diff options
Diffstat (limited to 'arch/powerpc/mm')
33 files changed, 3090 insertions, 1052 deletions
diff --git a/arch/powerpc/mm/44x_mmu.c b/arch/powerpc/mm/44x_mmu.c index 2c9441ee6bb..82b1ff759e2 100644 --- a/arch/powerpc/mm/44x_mmu.c +++ b/arch/powerpc/mm/44x_mmu.c @@ -41,7 +41,7 @@ int icache_44x_need_flush; unsigned long tlb_47x_boltmap[1024/8]; -static void __cpuinit ppc44x_update_tlb_hwater(void) +static void ppc44x_update_tlb_hwater(void) { extern unsigned int tlb_44x_patch_hwater_D[]; extern unsigned int tlb_44x_patch_hwater_I[]; @@ -134,7 +134,7 @@ static void __init ppc47x_update_boltmap(void) /* * "Pins" a 256MB TLB entry in AS0 for kernel lowmem for 47x type MMU */ -static void __cpuinit ppc47x_pin_tlb(unsigned int virt, unsigned int phys) +static void ppc47x_pin_tlb(unsigned int virt, unsigned int phys) { unsigned int rA; int bolted; @@ -229,7 +229,7 @@ void setup_initial_memory_limit(phys_addr_t first_memblock_base, } #ifdef CONFIG_SMP -void __cpuinit mmu_init_secondary(int cpu) +void mmu_init_secondary(int cpu) { unsigned long addr; unsigned long memstart = memstart_addr & ~(PPC_PIN_SIZE - 1); diff --git a/arch/powerpc/mm/Makefile b/arch/powerpc/mm/Makefile index 3787b61f7d2..51230ee6a40 100644 --- a/arch/powerpc/mm/Makefile +++ b/arch/powerpc/mm/Makefile @@ -4,19 +4,18 @@ subdir-ccflags-$(CONFIG_PPC_WERROR) := -Werror -ccflags-$(CONFIG_PPC64) := -mno-minimal-toc +ccflags-$(CONFIG_PPC64) := $(NO_MINIMAL_TOC) -obj-y := fault.o mem.o pgtable.o gup.o \ +obj-y := fault.o mem.o pgtable.o gup.o mmap.o \ init_$(CONFIG_WORD_SIZE).o \ pgtable_$(CONFIG_WORD_SIZE).o obj-$(CONFIG_PPC_MMU_NOHASH) += mmu_context_nohash.o tlb_nohash.o \ tlb_nohash_low.o obj-$(CONFIG_PPC_BOOK3E) += tlb_low_$(CONFIG_WORD_SIZE)e.o -obj-$(CONFIG_PPC64) += mmap_64.o hash64-$(CONFIG_PPC_NATIVE) := hash_native_64.o obj-$(CONFIG_PPC_STD_MMU_64) += hash_utils_64.o \ slb_low.o slb.o stab.o \ - mmap_64.o $(hash64-y) + $(hash64-y) obj-$(CONFIG_PPC_STD_MMU_32) += ppc_mmu_32.o obj-$(CONFIG_PPC_STD_MMU) += hash_low_$(CONFIG_WORD_SIZE).o \ tlb_hash$(CONFIG_WORD_SIZE).o \ @@ -28,11 +27,12 @@ obj-$(CONFIG_44x) += 44x_mmu.o obj-$(CONFIG_PPC_FSL_BOOK3E) += fsl_booke_mmu.o obj-$(CONFIG_NEED_MULTIPLE_NODES) += numa.o obj-$(CONFIG_PPC_MM_SLICES) += slice.o -ifeq ($(CONFIG_HUGETLB_PAGE),y) obj-y += hugetlbpage.o +ifeq ($(CONFIG_HUGETLB_PAGE),y) obj-$(CONFIG_PPC_STD_MMU_64) += hugetlbpage-hash64.o obj-$(CONFIG_PPC_BOOK3E_MMU) += hugetlbpage-book3e.o endif +obj-$(CONFIG_TRANSPARENT_HUGEPAGE) += hugepage-hash64.o obj-$(CONFIG_PPC_SUBPAGE_PROT) += subpage-prot.o obj-$(CONFIG_NOT_COHERENT_CACHE) += dma-noncoherent.o obj-$(CONFIG_HIGHMEM) += highmem.o diff --git a/arch/powerpc/mm/dma-noncoherent.c b/arch/powerpc/mm/dma-noncoherent.c index 6747eece84a..7b6c1075017 100644 --- a/arch/powerpc/mm/dma-noncoherent.c +++ b/arch/powerpc/mm/dma-noncoherent.c @@ -287,9 +287,7 @@ void __dma_free_coherent(size_t size, void *vaddr) pte_clear(&init_mm, addr, ptep); if (pfn_valid(pfn)) { struct page *page = pfn_to_page(pfn); - - ClearPageReserved(page); - __free_page(page); + __free_reserved_page(page); } } addr += PAGE_SIZE; diff --git a/arch/powerpc/mm/fault.c b/arch/powerpc/mm/fault.c index 3a8489a354e..51ab9e7e6c3 100644 --- a/arch/powerpc/mm/fault.c +++ b/arch/powerpc/mm/fault.c @@ -32,6 +32,7 @@ #include <linux/perf_event.h> #include <linux/magic.h> #include <linux/ratelimit.h> +#include <linux/context_tracking.h> #include <asm/firmware.h> #include <asm/page.h> @@ -196,6 +197,7 @@ static int mm_fault_error(struct pt_regs *regs, unsigned long addr, int fault) int __kprobes do_page_fault(struct pt_regs *regs, unsigned long address, unsigned long error_code) { + enum ctx_state prev_state = exception_enter(); struct vm_area_struct * vma; struct mm_struct *mm = current->mm; unsigned int flags = FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_KILLABLE; @@ -204,6 +206,7 @@ int __kprobes do_page_fault(struct pt_regs *regs, unsigned long address, int trap = TRAP(regs); int is_exec = trap == 0x400; int fault; + int rc = 0, store_update_sp = 0; #if !(defined(CONFIG_4xx) || defined(CONFIG_BOOKE)) /* @@ -220,9 +223,6 @@ int __kprobes do_page_fault(struct pt_regs *regs, unsigned long address, is_write = error_code & ESR_DST; #endif /* CONFIG_4xx || CONFIG_BOOKE */ - if (is_write) - flags |= FAULT_FLAG_WRITE; - #ifdef CONFIG_PPC_ICSWX /* * we need to do this early because this "data storage @@ -230,28 +230,30 @@ int __kprobes do_page_fault(struct pt_regs *regs, unsigned long address, * look at it */ if (error_code & ICSWX_DSI_UCT) { - int rc = acop_handle_fault(regs, address, error_code); + rc = acop_handle_fault(regs, address, error_code); if (rc) - return rc; + goto bail; } #endif /* CONFIG_PPC_ICSWX */ if (notify_page_fault(regs)) - return 0; + goto bail; if (unlikely(debugger_fault_handler(regs))) - return 0; + goto bail; /* On a kernel SLB miss we can only check for a valid exception entry */ - if (!user_mode(regs) && (address >= TASK_SIZE)) - return SIGSEGV; + if (!user_mode(regs) && (address >= TASK_SIZE)) { + rc = SIGSEGV; + goto bail; + } #if !(defined(CONFIG_4xx) || defined(CONFIG_BOOKE) || \ defined(CONFIG_PPC_BOOK3S_64)) if (error_code & DSISR_DABRMATCH) { - /* DABR match */ - do_dabr(regs, address, error_code); - return 0; + /* breakpoint match */ + do_break(regs, address, error_code); + goto bail; } #endif @@ -260,8 +262,10 @@ int __kprobes do_page_fault(struct pt_regs *regs, unsigned long address, local_irq_enable(); if (in_atomic() || mm == NULL) { - if (!user_mode(regs)) - return SIGSEGV; + if (!user_mode(regs)) { + rc = SIGSEGV; + goto bail; + } /* in_atomic() in user mode is really bad, as is current->mm == NULL. */ printk(KERN_EMERG "Page fault in user mode with " @@ -273,6 +277,17 @@ int __kprobes do_page_fault(struct pt_regs *regs, unsigned long address, perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, regs, address); + /* + * We want to do this outside mmap_sem, because reading code around nip + * can result in fault, which will cause a deadlock when called with + * mmap_sem held + */ + if (user_mode(regs)) + store_update_sp = store_updates_sp(regs); + + if (user_mode(regs)) + flags |= FAULT_FLAG_USER; + /* When running in the kernel we expect faults to occur only to * addresses in user space. All other faults represent errors in the * kernel and should generate an OOPS. Unfortunately, in the case of an @@ -338,8 +353,7 @@ retry: * between the last mapped region and the stack will * expand the stack rather than segfaulting. */ - if (address + 2048 < uregs->gpr[1] - && (!user_mode(regs) || !store_updates_sp(regs))) + if (address + 2048 < uregs->gpr[1] && !store_update_sp) goto bad_area; } if (expand_stack(vma, address)) @@ -401,6 +415,7 @@ good_area: } else if (is_write) { if (!(vma->vm_flags & VM_WRITE)) goto bad_area; + flags |= FAULT_FLAG_WRITE; /* a read */ } else { /* protection fault */ @@ -417,9 +432,11 @@ good_area: */ fault = handle_mm_fault(mm, vma, address, flags); if (unlikely(fault & (VM_FAULT_RETRY|VM_FAULT_ERROR))) { - int rc = mm_fault_error(regs, address, fault); + rc = mm_fault_error(regs, address, fault); if (rc >= MM_FAULT_RETURN) - return rc; + goto bail; + else + rc = 0; } /* @@ -434,8 +451,12 @@ good_area: regs, address); #ifdef CONFIG_PPC_SMLPAR if (firmware_has_feature(FW_FEATURE_CMO)) { + u32 page_ins; + preempt_disable(); - get_lppaca()->page_ins += (1 << PAGE_FACTOR); + page_ins = be32_to_cpu(get_lppaca()->page_ins); + page_ins += 1 << PAGE_FACTOR; + get_lppaca()->page_ins = cpu_to_be32(page_ins); preempt_enable(); } #endif /* CONFIG_PPC_SMLPAR */ @@ -454,7 +475,7 @@ good_area: } up_read(&mm->mmap_sem); - return 0; + goto bail; bad_area: up_read(&mm->mmap_sem); @@ -463,7 +484,7 @@ bad_area_nosemaphore: /* User mode accesses cause a SIGSEGV */ if (user_mode(regs)) { _exception(SIGSEGV, regs, code, address); - return 0; + goto bail; } if (is_exec && (error_code & DSISR_PROTFAULT)) @@ -471,7 +492,11 @@ bad_area_nosemaphore: " page (%lx) - exploit attempt? (uid: %d)\n", address, from_kuid(&init_user_ns, current_uid())); - return SIGSEGV; + rc = SIGSEGV; + +bail: + exception_exit(prev_state); + return rc; } diff --git a/arch/powerpc/mm/fsl_booke_mmu.c b/arch/powerpc/mm/fsl_booke_mmu.c index 07ba45b0f07..94cd728166d 100644 --- a/arch/powerpc/mm/fsl_booke_mmu.c +++ b/arch/powerpc/mm/fsl_booke_mmu.c @@ -52,6 +52,7 @@ #include <asm/smp.h> #include <asm/machdep.h> #include <asm/setup.h> +#include <asm/paca.h> #include "mmu_decl.h" @@ -171,11 +172,10 @@ unsigned long calc_cam_sz(unsigned long ram, unsigned long virt, return 1UL << camsize; } -unsigned long map_mem_in_cams(unsigned long ram, int max_cam_idx) +static unsigned long map_mem_in_cams_addr(phys_addr_t phys, unsigned long virt, + unsigned long ram, int max_cam_idx) { int i; - unsigned long virt = PAGE_OFFSET; - phys_addr_t phys = memstart_addr; unsigned long amount_mapped = 0; /* Calculate CAM values */ @@ -192,9 +192,23 @@ unsigned long map_mem_in_cams(unsigned long ram, int max_cam_idx) } tlbcam_index = i; +#ifdef CONFIG_PPC64 + get_paca()->tcd.esel_next = i; + get_paca()->tcd.esel_max = mfspr(SPRN_TLB1CFG) & TLBnCFG_N_ENTRY; + get_paca()->tcd.esel_first = i; +#endif + return amount_mapped; } +unsigned long map_mem_in_cams(unsigned long ram, int max_cam_idx) +{ + unsigned long virt = PAGE_OFFSET; + phys_addr_t phys = memstart_addr; + + return map_mem_in_cams_addr(phys, virt, ram, max_cam_idx); +} + #ifdef CONFIG_PPC32 #if defined(CONFIG_LOWMEM_CAM_NUM_BOOL) && (CONFIG_LOWMEM_CAM_NUM >= NUM_TLBCAMS) @@ -222,7 +236,9 @@ void __init adjust_total_lowmem(void) /* adjust lowmem size to __max_low_memory */ ram = min((phys_addr_t)__max_low_memory, (phys_addr_t)total_lowmem); + i = switch_to_as1(); __max_low_memory = map_mem_in_cams(ram, CONFIG_LOWMEM_CAM_NUM); + restore_to_as0(i, 0, 0, 1); pr_info("Memory CAM mapping: "); for (i = 0; i < tlbcam_index - 1; i++) @@ -241,4 +257,62 @@ void setup_initial_memory_limit(phys_addr_t first_memblock_base, /* 64M mapped initially according to head_fsl_booke.S */ memblock_set_current_limit(min_t(u64, limit, 0x04000000)); } + +#ifdef CONFIG_RELOCATABLE +int __initdata is_second_reloc; +notrace void __init relocate_init(u64 dt_ptr, phys_addr_t start) +{ + unsigned long base = KERNELBASE; + + kernstart_addr = start; + if (is_second_reloc) { + virt_phys_offset = PAGE_OFFSET - memstart_addr; + return; + } + + /* + * Relocatable kernel support based on processing of dynamic + * relocation entries. Before we get the real memstart_addr, + * We will compute the virt_phys_offset like this: + * virt_phys_offset = stext.run - kernstart_addr + * + * stext.run = (KERNELBASE & ~0x3ffffff) + + * (kernstart_addr & 0x3ffffff) + * When we relocate, we have : + * + * (kernstart_addr & 0x3ffffff) = (stext.run & 0x3ffffff) + * + * hence: + * virt_phys_offset = (KERNELBASE & ~0x3ffffff) - + * (kernstart_addr & ~0x3ffffff) + * + */ + start &= ~0x3ffffff; + base &= ~0x3ffffff; + virt_phys_offset = base - start; + early_get_first_memblock_info(__va(dt_ptr), NULL); + /* + * We now get the memstart_addr, then we should check if this + * address is the same as what the PAGE_OFFSET map to now. If + * not we have to change the map of PAGE_OFFSET to memstart_addr + * and do a second relocation. + */ + if (start != memstart_addr) { + int n; + long offset = start - memstart_addr; + + is_second_reloc = 1; + n = switch_to_as1(); + /* map a 64M area for the second relocation */ + if (memstart_addr > start) + map_mem_in_cams(0x4000000, CONFIG_LOWMEM_CAM_NUM); + else + map_mem_in_cams_addr(start, PAGE_OFFSET + offset, + 0x4000000, CONFIG_LOWMEM_CAM_NUM); + restore_to_as0(n, offset, __va(dt_ptr), 1); + /* We should never reach here */ + panic("Relocation error"); + } +} +#endif #endif diff --git a/arch/powerpc/mm/gup.c b/arch/powerpc/mm/gup.c index d7efdbf640c..d8746684f60 100644 --- a/arch/powerpc/mm/gup.c +++ b/arch/powerpc/mm/gup.c @@ -34,8 +34,13 @@ static noinline int gup_pte_range(pmd_t pmd, unsigned long addr, ptep = pte_offset_kernel(&pmd, addr); do { - pte_t pte = *ptep; + pte_t pte = ACCESS_ONCE(*ptep); struct page *page; + /* + * Similar to the PMD case, NUMA hinting must take slow path + */ + if (pte_numa(pte)) + return 0; if ((pte_val(pte) & mask) != result) return 0; @@ -63,12 +68,30 @@ static int gup_pmd_range(pud_t pud, unsigned long addr, unsigned long end, pmdp = pmd_offset(&pud, addr); do { - pmd_t pmd = *pmdp; + pmd_t pmd = ACCESS_ONCE(*pmdp); next = pmd_addr_end(addr, end); - if (pmd_none(pmd)) + /* + * If we find a splitting transparent hugepage we + * return zero. That will result in taking the slow + * path which will call wait_split_huge_page() + * if the pmd is still in splitting state + */ + if (pmd_none(pmd) || pmd_trans_splitting(pmd)) return 0; - if (is_hugepd(pmdp)) { + if (pmd_huge(pmd) || pmd_large(pmd)) { + /* + * NUMA hinting faults need to be handled in the GUP + * slowpath for accounting purposes and so that they + * can be serialised against THP migration. + */ + if (pmd_numa(pmd)) + return 0; + + if (!gup_hugepte((pte_t *)pmdp, PMD_SIZE, addr, next, + write, pages, nr)) + return 0; + } else if (is_hugepd(pmdp)) { if (!gup_hugepd((hugepd_t *)pmdp, PMD_SHIFT, addr, next, write, pages, nr)) return 0; @@ -87,12 +110,16 @@ static int gup_pud_range(pgd_t pgd, unsigned long addr, unsigned long end, pudp = pud_offset(&pgd, addr); do { - pud_t pud = *pudp; + pud_t pud = ACCESS_ONCE(*pudp); next = pud_addr_end(addr, end); if (pud_none(pud)) return 0; - if (is_hugepd(pudp)) { + if (pud_huge(pud)) { + if (!gup_hugepte((pte_t *)pudp, PUD_SIZE, addr, next, + write, pages, nr)) + return 0; + } else if (is_hugepd(pudp)) { if (!gup_hugepd((hugepd_t *)pudp, PUD_SHIFT, addr, next, write, pages, nr)) return 0; @@ -103,12 +130,13 @@ static int gup_pud_range(pgd_t pgd, unsigned long addr, unsigned long end, return 1; } -int get_user_pages_fast(unsigned long start, int nr_pages, int write, - struct page **pages) +int __get_user_pages_fast(unsigned long start, int nr_pages, int write, + struct page **pages) { struct mm_struct *mm = current->mm; unsigned long addr, len, end; unsigned long next; + unsigned long flags; pgd_t *pgdp; int nr = 0; @@ -121,7 +149,7 @@ int get_user_pages_fast(unsigned long start, int nr_pages, int write, if (unlikely(!access_ok(write ? VERIFY_WRITE : VERIFY_READ, start, len))) - goto slow_irqon; + return 0; pr_devel(" aligned: %lx .. %lx\n", start, end); @@ -142,36 +170,45 @@ int get_user_pages_fast(unsigned long start, int nr_pages, int write, * So long as we atomically load page table pointers versus teardown, * we can follow the address down to the the page and take a ref on it. */ - local_irq_disable(); + local_irq_save(flags); pgdp = pgd_offset(mm, addr); do { - pgd_t pgd = *pgdp; + pgd_t pgd = ACCESS_ONCE(*pgdp); pr_devel(" %016lx: normal pgd %p\n", addr, (void *)pgd_val(pgd)); next = pgd_addr_end(addr, end); if (pgd_none(pgd)) - goto slow; - if (is_hugepd(pgdp)) { + break; + if (pgd_huge(pgd)) { + if (!gup_hugepte((pte_t *)pgdp, PGDIR_SIZE, addr, next, + write, pages, &nr)) + break; + } else if (is_hugepd(pgdp)) { if (!gup_hugepd((hugepd_t *)pgdp, PGDIR_SHIFT, addr, next, write, pages, &nr)) - goto slow; + break; } else if (!gup_pud_range(pgd, addr, next, write, pages, &nr)) - goto slow; + break; } while (pgdp++, addr = next, addr != end); - local_irq_enable(); + local_irq_restore(flags); - VM_BUG_ON(nr != (end - start) >> PAGE_SHIFT); return nr; +} - { - int ret; +int get_user_pages_fast(unsigned long start, int nr_pages, int write, + struct page **pages) +{ + struct mm_struct *mm = current->mm; + int nr, ret; -slow: - local_irq_enable(); -slow_irqon: + start &= PAGE_MASK; + nr = __get_user_pages_fast(start, nr_pages, write, pages); + ret = nr; + + if (nr < nr_pages) { pr_devel(" slow path ! nr = %d\n", nr); /* Try to get the remaining pages with get_user_pages */ @@ -180,7 +217,7 @@ slow_irqon: down_read(&mm->mmap_sem); ret = get_user_pages(current, mm, start, - (end - start) >> PAGE_SHIFT, write, 0, pages, NULL); + nr_pages - nr, write, 0, pages, NULL); up_read(&mm->mmap_sem); /* Have to be a bit careful with return values */ @@ -190,9 +227,9 @@ slow_irqon: else ret += nr; } - - return ret; } + + return ret; } #endif /* __HAVE_ARCH_PTE_SPECIAL */ diff --git a/arch/powerpc/mm/hash_low_64.S b/arch/powerpc/mm/hash_low_64.S index 56585086413..057cbbb4c57 100644 --- a/arch/powerpc/mm/hash_low_64.S +++ b/arch/powerpc/mm/hash_low_64.S @@ -115,11 +115,13 @@ END_MMU_FTR_SECTION_IFSET(MMU_FTR_1T_SEGMENT) sldi r29,r5,SID_SHIFT - VPN_SHIFT rldicl r28,r3,64 - VPN_SHIFT,64 - (SID_SHIFT - VPN_SHIFT) or r29,r28,r29 - - /* Calculate hash value for primary slot and store it in r28 */ - rldicl r5,r5,0,25 /* vsid & 0x0000007fffffffff */ - rldicl r0,r3,64-12,48 /* (ea >> 12) & 0xffff */ - xor r28,r5,r0 + /* + * Calculate hash value for primary slot and store it in r28 + * r3 = va, r5 = vsid + * r0 = (va >> 12) & ((1ul << (28 - 12)) -1) + */ + rldicl r0,r3,64-12,48 + xor r28,r5,r0 /* hash */ b 4f 3: /* Calc vpn and put it in r29 */ @@ -130,11 +132,12 @@ END_MMU_FTR_SECTION_IFSET(MMU_FTR_1T_SEGMENT) /* * calculate hash value for primary slot and * store it in r28 for 1T segment + * r3 = va, r5 = vsid */ - rldic r28,r5,25,25 /* (vsid << 25) & 0x7fffffffff */ - clrldi r5,r5,40 /* vsid & 0xffffff */ - rldicl r0,r3,64-12,36 /* (ea >> 12) & 0xfffffff */ - xor r28,r28,r5 + sldi r28,r5,25 /* vsid << 25 */ + /* r0 = (va >> 12) & ((1ul << (40 - 12)) -1) */ + rldicl r0,r3,64-12,36 + xor r28,r28,r5 /* vsid ^ ( vsid << 25) */ xor r28,r28,r0 /* hash */ /* Convert linux PTE bits into HW equivalents */ @@ -145,7 +148,10 @@ END_MMU_FTR_SECTION_IFSET(MMU_FTR_1T_SEGMENT) and r0,r0,r4 /* _PAGE_RW & _PAGE_DIRTY ->r0 bit 30*/ andc r0,r30,r0 /* r0 = pte & ~r0 */ rlwimi r3,r0,32-1,31,31 /* Insert result into PP lsb */ - ori r3,r3,HPTE_R_C /* Always add "C" bit for perf. */ + /* + * Always add "C" bit for perf. Memory coherence is always enabled + */ + ori r3,r3,HPTE_R_C | HPTE_R_M /* We eventually do the icache sync here (maybe inline that * code rather than call a C function...) @@ -153,7 +159,7 @@ END_MMU_FTR_SECTION_IFSET(MMU_FTR_1T_SEGMENT) BEGIN_FTR_SECTION mr r4,r30 mr r5,r7 - bl .hash_page_do_lazy_icache + bl hash_page_do_lazy_icache END_FTR_SECTION(CPU_FTR_NOEXECUTE|CPU_FTR_COHERENT_ICACHE, CPU_FTR_NOEXECUTE) /* At this point, r3 contains new PP bits, save them in @@ -193,8 +199,10 @@ htab_insert_pte: mr r4,r29 /* Retrieve vpn */ li r7,0 /* !bolted, !secondary */ li r8,MMU_PAGE_4K /* page size */ - ld r9,STK_PARAM(R9)(r1) /* segment size */ -_GLOBAL(htab_call_hpte_insert1) + li r9,MMU_PAGE_4K /* actual page size */ + ld r10,STK_PARAM(R9)(r1) /* segment size */ +.globl htab_call_hpte_insert1 +htab_call_hpte_insert1: bl . /* Patched by htab_finish_init() */ cmpdi 0,r3,0 bge htab_pte_insert_ok /* Insertion successful */ @@ -216,8 +224,10 @@ _GLOBAL(htab_call_hpte_insert1) mr r4,r29 /* Retrieve vpn */ li r7,HPTE_V_SECONDARY /* !bolted, secondary */ li r8,MMU_PAGE_4K /* page size */ - ld r9,STK_PARAM(R9)(r1) /* segment size */ -_GLOBAL(htab_call_hpte_insert2) + li r9,MMU_PAGE_4K /* actual page size */ + ld r10,STK_PARAM(R9)(r1) /* segment size */ +.globl htab_call_hpte_insert2 +htab_call_hpte_insert2: bl . /* Patched by htab_finish_init() */ cmpdi 0,r3,0 bge+ htab_pte_insert_ok /* Insertion successful */ @@ -234,7 +244,8 @@ _GLOBAL(htab_call_hpte_insert2) 2: and r0,r5,r27 rldicr r3,r0,3,63-3 /* r0 = (hash & mask) << 3 */ /* Call ppc_md.hpte_remove */ -_GLOBAL(htab_call_hpte_remove) +.globl htab_call_hpte_remove +htab_call_hpte_remove: bl . /* Patched by htab_finish_init() */ /* Try all again */ @@ -284,10 +295,12 @@ htab_modify_pte: /* Call ppc_md.hpte_updatepp */ mr r5,r29 /* vpn */ - li r6,MMU_PAGE_4K /* page size */ - ld r7,STK_PARAM(R9)(r1) /* segment size */ - ld r8,STK_PARAM(R8)(r1) /* get "local" param */ -_GLOBAL(htab_call_hpte_updatepp) + li r6,MMU_PAGE_4K /* base page size */ + li r7,MMU_PAGE_4K /* actual page size */ + ld r8,STK_PARAM(R9)(r1) /* segment size */ + ld r9,STK_PARAM(R8)(r1) /* get "local" param */ +.globl htab_call_hpte_updatepp +htab_call_hpte_updatepp: bl . /* Patched by htab_finish_init() */ /* if we failed because typically the HPTE wasn't really here @@ -407,11 +420,13 @@ END_MMU_FTR_SECTION_IFSET(MMU_FTR_1T_SEGMENT) */ rldicl r28,r3,64 - VPN_SHIFT,64 - (SID_SHIFT - VPN_SHIFT) or r29,r28,r29 - - /* Calculate hash value for primary slot and store it in r28 */ - rldicl r5,r5,0,25 /* vsid & 0x0000007fffffffff */ - rldicl r0,r3,64-12,48 /* (ea >> 12) & 0xffff */ - xor r28,r5,r0 + /* + * Calculate hash value for primary slot and store it in r28 + * r3 = va, r5 = vsid + * r0 = (va >> 12) & ((1ul << (28 - 12)) -1) + */ + rldicl r0,r3,64-12,48 + xor r28,r5,r0 /* hash */ b 4f 3: /* Calc vpn and put it in r29 */ @@ -426,11 +441,12 @@ END_MMU_FTR_SECTION_IFSET(MMU_FTR_1T_SEGMENT) /* * Calculate hash value for primary slot and * store it in r28 for 1T segment + * r3 = va, r5 = vsid */ - rldic r28,r5,25,25 /* (vsid << 25) & 0x7fffffffff */ - clrldi r5,r5,40 /* vsid & 0xffffff */ - rldicl r0,r3,64-12,36 /* (ea >> 12) & 0xfffffff */ - xor r28,r28,r5 + sldi r28,r5,25 /* vsid << 25 */ + /* r0 = (va >> 12) & ((1ul << (40 - 12)) -1) */ + rldicl r0,r3,64-12,36 + xor r28,r28,r5 /* vsid ^ ( vsid << 25) */ xor r28,r28,r0 /* hash */ /* Convert linux PTE bits into HW equivalents */ @@ -448,7 +464,10 @@ END_MMU_FTR_SECTION_IFSET(MMU_FTR_1T_SEGMENT) and r0,r0,r4 /* _PAGE_RW & _PAGE_DIRTY ->r0 bit 30*/ andc r0,r3,r0 /* r0 = pte & ~r0 */ rlwimi r3,r0,32-1,31,31 /* Insert result into PP lsb */ - ori r3,r3,HPTE_R_C /* Always add "C" bit for perf. */ + /* + * Always add "C" bit for perf. Memory coherence is always enabled + */ + ori r3,r3,HPTE_R_C | HPTE_R_M /* We eventually do the icache sync here (maybe inline that * code rather than call a C function...) @@ -456,7 +475,7 @@ END_MMU_FTR_SECTION_IFSET(MMU_FTR_1T_SEGMENT) BEGIN_FTR_SECTION mr r4,r30 mr r5,r7 - bl .hash_page_do_lazy_icache + bl hash_page_do_lazy_icache END_FTR_SECTION(CPU_FTR_NOEXECUTE|CPU_FTR_COHERENT_ICACHE, CPU_FTR_NOEXECUTE) /* At this point, r3 contains new PP bits, save them in @@ -484,7 +503,7 @@ END_FTR_SECTION(CPU_FTR_NOEXECUTE|CPU_FTR_COHERENT_ICACHE, CPU_FTR_NOEXECUTE) beq htab_inval_old_hpte ld r6,STK_PARAM(R6)(r1) - ori r26,r6,0x8000 /* Load the hidx mask */ + ori r26,r6,PTE_PAGE_HIDX_OFFSET /* Load the hidx mask. */ ld r26,0(r26) addi r5,r25,36 /* Check actual HPTE_SUB bit, this */ rldcr. r0,r31,r5,0 /* must match pgtable.h definition */ @@ -509,8 +528,10 @@ htab_special_pfn: mr r4,r29 /* Retrieve vpn */ li r7,0 /* !bolted, !secondary */ li r8,MMU_PAGE_4K /* page size */ - ld r9,STK_PARAM(R9)(r1) /* segment size */ -_GLOBAL(htab_call_hpte_insert1) + li r9,MMU_PAGE_4K /* actual page size */ + ld r10,STK_PARAM(R9)(r1) /* segment size */ +.globl htab_call_hpte_insert1 +htab_call_hpte_insert1: bl . /* patched by htab_finish_init() */ cmpdi 0,r3,0 bge htab_pte_insert_ok /* Insertion successful */ @@ -536,8 +557,10 @@ _GLOBAL(htab_call_hpte_insert1) mr r4,r29 /* Retrieve vpn */ li r7,HPTE_V_SECONDARY /* !bolted, secondary */ li r8,MMU_PAGE_4K /* page size */ - ld r9,STK_PARAM(R9)(r1) /* segment size */ -_GLOBAL(htab_call_hpte_insert2) + li r9,MMU_PAGE_4K /* actual page size */ + ld r10,STK_PARAM(R9)(r1) /* segment size */ +.globl htab_call_hpte_insert2 +htab_call_hpte_insert2: bl . /* patched by htab_finish_init() */ cmpdi 0,r3,0 bge+ htab_pte_insert_ok /* Insertion successful */ @@ -554,7 +577,8 @@ _GLOBAL(htab_call_hpte_insert2) 2: and r0,r5,r27 rldicr r3,r0,3,63-3 /* r0 = (hash & mask) << 3 */ /* Call ppc_md.hpte_remove */ -_GLOBAL(htab_call_hpte_remove) +.globl htab_call_hpte_remove +htab_call_hpte_remove: bl . /* patched by htab_finish_init() */ /* Try all again */ @@ -571,7 +595,7 @@ htab_inval_old_hpte: li r6,MMU_PAGE_64K /* psize */ ld r7,STK_PARAM(R9)(r1) /* ssize */ ld r8,STK_PARAM(R8)(r1) /* local */ - bl .flush_hash_page + bl flush_hash_page /* Clear out _PAGE_HPTE_SUB bits in the new linux PTE */ lis r0,_PAGE_HPTE_SUB@h ori r0,r0,_PAGE_HPTE_SUB@l @@ -601,7 +625,7 @@ htab_pte_insert_ok: sld r4,r4,r5 andc r26,r26,r4 or r26,r26,r3 - ori r5,r6,0x8000 + ori r5,r6,PTE_PAGE_HIDX_OFFSET std r26,0(r5) lwsync std r30,0(r6) @@ -639,10 +663,12 @@ htab_modify_pte: /* Call ppc_md.hpte_updatepp */ mr r5,r29 /* vpn */ - li r6,MMU_PAGE_4K /* page size */ - ld r7,STK_PARAM(R9)(r1) /* segment size */ - ld r8,STK_PARAM(R8)(r1) /* get "local" param */ -_GLOBAL(htab_call_hpte_updatepp) + li r6,MMU_PAGE_4K /* base page size */ + li r7,MMU_PAGE_4K /* actual page size */ + ld r8,STK_PARAM(R9)(r1) /* segment size */ + ld r9,STK_PARAM(R8)(r1) /* get "local" param */ +.globl htab_call_hpte_updatepp +htab_call_hpte_updatepp: bl . /* patched by htab_finish_init() */ /* if we failed because typically the HPTE wasn't really here @@ -752,25 +778,27 @@ END_MMU_FTR_SECTION_IFSET(MMU_FTR_1T_SEGMENT) rldicl r28,r3,64 - VPN_SHIFT,64 - (SID_SHIFT - VPN_SHIFT) or r29,r28,r29 - /* Calculate hash value for primary slot and store it in r28 */ - rldicl r5,r5,0,25 /* vsid & 0x0000007fffffffff */ - rldicl r0,r3,64-16,52 /* (ea >> 16) & 0xfff */ - xor r28,r5,r0 + /* Calculate hash value for primary slot and store it in r28 + * r3 = va, r5 = vsid + * r0 = (va >> 16) & ((1ul << (28 - 16)) -1) + */ + rldicl r0,r3,64-16,52 + xor r28,r5,r0 /* hash */ b 4f 3: /* Calc vpn and put it in r29 */ sldi r29,r5,SID_SHIFT_1T - VPN_SHIFT rldicl r28,r3,64 - VPN_SHIFT,64 - (SID_SHIFT_1T - VPN_SHIFT) or r29,r28,r29 - /* * calculate hash value for primary slot and * store it in r28 for 1T segment + * r3 = va, r5 = vsid */ - rldic r28,r5,25,25 /* (vsid << 25) & 0x7fffffffff */ - clrldi r5,r5,40 /* vsid & 0xffffff */ - rldicl r0,r3,64-16,40 /* (ea >> 16) & 0xffffff */ - xor r28,r28,r5 + sldi r28,r5,25 /* vsid << 25 */ + /* r0 = (va >> 16) & ((1ul << (40 - 16)) -1) */ + rldicl r0,r3,64-16,40 + xor r28,r28,r5 /* vsid ^ ( vsid << 25) */ xor r28,r28,r0 /* hash */ /* Convert linux PTE bits into HW equivalents */ @@ -781,7 +809,10 @@ END_MMU_FTR_SECTION_IFSET(MMU_FTR_1T_SEGMENT) and r0,r0,r4 /* _PAGE_RW & _PAGE_DIRTY ->r0 bit 30*/ andc r0,r30,r0 /* r0 = pte & ~r0 */ rlwimi r3,r0,32-1,31,31 /* Insert result into PP lsb */ - ori r3,r3,HPTE_R_C /* Always add "C" bit for perf. */ + /* + * Always add "C" bit for perf. Memory coherence is always enabled + */ + ori r3,r3,HPTE_R_C | HPTE_R_M /* We eventually do the icache sync here (maybe inline that * code rather than call a C function...) @@ -789,7 +820,7 @@ END_MMU_FTR_SECTION_IFSET(MMU_FTR_1T_SEGMENT) BEGIN_FTR_SECTION mr r4,r30 mr r5,r7 - bl .hash_page_do_lazy_icache + bl hash_page_do_lazy_icache END_FTR_SECTION(CPU_FTR_NOEXECUTE|CPU_FTR_COHERENT_ICACHE, CPU_FTR_NOEXECUTE) /* At this point, r3 contains new PP bits, save them in @@ -832,8 +863,10 @@ ht64_insert_pte: mr r4,r29 /* Retrieve vpn */ li r7,0 /* !bolted, !secondary */ li r8,MMU_PAGE_64K - ld r9,STK_PARAM(R9)(r1) /* segment size */ -_GLOBAL(ht64_call_hpte_insert1) + li r9,MMU_PAGE_64K /* actual page size */ + ld r10,STK_PARAM(R9)(r1) /* segment size */ +.globl ht64_call_hpte_insert1 +ht64_call_hpte_insert1: bl . /* patched by htab_finish_init() */ cmpdi 0,r3,0 bge ht64_pte_insert_ok /* Insertion successful */ @@ -855,8 +888,10 @@ _GLOBAL(ht64_call_hpte_insert1) mr r4,r29 /* Retrieve vpn */ li r7,HPTE_V_SECONDARY /* !bolted, secondary */ li r8,MMU_PAGE_64K - ld r9,STK_PARAM(R9)(r1) /* segment size */ -_GLOBAL(ht64_call_hpte_insert2) + li r9,MMU_PAGE_64K /* actual page size */ + ld r10,STK_PARAM(R9)(r1) /* segment size */ +.globl ht64_call_hpte_insert2 +ht64_call_hpte_insert2: bl . /* patched by htab_finish_init() */ cmpdi 0,r3,0 bge+ ht64_pte_insert_ok /* Insertion successful */ @@ -873,7 +908,8 @@ _GLOBAL(ht64_call_hpte_insert2) 2: and r0,r5,r27 rldicr r3,r0,3,63-3 /* r0 = (hash & mask) << 3 */ /* Call ppc_md.hpte_remove */ -_GLOBAL(ht64_call_hpte_remove) +.globl ht64_call_hpte_remove +ht64_call_hpte_remove: bl . /* patched by htab_finish_init() */ /* Try all again */ @@ -923,10 +959,12 @@ ht64_modify_pte: /* Call ppc_md.hpte_updatepp */ mr r5,r29 /* vpn */ - li r6,MMU_PAGE_64K - ld r7,STK_PARAM(R9)(r1) /* segment size */ - ld r8,STK_PARAM(R8)(r1) /* get "local" param */ -_GLOBAL(ht64_call_hpte_updatepp) + li r6,MMU_PAGE_64K /* base page size */ + li r7,MMU_PAGE_64K /* actual page size */ + ld r8,STK_PARAM(R9)(r1) /* segment size */ + ld r9,STK_PARAM(R8)(r1) /* get "local" param */ +.globl ht64_call_hpte_updatepp +ht64_call_hpte_updatepp: bl . /* patched by htab_finish_init() */ /* if we failed because typically the HPTE wasn't really here diff --git a/arch/powerpc/mm/hash_native_64.c b/arch/powerpc/mm/hash_native_64.c index ffc1e00f7a2..cf1d325eae8 100644 --- a/arch/powerpc/mm/hash_native_64.c +++ b/arch/powerpc/mm/hash_native_64.c @@ -35,14 +35,19 @@ #define DBG_LOW(fmt...) #endif +#ifdef __BIG_ENDIAN__ #define HPTE_LOCK_BIT 3 +#else +#define HPTE_LOCK_BIT (56+3) +#endif DEFINE_RAW_SPINLOCK(native_tlbie_lock); -static inline void __tlbie(unsigned long vpn, int psize, int ssize) +static inline void __tlbie(unsigned long vpn, int psize, int apsize, int ssize) { unsigned long va; unsigned int penc; + unsigned long sllp; /* * We need 14 to 65 bits of va for a tlibe of 4K page @@ -61,17 +66,30 @@ static inline void __tlbie(unsigned long vpn, int psize, int ssize) switch (psize) { case MMU_PAGE_4K: + /* clear out bits after (52) [0....52.....63] */ + va &= ~((1ul << (64 - 52)) - 1); va |= ssize << 8; + sllp = ((mmu_psize_defs[apsize].sllp & SLB_VSID_L) >> 6) | + ((mmu_psize_defs[apsize].sllp & SLB_VSID_LP) >> 4); + va |= sllp << 5; asm volatile(ASM_FTR_IFCLR("tlbie %0,0", PPC_TLBIE(%1,%0), %2) : : "r" (va), "r"(0), "i" (CPU_FTR_ARCH_206) : "memory"); break; default: /* We need 14 to 14 + i bits of va */ - penc = mmu_psize_defs[psize].penc; - va &= ~((1ul << mmu_psize_defs[psize].shift) - 1); + penc = mmu_psize_defs[psize].penc[apsize]; + va &= ~((1ul << mmu_psize_defs[apsize].shift) - 1); va |= penc << 12; va |= ssize << 8; + /* + * AVAL bits: + * We don't need all the bits, but rest of the bits + * must be ignored by the processor. + * vpn cover upto 65 bits of va. (0...65) and we need + * 58..64 bits of va. + */ + va |= (vpn & 0xfe); /* AVAL */ va |= 1; /* L */ asm volatile(ASM_FTR_IFCLR("tlbie %0,1", PPC_TLBIE(%1,%0), %2) : : "r" (va), "r"(0), "i" (CPU_FTR_ARCH_206) @@ -80,10 +98,11 @@ static inline void __tlbie(unsigned long vpn, int psize, int ssize) } } -static inline void __tlbiel(unsigned long vpn, int psize, int ssize) +static inline void __tlbiel(unsigned long vpn, int psize, int apsize, int ssize) { unsigned long va; unsigned int penc; + unsigned long sllp; /* VPN_SHIFT can be atmost 12 */ va = vpn << VPN_SHIFT; @@ -96,16 +115,29 @@ static inline void __tlbiel(unsigned long vpn, int psize, int ssize) switch (psize) { case MMU_PAGE_4K: + /* clear out bits after(52) [0....52.....63] */ + va &= ~((1ul << (64 - 52)) - 1); va |= ssize << 8; + sllp = ((mmu_psize_defs[apsize].sllp & SLB_VSID_L) >> 6) | + ((mmu_psize_defs[apsize].sllp & SLB_VSID_LP) >> 4); + va |= sllp << 5; asm volatile(".long 0x7c000224 | (%0 << 11) | (0 << 21)" : : "r"(va) : "memory"); break; default: /* We need 14 to 14 + i bits of va */ - penc = mmu_psize_defs[psize].penc; - va &= ~((1ul << mmu_psize_defs[psize].shift) - 1); + penc = mmu_psize_defs[psize].penc[apsize]; + va &= ~((1ul << mmu_psize_defs[apsize].shift) - 1); va |= penc << 12; va |= ssize << 8; + /* + * AVAL bits: + * We don't need all the bits, but rest of the bits + * must be ignored by the processor. + * vpn cover upto 65 bits of va. (0...65) and we need + * 58..64 bits of va. + */ + va |= (vpn & 0xfe); va |= 1; /* L */ asm volatile(".long 0x7c000224 | (%0 << 11) | (1 << 21)" : : "r"(va) : "memory"); @@ -114,7 +146,8 @@ static inline void __tlbiel(unsigned long vpn, int psize, int ssize) } -static inline void tlbie(unsigned long vpn, int psize, int ssize, int local) +static inline void tlbie(unsigned long vpn, int psize, int apsize, + int ssize, int local) { unsigned int use_local = local && mmu_has_feature(MMU_FTR_TLBIEL); int lock_tlbie = !mmu_has_feature(MMU_FTR_LOCKLESS_TLBIE); @@ -125,10 +158,10 @@ static inline void tlbie(unsigned long vpn, int psize, int ssize, int local) raw_spin_lock(&native_tlbie_lock); asm volatile("ptesync": : :"memory"); if (use_local) { - __tlbiel(vpn, psize, ssize); + __tlbiel(vpn, psize, apsize, ssize); asm volatile("ptesync": : :"memory"); } else { - __tlbie(vpn, psize, ssize); + __tlbie(vpn, psize, apsize, ssize); asm volatile("eieio; tlbsync; ptesync": : :"memory"); } if (lock_tlbie && !use_local) @@ -137,7 +170,7 @@ static inline void tlbie(unsigned long vpn, int psize, int ssize, int local) static inline void native_lock_hpte(struct hash_pte *hptep) { - unsigned long *word = &hptep->v; + unsigned long *word = (unsigned long *)&hptep->v; while (1) { if (!test_and_set_bit_lock(HPTE_LOCK_BIT, word)) @@ -149,14 +182,14 @@ static inline void native_lock_hpte(struct hash_pte *hptep) static inline void native_unlock_hpte(struct hash_pte *hptep) { - unsigned long *word = &hptep->v; + unsigned long *word = (unsigned long *)&hptep->v; clear_bit_unlock(HPTE_LOCK_BIT, word); } static long native_hpte_insert(unsigned long hpte_group, unsigned long vpn, unsigned long pa, unsigned long rflags, - unsigned long vflags, int psize, int ssize) + unsigned long vflags, int psize, int apsize, int ssize) { struct hash_pte *hptep = htab_address + hpte_group; unsigned long hpte_v, hpte_r; @@ -169,10 +202,10 @@ static long native_hpte_insert(unsigned long hpte_group, unsigned long vpn, } for (i = 0; i < HPTES_PER_GROUP; i++) { - if (! (hptep->v & HPTE_V_VALID)) { + if (! (be64_to_cpu(hptep->v) & HPTE_V_VALID)) { /* retry with lock held */ native_lock_hpte(hptep); - if (! (hptep->v & HPTE_V_VALID)) + if (! (be64_to_cpu(hptep->v) & HPTE_V_VALID)) break; native_unlock_hpte(hptep); } @@ -183,22 +216,22 @@ static long native_hpte_insert(unsigned long hpte_group, unsigned long vpn, if (i == HPTES_PER_GROUP) return -1; - hpte_v = hpte_encode_v(vpn, psize, ssize) | vflags | HPTE_V_VALID; - hpte_r = hpte_encode_r(pa, psize) | rflags; + hpte_v = hpte_encode_v(vpn, psize, apsize, ssize) | vflags | HPTE_V_VALID; + hpte_r = hpte_encode_r(pa, psize, apsize) | rflags; if (!(vflags & HPTE_V_BOLTED)) { DBG_LOW(" i=%x hpte_v=%016lx, hpte_r=%016lx\n", i, hpte_v, hpte_r); } - hptep->r = hpte_r; + hptep->r = cpu_to_be64(hpte_r); /* Guarantee the second dword is visible before the valid bit */ eieio(); /* * Now set the first dword including the valid bit * NOTE: this also unlocks the hpte */ - hptep->v = hpte_v; + hptep->v = cpu_to_be64(hpte_v); __asm__ __volatile__ ("ptesync" : : : "memory"); @@ -219,12 +252,12 @@ static long native_hpte_remove(unsigned long hpte_group) for (i = 0; i < HPTES_PER_GROUP; i++) { hptep = htab_address + hpte_group + slot_offset; - hpte_v = hptep->v; + hpte_v = be64_to_cpu(hptep->v); if ((hpte_v & HPTE_V_VALID) && !(hpte_v & HPTE_V_BOLTED)) { /* retry with lock held */ native_lock_hpte(hptep); - hpte_v = hptep->v; + hpte_v = be64_to_cpu(hptep->v); if ((hpte_v & HPTE_V_VALID) && !(hpte_v & HPTE_V_BOLTED)) break; @@ -245,36 +278,41 @@ static long native_hpte_remove(unsigned long hpte_group) } static long native_hpte_updatepp(unsigned long slot, unsigned long newpp, - unsigned long vpn, int psize, int ssize, - int local) + unsigned long vpn, int bpsize, + int apsize, int ssize, int local) { struct hash_pte *hptep = htab_address + slot; unsigned long hpte_v, want_v; int ret = 0; - want_v = hpte_encode_v(vpn, psize, ssize); + want_v = hpte_encode_avpn(vpn, bpsize, ssize); DBG_LOW(" update(vpn=%016lx, avpnv=%016lx, group=%lx, newpp=%lx)", vpn, want_v & HPTE_V_AVPN, slot, newpp); native_lock_hpte(hptep); - hpte_v = hptep->v; - - /* Even if we miss, we need to invalidate the TLB */ + hpte_v = be64_to_cpu(hptep->v); + /* + * We need to invalidate the TLB always because hpte_remove doesn't do + * a tlb invalidate. If a hash bucket gets full, we "evict" a more/less + * random entry from it. When we do that we don't invalidate the TLB + * (hpte_remove) because we assume the old translation is still + * technically "valid". + */ if (!HPTE_V_COMPARE(hpte_v, want_v) || !