper 32 bits that are used
+ * for the cpu number
+ */
+ BUILD_BUG_ON(NR_PAGEFLAGS > 32);
+
+ /*
+ * The bit fields placed in the high range must not reach below
+ * the 32 bit boundary. Otherwise we cannot place the cpu field
+ * at the 32 bit boundary.
+ */
+ BUILD_BUG_ON(SECTIONS_WIDTH + NODES_WIDTH + ZONES_WIDTH +
+ ilog2(roundup_pow_of_two(NR_CPUS)) > 32);
+
+ BUILD_BUG_ON(NR_CPUS > 4096);
+
kern_base = (prom_boot_mapping_phys_low >> 22UL) << 22UL;
kern_size = (unsigned long)&_end - (unsigned long)KERNBASE;
+ sstate_booting();
+
/* Invalidate both kernel TSBs. */
memset(swapper_tsb, 0x40, sizeof(swapper_tsb));
+#ifndef CONFIG_DEBUG_PAGEALLOC
memset(swapper_4m_tsb, 0x40, sizeof(swapper_4m_tsb));
+#endif
if (tlb_type == hypervisor)
sun4v_pgprot_init();
sun4v_ktsb_init();
}
- /* Find available physical memory... */
+ lmb_init();
+
+ /* Find available physical memory...
+ *
+ * Read it twice in order to work around a bug in openfirmware.
+ * The call to grab this table itself can cause openfirmware to
+ * allocate memory, which in turn can take away some space from
+ * the list of available memory. Reading it twice makes sure
+ * we really do get the final value.
+ */
+ read_obp_translations();
+ read_obp_memory("reg", &pall[0], &pall_ents);
+ read_obp_memory("available", &pavail[0], &pavail_ents);
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);
shift = kern_base + PAGE_OFFSET - ((unsigned long)KERNBASE);
real_end = (unsigned long)_end;
- if ((real_end > ((unsigned long)KERNBASE + 0x400000)))
- bigkernel = 1;
- if ((real_end > ((unsigned long)KERNBASE + 0x800000))) {
- prom_printf("paging_init: Kernel > 8MB, too large.\n");
- prom_halt();
- }
+ num_kernel_image_mappings = DIV_ROUND_UP(real_end - KERNBASE, 1 << 22);
+ printk("Kernel: Using %d locked TLB entries for main kernel image.\n",
+ num_kernel_image_mappings);
/* Set kernel pgd to upper alias so physical page computations
* work.
inherit_prom_mappings();
+ init_kpte_bitmap();
+
/* Ok, we can use our TLB miss and window trap handlers safely. */
setup_tba();
if (tlb_type == hypervisor)
sun4v_ktsb_register();
+ /* 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();
+
+ if (tlb_type == hypervisor)
+ sun4v_mdesc_init();
+
/* Setup bootmem... */
- pages_avail = 0;
- last_valid_pfn = end_pfn = bootmem_init(&pages_avail, phys_base);
+ 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_DMA] = end_pfn;
- zholes_size[ZONE_DMA] = 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);
}
- device_scan();
+ printk("Booting Linux...\n");
+
+ central_probe();
+ 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;
- for (n = 0; pavail_rescan_ents; n++) {
+ for (n = 0; n < pavail_rescan_ents; n++) {
unsigned long new_start, new_end;
new_start = pavail_rescan[n].phys_addr;
goto do_next_page;
}
}
- reserve_bootmem(old_start, PAGE_SIZE);
+
+ 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;
addr += PAGE_SIZE;
}
- taint_real_pages();
+ setup_valid_addr_bitmap_from_pavail();
high_memory = __va(last_valid_pfn << PAGE_SHIFT);
-#ifdef CONFIG_DEBUG_BOOTMEM
- prom_printf("mem_init: Calling free_all_bootmem().\n");
+#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
- totalram_pages = num_physpages = free_all_bootmem() - 1;
+
+ /* We subtract one to account for the mem_map_zero page
+ * allocated below.