(hpte_v & HPTE_V_VALID)) { DBG_LOW(" -> miss\n"); ret = -1; } else { DBG_LOW(" -> hit\n"); /* Update the HPTE */ - hptep->r = (hptep->r & ~(HPTE_R_PP | HPTE_R_N)) | - (newpp & (HPTE_R_PP | HPTE_R_N | HPTE_R_C)); + hptep->r = cpu_to_be64((be64_to_cpu(hptep->r) & ~(HPTE_R_PP | HPTE_R_N)) | + (newpp & (HPTE_R_PP | HPTE_R_N | HPTE_R_C))); } native_unlock_hpte(hptep); /* Ensure it is out of the tlb too. */ - tlbie(vpn, psize, ssize, local); + tlbie(vpn, bpsize, apsize, ssize, local); return ret; } @@ -288,13 +326,13 @@ static long native_hpte_find(unsigned long vpn, int psize, int ssize) unsigned long want_v, hpte_v; hash = hpt_hash(vpn, mmu_psize_defs[psize].shift, ssize); - want_v = hpte_encode_v(vpn, psize, ssize); + want_v = hpte_encode_avpn(vpn, psize, ssize); /* Bolted mappings are only ever in the primary group */ slot = (hash & htab_hash_mask) * HPTES_PER_GROUP; for (i = 0; i < HPTES_PER_GROUP; i++) { hptep = htab_address + slot; - hpte_v = hptep->v; + hpte_v = be64_to_cpu(hptep->v); if (HPTE_V_COMPARE(hpte_v, want_v) && (hpte_v & HPTE_V_VALID)) /* HPTE matches */ @@ -329,15 +367,18 @@ static void native_hpte_updateboltedpp(unsigned long newpp, unsigned long ea, hptep = htab_address + slot; /* Update the HPTE */ - hptep->r = (hptep->r & ~(HPTE_R_PP | HPTE_R_N)) | - (newpp & (HPTE_R_PP | HPTE_R_N)); - - /* Ensure it is out of the tlb too. */ - tlbie(vpn, psize, ssize, 0); + hptep->r = cpu_to_be64((be64_to_cpu(hptep->r) & + ~(HPTE_R_PP | HPTE_R_N)) | + (newpp & (HPTE_R_PP | HPTE_R_N))); + /* + * Ensure it is out of the tlb too. Bolted entries base and + * actual page size will be same. + */ + tlbie(vpn, psize, psize, ssize, 0); } static void native_hpte_invalidate(unsigned long slot, unsigned long vpn, - int psize, int ssize, int local) + int bpsize, int apsize, int ssize, int local) { struct hash_pte *hptep = htab_address + slot; unsigned long hpte_v; @@ -348,11 +389,17 @@ static void native_hpte_invalidate(unsigned long slot, unsigned long vpn, DBG_LOW(" invalidate(vpn=%016lx, hash: %lx)\n", vpn, slot); - want_v = hpte_encode_v(vpn, psize, ssize); + want_v = hpte_encode_avpn(vpn, bpsize, ssize); native_lock_hpte(hptep); - hpte_v = hptep->v; + hpte_v = be64_to_cpu(hptep->v); - /* Even if we miss, we need to invalidate the TLB */ + /* + * We need to invalidate the TLB always because hpte_remove doesn't do + * a tlb invalidate. If a hash bucket gets full, we "evict" a more/less + * random entry from it. When we do that we don't invalidate the TLB + * (hpte_remove) because we assume the old translation is still + * technically "valid". + */ if (!HPTE_V_COMPARE(hpte_v, want_v) || !(hpte_v & HPTE_V_VALID)) native_unlock_hpte(hptep); else @@ -360,47 +407,139 @@ static void native_hpte_invalidate(unsigned long slot, unsigned long vpn, hptep->v = 0; /* Invalidate the TLB */ - tlbie(vpn, psize, ssize, local); + tlbie(vpn, bpsize, apsize, ssize, local); + + local_irq_restore(flags); +} + +static void native_hugepage_invalidate(struct mm_struct *mm, + unsigned char *hpte_slot_array, + unsigned long addr, int psize) +{ + int ssize = 0, i; + int lock_tlbie; + struct hash_pte *hptep; + int actual_psize = MMU_PAGE_16M; + unsigned int max_hpte_count, valid; + unsigned long flags, s_addr = addr; + unsigned long hpte_v, want_v, shift; + unsigned long hidx, vpn = 0, vsid, hash, slot; + + shift = mmu_psize_defs[psize].shift; + max_hpte_count = 1U << (PMD_SHIFT - shift); + + local_irq_save(flags); + for (i = 0; i < max_hpte_count; i++) { + valid = hpte_valid(hpte_slot_array, i); + if (!valid) + continue; + hidx = hpte_hash_index(hpte_slot_array, i); + + /* get the vpn */ + addr = s_addr + (i * (1ul << shift)); + if (!is_kernel_addr(addr)) { + ssize = user_segment_size(addr); + vsid = get_vsid(mm->context.id, addr, ssize); + WARN_ON(vsid == 0); + } else { + vsid = get_kernel_vsid(addr, mmu_kernel_ssize); + ssize = mmu_kernel_ssize; + } + + vpn = hpt_vpn(addr, vsid, ssize); + hash = hpt_hash(vpn, shift, ssize); + if (hidx & _PTEIDX_SECONDARY) + hash = ~hash; + + slot = (hash & htab_hash_mask) * HPTES_PER_GROUP; + slot += hidx & _PTEIDX_GROUP_IX; + + hptep = htab_address + slot; + want_v = hpte_encode_avpn(vpn, psize, ssize); + native_lock_hpte(hptep); + hpte_v = be64_to_cpu(hptep->v); + + /* Even if we miss, we need to invalidate the TLB */ + if (!HPTE_V_COMPARE(hpte_v, want_v) || !(hpte_v & HPTE_V_VALID)) + native_unlock_hpte(hptep); + else + /* Invalidate the hpte. NOTE: this also unlocks it */ + hptep->v = 0; + } + /* + * Since this is a hugepage, we just need a single tlbie. + * use the last vpn. + */ + lock_tlbie = !mmu_has_feature(MMU_FTR_LOCKLESS_TLBIE); + if (lock_tlbie) + raw_spin_lock(&native_tlbie_lock); + + asm volatile("ptesync":::"memory"); + __tlbie(vpn, psize, actual_psize, ssize); + asm volatile("eieio; tlbsync; ptesync":::"memory"); + + if (lock_tlbie) + raw_spin_unlock(&native_tlbie_lock); local_irq_restore(flags); } -#define LP_SHIFT 12 -#define LP_BITS 8 -#define LP_MASK(i) ((0xFF >> (i)) << LP_SHIFT) +static inline int __hpte_actual_psize(unsigned int lp, int psize) +{ + int i, shift; + unsigned int mask; + + /* start from 1 ignoring MMU_PAGE_4K */ + for (i = 1; i < MMU_PAGE_COUNT; i++) { + + /* invalid penc */ + if (mmu_psize_defs[psize].penc[i] == -1) + continue; + /* + * encoding bits per actual page size + * PTE LP actual page size + * rrrr rrrz >=8KB + * rrrr rrzz >=16KB + * rrrr rzzz >=32KB + * rrrr zzzz >=64KB + * ....... + */ + shift = mmu_psize_defs[i].shift - LP_SHIFT; + if (shift > LP_BITS) + shift = LP_BITS; + mask = (1 << shift) - 1; + if ((lp & mask) == mmu_psize_defs[psize].penc[i]) + return i; + } + return -1; +} static void hpte_decode(struct hash_pte *hpte, unsigned long slot, - int *psize, int *ssize, unsigned long *vpn) + int *psize, int *apsize, int *ssize, unsigned long *vpn) { unsigned long avpn, pteg, vpi; - unsigned long hpte_r = hpte->r; - unsigned long hpte_v = hpte->v; + unsigned long hpte_v = be64_to_cpu(hpte->v); + unsigned long hpte_r = be64_to_cpu(hpte->r); unsigned long vsid, seg_off; - int i, size, shift, penc; + int size, a_size, shift; + /* Look at the 8 bit LP value */ + unsigned int lp = (hpte_r >> LP_SHIFT) & ((1 << LP_BITS) - 1); - if (!(hpte_v & HPTE_V_LARGE)) - size = MMU_PAGE_4K; - else { - for (i = 0; i < LP_BITS; i++) { - if ((hpte_r & LP_MASK(i+1)) == LP_MASK(i+1)) - break; - } - penc = LP_MASK(i+1) >> LP_SHIFT; + if (!(hpte_v & HPTE_V_LARGE)) { + size = MMU_PAGE_4K; + a_size = MMU_PAGE_4K; + } else { for (size = 0; size < MMU_PAGE_COUNT; size++) { - /* 4K pages are not represented by LP */ - if (size == MMU_PAGE_4K) - continue; - /* valid entries have a shift value */ if (!mmu_psize_defs[size].shift) continue; - if (penc == mmu_psize_defs[size].penc) + a_size = __hpte_actual_psize(lp, size); + if (a_size != -1) break; } } - /* This works for all page sizes, and for 256M and 1T segments */ *ssize = hpte_v >> HPTE_V_SSIZE_SHIFT; shift = mmu_psize_defs[size].shift; @@ -421,6 +560,7 @@ static void hpte_decode(struct hash_pte *hpte, unsigned long slot, seg_off |= vpi << shift; } *vpn = vsid << (SID_SHIFT - VPN_SHIFT) | seg_off >> VPN_SHIFT; + break; case MMU_SEGSIZE_1T: /* We only have 40 - 23 bits of seg_off in avpn */ seg_off = (avpn & 0x1ffff) << 23; @@ -430,10 +570,12 @@ static void hpte_decode(struct hash_pte *hpte, unsigned long slot, seg_off |= vpi << shift; } *vpn = vsid << (SID_SHIFT_1T - VPN_SHIFT) | seg_off >> VPN_SHIFT; + break; default: *vpn = size = 0; } - *psize = size; + *psize = size; + *apsize = a_size; } /* @@ -451,7 +593,7 @@ static void native_hpte_clear(void) struct hash_pte *hptep = htab_address; unsigned long hpte_v; unsigned long pteg_count; - int psize, ssize; + int psize, apsize, ssize; pteg_count = htab_hash_mask + 1; @@ -470,16 +612,16 @@ static void native_hpte_clear(void) * running, right? and for crash dump, we probably * don't want to wait for a maybe bad cpu. */ - hpte_v = hptep->v; + hpte_v = be64_to_cpu(hptep->v); /* * Call __tlbie() here rather than tlbie() since we * already hold the native_tlbie_lock. */ if (hpte_v & HPTE_V_VALID) { - hpte_decode(hptep, slot, &psize, &ssize, &vpn); + hpte_decode(hptep, slot, &psize, &apsize, &ssize, &vpn); hptep->v = 0; - __tlbie(vpn, psize, ssize); + __tlbie(vpn, psize, apsize, ssize); } } @@ -520,9 +662,9 @@ static void native_flush_hash_range(unsigned long number, int local) slot = (hash & htab_hash_mask) * HPTES_PER_GROUP; slot += hidx & _PTEIDX_GROUP_IX; hptep = htab_address + slot; - want_v = hpte_encode_v(vpn, psize, ssize); + want_v = hpte_encode_avpn(vpn, psize, ssize); native_lock_hpte(hptep); - hpte_v = hptep->v; + hpte_v = be64_to_cpu(hptep->v); if (!HPTE_V_COMPARE(hpte_v, want_v) || !(hpte_v & HPTE_V_VALID)) native_unlock_hpte(hptep); @@ -540,7 +682,7 @@ static void native_flush_hash_range(unsigned long number, int local) pte_iterate_hashed_subpages(pte, psize, vpn, index, shift) { - __tlbiel(vpn, psize, ssize); + __tlbiel(vpn, psize, psize, ssize); } pte_iterate_hashed_end(); } asm volatile("ptesync":::"memory"); @@ -557,7 +699,7 @@ static void native_flush_hash_range(unsigned long number, int local) pte_iterate_hashed_subpages(pte, psize, vpn, index, shift) { - __tlbie(vpn, psize, ssize); + __tlbie(vpn, psize, psize, ssize); } pte_iterate_hashed_end(); } asm volatile("eieio; tlbsync; ptesync":::"memory"); @@ -578,4 +720,5 @@ void __init hpte_init_native(void) ppc_md.hpte_remove = native_hpte_remove; ppc_md.hpte_clear_all = native_hpte_clear; ppc_md.flush_hash_range = native_flush_hash_range; + ppc_md.hugepage_invalidate = native_hugepage_invalidate; } diff --git a/arch/powerpc/mm/hash_utils_64.c b/arch/powerpc/mm/hash_utils_64.c index 3a292be2e07..88fdd9d2507 100644 --- a/arch/powerpc/mm/hash_utils_64.c +++ b/arch/powerpc/mm/hash_utils_64.c @@ -33,6 +33,7 @@ #include <linux/init.h> #include <linux/signal.h> #include <linux/memblock.h> +#include <linux/context_tracking.h> #include <asm/processor.h> #include <asm/pgtable.h> @@ -55,6 +56,7 @@ #include <asm/code-patching.h> #include <asm/fadump.h> #include <asm/firmware.h> +#include <asm/tm.h> #ifdef DEBUG #define DBG(fmt...) udbg_printf(fmt) @@ -125,7 +127,7 @@ static struct mmu_psize_def mmu_psize_defaults_old[] = { [MMU_PAGE_4K] = { .shift = 12, .sllp = 0, - .penc = 0, + .penc = {[MMU_PAGE_4K] = 0, [1 ... MMU_PAGE_COUNT - 1] = -1}, .avpnm = 0, .tlbiel = 0, }, @@ -139,14 +141,15 @@ static struct mmu_psize_def mmu_psize_defaults_gp[] = { [MMU_PAGE_4K] = { .shift = 12, .sllp = 0, - .penc = 0, + .penc = {[MMU_PAGE_4K] = 0, [1 ... MMU_PAGE_COUNT - 1] = -1}, .avpnm = 0, .tlbiel = 1, }, [MMU_PAGE_16M] = { .shift = 24, .sllp = SLB_VSID_L, - .penc = 0, + .penc = {[0 ... MMU_PAGE_16M - 1] = -1, [MMU_PAGE_16M] = 0, + [MMU_PAGE_16M + 1 ... MMU_PAGE_COUNT - 1] = -1 }, .avpnm = 0x1UL, .tlbiel = 0, }, @@ -166,9 +169,10 @@ static unsigned long htab_convert_pte_flags(unsigned long pteflags) if ((pteflags & _PAGE_USER) && !((pteflags & _PAGE_RW) && (pteflags & _PAGE_DIRTY))) rflags |= 1; - - /* Always add C */ - return rflags | HPTE_R_C; + /* + * Always add "C" bit for perf. Memory coherence is always enabled + */ + return rflags | HPTE_R_C | HPTE_R_M; } int htab_bolt_mapping(unsigned long vstart, unsigned long vend, @@ -194,16 +198,39 @@ int htab_bolt_mapping(unsigned long vstart, unsigned long vend, unsigned long vpn = hpt_vpn(vaddr, vsid, ssize); unsigned long tprot = prot; + /* + * If we hit a bad address return error. + */ + if (!vsid) + return -1; /* Make kernel text executable */ if (overlaps_kernel_text(vaddr, vaddr + step)) tprot &= ~HPTE_R_N; + /* Make kvm guest trampolines executable */ + if (overlaps_kvm_tmp(vaddr, vaddr + step)) + tprot &= ~HPTE_R_N; + + /* + * If relocatable, check if it overlaps interrupt vectors that + * are copied down to real 0. For relocatable kernel + * (e.g. kdump case) we copy interrupt vectors down to real + * address 0. Mark that region as executable. This is + * because on p8 system with relocation on exception feature + * enabled, exceptions are raised with MMU (IR=DR=1) ON. Hence + * in order to execute the interrupt handlers in virtual + * mode the vector region need to be marked as executable. + */ + if ((PHYSICAL_START > MEMORY_START) && + overlaps_interrupt_vector_text(vaddr, vaddr + step)) + tprot &= ~HPTE_R_N; + hash = hpt_hash(vpn, shift, ssize); hpteg = ((hash & htab_hash_mask) * HPTES_PER_GROUP); BUG_ON(!ppc_md.hpte_insert); ret = ppc_md.hpte_insert(hpteg, vpn, paddr, tprot, - HPTE_V_BOLTED, psize, ssize); + HPTE_V_BOLTED, psize, psize, ssize); if (ret < 0) break; @@ -242,20 +269,19 @@ static int __init htab_dt_scan_seg_sizes(unsigned long node, const char *uname, int depth, void *data) { - char *type = of_get_flat_dt_prop(node, "device_type", NULL); - u32 *prop; - unsigned long size = 0; + const char *type = of_get_flat_dt_prop(node, "device_type", NULL); + const __be32 *prop; + int size = 0; /* We are scanning "cpu" nodes only */ if (type == NULL || strcmp(type, "cpu") != 0) return 0; - prop = (u32 *)of_get_flat_dt_prop(node, "ibm,processor-segment-sizes", - &size); + prop = of_get_flat_dt_prop(node, "ibm,processor-segment-sizes", &size); if (prop == NULL) return 0; for (; size >= 4; size -= 4, ++prop) { - if (prop[0] == 40) { + if (be32_to_cpu(prop[0]) == 40) { DBG("1T segment support detected\n"); cur_cpu_spec->mmu_features |= MMU_FTR_1T_SEGMENT; return 1; @@ -270,79 +296,103 @@ static void __init htab_init_seg_sizes(void) of_scan_flat_dt(htab_dt_scan_seg_sizes, NULL); } +static int __init get_idx_from_shift(unsigned int shift) +{ + int idx = -1; + + switch (shift) { + case 0xc: + idx = MMU_PAGE_4K; + break; + case 0x10: + idx = MMU_PAGE_64K; + break; + case 0x14: + idx = MMU_PAGE_1M; + break; + case 0x18: + idx = MMU_PAGE_16M; + break; + case 0x22: + idx = MMU_PAGE_16G; + break; + } + return idx; +} + static int __init htab_dt_scan_page_sizes(unsigned long node, const char *uname, int depth, void *data) { - char *type = of_get_flat_dt_prop(node, "device_type", NULL); - u32 *prop; - unsigned long size = 0; + const char *type = of_get_flat_dt_prop(node, "device_type", NULL); + const __be32 *prop; + int size = 0; /* We are scanning "cpu" nodes only */ if (type == NULL || strcmp(type, "cpu") != 0) return 0; - prop = (u32 *)of_get_flat_dt_prop(node, - "ibm,segment-page-sizes", &size); + prop = of_get_flat_dt_prop(node, "ibm,segment-page-sizes", &size); if (prop != NULL) { - DBG("Page sizes from device-tree:\n"); + pr_info("Page sizes from device-tree:\n"); size /= 4; cur_cpu_spec->mmu_features &= ~(MMU_FTR_16M_PAGE); while(size > 0) { - unsigned int shift = prop[0]; - unsigned int slbenc = prop[1]; - unsigned int lpnum = prop[2]; - unsigned int lpenc = 0; + unsigned int base_shift = be32_to_cpu(prop[0]); + unsigned int slbenc = be32_to_cpu(prop[1]); + unsigned int lpnum = be32_to_cpu(prop[2]); struct mmu_psize_def *def; - int idx = -1; + int idx, base_idx; size -= 3; prop += 3; - while(size > 0 && lpnum) { - if (prop[0] == shift) - lpenc = prop[1]; - prop += 2; size -= 2; - lpnum--; + base_idx = get_idx_from_shift(base_shift); + if (base_idx < 0) { + /* + * skip the pte encoding also + */ + prop += lpnum * 2; size -= lpnum * 2; + continue; } - switch(shift) { - case 0xc: - idx = MMU_PAGE_4K; - break; - case 0x10: - idx = MMU_PAGE_64K; - break; - case 0x14: - idx = MMU_PAGE_1M; - break; - case 0x18: - idx = MMU_PAGE_16M; + def = &mmu_psize_defs[base_idx]; + if (base_idx == MMU_PAGE_16M) cur_cpu_spec->mmu_features |= MMU_FTR_16M_PAGE; - break; - case 0x22: - idx = MMU_PAGE_16G; - break; - } - if (idx < 0) - continue; - def = &mmu_psize_defs[idx]; - def->shift = shift; - if (shift <= 23) + + def->shift = base_shift; + if (base_shift <= 23) def->avpnm = 0; else - def->avpnm = (1 << (shift - 23)) - 1; + def->avpnm = (1 << (base_shift - 23)) - 1; def->sllp = slbenc; - def->penc = lpenc; - /* We don't know for sure what's up with tlbiel, so + /* + * We don't know for sure what's up with tlbiel, so * for now we only set it for 4K and 64K pages */ - if (idx == MMU_PAGE_4K || idx == MMU_PAGE_64K) + if (base_idx == MMU_PAGE_4K || base_idx == MMU_PAGE_64K) def->tlbiel = 1; else def->tlbiel = 0; - DBG(" %d: shift=%02x, sllp=%04lx, avpnm=%08lx, " - "tlbiel=%d, penc=%d\n", - idx, shift, def->sllp, def->avpnm, def->tlbiel, - def->penc); + while (size > 0 && lpnum) { + unsigned int shift = be32_to_cpu(prop[0]); + int penc = be32_to_cpu(prop[1]); + + prop += 2; size -= 2; + lpnum--; + + idx = get_idx_from_shift(shift); + if (idx < 0) + continue; + + if (penc == -1) + pr_err("Invalid penc for base_shift=%d " + "shift=%d\n", base_shift, shift); + + def->penc[idx] = penc; + pr_info("base_shift=%d: shift=%d, sllp=0x%04lx," + " avpnm=0x%08lx, tlbiel=%d, penc=%d\n", + base_shift, shift, def->sllp, + def->avpnm, def->tlbiel, def->penc[idx]); + } } return 1; } @@ -356,9 +406,9 @@ static int __init htab_dt_scan_page_sizes(unsigned long node, static int __init htab_dt_scan_hugepage_blocks(unsigned long node, const char *uname, int depth, void *data) { - char *type = of_get_flat_dt_prop(node, "device_type", NULL); - unsigned long *addr_prop; - u32 *page_count_prop; + const char *type = of_get_flat_dt_prop(node, "device_type", NULL); + const __be64 *addr_prop; + const __be32 *page_count_prop; unsigned int expected_pages; long unsigned int phys_addr; long unsigned int block_size; @@ -372,12 +422,12 @@ static int __init htab_dt_scan_hugepage_blocks(unsigned long node, page_count_prop = of_get_flat_dt_prop(node, "ibm,expected#pages", NULL); if (page_count_prop == NULL) return 0; - expected_pages = (1 << page_count_prop[0]); + expected_pages = (1 << be32_to_cpu(page_count_prop[0])); addr_prop = of_get_flat_dt_prop(node, "reg", NULL); if (addr_prop == NULL) return 0; - phys_addr = addr_prop[0]; - block_size = addr_prop[1]; + phys_addr = be64_to_cpu(addr_prop[0]); + block_size = be64_to_cpu(addr_prop[1]); if (block_size != (16 * GB)) return 0; printk(KERN_INFO "Huge page(16GB) memory: " @@ -391,10 +441,39 @@ static int __init htab_dt_scan_hugepage_blocks(unsigned long node, } #endif /* CONFIG_HUGETLB_PAGE */ +static void mmu_psize_set_default_penc(void) +{ + int bpsize, apsize; + for (bpsize = 0; bpsize < MMU_PAGE_COUNT; bpsize++) + for (apsize = 0; apsize < MMU_PAGE_COUNT; apsize++) + mmu_psize_defs[bpsize].penc[apsize] = -1; +} + +#ifdef CONFIG_PPC_64K_PAGES + +static bool might_have_hea(void) +{ + /* + * The HEA ethernet adapter requires awareness of the + * GX bus. Without that awareness we can easily assume + * we will never see an HEA ethernet device. + */ +#ifdef CONFIG_IBMEBUS + return !cpu_has_feature(CPU_FTR_ARCH_207S); +#else + return false; +#endif +} + +#endif /* #ifdef CONFIG_PPC_64K_PAGES */ + static void __init htab_init_page_sizes(void) { int rc; + /* se the invalid penc to -1 */ + mmu_psize_set_default_penc(); + /* Default to 4K pages only */ memcpy(mmu_psize_defs, mmu_psize_defaults_old, sizeof(mmu_psize_defaults_old)); @@ -442,10 +521,11 @@ static void __init htab_init_page_sizes(void) mmu_linear_psize = MMU_PAGE_64K; if (mmu_has_feature(MMU_FTR_CI_LARGE_PAGE)) { /* - * Don't use 64k pages for ioremap on pSeries, since - * that would stop us accessing the HEA ethernet. + * When running on pSeries using 64k pages for ioremap + * would stop us accessing the HEA ethernet. So if we + * have the chance of ever seeing one, stay at 4k. */ - if (!machine_is(pseries)) + if (!might_have_hea() || !machine_is(pseries)) mmu_io_psize = MMU_PAGE_64K; } else mmu_ci_restrictions = 1; @@ -489,17 +569,17 @@ static int __init htab_dt_scan_pftsize(unsigned long node, const char *uname, int depth, void *data) { - char *type = of_get_flat_dt_prop(node, "device_type", NULL); - u32 *prop; + const char *type = of_get_flat_dt_prop(node, "device_type", NULL); + const __be32 *prop; /* We are scanning "cpu" nodes only */ if (type == NULL || strcmp(type, "cpu") != 0) return 0; - prop = (u32 *)of_get_flat_dt_prop(node, "ibm,pft-size", NULL); + prop = of_get_flat_dt_prop(node, "ibm,pft-size", NULL); if (prop != NULL) { /* pft_size[0] is the NUMA CEC cookie */ - ppc64_pft_size = prop[1]; + ppc64_pft_size = be32_to_cpu(prop[1]); return 1; } return 0; @@ -546,47 +626,43 @@ int remove_section_mapping(unsigned long start, unsigned long end) } #endif /* CONFIG_MEMORY_HOTPLUG */ -#define FUNCTION_TEXT(A) ((*(unsigned long *)(A))) +extern u32 htab_call_hpte_insert1[]; +extern u32 htab_call_hpte_insert2[]; +extern u32 htab_call_hpte_remove[]; +extern u32 htab_call_hpte_updatepp[]; +extern u32 ht64_call_hpte_insert1[]; +extern u32 ht64_call_hpte_insert2[]; +extern u32 ht64_call_hpte_remove[]; +extern u32 ht64_call_hpte_updatepp[]; static void __init htab_finish_init(void) { - extern unsigned int *htab_call_hpte_insert1; - extern unsigned int *htab_call_hpte_insert2; - extern unsigned int *htab_call_hpte_remove; - extern unsigned int *htab_call_hpte_updatepp; - #ifdef CONFIG_PPC_HAS_HASH_64K - extern unsigned int *ht64_call_hpte_insert1; - extern unsigned int *ht64_call_hpte_insert2; - extern unsigned int *ht64_call_hpte_remove; - extern unsigned int *ht64_call_hpte_updatepp; - patch_branch(ht64_call_hpte_insert1, - FUNCTION_TEXT(ppc_md.hpte_insert), + ppc_function_entry(ppc_md.hpte_insert), BRANCH_SET_LINK); patch_branch(ht64_call_hpte_insert2, - FUNCTION_TEXT(ppc_md.hpte_insert), + ppc_function_entry(ppc_md.hpte_insert), BRANCH_SET_LINK); patch_branch(ht64_call_hpte_remove, - FUNCTION_TEXT(ppc_md.hpte_remove), + ppc_function_entry(ppc_md.hpte_remove), BRANCH_SET_LINK); patch_branch(ht64_call_hpte_updatepp, - FUNCTION_TEXT(ppc_md.hpte_updatepp), + ppc_function_entry(ppc_md.hpte_updatepp), BRANCH_SET_LINK); - #endif /* CONFIG_PPC_HAS_HASH_64K */ patch_branch(htab_call_hpte_insert1, - FUNCTION_TEXT(ppc_md.hpte_insert), + ppc_function_entry(ppc_md.hpte_insert), BRANCH_SET_LINK); patch_branch(htab_call_hpte_insert2, - FUNCTION_TEXT(ppc_md.hpte_insert), + ppc_function_entry(ppc_md.hpte_insert), BRANCH_SET_LINK); patch_branch(htab_call_hpte_remove, - FUNCTION_TEXT(ppc_md.hpte_remove), + ppc_function_entry(ppc_md.hpte_remove), BRANCH_SET_LINK); patch_branch(htab_call_hpte_updatepp, - FUNCTION_TEXT(ppc_md.hpte_updatepp), + ppc_function_entry(ppc_md.hpte_updatepp), BRANCH_SET_LINK); } @@ -758,10 +834,12 @@ void __init early_init_mmu(void) /* Initialize stab / SLB management */ if (mmu_has_feature(MMU_FTR_SLB)) slb_initialize(); + else + stab_initialize(get_paca()->stab_real); } #ifdef CONFIG_SMP -void __cpuinit early_init_mmu_secondary(void) +void early_init_mmu_secondary(void) { /* Initialize hash table for that CPU */ if (!firmware_has_feature(FW_FEATURE_LPAR)) @@ -861,7 +939,7 @@ static int subpage_protection(struct mm_struct *mm, unsigned long ea) if (ea >= spt->maxaddr) return 0; - if (ea < 0x100000000) { + if (ea < 0x100000000UL) { /* addresses below 4GB use spt->low_prot */ sbpm = spt->low_prot; } else { @@ -891,14 +969,30 @@ static inline int subpage_protection(struct mm_struct *mm, unsigned long ea) void hash_failure_debug(unsigned long ea, unsigned long access, unsigned long vsid, unsigned long trap, - int ssize, int psize, unsigned long pte) + int ssize, int psize, int lpsize, unsigned long pte) { if (!printk_ratelimit()) return; pr_info("mm: Hashing failure ! EA=0x%lx access=0x%lx current=%s\n", ea, access, current->comm); - pr_info(" trap=0x%lx vsid=0x%lx ssize=%d psize=%d pte=0x%lx\n", - trap, vsid, ssize, psize, pte); + pr_info(" trap=0x%lx vsid=0x%lx ssize=%d base psize=%d psize %d pte=0x%lx\n", + trap, vsid, ssize, psize, lpsize, pte); +} + +static void check_paca_psize(unsigned long ea, struct mm_struct *mm, + int psize, bool user_region) +{ + if (user_region) { + if (psize != get_paca_psize(ea)) { + get_paca()->context = mm->context; + slb_flush_and_rebolt(); + } + } else if (get_paca()->vmalloc_sllp != + mmu_psize_defs[mmu_vmalloc_psize].sllp) { + get_paca()->vmalloc_sllp = + mmu_psize_defs[mmu_vmalloc_psize].sllp; + slb_vmalloc_update(); + } } /* Result code is: @@ -909,6 +1003,7 @@ void hash_failure_debug(unsigned long ea, unsigned long access, */ int hash_page(unsigned long ea, unsigned long access, unsigned long trap) { + enum ctx_state prev_state = exception_enter(); pgd_t *pgdir; unsigned long vsid; struct mm_struct *mm; @@ -921,11 +1016,6 @@ int hash_page(unsigned long ea, unsigned long access, unsigned long trap) DBG_LOW("hash_page(ea=%016lx, access=%lx, trap=%lx\n", ea, access, trap); - if ((ea & ~REGION_MASK) >= PGTABLE_RANGE) { - DBG_LOW(" out of pgtable range !\n"); - return 1; - } - /* Get region & vsid */ switch (REGION_ID(ea)) { case USER_REGION_ID: @@ -933,7 +1023,8 @@ int hash_page(unsigned long ea, unsigned long access, unsigned long trap) mm = current->mm; if (! mm) { DBG_LOW(" user region with no mm !\n"); - return 1; + rc = 1; + goto bail; } psize = get_slice_psize(mm, ea); ssize = user_segment_size(ea); @@ -952,14 +1043,23 @@ int hash_page(unsigned long ea, unsigned long access, unsigned long trap) /* Not a valid range * Send the problem up to do_page_fault */ - return 1; + rc = 1; + goto bail; } DBG_LOW(" mm=%p, mm->pgdir=%p, vsid=%016lx\n", mm, mm->pgd, vsid); + /* Bad address. */ + if (!vsid) { + DBG_LOW("Bad address!\n"); + rc = 1; + goto bail; + } /* Get pgdir */ pgdir = mm->pgd; - if (pgdir == NULL) - return 1; + if (pgdir == NULL) { + rc = 1; + goto bail; + } /* Check CPU locality */ tmp = cpumask_of(smp_processor_id()); @@ -982,7 +1082,8 @@ int hash_page(unsigned long ea, unsigned long access, unsigned long trap) ptep = find_linux_pte_or_hugepte(pgdir, ea, &hugeshift); if (ptep == NULL || !pte_present(*ptep)) { DBG_LOW(" no PTE !\n"); - return 1; + rc = 1; + goto bail; } /* Add _PAGE_PRESENT to the required access perm */ @@ -993,14 +1094,32 @@ int hash_page(unsigned long ea, unsigned long access, unsigned long trap) */ if (access & ~pte_val(*ptep)) { DBG_LOW(" no access !\n"); - return 1; + rc = 1; + goto bail; } + if (hugeshift) { + if (pmd_trans_huge(*(pmd_t *)ptep)) + rc = __hash_page_thp(ea, access, vsid, (pmd_t *)ptep, + trap, local, ssize, psize); #ifdef CONFIG_HUGETLB_PAGE - if (hugeshift) - return __hash_page_huge(ea, access, vsid, ptep, trap, local, - ssize, hugeshift, psize); -#endif /* CONFIG_HUGETLB_PAGE */ + else + rc = __hash_page_huge(ea, access, vsid, ptep, trap, + local, ssize, hugeshift, psize); +#else + else { + /* + * if we have hugeshift, and is not transhuge with + * hugetlb disabled, something is really wrong. + */ + rc = 1; + WARN_ON(1); + } +#endif + check_paca_psize(ea, mm, psize, user_region); + + goto bail; + } #ifndef CONFIG_PPC_64K_PAGES DBG_LOW(" i-pte: %016lx\n", pte_val(*ptep)); @@ -1039,17 +1158,8 @@ int hash_page(unsigned long ea, unsigned long access, unsigned long trap) #endif } } - if (user_region) { - if (psize != get_paca_psize(ea)) { - get_paca()->context = mm->context; - slb_flush_and_rebolt(); - } - } else if (get_paca()->vmalloc_sllp != - mmu_psize_defs[mmu_vmalloc_psize].sllp) { - get_paca()->vmalloc_sllp = - mmu_psize_defs[mmu_vmalloc_psize].sllp; - slb_vmalloc_update(); - } + + check_paca_psize(ea, mm, psize, user_region); #endif /* CONFIG_PPC_64K_PAGES */ #ifdef CONFIG_PPC_HAS_HASH_64K @@ -1071,7 +1181,7 @@ int hash_page(unsigned long ea, unsigned long access, unsigned long trap) */ if (rc == -1) hash_failure_debug(ea, access, vsid, trap, ssize, psize, - pte_val(*ptep)); + psize, pte_val(*ptep)); #ifndef CONFIG_PPC_64K_PAGES DBG_LOW(" o-pte: %016lx\n", pte_val(*ptep)); #else @@ -1079,6 +1189,9 @@ int hash_page(unsigned long ea, unsigned long access, unsigned long trap) pte_val(*(ptep + PTRS_PER_PTE))); #endif DBG_LOW(" -> rc=%d\n", rc); + +bail: + exception_exit(prev_state); return rc; } EXPORT_SYMBOL_GPL(hash_page); @@ -1086,6 +1199,7 @@ EXPORT_SYMBOL_GPL(hash_page); void hash_preload(struct mm_struct *mm, unsigned long ea, unsigned long access, unsigned long trap) { + int hugepage_shift; unsigned long vsid; pgd_t *pgdir; pte_t *ptep; @@ -1107,10 +1221,27 @@ void hash_preload(struct mm_struct *mm, unsigned long ea, pgdir = mm->pgd; if (pgdir == NULL) return; - ptep = find_linux_pte(pgdir, ea); - if (!ptep) + + /* Get VSID */ + ssize = user_segment_size(ea); + vsid = get_vsid(mm->context.id, ea, ssize); + if (!vsid) return; + /* + * Hash doesn't like irqs. Walking linux page table with irq disabled + * saves us from holding multiple locks. + */ + local_irq_save(flags); + /* + * THP pages use update_mmu_cache_pmd. We don't do + * hash preload there. Hence can ignore THP here + */ + ptep = find_linux_pte_or_hugepte(pgdir, ea, &hugepage_shift); + if (!ptep) + goto out_exit; + + WARN_ON(hugepage_shift); #ifdef CONFIG_PPC_64K_PAGES /* If either _PAGE_4K_PFN or _PAGE_NO_CACHE is set (and we are on * a 64K kernel), then we don't preload, hash_page() will take @@ -1119,16 +1250,9 @@ void hash_preload(struct mm_struct *mm, unsigned long ea, * page size demotion here */ if (pte_val(*ptep) & (_PAGE_4K_PFN | _PAGE_NO_CACHE)) - return; + goto out_exit; #endif /* CONFIG_PPC_64K_PAGES */ - /* Get VSID */ - ssize = user_segment_size(ea); - vsid = get_vsid(mm->context.id, ea, ssize); - - /* Hash doesn't like irqs */ - local_irq_save(flags); - /* Is that local to this CPU ? */ if (cpumask_equal(mm_cpumask(mm), cpumask_of(smp_processor_id()))) local = 1; @@ -1147,8 +1271,10 @@ void hash_preload(struct mm_struct *mm, unsigned long ea, */ if (rc == -1) hash_failure_debug(ea, access, vsid, trap, ssize, - mm->context.user_psize, pte_val(*ptep)); - + mm->context.user_psize, + mm->context.user_psize, + pte_val(*ptep)); +out_exit: local_irq_restore(flags); } @@ -1169,8 +1295,28 @@ void flush_hash_page(unsigned long vpn, real_pte_t pte, int psize, int ssize, slot = (hash & htab_hash_mask) * HPTES_PER_GROUP; slot += hidx & _PTEIDX_GROUP_IX; DBG_LOW(" sub %ld: hash=%lx, hidx=%lx\n", index, slot, hidx); - ppc_md.hpte_invalidate(slot, vpn, psize, ssize, local); + /* + * We use same base page size and actual psize, because we don't + * use these functions for hugepage + */ + ppc_md.hpte_invalidate(slot, vpn, psize, psize, ssize, local); } pte_iterate_hashed_end(); + +#ifdef CONFIG_PPC_TRANSACTIONAL_MEM + /* Transactions are not aborted by tlbiel, only tlbie. + * Without, syncing a page back to a block device w/ PIO could pick up + * transactional data (bad!) so we force an abort here. Before the + * sync the page will be made read-only, which will flush_hash_page. + * BIG ISSUE here: if the kernel uses a page from userspace without + * unmapping it first, it may see the speculated version. + */ + if (local && cpu_has_feature(CPU_FTR_TM) && + current->thread.regs && + MSR_TM_ACTIVE(current->thread.regs->msr)) { + tm_enable(); + tm_abort(TM_CAUSE_TLBI); + } +#endif } void flush_hash_range(unsigned long number, int local) @@ -1194,6 +1340,8 @@ void flush_hash_range(unsigned long number, int local) */ void low_hash_fault(struct pt_regs *regs, unsigned long address, int rc) { + enum ctx_state prev_state = exception_enter(); + if (user_mode(regs)) { #ifdef CONFIG_PPC_SUBPAGE_PROT if (rc == -2) @@ -1203,23 +1351,64 @@ void low_hash_fault(struct pt_regs *regs, unsigned long address, int rc) _exception(SIGBUS, regs, BUS_ADRERR, address); } else bad_page_fault(regs, address, SIGBUS); + + exception_exit(prev_state); +} + +long hpte_insert_repeating(unsigned long hash, unsigned long vpn, + unsigned long pa, unsigned long rflags, + unsigned long vflags, int psize, int ssize) +{ + unsigned long hpte_group; + long slot; + +repeat: + hpte_group = ((hash & htab_hash_mask) * + HPTES_PER_GROUP) & ~0x7UL; + + /* Insert into the hash table, primary slot */ + slot = ppc_md.hpte_insert(hpte_group, vpn, pa, rflags, vflags, + psize, psize, ssize); + + /* Primary is full, try the secondary */ + if (unlikely(slot == -1)) { + hpte_group = ((~hash & htab_hash_mask) * + HPTES_PER_GROUP) & ~0x7UL; + slot = ppc_md.hpte_insert(hpte_group, vpn, pa, rflags, + vflags | HPTE_V_SECONDARY, + psize, psize, ssize); + if (slot == -1) { + if (mftb() & 0x1) + hpte_group = ((hash & htab_hash_mask) * + HPTES_PER_GROUP)&~0x7UL; + + ppc_md.hpte_remove(hpte_group); + goto repeat; + } + } + + return slot; } #ifdef CONFIG_DEBUG_PAGEALLOC static void kernel_map_linear_page(unsigned long vaddr, unsigned long lmi) { - unsigned long hash, hpteg; + unsigned long hash; unsigned long vsid = get_kernel_vsid(vaddr, mmu_kernel_ssize); unsigned long vpn = hpt_vpn(vaddr, vsid, mmu_kernel_ssize); unsigned long mode = htab_convert_pte_flags(PAGE_KERNEL); - int ret; + long ret; hash = hpt_hash(vpn, PAGE_SHIFT, mmu_kernel_ssize); - hpteg = ((hash & htab_hash_mask) * HPTES_PER_GROUP); - ret = ppc_md.hpte_insert(hpteg, vpn, __pa(vaddr), - mode, HPTE_V_BOLTED, - mmu_linear_psize, mmu_kernel_ssize); + /* Don't create HPTE entries for bad address */ + if (!vsid) + return; + + ret = hpte_insert_repeating(hash, vpn, __pa(vaddr), mode, + HPTE_V_BOLTED, + mmu_linear_psize, mmu_kernel_ssize); + BUG_ON (ret < 0); spin_lock(&linear_map_hash_lock); BUG_ON(linear_map_hash_slots[lmi] & 0x80); @@ -1243,7 +1432,8 @@ static void kernel_unmap_linear_page(unsigned long vaddr, unsigned long lmi) hash = ~hash; slot = (hash & htab_hash_mask) * HPTES_PER_GROUP; slot += hidx & _PTEIDX_GROUP_IX; - ppc_md.hpte_invalidate(slot, vpn, mmu_linear_psize, mmu_kernel_ssize, 0); + ppc_md.hpte_invalidate(slot, vpn, mmu_linear_psize, mmu_linear_psize, + mmu_kernel_ssize, 0); } void kernel_map_pages(struct page *page, int numpages, int enable) diff --git a/arch/powerpc/mm/hugepage-hash64.c b/arch/powerpc/mm/hugepage-hash64.c new file mode 100644 index 00000000000..826893fcb3a --- /dev/null +++ b/arch/powerpc/mm/hugepage-hash64.c @@ -0,0 +1,179 @@ +/* + * Copyright IBM Corporation, 2013 + * Author Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of version 2.1 of the GNU Lesser General Public License + * as published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * + */ + +/* + * PPC64 THP Support for hash based MMUs + */ +#include <linux/mm.h> +#include <asm/machdep.h> + +int __hash_page_thp(unsigned long ea, unsigned long access, unsigned long vsid, + pmd_t *pmdp, unsigned long trap, int local, int ssize, + unsigned int psize) +{ + unsigned int index, valid; + unsigned char *hpte_slot_array; + unsigned long rflags, pa, hidx; + unsigned long old_pmd, new_pmd; + int ret, lpsize = MMU_PAGE_16M; + unsigned long vpn, hash, shift, slot; + + /* + * atomically mark the linux large page PMD busy and dirty + */ + do { + old_pmd = pmd_val(*pmdp); + /* If PMD busy, retry the access */ + if (unlikely(old_pmd & _PAGE_BUSY)) + return 0; + /* If PMD is trans splitting retry the access */ + if (unlikely(old_pmd & _PAGE_SPLITTING)) + return 0; + /* If PMD permissions don't match, take page fault */ + if (unlikely(access & ~old_pmd)) + return 1; + /* + * Try to lock the PTE, add ACCESSED and DIRTY if it was + * a write access + */ + new_pmd = old_pmd | _PAGE_BUSY | _PAGE_ACCESSED; + if (access & _PAGE_RW) + new_pmd |= _PAGE_DIRTY; + } while (old_pmd != __cmpxchg_u64((unsigned long *)pmdp, + old_pmd, new_pmd)); + /* + * PP bits. _PAGE_USER is already PP bit 0x2, so we only + * need to add in 0x1 if it's a read-only user page + */ + rflags = new_pmd & _PAGE_USER; + if ((new_pmd & _PAGE_USER) && !((new_pmd & _PAGE_RW) && + (new_pmd & _PAGE_DIRTY))) + rflags |= 0x1; + /* + * _PAGE_EXEC -> HW_NO_EXEC since it's inverted + */ + rflags |= ((new_pmd & _PAGE_EXEC) ? 