+ */
+ totalram_pages -= 1;
+ num_physpages = totalram_pages;
/*
* Set up the zero page, mark it reserved, so that page count
initpages = (((unsigned long) __init_end) - ((unsigned long) __init_begin));
initpages = PAGE_ALIGN(initpages) >> PAGE_SHIFT;
- printk("Memory: %uk available (%ldk kernel code, %ldk data, %ldk init) [%016lx,%016lx]\n",
+ printk("Memory: %luk available (%ldk kernel code, %ldk data, %ldk init) [%016lx,%016lx]\n",
nr_free_pages() << (PAGE_SHIFT-10),
codepages << (PAGE_SHIFT-10),
datapages << (PAGE_SHIFT-10),
page = (addr +
((unsigned long) __va(kern_base)) -
((unsigned long) KERNBASE));
- memset((void *)addr, 0xcc, PAGE_SIZE);
+ memset((void *)addr, POISON_FREE_INITMEM, PAGE_SIZE);
p = virt_to_page(page);
ClearPageReserved(p);
- set_page_count(p, 1);
+ init_page_count(p);
__free_page(p);
num_physpages++;
totalram_pages++;
struct page *p = virt_to_page(start);
ClearPageReserved(p);
- set_page_count(p, 1);
+ init_page_count(p);
__free_page(p);
num_physpages++;
totalram_pages++;
unsigned long pg_iobits __read_mostly;
unsigned long _PAGE_IE __read_mostly;
+EXPORT_SYMBOL(_PAGE_IE);
unsigned long _PAGE_E __read_mostly;
EXPORT_SYMBOL(_PAGE_E);
unsigned long _PAGE_CACHE __read_mostly;
EXPORT_SYMBOL(_PAGE_CACHE);
+#ifdef CONFIG_SPARSEMEM_VMEMMAP
+
+#define VMEMMAP_CHUNK_SHIFT 22
+#define VMEMMAP_CHUNK (1UL << VMEMMAP_CHUNK_SHIFT)
+#define VMEMMAP_CHUNK_MASK ~(VMEMMAP_CHUNK - 1UL)
+#define VMEMMAP_ALIGN(x) (((x)+VMEMMAP_CHUNK-1UL)&VMEMMAP_CHUNK_MASK)
+
+#define VMEMMAP_SIZE ((((1UL << MAX_PHYSADDR_BITS) >> PAGE_SHIFT) * \
+ sizeof(struct page *)) >> VMEMMAP_CHUNK_SHIFT)
+unsigned long vmemmap_table[VMEMMAP_SIZE];
+
+int __meminit vmemmap_populate(struct page *start, unsigned long nr, int node)
+{
+ unsigned long vstart = (unsigned long) start;
+ unsigned long vend = (unsigned long) (start + nr);
+ unsigned long phys_start = (vstart - VMEMMAP_BASE);
+ unsigned long phys_end = (vend - VMEMMAP_BASE);
+ unsigned long addr = phys_start & VMEMMAP_CHUNK_MASK;
+ unsigned long end = VMEMMAP_ALIGN(phys_end);
+ unsigned long pte_base;
+
+ pte_base = (_PAGE_VALID | _PAGE_SZ4MB_4U |
+ _PAGE_CP_4U | _PAGE_CV_4U |
+ _PAGE_P_4U | _PAGE_W_4U);
+ if (tlb_type == hypervisor)
+ pte_base = (_PAGE_VALID | _PAGE_SZ4MB_4V |
+ _PAGE_CP_4V | _PAGE_CV_4V |
+ _PAGE_P_4V | _PAGE_W_4V);
+
+ for (; addr < end; addr += VMEMMAP_CHUNK) {
+ unsigned long *vmem_pp =
+ vmemmap_table + (addr >> VMEMMAP_CHUNK_SHIFT);
+ void *block;
+
+ if (!(*vmem_pp & _PAGE_VALID)) {
+ block = vmemmap_alloc_block(1UL << 22, node);
+ if (!block)
+ return -ENOMEM;
+
+ *vmem_pp = pte_base | __pa(block);
+
+ printk(KERN_INFO "[%p-%p] page_structs=%lu "
+ "node=%d entry=%lu/%lu\n", start, block, nr,
+ node,
+ addr >> VMEMMAP_CHUNK_SHIFT,
+ VMEMMAP_SIZE >> VMEMMAP_CHUNK_SHIFT);
+ }
+ }
+ return 0;
+}
+#endif /* CONFIG_SPARSEMEM_VMEMMAP */