0 : HPTE_R_N); + +#if 0 + if (!cpu_has_feature(CPU_FTR_COHERENT_ICACHE)) { + + /* + * No CPU has hugepages but lacks no execute, so we + * don't need to worry about that case + */ + rflags = hash_page_do_lazy_icache(rflags, __pte(old_pte), trap); + } +#endif + /* + * Find the slot index details for this ea, using base page size. + */ + shift = mmu_psize_defs[psize].shift; + index = (ea & ~HPAGE_PMD_MASK) >> shift; + BUG_ON(index >= 4096); + + vpn = hpt_vpn(ea, vsid, ssize); + hash = hpt_hash(vpn, shift, ssize); + hpte_slot_array = get_hpte_slot_array(pmdp); + + valid = hpte_valid(hpte_slot_array, index); + if (valid) { + /* update the hpte bits */ + hidx = hpte_hash_index(hpte_slot_array, index); + if (hidx & _PTEIDX_SECONDARY) + hash = ~hash; + slot = (hash & htab_hash_mask) * HPTES_PER_GROUP; + slot += hidx & _PTEIDX_GROUP_IX; + + ret = ppc_md.hpte_updatepp(slot, rflags, vpn, + psize, lpsize, ssize, local); + /* + * We failed to update, try to insert a new entry. + */ + if (ret == -1) { + /* + * large pte is marked busy, so we can be sure + * nobody is looking at hpte_slot_array. hence we can + * safely update this here. + */ + valid = 0; + new_pmd &= ~_PAGE_HPTEFLAGS; + hpte_slot_array[index] = 0; + } else + /* clear the busy bits and set the hash pte bits */ + new_pmd = (new_pmd & ~_PAGE_HPTEFLAGS) | _PAGE_HASHPTE; + } + + if (!valid) { + unsigned long hpte_group; + + /* insert new entry */ + pa = pmd_pfn(__pmd(old_pmd)) << PAGE_SHIFT; +repeat: + hpte_group = ((hash & htab_hash_mask) * HPTES_PER_GROUP) & ~0x7UL; + + /* clear the busy bits and set the hash pte bits */ + new_pmd = (new_pmd & ~_PAGE_HPTEFLAGS) | _PAGE_HASHPTE; + + /* Add in WIMG bits */ + rflags |= (new_pmd & (_PAGE_WRITETHRU | _PAGE_NO_CACHE | + _PAGE_GUARDED)); + /* + * enable the memory coherence always + */ + rflags |= HPTE_R_M; + + /* Insert into the hash table, primary slot */ + slot = ppc_md.hpte_insert(hpte_group, vpn, pa, rflags, 0, + psize, lpsize, ssize); + /* + * Primary is full, try the secondary + */ + if (unlikely(slot == -1)) { + hpte_group = ((~hash & htab_hash_mask) * + HPTES_PER_GROUP) & ~0x7UL; + slot = ppc_md.hpte_insert(hpte_group, vpn, pa, + rflags, HPTE_V_SECONDARY, + psize, lpsize, ssize); + if (slot == -1) { + if (mftb() & 0x1) + hpte_group = ((hash & htab_hash_mask) * + HPTES_PER_GROUP) & ~0x7UL; + + ppc_md.hpte_remove(hpte_group); + goto repeat; + } + } + /* + * Hypervisor failure. Restore old pmd and return -1 + * similar to __hash_page_* + */ + if (unlikely(slot == -2)) { + *pmdp = __pmd(old_pmd); + hash_failure_debug(ea, access, vsid, trap, ssize, + psize, lpsize, old_pmd); + return -1; + } + /* + * large pte is marked busy, so we can be sure + * nobody is looking at hpte_slot_array. hence we can + * safely update this here. + */ + mark_hpte_slot_valid(hpte_slot_array, index, slot); + } + /* + * No need to use ldarx/stdcx here + */ + *pmdp = __pmd(new_pmd & ~_PAGE_BUSY); + return 0; +} diff --git a/arch/powerpc/mm/hugetlbpage-book3e.c b/arch/powerpc/mm/hugetlbpage-book3e.c index 3bc700655fc..5e4ee257390 100644 --- a/arch/powerpc/mm/hugetlbpage-book3e.c +++ b/arch/powerpc/mm/hugetlbpage-book3e.c @@ -8,6 +8,44 @@ #include <linux/mm.h> #include <linux/hugetlb.h> +#ifdef CONFIG_PPC_FSL_BOOK3E +#ifdef CONFIG_PPC64 +static inline int tlb1_next(void) +{ + struct paca_struct *paca = get_paca(); + struct tlb_core_data *tcd; + int this, next; + + tcd = paca->tcd_ptr; + this = tcd->esel_next; + + next = this + 1; + if (next >= tcd->esel_max) + next = tcd->esel_first; + + tcd->esel_next = next; + return this; +} +#else +static inline int tlb1_next(void) +{ + int index, ncams; + + ncams = mfspr(SPRN_TLB1CFG) & TLBnCFG_N_ENTRY; + + index = __get_cpu_var(next_tlbcam_idx); + + /* Just round-robin the entries and wrap when we hit the end */ + if (unlikely(index == ncams - 1)) + __get_cpu_var(next_tlbcam_idx) = tlbcam_index; + else + __get_cpu_var(next_tlbcam_idx)++; + + return index; +} +#endif /* !PPC64 */ +#endif /* FSL */ + static inline int mmu_get_tsize(int psize) { return mmu_psize_defs[psize].enc; @@ -47,7 +85,7 @@ void book3e_hugetlb_preload(struct vm_area_struct *vma, unsigned long ea, struct mm_struct *mm; #ifdef CONFIG_PPC_FSL_BOOK3E - int index, ncams; + int index; #endif if (unlikely(is_kernel_addr(ea))) @@ -77,18 +115,11 @@ void book3e_hugetlb_preload(struct vm_area_struct *vma, unsigned long ea, } #ifdef CONFIG_PPC_FSL_BOOK3E - ncams = mfspr(SPRN_TLB1CFG) & TLBnCFG_N_ENTRY; - /* We have to use the CAM(TLB1) on FSL parts for hugepages */ - index = __get_cpu_var(next_tlbcam_idx); + index = tlb1_next(); mtspr(SPRN_MAS0, MAS0_ESEL(index) | MAS0_TLBSEL(1)); - - /* Just round-robin the entries and wrap when we hit the end */ - if (unlikely(index == ncams - 1)) - __get_cpu_var(next_tlbcam_idx) = tlbcam_index; - else - __get_cpu_var(next_tlbcam_idx)++; #endif + mas1 = MAS1_VALID | MAS1_TID(mm->context.id) | MAS1_TSIZE(tsize); mas2 = ea & ~((1UL << shift) - 1); mas2 |= (pte_val(pte) >> PTE_WIMGE_SHIFT) & MAS2_WIMGE_MASK; @@ -103,7 +134,8 @@ void book3e_hugetlb_preload(struct vm_area_struct *vma, unsigned long ea, if (mmu_has_feature(MMU_FTR_USE_PAIRED_MAS)) { mtspr(SPRN_MAS7_MAS3, mas7_3); } else { - mtspr(SPRN_MAS7, upper_32_bits(mas7_3)); + if (mmu_has_feature(MMU_FTR_BIG_PHYS)) + mtspr(SPRN_MAS7, upper_32_bits(mas7_3)); mtspr(SPRN_MAS3, lower_32_bits(mas7_3)); } @@ -117,6 +149,5 @@ void flush_hugetlb_page(struct vm_area_struct *vma, unsigned long vmaddr) struct hstate *hstate = hstate_file(vma->vm_file); unsigned long tsize = huge_page_shift(hstate) - 10; - __flush_tlb_page(vma ? vma->vm_mm : NULL, vmaddr, tsize, 0); - + __flush_tlb_page(vma->vm_mm, vmaddr, tsize, 0); } diff --git a/arch/powerpc/mm/hugetlbpage-hash64.c b/arch/powerpc/mm/hugetlbpage-hash64.c index cecad348f60..a5bcf930119 100644 --- a/arch/powerpc/mm/hugetlbpage-hash64.c +++ b/arch/powerpc/mm/hugetlbpage-hash64.c @@ -14,6 +14,10 @@ #include <asm/cacheflush.h> #include <asm/machdep.h> +extern long hpte_insert_repeating(unsigned long hash, unsigned long vpn, + unsigned long pa, unsigned long rlags, + unsigned long vflags, int psize, int ssize); + int __hash_page_huge(unsigned long ea, unsigned long access, unsigned long vsid, pte_t *ptep, unsigned long trap, int local, int ssize, unsigned int shift, unsigned int mmu_psize) @@ -77,20 +81,15 @@ int __hash_page_huge(unsigned long ea, unsigned long access, unsigned long vsid, slot += (old_pte & _PAGE_F_GIX) >> 12; if (ppc_md.hpte_updatepp(slot, rflags, vpn, mmu_psize, - ssize, local) == -1) + mmu_psize, ssize, local) == -1) old_pte &= ~_PAGE_HPTEFLAGS; } if (likely(!(old_pte & _PAGE_HASHPTE))) { unsigned long hash = hpt_hash(vpn, shift, ssize); - unsigned long hpte_group; pa = pte_pfn(__pte(old_pte)) << PAGE_SHIFT; -repeat: - hpte_group = ((hash & htab_hash_mask) * - HPTES_PER_GROUP) & ~0x7UL; - /* clear HPTE slot informations in new PTE */ #ifdef CONFIG_PPC_64K_PAGES new_pte = (new_pte & ~_PAGE_HPTEFLAGS) | _PAGE_HPTE_SUB0; @@ -100,27 +99,13 @@ repeat: /* Add in WIMG bits */ rflags |= (new_pte & (_PAGE_WRITETHRU | _PAGE_NO_CACHE | _PAGE_COHERENT | _PAGE_GUARDED)); + /* + * enable the memory coherence always + */ + rflags |= HPTE_R_M; - /* Insert into the hash table, primary slot */ - slot = ppc_md.hpte_insert(hpte_group, vpn, pa, rflags, 0, - mmu_psize, ssize); - - /* Primary is full, try the secondary */ - if (unlikely(slot == -1)) { - hpte_group = ((~hash & htab_hash_mask) * - HPTES_PER_GROUP) & ~0x7UL; - slot = ppc_md.hpte_insert(hpte_group, vpn, pa, rflags, - HPTE_V_SECONDARY, - mmu_psize, ssize); - if (slot == -1) { - if (mftb() & 0x1) - hpte_group = ((hash & htab_hash_mask) * - HPTES_PER_GROUP)&~0x7UL; - - ppc_md.hpte_remove(hpte_group); - goto repeat; - } - } + slot = hpte_insert_repeating(hash, vpn, pa, rflags, 0, + mmu_psize, ssize); /* * Hypervisor failure. Restore old pte and return -1 @@ -129,7 +114,7 @@ repeat: if (unlikely(slot == -2)) { *ptep = __pte(old_pte); hash_failure_debug(ea, access, vsid, trap, ssize, - mmu_psize, old_pte); + mmu_psize, mmu_psize, old_pte); return -1; } diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c index 1a6de0a7d8e..7e70ae968e5 100644 --- a/arch/powerpc/mm/hugetlbpage.c +++ b/arch/powerpc/mm/hugetlbpage.c @@ -21,6 +21,9 @@ #include <asm/pgalloc.h> #include <asm/tlb.h> #include <asm/setup.h> +#include <asm/hugetlb.h> + +#ifdef CONFIG_HUGETLB_PAGE #define PAGE_SHIFT_64K 16 #define PAGE_SHIFT_16M 24 @@ -48,66 +51,61 @@ static u64 gpage_freearray[MAX_NUMBER_GPAGES]; static unsigned nr_gpages; #endif -static inline int shift_to_mmu_psize(unsigned int shift) -{ - int psize; +#define hugepd_none(hpd) ((hpd).pd == 0) - for (psize = 0; psize < MMU_PAGE_COUNT; ++psize) - if (mmu_psize_defs[psize].shift == shift) - return psize; - return -1; -} +#ifdef CONFIG_PPC_BOOK3S_64 +/* + * At this point we do the placement change only for BOOK3S 64. This would + * possibly work on other subarchs. + */ -static inline unsigned int mmu_psize_to_shift(unsigned int mmu_psize) +/* + * We have PGD_INDEX_SIZ = 12 and PTE_INDEX_SIZE = 8, so that we can have + * 16GB hugepage pte in PGD and 16MB hugepage pte at PMD; + */ +int pmd_huge(pmd_t pmd) { - if (mmu_psize_defs[mmu_psize].shift) - return mmu_psize_defs[mmu_psize].shift; - BUG(); + /* + * leaf pte for huge page, bottom two bits != 00 + */ + return ((pmd_val(pmd) & 0x3) != 0x0); } -#define hugepd_none(hpd) ((hpd).pd == 0) - -pte_t *find_linux_pte_or_hugepte(pgd_t *pgdir, unsigned long ea, unsigned *shift) +int pud_huge(pud_t pud) { - pgd_t *pg; - pud_t *pu; - pmd_t *pm; - hugepd_t *hpdp = NULL; - unsigned pdshift = PGDIR_SHIFT; - - if (shift) - *shift = 0; + /* + * leaf pte for huge page, bottom two bits != 00 + */ + return ((pud_val(pud) & 0x3) != 0x0); +} - pg = pgdir + pgd_index(ea); - if (is_hugepd(pg)) { - hpdp = (hugepd_t *)pg; - } else if (!pgd_none(*pg)) { - pdshift = PUD_SHIFT; - pu = pud_offset(pg, ea); - if (is_hugepd(pu)) - hpdp = (hugepd_t *)pu; - else if (!pud_none(*pu)) { - pdshift = PMD_SHIFT; - pm = pmd_offset(pu, ea); - if (is_hugepd(pm)) - hpdp = (hugepd_t *)pm; - else if (!pmd_none(*pm)) { - return pte_offset_kernel(pm, ea); - } - } - } +int pgd_huge(pgd_t pgd) +{ + /* + * leaf pte for huge page, bottom two bits != 00 + */ + return ((pgd_val(pgd) & 0x3) != 0x0); +} +#else +int pmd_huge(pmd_t pmd) +{ + return 0; +} - if (!hpdp) - return NULL; +int pud_huge(pud_t pud) +{ + return 0; +} - if (shift) - *shift = hugepd_shift(*hpdp); - return hugepte_offset(hpdp, ea, pdshift); +int pgd_huge(pgd_t pgd) +{ + return 0; } -EXPORT_SYMBOL_GPL(find_linux_pte_or_hugepte); +#endif pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) { + /* Only called for hugetlbfs pages, hence can ignore THP */ return find_linux_pte_or_hugepte(mm->pgd, addr, NULL); } @@ -145,6 +143,7 @@ static int __hugepte_alloc(struct mm_struct *mm, hugepd_t *hpdp, if (unlikely(!hugepd_none(*hpdp))) break; else + /* We use the old format for PPC_FSL_BOOK3E */ hpdp->pd = ((unsigned long)new & ~PD_HUGE) | pshift; } /* If we bailed from the for loop early, an error occurred, clean up */ @@ -156,9 +155,15 @@ static int __hugepte_alloc(struct mm_struct *mm, hugepd_t *hpdp, #else if (!hugepd_none(*hpdp)) kmem_cache_free(cachep, new); - else + else { +#ifdef CONFIG_PPC_BOOK3S_64 + hpdp->pd = (unsigned long)new | + (shift_to_mmu_psize(pshift) << 2); +#else hpdp->pd = ((unsigned long)new & ~PD_HUGE) | pshift; #endif + } +#endif spin_unlock(&mm->page_table_lock); return 0; } @@ -175,6 +180,61 @@ static int __hugepte_alloc(struct mm_struct *mm, hugepd_t *hpdp, #define HUGEPD_PUD_SHIFT PMD_SHIFT #endif +#ifdef CONFIG_PPC_BOOK3S_64 +/* + * At this point we do the placement change only for BOOK3S 64. This would + * possibly work on other subarchs. + */ +pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr, unsigned long sz) +{ + pgd_t *pg; + pud_t *pu; + pmd_t *pm; + hugepd_t *hpdp = NULL; + unsigned pshift = __ffs(sz); + unsigned pdshift = PGDIR_SHIFT; + + addr &= ~(sz-1); + pg = pgd_offset(mm, addr); + + if (pshift == PGDIR_SHIFT) + /* 16GB huge page */ + return (pte_t *) pg; + else if (pshift > PUD_SHIFT) + /* + * We need to use hugepd table + */ + hpdp = (hugepd_t *)pg; + else { + pdshift = PUD_SHIFT; + pu = pud_alloc(mm, pg, addr); + if (pshift == PUD_SHIFT) + return (pte_t *)pu; + else if (pshift > PMD_SHIFT) + hpdp = (hugepd_t *)pu; + else { + pdshift = PMD_SHIFT; + pm = pmd_alloc(mm, pu, addr); + if (pshift == PMD_SHIFT) + /* 16MB hugepage */ + return (pte_t *)pm; + else + hpdp = (hugepd_t *)pm; + } + } + if (!hpdp) + return NULL; + + BUG_ON(!hugepd_none(*hpdp) && !hugepd_ok(*hpdp)); + + if (hugepd_none(*hpdp) && __hugepte_alloc(mm, hpdp, addr, pdshift, pshift)) + return NULL; + + return hugepte_offset(hpdp, addr, pdshift); +} + +#else + pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr, unsigned long sz) { pgd_t *pg; @@ -212,6 +272,7 @@ pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr, unsigned long sz return hugepte_offset(hpdp, addr, pdshift); } +#endif #ifdef CONFIG_PPC_FSL_BOOK3E /* Build list of addresses of gigantic pages. This function is used in early @@ -240,7 +301,7 @@ void add_gpage(u64 addr, u64 page_size, unsigned long number_of_pages) int alloc_bootmem_huge_page(struct hstate *hstate) { struct huge_bootmem_page *m; - int idx = shift_to_mmu_psize(hstate->order + PAGE_SHIFT); + int idx = shift_to_mmu_psize(huge_page_shift(hstate)); int nr_gpages = gpage_freearray[idx].nr_gpages; if (nr_gpages == 0) @@ -401,12 +462,13 @@ static void hugepd_free(struct mmu_gather *tlb, void *hugepte) { struct hugepd_freelist **batchp; - batchp = &__get_cpu_var(hugepd_freelist_cur); + batchp = &get_cpu_var(hugepd_freelist_cur); if (atomic_read(&tlb->mm->mm_users) < 2 || cpumask_equal(mm_cpumask(tlb->mm), cpumask_of(smp_processor_id()))) { kmem_cache_free(hugepte_cache, hugepte); + put_cpu_var(hugepd_freelist_cur); return; } @@ -420,6 +482,7 @@ static void hugepd_free(struct mmu_gather *tlb, void *hugepte) call_rcu_sched(&(*batchp)->rcu, hugepd_free_rcu_callback); *batchp = NULL; } + put_cpu_var(hugepd_freelist_cur); } #endif @@ -475,8 +538,14 @@ static void hugetlb_free_pmd_range(struct mmu_gather *tlb, pud_t *pud, do { pmd = pmd_offset(pud, addr); next = pmd_addr_end(addr, end); - if (pmd_none(*pmd)) + if (!is_hugepd(pmd)) { + /* + * if it is not hugepd pointer, we should already find + * it cleared. + */ + WARN_ON(!pmd_none_or_clear_bad(pmd)); continue; + } #ifdef CONFIG_PPC_FSL_BOOK3E /* * Increment next by the size of the huge mapping since @@ -556,8 +625,6 @@ static void hugetlb_free_pud_range(struct mmu_gather *tlb, pgd_t *pgd, /* * This function frees user-level page tables of a process. - * - * Must be called with pagetable lock held. */ void hugetlb_free_pgd_range(struct mmu_gather *tlb, unsigned long addr, unsigned long end, @@ -613,11 +680,14 @@ follow_huge_addr(struct mm_struct *mm, unsigned long address, int write) struct page *page; unsigned shift; unsigned long mask; - + /* + * Transparent hugepages are handled by generic code. We can skip them + * here. + */ ptep = find_linux_pte_or_hugepte(mm->pgd, address, &shift); /* Verify it is a huge page else bail. */ - if (!ptep || !shift) + if (!ptep || !shift || pmd_trans_huge(*(pmd_t *)ptep)) return ERR_PTR(-EINVAL); mask = (1UL << shift) - 1; @@ -628,16 +698,6 @@ follow_huge_addr(struct mm_struct *mm, unsigned long address, int write) return page; } -int pmd_huge(pmd_t pmd) -{ - return 0; -} - -int pud_huge(pud_t pud) -{ - return 0; -} - struct page * follow_huge_pmd(struct mm_struct *mm, unsigned long address, pmd_t *pmd, int write) @@ -646,69 +706,6 @@ follow_huge_pmd(struct mm_struct *mm, unsigned long address, return NULL; } -static noinline int gup_hugepte(pte_t *ptep, unsigned long sz, unsigned long addr, - unsigned long end, int write, struct page **pages, int *nr) -{ - unsigned long mask; - unsigned long pte_end; - struct page *head, *page, *tail; - pte_t pte; - int refs; - - pte_end = (addr + sz) & ~(sz-1); - if (pte_end < end) - end = pte_end; - - pte = *ptep; - mask = _PAGE_PRESENT | _PAGE_USER; - if (write) - mask |= _PAGE_RW; - - if ((pte_val(pte) & mask) != mask) - return 0; - - /* hugepages are never "special" */ - VM_BUG_ON(!pfn_valid(pte_pfn(pte))); - - refs = 0; - head = pte_page(pte); - - page = head + ((addr & (sz-1)) >> PAGE_SHIFT); - tail = page; - do { - VM_BUG_ON(compound_head(page) != head); - pages[*nr] = page; - (*nr)++; - page++; - refs++; - } while (addr += PAGE_SIZE, addr != end); - - if (!page_cache_add_speculative(head, refs)) { - *nr -= refs; - return 0; - } - - if (unlikely(pte_val(pte) != pte_val(*ptep))) { - /* Could be optimized better */ - *nr -= refs; - while (refs--) - put_page(head); - return 0; - } - - /* - * Any tail page need their mapcount reference taken before we - * return. - */ - while (refs--) { - if (PageTail(tail)) - get_huge_page_tail(tail); - tail++; - } - - return 1; -} - static unsigned long hugepte_addr_end(unsigned long addr, unsigned long end, unsigned long sz) { @@ -742,7 +739,7 @@ unsigned long hugetlb_get_unmapped_area(struct file *file, unsigned long addr, struct hstate *hstate = hstate_file(file); int mmu_psize = shift_to_mmu_psize(huge_page_shift(hstate)); - return slice_get_unmapped_area(addr, len, flags, mmu_psize, 1, 0); + return slice_get_unmapped_area(addr, len, flags, mmu_psize, 1); } #endif @@ -883,11 +880,16 @@ static int __init hugetlbpage_init(void) pdshift = PUD_SHIFT; else pdshift = PGDIR_SHIFT; - - pgtable_cache_add(pdshift - shift, NULL); - if (!PGT_CACHE(pdshift - shift)) - panic("hugetlbpage_init(): could not create " - "pgtable cache for %d bit pagesize\n", shift); + /* + * if we have pdshift and shift value same, we don't + * use pgt cache for hugepd. + */ + if (pdshift != shift) { + pgtable_cache_add(pdshift - shift, NULL); + if (!PGT_CACHE(pdshift - shift)) + panic("hugetlbpage_init(): could not create " + "pgtable cache for %d bit pagesize\n", shift); + } } /* Set default large page size. Currently, we pick 16M or 1M @@ -920,3 +922,168 @@ void flush_dcache_icache_hugepage(struct page *page) } } } + +#endif /* CONFIG_HUGETLB_PAGE */ + +/* + * We have 4 cases for pgds and pmds: + * (1) invalid (all zeroes) + * (2) pointer to next table, as normal; bottom 6 bits == 0 + * (3) leaf pte for huge page, bottom two bits != 00 + * (4) hugepd pointer, bottom two bits == 00, next 4 bits indicate size of table + * + * So long as we atomically load page table pointers we are safe against teardown, + * we can follow the address down to the the page and take a ref on it. + */ + +pte_t *find_linux_pte_or_hugepte(pgd_t *pgdir, unsigned long ea, unsigned *shift) +{ + pgd_t pgd, *pgdp; + pud_t pud, *pudp; + pmd_t pmd, *pmdp; + pte_t *ret_pte; + hugepd_t *hpdp = NULL; + unsigned pdshift = PGDIR_SHIFT; + + if (shift) + *shift = 0; + + pgdp = pgdir + pgd_index(ea); + pgd = ACCESS_ONCE(*pgdp); + /* + * Always operate on the local stack value. This make sure the + * value don't get updated by a parallel THP split/collapse, + * page fault or a page unmap. The return pte_t * is still not + * stable. So should be checked there for above conditions. + */ + if (pgd_none(pgd)) + return NULL; + else if (pgd_huge(pgd)) { + ret_pte = (pte_t *) pgdp; + goto out; + } else if (is_hugepd(&pgd)) + hpdp = (hugepd_t *)&pgd; + else { + /* + * Even if we end up with an unmap, the pgtable will not + * be freed, because we do an rcu free and here we are + * irq disabled + */ + pdshift = PUD_SHIFT; + pudp = pud_offset(&pgd, ea); + pud = ACCESS_ONCE(*pudp); + + if (pud_none(pud)) + return NULL; + else if (pud_huge(pud)) { + ret_pte = (pte_t *) pudp; + goto out; + } else if (is_hugepd(&pud)) + hpdp = (hugepd_t *)&pud; + else { + pdshift = PMD_SHIFT; + pmdp = pmd_offset(&pud, ea); + pmd = ACCESS_ONCE(*pmdp); + /* + * A hugepage collapse is captured by pmd_none, because + * it mark the pmd none and do a hpte invalidate. + * + * A hugepage split is captured by pmd_trans_splitting + * because we mark the pmd trans splitting and do a + * hpte invalidate + * + */ + if (pmd_none(pmd) || pmd_trans_splitting(pmd)) + return NULL; + + if (pmd_huge(pmd) || pmd_large(pmd)) { + ret_pte = (pte_t *) pmdp; + goto out; + } else if (is_hugepd(&pmd)) + hpdp = (hugepd_t *)&pmd; + else + return pte_offset_kernel(&pmd, ea); + } + } + if (!hpdp) + return NULL; + + ret_pte = hugepte_offset(hpdp, ea, pdshift); + pdshift = hugepd_shift(*hpdp); +out: + if (shift) + *shift = pdshift; + return ret_pte; +} +EXPORT_SYMBOL_GPL(find_linux_pte_or_hugepte); + +int gup_hugepte(pte_t *ptep, unsigned long sz, unsigned long addr, + unsigned long end, int write, struct page **pages, int *nr) +{ + unsigned long mask; + unsigned long pte_end; + struct page *head, *page, *tail; + pte_t pte; + int refs; + + pte_end = (addr + sz) & ~(sz-1); + if (pte_end < end) + end = pte_end; + + pte = ACCESS_ONCE(*ptep); + mask = _PAGE_PRESENT | _PAGE_USER; + if (write) + mask |= _PAGE_RW; + + if ((pte_val(pte) & mask) != mask) + return 0; + +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + /* + * check for splitting here + */ + if (pmd_trans_splitting(pte_pmd(pte))) + return 0; +#endif + + /* hugepages are never "special" */ + VM_BUG_ON(!pfn_valid(pte_pfn(pte))); + + refs = 0; + head = pte_page(pte); + + page = head + ((addr & (sz-1)) >> PAGE_SHIFT); + tail = page; + do { + VM_BUG_ON(compound_head(page) != head); + pages[*nr] = page; + (*nr)++; + page++; + refs++; + } while (addr += PAGE_SIZE, addr != end); + + if (!page_cache_add_speculative(head, refs)) { + *nr -= refs; + return 0; + } + + if (unlikely(pte_val(pte) != pte_val(*ptep))) { + /* Could be optimized better */ + *nr -= refs; + while (refs--) + put_page(head); + return 0; + } + + /* + * Any tail page need their mapcount reference taken before we + * return. + */ + while (refs--) { + if (PageTail(tail)) + get_huge_page_tail(tail); + tail++; + } + + return 1; +} diff --git a/arch/powerpc/mm/icswx.c b/arch/powerpc/mm/icswx.c index 8cdbd8634a5..915412e4d5b 100644 --- a/arch/powerpc/mm/icswx.c +++ b/arch/powerpc/mm/icswx.c @@ -67,7 +67,7 @@ void switch_cop(struct mm_struct *next) { -#ifdef CONFIG_ICSWX_PID +#ifdef CONFIG_PPC_ICSWX_PID mtspr(SPRN_PID, next->context.cop_pid); #endif mtspr(SPRN_ACOP, next->context.acop); diff --git a/arch/powerpc/mm/init_32.c b/arch/powerpc/mm/init_32.c index 01e2db97a21..cff59f1bec2 100644 --- a/arch/powerpc/mm/init_32.c +++ b/arch/powerpc/mm/init_32.c @@ -52,7 +52,7 @@ #if defined(CONFIG_KERNEL_START_BOOL) || defined(CONFIG_LOWMEM_SIZE_BOOL) /* The amount of lowmem must be within 0xF0000000 - KERNELBASE. */ #if (CONFIG_LOWMEM_SIZE > (0xF0000000 - PAGE_OFFSET)) -#error "You must adjust CONFIG_LOWMEM_SIZE or CONFIG_START_KERNEL" +#error "You must adjust CONFIG_LOWMEM_SIZE or CONFIG_KERNEL_START" #endif #endif #define MAX_LOW_MEM CONFIG_LOWMEM_SIZE @@ -213,7 +213,12 @@ void setup_initial_memory_limit(phys_addr_t first_memblock_base, */ BUG_ON(first_memblock_base != 0); +#ifdef CONFIG_PIN_TLB + /* 8xx can only access 24MB at the moment */ + memblock_set_current_limit(min_t(u64, first_memblock_size, 0x01800000)); +#else /* 8xx can only access 8MB at the moment */ memblock_set_current_limit(min_t(u64, first_memblock_size, 0x00800000)); +#endif } #endif /* CONFIG_8xx */ diff --git a/arch/powerpc/mm/init_64.c b/arch/powerpc/mm/init_64.c index 95a45293e5a..e3734edffa6 100644 --- a/arch/powerpc/mm/init_64.c +++ b/arch/powerpc/mm/init_64.c @@ -88,7 +88,11 @@ static void pgd_ctor(void *addr) static void pmd_ctor(void *addr) { +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + memset(addr, 0, PMD_TABLE_SIZE * 2); +#else memset(addr, 0, PMD_TABLE_SIZE); +#endif } struct kmem_cache *pgtable_cache[MAX_PGTABLE_INDEX_SIZE]; @@ -129,8 +133,7 @@ void pgtable_cache_add(unsigned shift, void (*ctor)(void *)) align = max_t(unsigned long, align, minalign); name = kasprintf(GFP_KERNEL, "pgtable-2^%d", shift); new = kmem_cache_create(name, table_size, align, 0, ctor); - PGT_CACHE(shift) = new; - + pgtable_cache[shift - 1] = new; pr_debug("Allocated pgtable cache for order %d\n", shift); } @@ -138,10 +141,9 @@ void pgtable_cache_add(unsigned shift, void (*ctor)(void *)) void pgtable_cache_init(void) { pgtable_cache_add(PGD_INDEX_SIZE, pgd_ctor); - pgtable_cache_add(PMD_INDEX_SIZE, pmd_ctor); - if (!PGT_CACHE(PGD_INDEX_SIZE) || !PGT_CACHE(PMD_INDEX_SIZE)) + pgtable_cache_add(PMD_CACHE_INDEX, pmd_ctor); + if (!PGT_CACHE(PGD_INDEX_SIZE) || !PGT_CACHE(PMD_CACHE_INDEX)) panic("Couldn't allocate pgtable caches"); - /* In all current configs, when the PUD index exists it's the * same size as either the pgd or pmd index. Verify that the * initialization above has also created a PUD cache. This @@ -216,7 +218,8 @@ static void __meminit vmemmap_create_mapping(unsigned long start, unsigned long phys) { int mapped = htab_bolt_mapping(start, start + page_size, phys, - PAGE_KERNEL, mmu_vmemmap_psize, + pgprot_val(PAGE_KERNEL), + mmu_vmemmap_psize, mmu_kernel_ssize); BUG_ON(mapped < 0); } @@ -263,19 +266,14 @@ static __meminit void vmemmap_list_populate(unsigned long phys, vmemmap_list = vmem_back; } -int __meminit vmemmap_populate(struct page *start_page, - unsigned long nr_pages, int node) +int __meminit vmemmap_populate(unsigned long start, unsigned long end, int node) { - unsigned long start = (unsigned long)start_page; - unsigned long end = (unsigned long)(start_page + nr_pages); unsigned long page_size = 1 << mmu_psize_defs[mmu_vmemmap_psize].shift; /* Align to the page size of the linear mapping. */ start = _ALIGN_DOWN(start, page_size); - pr_debug("vmemmap_populate page %p, %ld pages, node %d\n", - start_page, nr_pages, node); - pr_debug(" -> map %lx..%lx\n", start, end); + pr_debug("vmemmap_populate %lx..%lx, node %d\n", start, end, node); for (; start < end; start += page_size) { void *p; @@ -297,5 +295,63 @@ int __meminit vmemmap_populate(struct page *start_page, return 0; } -#endif /* CONFIG_SPARSEMEM_VMEMMAP */ +void vmemmap_free(unsigned long start, unsigned long end) +{ +} + +void register_page_bootmem_memmap(unsigned long section_nr, + struct page *start_page, unsigned long size) +{ +} + +/* + * We do not have access to the sparsemem vmemmap, so we fallback to + * walking the list of sparsemem blocks which we already maintain for + * the sake of crashdump. In the long run, we might want to maintain + * a tree if performance of that linear walk becomes a problem. + * + * realmode_pfn_to_page functions can fail due to: + * 1) As real sparsemem blocks do not lay in RAM continously (they + * are in virtual address space which is not available in the real mode), + * the requested page struct can be split between blocks so get_page/put_page + * may fail. + * 2) When huge pages are used, the get_page/put_page API will fail + * in real mode as the linked addresses in the page struct are virtual + * too. + */ +struct page *realmode_pfn_to_page(unsigned long pfn) +{ + struct vmemmap_backing *vmem_back; + struct page *page; + unsigned long page_size = 1 << mmu_psize_defs[mmu_vmemmap_psize].shift; + unsigned long pg_va = (unsigned long) pfn_to_page(pfn); + + for (vmem_back = vmemmap_list; vmem_back; vmem_back = vmem_back->list) { + if (pg_va < vmem_back->virt_addr) + continue; + + /* Check that page struct is not split between real pages */ + if ((pg_va + sizeof(struct page)) > + (vmem_back->virt_addr + page_size)) + return NULL; + + page = (struct page *) (vmem_back->phys + pg_va - + vmem_back->virt_addr); + return page; + } + + return NULL; +} +EXPORT_SYMBOL_GPL(realmode_pfn_to_page); + +#elif defined(CONFIG_FLATMEM) + +struct page *realmode_pfn_to_page(unsigned long pfn) +{ + struct page *page = pfn_to_page(pfn); + return page; +} +EXPORT_SYMBOL_GPL(realmode_pfn_to_page); + +#endif /* CONFIG_SPARSEMEM_VMEMMAP/CONFIG_FLATMEM */ diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c index 0dba5066c22..2c8e90f5789 100644 --- a/arch/powerpc/mm/mem.c +++ b/arch/powerpc/mm/mem.c @@ -66,10 +66,9 @@ unsigned long long memory_limit; #ifdef CONFIG_HIGHMEM pte_t *kmap_pte; +EXPORT_SYMBOL(kmap_pte); pgprot_t kmap_prot; - EXPORT_SYMBOL(kmap_prot); -EXPORT_SYMBOL(kmap_pte); static inline pte_t *virt_to_kpte(unsigned long vaddr) { @@ -133,6 +132,23 @@ int arch_add_memory(int nid, u64 start, u64 size) return __add_pages(nid, zone, start_pfn, nr_pages); } + +#ifdef CONFIG_MEMORY_HOTREMOVE +int arch_remove_memory(u64 start, u64 size) +{ + unsigned long start_pfn = start >> PAGE_SHIFT; + unsigned long nr_pages = size >> PAGE_SHIFT; + struct zone *zone; + int ret; + + zone = page_zone(pfn_to_page(start_pfn)); + ret = __remove_pages(zone, start_pfn, nr_pages); + if (!ret && (ppc_md.remove_memory)) + ret = ppc_md.remove_memory(start, size); + + return ret; +} +#endif #endif /* CONFIG_MEMORY_HOTPLUG */ /* @@ -195,13 +211,10 @@ void __init do_init_bootmem(void) min_low_pfn = MEMORY_START >> PAGE_SHIFT; boot_mapsize = init_bootmem_node(NODE_DATA(0), start >> PAGE_SHIFT, min_low_pfn, max_low_pfn); - /* Add active regions with valid PFNs */ - for_each_memblock(memory, reg) { - unsigned long start_pfn, end_pfn; - start_pfn = memblock_region_memory_base_pfn(reg); - end_pfn = memblock_region_memory_end_pfn(reg); - memblock_set_node(0, (phys_addr_t)ULLONG_MAX, 0); - } + /* Place all memblock_regions in the same node and merge contiguous + * memblock_regions + */ + memblock_set_node(0, (phys_addr_t)ULLONG_MAX, &memblock.memory, 0); /* Add all physical memory to the bootmem map, mark each area * present. @@ -289,49 +302,30 @@ void __init paging_init(void) } #endif /* ! CONFIG_NEED_MULTIPLE_NODES */ +static void __init register_page_bootmem_info(void) +{ + int i; + + for_each_online_node(i) + register_page_bootmem_info_node(NODE_DATA(i)); +} + void __init mem_init(void) { -#ifdef CONFIG_NEED_MULTIPLE_NODES - int nid; -#endif - pg_data_t *pgdat; - unsigned long i; - struct page *page; - unsigned long reservedpages = 0, codesize, initsize, datasize, bsssize; + /* + * book3s is limited to 16 page sizes due to encoding this in + * a 4-bit field for slices. + */ + BUILD_BUG_ON(MMU_PAGE_COUNT > 16); #ifdef CONFIG_SWIOTLB swiotlb_init(0); #endif - num_physpages = memblock_phys_mem_size() >> PAGE_SHIFT; + register_page_bootmem_info(); high_memory = (void *) __va(max_low_pfn * PAGE_SIZE); - -#ifdef CONFIG_NEED_MULTIPLE_NODES - for_each_online_node(nid) { - if (NODE_DATA(nid)->node_spanned_pages != 0) { - printk("freeing bootmem node %d\n", nid); - totalram_pages += - free_all_bootmem_node(NODE_DATA(nid)); - } - } -#else - max_mapnr = max_pfn; - totalram_pages += free_all_bootmem(); -#endif - for_each_online_pgdat(pgdat) { - for (i = 0; i < pgdat->node_spanned_pages; i++) { - if (!pfn_valid(pgdat->node_start_pfn + i)) - continue; - page = pgdat_page_nr(pgdat, i); - if (PageReserved(page)) - reservedpages++; - } - } - - codesize = (unsigned long)&_sdata - (unsigned long)&_stext; - datasize = (unsigned long)&_edata - (unsigned long)&_sdata; - initsize = (unsigned long)&__init_end - (unsigned long)&__init_begin; - bsssize = (unsigned long)&__bss_stop - (unsigned long)&__bss_start; + set_max_mapnr(max_pfn); + free_all_bootmem(); #ifdef CONFIG_HIGHMEM { @@ -341,17 +335,9 @@ void __init mem_init(void) for (pfn = highmem_mapnr; pfn < max_mapnr; ++pfn) { phys_addr_t paddr = (phys_addr_t)pfn << PAGE_SHIFT; struct page *page = pfn_to_page(pfn); - if (memblock_is_reserved(paddr)) - continue; - ClearPageReserved(page); - init_page_count(page); - __free_page(page); - totalhigh_pages++; - reservedpages--; + if (!memblock_is_reserved(paddr)) + free_highmem_page(page); } - totalram_pages += totalhigh_pages; - printk(KERN_DEBUG "High memory: %luk\n", - totalhigh_pages << (PAGE_SHIFT-10)); } #endif /* CONFIG_HIGHMEM */ @@ -364,16 +350,7 @@ void __init mem_init(void) (mfspr(SPRN_TLB1CFG) & TLBnCFG_N_ENTRY) - 1; #endif - printk(KERN_INFO "Memory: %luk/%luk available (%luk kernel code, " - "%luk reserved, %luk data, %luk bss, %luk init)\n", - nr_free_pages() << (PAGE_SHIFT-10), - num_physpages << (PAGE_SHIFT-10), - codesize >> 10, - reservedpages << (PAGE_SHIFT-10), - datasize >> 10, - bsssize >> 10, - initsize >> 10); - + mem_init_print_info(NULL); #ifdef CONFIG_PPC32 pr_info("Kernel virtual memory layout:\n"); pr_info(" * 0x%08lx..0x%08lx : fixmap\n", FIXADDR_START, FIXADDR_TOP); @@ -396,39 +373,14 @@ void __init mem_init(void) void free_initmem(void) { - unsigned long addr; - ppc_md.progress = ppc_printk_progress; - - addr = (unsigned long)__init_begin; - for (; addr < (unsigned long)__init_end; addr += PAGE_SIZE) { - memset((void *)addr, POISON_FREE_INITMEM, PAGE_SIZE); - ClearPageReserved(virt_to_page(addr)); - init_page_count(virt_to_page(addr)); - free_page(addr); - totalram_pages++; - } - pr_info("Freeing unused kernel memory: %luk freed\n", - ((unsigned long)__init_end - - (unsigned long)__init_begin) >> 10); + free_initmem_default(POISON_FREE_INITMEM); } #ifdef CONFIG_BLK_DEV_INITRD void __init free_initrd_mem(unsigned long start, unsigned long end) { - if (start >= end) - return; - - start = _ALIGN_DOWN(start, PAGE_SIZE); - end = _ALIGN_UP(end, PAGE_SIZE); - pr_info("Freeing initrd memory: %ldk freed\n", (end - start) >> 10); - - for (; start < end; start += PAGE_SIZE) { - ClearPageReserved(virt_to_page(start)); - init_page_count(virt_to_page(start)); - free_page(start); - totalram_pages++; - } + free_reserved_area((void *)start, (void *)end, -1, "initrd"); } #endif @@ -529,6 +481,10 @@ void update_mmu_cache(struct vm_area_struct *vma, unsigned long address, pte_t *ptep) { #ifdef CONFIG_PPC_STD_MMU + /* + * We don't need to worry about _PAGE_PRESENT here because we are + * called with either mm->page_table_lock held or ptl lock held + */ unsigned long access = 0, trap; /* We only want HPTEs for linux PTEs that have _PAGE_ACCESSED set */ @@ -562,7 +518,7 @@ void update_mmu_cache(struct vm_area_struct *vma, unsigned long address, * System memory should not be in /proc/iomem but various tools expect it * (eg kdump). */ -static int add_system_ram_resources(void) +static int __init add_system_ram_resources(void) { struct memblock_region *reg; @@ -578,7 +534,7 @@ static int add_system_ram_resources(void) res->name = "System RAM"; res->start = base; res->end = base + size - 1; - res->flags = IORESOURCE_MEM; + res->flags = IORESOURCE_MEM | IORESOURCE_BUSY; WARN_ON(request_resource(&iomem_resource, res) < 0); } } diff --git a/arch/powerpc/mm/mmap_64.c b/arch/powerpc/mm/mmap.c index 67a42ed0d2f..cb8bdbe4972 100644 --- a/arch/powerpc/mm/mmap_64.c +++ b/arch/powerpc/mm/mmap.c @@ -92,10 +92,8 @@ void arch_pick_mmap_layout(struct mm_struct *mm) if (mmap_is_legacy()) { mm->mmap_base = TASK_UNMAPPED_BASE; mm->get_unmapped_area = arch_get_unmapped_area; - mm->unmap_area = arch_unmap_area; } else { mm->mmap_base = mmap_base(); mm->get_unmapped_area = arch_get_unmapped_area_topdown; - mm->unmap_area = arch_unmap_area_topdown; } } diff --git a/arch/powerpc/mm/mmu_context_hash64.c b/arch/powerpc/mm/mmu_context_hash64.c index 40bc5b0ace5..178876aef40 100644 --- a/arch/powerpc/mm/mmu_context_hash64.c +++ b/arch/powerpc/mm/mmu_context_hash64.c @@ -23,21 +23,13 @@ #include <linux/slab.h> #include <asm/mmu_context.h> +#include <asm/pgalloc.h> #include "icswx.h" static DEFINE_SPINLOCK(mmu_context_lock); static DEFINE_IDA(mmu_context_ida); -/* - * 256MB segment - * The proto-VSID space has 2^(CONTEX_BITS + USER_ESID_BITS) - 1 segments - * available for user mappings. Each segment contains 2^28 bytes. Each - * context maps 2^46 bytes (64TB) so we can support 2^19-1 contexts - * (19 == 37 + 28 - 46). - */ -#define MAX_CONTEXT ((1UL << CONTEXT_BITS) - 1) - int __init_new_context(void) { int index; @@ -56,7 +48,7 @@ again: else if (err) return err; - if (index > MAX_CONTEXT) { + if (index > MAX_USER_CONTEXT) { spin_lock(&mmu_context_lock); ida_remove(&mmu_context_ida, index); spin_unlock(&mmu_context_lock); @@ -94,6 +86,9 @@ int init_new_context(struct task_struct *tsk, struct mm_struct *mm) spin_lock_init(mm->context.cop_lockp); #endif /* CONFIG_PPC_ICSWX */ +#ifdef CONFIG_PPC_64K_PAGES + mm->context.pte_frag = NULL; +#endif return 0; } @@ -105,13 +100,46 @@ void __destroy_context(int context_id) } EXPORT_SYMBOL_GPL(__destroy_context); +#ifdef CONFIG_PPC_64K_PAGES +static void destroy_pagetable_page(struct mm_struct *mm) +{ + int count; + void *pte_frag; + struct page *page; + + pte_frag = mm->context.pte_frag; + if (!pte_frag) + return; + + page = virt_to_page(pte_frag); + /* drop all the pending references */ + count = ((unsigned long)pte_frag & ~PAGE_MASK) >> PTE_FRAG_SIZE_SHIFT; + /* We allow PTE_FRAG_NR fragments from a PTE page */ + count = atomic_sub_return(PTE_FRAG_NR - count, &page->_count); + if (!count) { + pgtable_page_dtor(page); + free_hot_cold_page(page, 0); + } +} + +#else +static inline void destroy_pagetable_page(struct mm_struct *mm) +{ + return; +} +#endif + + void destroy_context(struct mm_struct *mm) { + #ifdef CONFIG_PPC_ICSWX drop_cop(mm->context.acop, mm); kfree(mm->context.cop_lockp); mm->context.cop_lockp = NULL; #endif /* CONFIG_PPC_ICSWX */ + + destroy_pagetable_page(mm); __destroy_context(mm->context.id); subpage_prot_free(mm); mm->context.id = MMU_NO_CONTEXT; diff --git a/arch/powerpc/mm/mmu_context_nohash.c b/arch/powerpc/mm/mmu_context_nohash.c index e779642c25e..928ebe79668 100644 --- a/arch/powerpc/mm/mmu_context_nohash.c +++ b/arch/powerpc/mm/mmu_context_nohash.c @@ -112,8 +112,10 @@ static unsigned int steal_context_smp(unsigned int id) */ for_each_cpu(cpu, mm_cpumask(mm)) { for (i = cpu_first_thread_sibling(cpu); - i <= cpu_last_thread_sibling(cpu); i++) - __set_bit(id, stale_map[i]); + i <= cpu_last_thread_sibling(cpu); i++) { + if (stale_map[i]) + __set_bit(id, stale_map[i]); + } cpu = i - 1; } return id; @@ -272,7 +274,8 @@ void switch_mmu_context(struct mm_struct *prev, struct mm_struct *next) /* XXX This clear should ultimately be part of local_flush_tlb_mm */ for (i = cpu_first_thread_sibling(cpu); i <= cpu_last_thread_sibling(cpu); i++) { - __clear_bit(id, stale_map[i]); + if (stale_map[i]) + __clear_bit(id, stale_map[i]); } } @@ -329,8 +332,8 @@ void destroy_context(struct mm_struct *mm) #ifdef CONFIG_SMP -static int __cpuinit mmu_context_cpu_notify(struct notifier_block *self, - unsigned long action, void *hcpu) +static int mmu_context_cpu_notify(struct notifier_block *self, + unsigned long action, void *hcpu) { unsigned int cpu = (unsigned int)(long)hcpu; @@ -363,7 +366,7 @@ static int __cpuinit mmu_context_cpu_notify(struct notifier_block *self, return NOTIFY_OK; } -static struct notifier_block __cpuinitdata mmu_context_cpu_nb = { +static struct notifier_block mmu_context_cpu_nb = { .notifier_call = mmu_context_cpu_notify, }; @@ -407,17 +410,7 @@ void __init mmu_context_init(void) } else if (mmu_has_feature(MMU_FTR_TYPE_47x)) { first_context = 1; last_context = 65535; - } else -#ifdef CONFIG_PPC_BOOK3E_MMU - if (mmu_has_feature(MMU_FTR_TYPE_3E)) { - u32 mmucfg = mfspr(SPRN_MMUCFG); - u32 pid_bits = (mmucfg & MMUCFG_PIDSIZE_MASK) - >> MMUCFG_PIDSIZE_SHIFT; - first_context = 1; - last_context = (1UL << (pid_bits + 1)) - 1; - } else -#endif - { + } else { first_context = 1; last_context = 255; } diff --git a/arch/powerpc/mm/mmu_decl.h b/arch/powerpc/mm/mmu_decl.h index 83eb5d5f53d..9615d82919b 100644 --- a/arch/powerpc/mm/mmu_decl.h +++ b/arch/powerpc/mm/mmu_decl.h @@ -148,6 +148,8 @@ extern unsigned long calc_cam_sz(unsigned long ram, unsigned long virt, extern void MMU_init_hw(void); extern unsigned long mmu_mapin_ram(unsigned long top); extern void adjust_total_lowmem(void); +extern int switch_to_as1(void); +extern void restore_to_as0(int esel, int offset, void *dt_ptr, int bootcpu); #endif extern void loadcam_entry(unsigned int index); diff --git a/arch/powerpc/mm/numa.c b/arch/powerpc/mm/numa.c index bba87ca2b4d..3b181b22cd4 100644 --- a/arch/powerpc/mm/numa.c +++ b/arch/powerpc/mm/numa.c @@ -22,13 +22,22 @@ #include <linux/pfn.h> #include <linux/cpuset.h> #include <linux/node.h> +#include <linux/stop_machine.h> +#include <linux/proc_fs.h> +#include <linux/seq_file.h> +#include <linux/uaccess.h> +#include <linux/slab.h> +#include <asm/cputhreads.h> #include <asm/sparsemem.h> #include <asm/prom.h> #include <asm/smp.h> +#include <asm/cputhreads.h> +#include <asm/topology.h> #include <asm/firmware.h> #include <asm/paca.h> #include <asm/hvcall.h> #include <asm/setup.h> +#include <asm/vdso.h> static int numa_enabled = 1; @@ -51,7 +60,7 @@ static int form1_affinity; #define MAX_DISTANCE_REF_POINTS 4 static int distance_ref_points_depth; -static const unsigned int *distance_ref_points; +static const __be32 *distance_ref_points; static int distance_lookup_table[MAX_NUMNODES][MAX_DISTANCE_REF_POINTS]; /* @@ -62,14 +71,11 @@ static int distance_lookup_table[MAX_NUMNODES][MAX_DISTANCE_REF_POINTS]; */ static void __init setup_node_to_cpumask_map(void) { - unsigned int node, num = 0; + unsigned int node; /* setup nr_node_ids if not done yet */ - if (nr_node_ids == MAX_NUMNODES) { - for_each_node_mask(node, node_possible_map) - num = node; - nr_node_ids = num + 1; - } + if (nr_node_ids == MAX_NUMNODES) + setup_nr_node_ids(); /* allocate the map */ for (node = 0; node < nr_node_ids; node++) @@ -79,7 +85,7 @@ static void __init setup_node_to_cpumask_map(void) dbg("Node to cpumask map for %d nodes\n", nr_node_ids); } -static int __cpuinit fake_numa_create_new_node(unsigned long end_pfn, +static int __init fake_numa_create_new_node(unsigned long end_pfn, unsigned int *nid) { unsigned long long mem; @@ -148,9 +154,22 @@ static void __init get_node_active_region(unsigned long pfn, } } -static void map_cpu_to_node(int cpu, int node) +static void reset_numa_cpu_lookup_table(void) +{ + unsigned int cpu; + + for_each_possible_cpu(cpu) + numa_cpu_lookup_table[cpu] = -1; +} + +static void update_numa_cpu_lookup_table(unsigned int cpu, int node) { numa_cpu_lookup_table[cpu] = node; +} + +static void map_cpu_to_node(int cpu, int node) +{ + update_numa_cpu_lookup_table(cpu, node); dbg("adding cpu %d to node %d\n", cpu, node); @@ -175,7 +194,7 @@ static void unmap_cpu_from_node(unsigned long cpu) #endif /* CONFIG_HOTPLUG_CPU || CONFIG_PPC_SPLPAR */ /* must hold reference to node during call */ -static const int *of_get_associativity(struct device_node *dev) +static const __be32 *of_get_associativity(struct device_node *dev) { return of_get_property(dev, "ibm,associativity", NULL); } @@ -185,13 +204,13 @@ static const int *of_get_associativity(struct device_node *dev) * it exists (the property exists only in kexec/kdump kernels, * added by kexec-tools) */ -static const u32 *of_get_usable_memory(struct device_node *memory) +static const __be32 *of_get_usable_memory(struct device_node *memory) { - const u32 *prop; + const __be32 *prop; u32 len; prop = of_get_property(memory, "linux,drconf-usable-memory", &len); if (!prop || len < sizeof(unsigned int)) - return 0; + return NULL; return prop; } @@ -201,7 +220,7 @@ int __node_distance(int a, int b) int distance = LOCAL_DISTANCE; if (!form1_affinity) - return distance; + return ((a == b) ? LOCAL_DISTANCE : REMOTE_DISTANCE); for (i = 0; i < distance_ref_points_depth; i++) { if (distance_lookup_table[a][i] == distance_lookup_table[b][i]) @@ -213,9 +232,10 @@ int __node_distance(int a, int b) return distance; } +EXPORT_SYMBOL(__node_distance); static void initialize_distance_lookup_table(int nid, - const unsigned int *associativity) + const __be32 *associativity) { int i; @@ -223,29 +243,32 @@ static void initialize_distance_lookup_table(int nid, return; for (i = 0; i < distance_ref_points_depth; i++) { - distance_lookup_table[nid][i] = - associativity[distance_ref_points[i]]; + const __be32 *entry; + + entry = &associativity[be32_to_cpu(distance_ref_points[i])]; + distance_lookup_table[nid][i] = of_read_number(entry, 1); } } /* Returns nid in the range [0..MAX_NUMNODES-1], or -1 if no useful numa * info is found. */ -static int associativity_to_nid(const unsigned int *associativity) +static int associativity_to_nid(const __be32 *associativity) { int nid = -1; if (min_common_depth == -1) goto out; - if (associativity[0] >= min_common_depth) - nid = associativity[min_common_depth]; + if (of_read_number(associativity, 1) >= min_common_depth) + nid = of_read_number(&associativity[min_common_depth], 1); /* POWER4 LPAR uses 0xffff as invalid node */ if (nid == 0xffff || nid >= MAX_NUMNODES) nid = -1; - if (nid > 0 && associativity[0] >= distance_ref_points_depth) + if (nid > 0 && + of_read_number(associativity, 1) >= distance_ref_points_depth) initialize_distance_lookup_table(nid, associativity); out: @@ -258,7 +281,7 @@ out: static int of_node_to_nid_single(struct device_node *device) { int nid = -1; - const unsigned int *tmp; + const __be32 *tmp; tmp = of_get_associativity(device); if (tmp) @@ -291,9 +314,7 @@ EXPORT_SYMBOL_GPL(of_node_to_nid); static int __init find_min_common_depth(void) { int depth; - struct device_node *chosen; struct device_node *root; - const char *vec5; if (firmware_has_feature(FW_FEATURE_OPAL)) root = of_find_node_by_path("/ibm,opal"); @@ -325,28 +346,14 @@ static int __init find_min_common_depth(void) distance_ref_points_depth /= sizeof(int); -#define VEC5_AFFINITY_BYTE 5 -#define VEC5_AFFINITY 0x80 - - if (firmware_has_feature(FW_FEATURE_OPAL)) + if (firmware_has_feature(FW_FEATURE_OPAL) || + firmware_has_feature(FW_FEATURE_TYPE1_AFFINITY)) { + dbg("Using form 1 affinity\n"); form1_affinity = 1; - else { - chosen = of_find_node_by_path("/chosen"); - if (chosen) { - vec5 = of_get_property(chosen, - "ibm,architecture-vec-5", NULL); - if (vec5 && (vec5[VEC5_AFFINITY_BYTE] & - VEC5_AFFINITY)) { - dbg("Using form 1 affinity\n"); - form1_affinity = 1; - } - - of_node_put(chosen); - } } if (form1_affinity) { - depth = distance_ref_points[0]; + depth = of_read_number(distance_ref_points, 1); } else { if (distance_ref_points_depth < 2) { printk(KERN_WARNING "NUMA: " @@ -354,7 +361,7 @@ static int __init find_min_common_depth(void) goto err; } - depth = distance_ref_points[1]; + depth = of_read_number(&distance_ref_points[1], 1); } /* @@ -388,12 +395,12 @@ static void __init get_n_mem_cells(int *n_addr_cells, int *n_size_cells) of_node_put(memory); } -static unsigned long read_n_cells(int n, const unsigned int **buf) +static unsigned long read_n_cells(int n, const __be32 **buf) { unsigned long result = 0; while (n--) { - result = (result << 32) | **buf; + result = (result << 32) | of_read_number(*buf, 1); (*buf)++; } return result; @@ -403,17 +410,17 @@ static unsigned long read_n_cells(int n, const unsigned int **buf) * Read the next memblock list entry from the ibm,dynamic-memory property * and return the information in the provided of_drconf_cell structure. */ -static void read_drconf_cell(struct of_drconf_cell *drmem, const u32 **cellp) +static void read_drconf_cell(struct of_drconf_cell *drmem, const __be32 **cellp) { - const u32 *cp; + const __be32 *cp; drmem->base_addr = read_n_cells(n_mem_addr_cells, cellp); cp = *cellp; - drmem->drc_index = cp[0]; - drmem->reserved = cp[1]; - drmem->aa_index = cp[2]; - drmem->flags = cp[3]; + drmem->drc_index = of_read_number(cp, 1); + drmem->reserved = of_read_number(&cp[1], 1); + drmem->aa_index = of_read_number(&cp[2], 1); + drmem->flags = of_read_number(&cp[3], 1); *cellp = cp + 4; } @@ -425,16 +432,16 @@ static void read_drconf_cell(struct of_drconf_cell *drmem, const u32 **cellp) * list entries followed by N memblock list entries. Each memblock list entry * contains information as laid out in the of_drconf_cell struct above. */ -static int of_get_drconf_memory(struct device_node *memory, const u32 **dm) +static int of_get_drconf_memory(struct device_node *memory, const __be32 **dm) { - const u32 *prop; + const __be32 *prop; u32 len, entries; prop = of_get_property(memory, "ibm,dynamic-memory", &len); if (!prop || len < sizeof(unsigned int)) return 0; - entries = *prop++; + entries = of_read_number(prop++, 1); /* Now that we know the number of entries, revalidate the size * of the property read in to ensure we have everything @@ -452,7 +459,7 @@ static int of_get_drconf_memory(struct device_node *memory, const u32 **dm) */ static u64 of_get_lmb_size(struct device_node *memory) { - const u32 *prop; + const __be32 *prop; u32 len; prop = of_get_property(memory, "ibm,lmb-size", &len); @@ -465,7 +472,7 @@ static u64 of_get_lmb_size(struct device_node *memory) struct assoc_arrays { u32 n_arrays; u32 array_sz; - const u32 *arrays; + const __be32 *arrays; }; /* @@ -481,15 +488,15 @@ struct assoc_arrays { static int of_get_assoc_arrays(struct device_node *memory, struct assoc_arrays *aa) { - const u32 *prop; + const __be32 *prop; u32 len; prop = of_get_property(memory, "ibm,associativity-lookup-arrays", &len); if (!prop || len < 2 * sizeof(unsigned int)) return -1; - aa->n_arrays = *prop++; - aa->array_sz = *prop++; + aa->n_arrays = of_read_number(prop++, 1); + aa->array_sz = of_read_number(prop++, 1); /* Now that we know the number of arrays and size of each array, * revalidate the size of the property read in. @@ -516,7 +523,7 @@ static int of_drconf_to_nid_single(struct of_drconf_cell *drmem, !(drmem->flags & DRCONF_MEM_AI_INVALID) && drmem->aa_index < aa->n_arrays) { index = drmem->aa_index * aa->array_sz + min_common_depth - 1; - nid = aa->arrays[index]; + nid = of_read_number(&aa->arrays[index], 1); if (nid == 0xffff || nid >= MAX_NUMNODES) nid = default_nid; @@ -529,13 +536,26 @@ static int of_drconf_to_nid_single(struct of_drconf_cell *drmem, * Figure out to which domain a cpu belongs and stick it there. * Return the id of the domain used. */ -static int __cpuinit numa_setup_cpu(unsigned long lcpu) +static int numa_setup_cpu(unsigned long lcpu) { - int nid = 0; - struct device_node *cpu = of_get_cpu_node(lcpu, NULL); + int nid; + struct device_node *cpu; + + /* + * If a valid cpu-to-node mapping is already available, use it + * directly instead of querying the firmware, since it represents + * the most recent mapping notified to us by the platform (eg: VPHN). + */ + if ((nid = numa_cpu_lookup_table[lcpu]) >= 0) { + map_cpu_to_node(lcpu, nid); + return nid; + } + + cpu = of_get_cpu_node(lcpu, NULL); if (!cpu) { WARN_ON(1); + nid = 0; goto out; } @@ -551,17 +571,38 @@ out: return nid; } -static int __cpuinit cpu_numa_callback(struct notifier_block *nfb, - unsigned long action, +static void verify_cpu_node_mapping(int cpu, int node) +{ + int base, sibling, i; + + /* Verify that all the threads in the core belong to the same node */ + base = cpu_first_thread_sibling(cpu); + + for (i = 0; i < threads_per_core; i++) { + sibling = base + i; + + if (sibling == cpu || cpu_is_offline(sibling)) + continue; + + if (cpu_to_node(sibling) != node) { + WARN(1, "CPU thread siblings %d and %d don't belong" + " to the same node!\n", cpu, sibling); + break; + } + } +} + +static int cpu_numa_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { unsigned long lcpu = (unsigned long)hcpu; - int ret = NOTIFY_DONE; + int ret = NOTIFY_DONE, nid; switch (action) { case CPU_UP_PREPARE: case CPU_UP_PREPARE_FROZEN: - numa_setup_cpu(lcpu); + nid = numa_setup_cpu(lcpu); + verify_cpu_node_mapping((int)lcpu, nid); ret = NOTIFY_OK; break; #ifdef CONFIG_HOTPLUG_CPU @@ -608,7 +649,7 @@ static unsigned long __init numa_enforce_memory_limit(unsigned long start, * Reads the counter for a given entry in * linux,drconf-usable-memory property */ -static inline int __init read_usm_ranges(const u32 **usm) +static inline int __init read_usm_ranges(const __be32 **usm) { /* * For each lmb in ibm,dynamic-memory a corresponding @@ -625,7 +666,7 @@ static inline int __init read_usm_ranges(const u32 **usm) */ static void __init parse_drconf_memory(struct device_node *memory) { - const u32 *uninitialized_var(dm), *usm; + const __be32 *uninitialized_var(dm), *usm; unsigned int n, rc, ranges, is_kexec_kdump = 0; unsigned long lmb_size, base, size, sz; int nid; @@ -680,7 +721,8 @@ static void __init parse_drconf_memory(struct device_node *memory) node_set_online(nid); sz = numa_enforce_memory_limit(base, size); if (sz) - memblock_set_node(base, sz, nid); + memblock_set_node(base, sz, + &memblock.memory, nid); } while (--ranges); } } @@ -734,7 +776,7 @@ static int __init parse_numa_properties(void) unsigned long size; int nid; int ranges; - const unsigned int *memcell_buf; + const __be32 *memcell_buf; unsigned int len; memcell_buf = of_get_property(memory, @@ -770,7 +812,7 @@ new_range: continue; } - memblock_set_node(start, size, nid); + memblock_set_node(start, size, &memblock.memory, nid); if (--ranges) goto new_range; @@ -807,7 +849,8 @@ static void __init setup_nonnuma(void) fake_numa_create_new_node(end_pfn, &nid); memblock_set_node(PFN_PHYS(start_pfn), - PFN_PHYS(end_pfn - start_pfn), nid); + PFN_PHYS(end_pfn - start_pfn), + &memblock.memory, nid); node_set_online(nid); } } @@ -932,7 +975,7 @@ static void __init *careful_zallocation(int nid, unsigned long size, return ret; } -static struct notifier_block __cpuinitdata ppc64_numa_nb = { +static struct notifier_block ppc64_numa_nb = { .notifier_call = cpu_numa_callback, .priority = 1 /* Must run before sched domains notifier. */ }; @@ -948,8 +991,7 @@ static void __init mark_reserved_regions_for_nid(int nid) unsigned long start_pfn = physbase >> PAGE_SHIFT; unsigned long end_pfn = PFN_UP(physbase + size); struct node_active_region node_ar; - unsigned long node_end_pfn = node->node_start_pfn + - node->node_spanned_pages; + unsigned long node_end_pfn = pgdat_end_pfn(node); /* * Check to make sure that this memblock.reserved area is @@ -1078,6 +1120,7 @@ void __init do_init_bootmem(void) */ setup_node_to_cpumask_map(); + reset_numa_cpu_lookup_table(); register_cpu_notifier(&ppc64_numa_nb); cpu_numa_callback(&ppc64_numa_nb, CPU_UP_PREPARE, (void *)(unsigned long)boot_cpuid); @@ -1119,7 +1162,7 @@ early_param("numa", early_numa); static int hot_add_drconf_scn_to_nid(struct device_node *memory, unsigned long scn_addr) { - const u32 *dm; + const __be32 *dm; unsigned int drconf_cell_cnt, rc; unsigned long lmb_size; struct assoc_arrays aa; @@ -1164,7 +1207,7 @@ static int hot_add_drconf_scn_to_nid(struct device_node *memory, * represented in the device tree as a node (i.e. memory@XXXX) for * each memblock. */ -int hot_add_node_scn_to_nid(unsigned long scn_addr) +static int hot_add_node_scn_to_nid(unsigned long scn_addr) { struct device_node *memory; int nid = -1; @@ -1172,7 +1215,7 @@ int hot_add_node_scn_to_nid(unsigned long scn_addr) for_each_node_by_type(memory, "memory") { unsigned long start, size; int ranges; - const unsigned int *memcell_buf; + const __be32 *memcell_buf; unsigned int len; memcell_buf = of_get_property(memory, "reg", &len); @@ -1245,7 +1288,7 @@ static u64 hot_add_drconf_memory_max(void) struct device_node *memory = NULL; unsigned int drconf_cell_cnt = 0; u64 lmb_size = 0; - const u32 *dm = 0; + const __be32 *dm = NULL; memory = of_find_node_by_path("/ibm,dynamic-reconfiguration-memory"); if (memory) { @@ -1270,10 +1313,18 @@ u64 memory_hotplug_max(void) /* Virtual Processor Home Node (VPHN) support */ #ifdef CONFIG_PPC_SPLPAR +struct topology_update_data { + struct topology_update_data *next; + unsigned int cpu; + int old_nid; + int new_nid; +}; + static u8 vphn_cpu_change_counts[NR_CPUS][MAX_DISTANCE_REF_POINTS]; static