+
static void prot_init_common(unsigned long page_none,
unsigned long page_shared,
unsigned long page_copy,
pg_iobits = (_PAGE_VALID | _PAGE_PRESENT_4U | __DIRTY_BITS_4U |
__ACCESS_BITS_4U | _PAGE_E_4U);
+#ifdef CONFIG_DEBUG_PAGEALLOC
+ kern_linear_pte_xor[0] = (_PAGE_VALID | _PAGE_SZBITS_4U) ^
+ 0xfffff80000000000;
+#else
kern_linear_pte_xor[0] = (_PAGE_VALID | _PAGE_SZ4MB_4U) ^
0xfffff80000000000;
+#endif
kern_linear_pte_xor[0] |= (_PAGE_CP_4U | _PAGE_CV_4U |
_PAGE_P_4U | _PAGE_W_4U);
_PAGE_E = _PAGE_E_4V;
_PAGE_CACHE = _PAGE_CACHE_4V;
+#ifdef CONFIG_DEBUG_PAGEALLOC
+ kern_linear_pte_xor[0] = (_PAGE_VALID | _PAGE_SZBITS_4V) ^
+ 0xfffff80000000000;
+#else
kern_linear_pte_xor[0] = (_PAGE_VALID | _PAGE_SZ4MB_4V) ^
0xfffff80000000000;
+#endif
kern_linear_pte_xor[0] |= (_PAGE_CP_4V | _PAGE_CV_4V |
_PAGE_P_4V | _PAGE_W_4V);
+#ifdef CONFIG_DEBUG_PAGEALLOC
+ kern_linear_pte_xor[1] = (_PAGE_VALID | _PAGE_SZBITS_4V) ^
+ 0xfffff80000000000;
+#else
kern_linear_pte_xor[1] = (_PAGE_VALID | _PAGE_SZ256MB_4V) ^
0xfffff80000000000;
+#endif
kern_linear_pte_xor[1] |= (_PAGE_CP_4V | _PAGE_CV_4V |
_PAGE_P_4V | _PAGE_W_4V);
return val | paddr;
}
-/*
- * Translate PROM's mapping we capture at boot time into physical address.
- * The second parameter is only set from prom_callback() invocations.
- */
-unsigned long prom_virt_to_phys(unsigned long promva, int *error)
-{
- unsigned long mask;
- int i;
-
- mask = _PAGE_PADDR_4U;
- if (tlb_type == hypervisor)
- mask = _PAGE_PADDR_4V;
-
- for (i = 0; i < prom_trans_ents; i++) {
- struct linux_prom_translation *p = &prom_trans[i];
-
- if (promva >= p->virt &&
- promva < (p->virt + p->size)) {
- unsigned long base = p->data & mask;
-
- if (error)
- *error = 0;
- return base + (promva & (8192 - 1));
- }
- }
- if (error)
- *error = 1;
- return 0UL;
-}
-
-/* XXX We should kill off this ugly thing at so me point. XXX */
-unsigned long sun4u_get_pte(unsigned long addr)
-{
- pgd_t *pgdp;
- pud_t *pudp;
- pmd_t *pmdp;
- pte_t *ptep;
- unsigned long mask = _PAGE_PADDR_4U;
-
- if (tlb_type == hypervisor)
- mask = _PAGE_PADDR_4V;
-
- if (addr >= PAGE_OFFSET)
- return addr & mask;
-
- if ((addr >= LOW_OBP_ADDRESS) && (addr < HI_OBP_ADDRESS))
- return prom_virt_to_phys(addr, NULL);
-
- pgdp = pgd_offset_k(addr);
- pudp = pud_offset(pgdp, addr);
- pmdp = pmd_offset(pudp, addr);
- ptep = pte_offset_kernel(pmdp, addr);
-
- return pte_val(*ptep) & mask;
-}
-
/* If not locked, zap it. */
void __flush_tlb_all(void)
{
"wrpr %0, %1, %%pstate"
: "=r" (pstate)
: "i" (PSTATE_IE));
- if (tlb_type == spitfire) {
+ if (tlb_type == hypervisor) {
+ sun4v_mmu_demap_all();
+ } else if (tlb_type == spitfire) {
for (i = 0; i < 64; i++) {
/* Spitfire Errata #32 workaround */
/* NOTE: Always runs on spitfire, so no
git://ftp.safe.ca / safe / jmp / linux-2.6 / blobdiff