cpumask_t cpu_associativity_changes_mask; static int vphn_enabled; -static void set_topology_timer(void); +static int prrn_enabled; +static void reset_topology_timer(void); /* * Store the current values of the associativity change counters in the @@ -1309,11 +1360,9 @@ static void setup_cpu_associativity_change_counters(void) */ static int update_cpu_associativity_changes_mask(void) { - int cpu, nr_cpus = 0; + int cpu; cpumask_t *changes = &cpu_associativity_changes_mask; - cpumask_clear(changes); - for_each_possible_cpu(cpu) { int i, changed = 0; u8 *counts = vphn_cpu_change_counts[cpu]; @@ -1326,12 +1375,12 @@ static int update_cpu_associativity_changes_mask(void) } } if (changed) { - cpumask_set_cpu(cpu, changes); - nr_cpus++; + cpumask_or(changes, changes, cpu_sibling_mask(cpu)); + cpu = cpu_last_thread_sibling(cpu); } } - return nr_cpus; + return cpumask_weight(changes); } /* @@ -1344,40 +1393,41 @@ static int update_cpu_associativity_changes_mask(void) * Convert the associativity domain numbers returned from the hypervisor * to the sequence they would appear in the ibm,associativity property. */ -static int vphn_unpack_associativity(const long *packed, unsigned int *unpacked) +static int vphn_unpack_associativity(const long *packed, __be32 *unpacked) { int i, nr_assoc_doms = 0; - const u16 *field = (const u16*) packed; + const __be16 *field = (const __be16 *) packed; #define VPHN_FIELD_UNUSED (0xffff) #define VPHN_FIELD_MSB (0x8000) #define VPHN_FIELD_MASK (~VPHN_FIELD_MSB) for (i = 1; i < VPHN_ASSOC_BUFSIZE; i++) { - if (*field == VPHN_FIELD_UNUSED) { + if (be16_to_cpup(field) == VPHN_FIELD_UNUSED) { /* All significant fields processed, and remaining * fields contain the reserved value of all 1's. * Just store them. */ - unpacked[i] = *((u32*)field); + unpacked[i] = *((__be32 *)field); field += 2; - } else if (*field & VPHN_FIELD_MSB) { + } else if (be16_to_cpup(field) & VPHN_FIELD_MSB) { /* Data is in the lower 15 bits of this field */ - unpacked[i] = *field & VPHN_FIELD_MASK; + unpacked[i] = cpu_to_be32( + be16_to_cpup(field) & VPHN_FIELD_MASK); field++; nr_assoc_doms++; } else { /* Data is in the lower 15 bits of this field * concatenated with the next 16 bit field */ - unpacked[i] = *((u32*)field); + unpacked[i] = *((__be32 *)field); field += 2; nr_assoc_doms++; } } /* The first cell contains the length of the property */ - unpacked[0] = nr_assoc_doms; + unpacked[0] = cpu_to_be32(nr_assoc_doms); return nr_assoc_doms; } @@ -1386,7 +1436,7 @@ static int vphn_unpack_associativity(const long *packed, unsigned int *unpacked) * Retrieve the new associativity information for a virtual processor's * home node. */ -static long hcall_vphn(unsigned long cpu, unsigned int *associativity) +static long hcall_vphn(unsigned long cpu, __be32 *associativity) { long rc; long retbuf[PLPAR_HCALL9_BUFSIZE] = {0}; @@ -1400,7 +1450,7 @@ static long hcall_vphn(unsigned long cpu, unsigned int *associativity) } static long vphn_get_associativity(unsigned long cpu, - unsigned int *associativity) + __be32 *associativity) { long rc; @@ -1423,40 +1473,162 @@ static long vphn_get_associativity(unsigned long cpu, } /* + * Update the CPU maps and sysfs entries for a single CPU when its NUMA + * characteristics change. This function doesn't perform any locking and is + * only safe to call from stop_machine(). + */ +static int update_cpu_topology(void *data) +{ + struct topology_update_data *update; + unsigned long cpu; + + if (!data) + return -EINVAL; + + cpu = smp_processor_id(); + + for (update = data; update; update = update->next) { + if (cpu != update->cpu) + continue; + + unmap_cpu_from_node(update->cpu); + map_cpu_to_node(update->cpu, update->new_nid); + vdso_getcpu_init(); + } + + return 0; +} + +static int update_lookup_table(void *data) +{ + struct topology_update_data *update; + + if (!data) + return -EINVAL; + + /* + * Upon topology update, the numa-cpu lookup table needs to be updated + * for all threads in the core, including offline CPUs, to ensure that + * future hotplug operations respect the cpu-to-node associativity + * properly. + */ + for (update = data; update; update = update->next) { + int nid, base, j; + + nid = update->new_nid; + base = cpu_first_thread_sibling(update->cpu); + + for (j = 0; j < threads_per_core; j++) { + update_numa_cpu_lookup_table(base + j, nid); + } + } + + return 0; +} + +/* * Update the node maps and sysfs entries for each cpu whose home node * has changed. Returns 1 when the topology has changed, and 0 otherwise. */ int arch_update_cpu_topology(void) { - int cpu, nid, old_nid, changed = 0; - unsigned int associativity[VPHN_ASSOC_BUFSIZE] = {0}; + unsigned int cpu, sibling, changed = 0; + struct topology_update_data *updates, *ud; + __be32 associativity[VPHN_ASSOC_BUFSIZE] = {0}; + cpumask_t updated_cpus; struct device *dev; + int weight, new_nid, i = 0; - for_each_cpu(cpu,&cpu_associativity_changes_mask) { - vphn_get_associativity(cpu, associativity); - nid = associativity_to_nid(associativity); + weight = cpumask_weight(&cpu_associativity_changes_mask); + if (!weight) + return 0; - if (nid < 0 || !node_online(nid)) - nid = first_online_node; + updates = kzalloc(weight * (sizeof(*updates)), GFP_KERNEL); + if (!updates) + return 0; - old_nid = numa_cpu_lookup_table[cpu]; + cpumask_clear(&updated_cpus); - /* Disable hotplug while we update the cpu - * masks and sysfs. + for_each_cpu(cpu, &cpu_associativity_changes_mask) { + /* + * If siblings aren't flagged for changes, updates list + * will be too short. Skip on this update and set for next + * update. */ - get_online_cpus(); - unregister_cpu_under_node(cpu, old_nid); - unmap_cpu_from_node(cpu); - map_cpu_to_node(cpu, nid); - register_cpu_under_node(cpu, nid); - put_online_cpus(); - - dev = get_cpu_device(cpu); + if (!cpumask_subset(cpu_sibling_mask(cpu), + &cpu_associativity_changes_mask)) { + pr_info("Sibling bits not set for associativity " + "change, cpu%d\n", cpu); + cpumask_or(&cpu_associativity_changes_mask, + &cpu_associativity_changes_mask, + cpu_sibling_mask(cpu)); + cpu = cpu_last_thread_sibling(cpu); + continue; + } + + /* Use associativity from first thread for all siblings */ + vphn_get_associativity(cpu, associativity); + new_nid = associativity_to_nid(associativity); + if (new_nid < 0 || !node_online(new_nid)) + new_nid = first_online_node; + + if (new_nid == numa_cpu_lookup_table[cpu]) { + cpumask_andnot(&cpu_associativity_changes_mask, + &cpu_associativity_changes_mask, + cpu_sibling_mask(cpu)); + cpu = cpu_last_thread_sibling(cpu); + continue; + } + + for_each_cpu(sibling, cpu_sibling_mask(cpu)) { + ud = &updates[i++]; + ud->cpu = sibling; + ud->new_nid = new_nid; + ud->old_nid = numa_cpu_lookup_table[sibling]; + cpumask_set_cpu(sibling, &updated_cpus); + if (i < weight) + ud->next = &updates[i]; + } + cpu = cpu_last_thread_sibling(cpu); + } + + /* + * In cases where we have nothing to update (because the updates list + * is too short or because the new topology is same as the old one), + * skip invoking update_cpu_topology() via stop-machine(). This is + * necessary (and not just a fast-path optimization) since stop-machine + * can end up electing a random CPU to run update_cpu_topology(), and + * thus trick us into setting up incorrect cpu-node mappings (since + * 'updates' is kzalloc()'ed). + * + * And for the similar reason, we will skip all the following updating. + */ + if (!cpumask_weight(&updated_cpus)) + goto out; + + stop_machine(update_cpu_topology, &updates[0], &updated_cpus); + + /* + * Update the numa-cpu lookup table with the new mappings, even for + * offline CPUs. It is best to perform this update from the stop- + * machine context. + */ + stop_machine(update_lookup_table, &updates[0], + cpumask_of(raw_smp_processor_id())); + + for (ud = &updates[0]; ud; ud = ud->next) { + unregister_cpu_under_node(ud->cpu, ud->old_nid); + register_cpu_under_node(ud->cpu, ud->new_nid); + + dev = get_cpu_device(ud->cpu); if (dev) kobject_uevent(&dev->kobj, KOBJ_CHANGE); + cpumask_clear_cpu(ud->cpu, &cpu_associativity_changes_mask); changed = 1; } +out: + kfree(updates); return changed; } @@ -1466,56 +1638,172 @@ static void topology_work_fn(struct work_struct *work) } static DECLARE_WORK(topology_work, topology_work_fn); -void topology_schedule_update(void) +static void topology_schedule_update(void) { schedule_work(&topology_work); } static void topology_timer_fn(unsigned long ignored) { - if (!vphn_enabled) - return; - if (update_cpu_associativity_changes_mask() > 0) + if (prrn_enabled && cpumask_weight(&cpu_associativity_changes_mask)) topology_schedule_update(); - set_topology_timer(); + else if (vphn_enabled) { + if (update_cpu_associativity_changes_mask() > 0) + topology_schedule_update(); + reset_topology_timer(); + } } static struct timer_list topology_timer = TIMER_INITIALIZER(topology_timer_fn, 0, 0); -static void set_topology_timer(void) +static void reset_topology_timer(void) { topology_timer.data = 0; topology_timer.expires = jiffies + 60 * HZ; - add_timer(&topology_timer); + mod_timer(&topology_timer, topology_timer.expires); } +#ifdef CONFIG_SMP + +static void stage_topology_update(int core_id) +{ + cpumask_or(&cpu_associativity_changes_mask, + &cpu_associativity_changes_mask, cpu_sibling_mask(core_id)); + reset_topology_timer(); +} + +static int dt_update_callback(struct notifier_block *nb, + unsigned long action, void *data) +{ + struct of_prop_reconfig *update; + int rc = NOTIFY_DONE; + + switch (action) { + case OF_RECONFIG_UPDATE_PROPERTY: + update = (struct of_prop_reconfig *)data; + if (!of_prop_cmp(update->dn->type, "cpu") && + !of_prop_cmp(update->prop->name, "ibm,associativity")) { + u32 core_id; + of_property_read_u32(update->dn, "reg", &core_id); + stage_topology_update(core_id); + rc = NOTIFY_OK; + } + break; + } + + return rc; +} + +static struct notifier_block dt_update_nb = { + .notifier_call = dt_update_callback, +}; + +#endif + /* - * Start polling for VPHN associativity changes. + * Start polling for associativity changes. */ int start_topology_update(void) { int rc = 0; - /* Disabled until races with load balancing are fixed */ - if (0 && firmware_has_feature(FW_FEATURE_VPHN) && - get_lppaca()->shared_proc) { - vphn_enabled = 1; - setup_cpu_associativity_change_counters(); - init_timer_deferrable(&topology_timer); - set_topology_timer(); - rc = 1; + if (firmware_has_feature(FW_FEATURE_PRRN)) { + if (!prrn_enabled) { + prrn_enabled = 1; + vphn_enabled = 0; +#ifdef CONFIG_SMP + rc = of_reconfig_notifier_register(&dt_update_nb); +#endif + } + } else if (firmware_has_feature(FW_FEATURE_VPHN) && + lppaca_shared_proc(get_lppaca())) { + if (!vphn_enabled) { + prrn_enabled = 0; + vphn_enabled = 1; + setup_cpu_associativity_change_counters(); + init_timer_deferrable(&topology_timer); + reset_topology_timer(); + } } return rc; } -__initcall(start_topology_update); /* * Disable polling for VPHN associativity changes. */ int stop_topology_update(void) { - vphn_enabled = 0; - return del_timer_sync(&topology_timer); + int rc = 0; + + if (prrn_enabled) { + prrn_enabled = 0; +#ifdef CONFIG_SMP + rc = of_reconfig_notifier_unregister(&dt_update_nb); +#endif + } else if (vphn_enabled) { + vphn_enabled = 0; + rc = del_timer_sync(&topology_timer); + } + + return rc; +} + +int prrn_is_enabled(void) +{ + return prrn_enabled; +} + +static int topology_read(struct seq_file *file, void *v) +{ + if (vphn_enabled || prrn_enabled) + seq_puts(file, "on\n"); + else + seq_puts(file, "off\n"); + + return 0; +} + +static int topology_open(struct inode *inode, struct file *file) +{ + return single_open(file, topology_read, NULL); +} + +static ssize_t topology_write(struct file *file, const char __user *buf, + size_t count, loff_t *off) +{ + char kbuf[4]; /* "on" or "off" plus null. */ + int read_len; + + read_len = count < 3 ? count : 3; + if (copy_from_user(kbuf, buf, read_len)) + return -EINVAL; + + kbuf[read_len] = '\0'; + + if (!strncmp(kbuf, "on", 2)) + start_topology_update(); + else if (!strncmp(kbuf, "off", 3)) + stop_topology_update(); + else + return -EINVAL; + + return count; +} + +static const struct file_operations topology_ops = { + .read = seq_read, + .write = topology_write, + .open = topology_open, + .release = single_release +}; + +static int topology_update_init(void) +{ + start_topology_update(); + proc_create("powerpc/topology_updates", 0644, NULL, &topology_ops); + + return 0; } +device_initcall(topology_update_init); #endif /* CONFIG_PPC_SPLPAR */ diff --git a/arch/powerpc/mm/pgtable.c b/arch/powerpc/mm/pgtable.c index 214130a4edc..c695943a513 100644 --- a/arch/powerpc/mm/pgtable.c +++ b/arch/powerpc/mm/pgtable.c @@ -24,7 +24,6 @@ #include <linux/kernel.h> #include <linux/gfp.h> #include <linux/mm.h> -#include <linux/init.h> #include <linux/percpu.h> #include <linux/hardirq.h> #include <linux/hugetlb.h> @@ -32,8 +31,6 @@ #include <asm/tlbflush.h> #include <asm/tlb.h> -#include "mmu_decl.h" - static inline int is_exec_fault(void) { return current->thread.regs && TRAP(current->thread.regs) == 0x400; @@ -72,7 +69,7 @@ struct page * maybe_pte_to_page(pte_t pte) * support falls into the same category. */ -static pte_t set_pte_filter(pte_t pte, unsigned long addr) +static pte_t set_pte_filter(pte_t pte) { pte = __pte(pte_val(pte) & ~_PAGE_HPTEFLAGS); if (pte_looks_normal(pte) && !(cpu_has_feature(CPU_FTR_COHERENT_ICACHE) || @@ -81,17 +78,6 @@ static pte_t set_pte_filter(pte_t pte, unsigned long addr) if (!pg) return pte; if (!test_bit(PG_arch_1, &pg->flags)) { -#ifdef CONFIG_8xx - /* On 8xx, cache control instructions (particularly - * "dcbst" from flush_dcache_icache) fault as write - * operation if there is an unpopulated TLB entry - * for the address in question. To workaround that, - * we invalidate the TLB here, thus avoiding dcbst - * misbehaviour. - */ - /* 8xx doesn't care about PID, size or ind args */ - _tlbil_va(addr, 0, 0, 0); -#endif /* CONFIG_8xx */ flush_dcache_icache_page(pg); set_bit(PG_arch_1, &pg->flags); } @@ -111,7 +97,7 @@ static pte_t set_access_flags_filter(pte_t pte, struct vm_area_struct *vma, * as we don't have two bits to spare for _PAGE_EXEC and _PAGE_HWEXEC so * instead we "filter out" the exec permission for non clean pages. */ -static pte_t set_pte_filter(pte_t pte, unsigned long addr) +static pte_t set_pte_filter(pte_t pte) { struct page *pg; @@ -187,13 +173,13 @@ void set_pte_at(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t pte) { #ifdef CONFIG_DEBUG_VM - WARN_ON(pte_present(*ptep)); + WARN_ON(pte_val(*ptep) & _PAGE_PRESENT); #endif /* Note: mm->context.id might not yet have been assigned as * this context might not have been activated yet when this * is called. */ - pte = set_pte_filter(pte, addr); + pte = set_pte_filter(pte); /* Perform the setting of the PTE */ __set_pte_at(mm, addr, ptep, pte, 0); @@ -235,6 +221,14 @@ void assert_pte_locked(struct mm_struct *mm, unsigned long addr) pud = pud_offset(pgd, addr); BUG_ON(pud_none(*pud)); pmd = pmd_offset(pud, addr); + /* + * khugepaged to collapse normal pages to hugepage, first set + * pmd to none to force page fault/gup to take mmap_sem. After + * pmd is set to none, we do a pte_clear which does this assertion + * so if we find pmd none, return. + */ + if (pmd_none(*pmd)) + return; BUG_ON(!pmd_present(*pmd)); assert_spin_locked(pte_lockptr(mm, pmd)); } diff --git a/arch/powerpc/mm/pgtable_32.c b/arch/powerpc/mm/pgtable_32.c index 6c856fb8c15..343a87fa78b 100644 --- a/arch/powerpc/mm/pgtable_32.c +++ b/arch/powerpc/mm/pgtable_32.c @@ -121,7 +121,10 @@ pgtable_t pte_alloc_one(struct mm_struct *mm, unsigned long address) ptepage = alloc_pages(flags, 0); if (!ptepage) return NULL; - pgtable_page_ctor(ptepage); + if (!pgtable_page_ctor(ptepage)) { + __free_page(ptepage); + return NULL; + } return ptepage; } @@ -296,6 +299,7 @@ int map_page(unsigned long va, phys_addr_t pa, int flags) set_pte_at(&init_mm, va, pg, pfn_pte(pa >> PAGE_SHIFT, __pgprot(flags))); } + smp_wmb(); return err; } diff --git a/arch/powerpc/mm/pgtable_64.c b/arch/powerpc/mm/pgtable_64.c index e212a271c7a..f6ce1f111f5 100644 --- a/arch/powerpc/mm/pgtable_64.c +++ b/arch/powerpc/mm/pgtable_64.c @@ -33,7 +33,6 @@ #include <linux/swap.h> #include <linux/stddef.h> #include <linux/vmalloc.h> -#include <linux/init.h> #include <linux/bootmem.h> #include <linux/memblock.h> #include <linux/slab.h> @@ -61,7 +60,7 @@ #endif #ifdef CONFIG_PPC_STD_MMU_64 -#if TASK_SIZE_USER64 > (1UL << (USER_ESID_BITS + SID_SHIFT)) +#if TASK_SIZE_USER64 > (1UL << (ESID_BITS + SID_SHIFT)) #error TASK_SIZE_USER64 exceeds user VSID range #endif #endif @@ -153,6 +152,18 @@ int map_kernel_page(unsigned long ea, unsigned long pa, int flags) } #endif /* !CONFIG_PPC_MMU_NOHASH */ } + +#ifdef CONFIG_PPC_BOOK3E_64 + /* + * With hardware tablewalk, a sync is needed to ensure that + * subsequent accesses see the PTE we just wrote. Unlike userspace + * mappings, we can't tolerate spurious faults, so make sure + * the new PTE will be seen the first time. + */ + mb(); +#else + smp_wmb(); +#endif return 0; } @@ -337,3 +348,543 @@ EXPORT_SYMBOL(__ioremap_at); EXPORT_SYMBOL(iounmap); EXPORT_SYMBOL(__iounmap); EXPORT_SYMBOL(__iounmap_at); + +/* + * For hugepage we have pfn in the pmd, we use PTE_RPN_SHIFT bits for flags + * For PTE page, we have a PTE_FRAG_SIZE (4K) aligned virtual address. + */ +struct page *pmd_page(pmd_t pmd) +{ +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + if (pmd_trans_huge(pmd)) + return pfn_to_page(pmd_pfn(pmd)); +#endif + return virt_to_page(pmd_page_vaddr(pmd)); +} + +#ifdef CONFIG_PPC_64K_PAGES +static pte_t *get_from_cache(struct mm_struct *mm) +{ + void *pte_frag, *ret; + + spin_lock(&mm->page_table_lock); + ret = mm->context.pte_frag; + if (ret) { + pte_frag = ret + PTE_FRAG_SIZE; + /* + * If we have taken up all the fragments mark PTE page NULL + */ + if (((unsigned long)pte_frag & ~PAGE_MASK) == 0) + pte_frag = NULL; + mm->context.pte_frag = pte_frag; + } + spin_unlock(&mm->page_table_lock); + return (pte_t *)ret; +} + +static pte_t *__alloc_for_cache(struct mm_struct *mm, int kernel) +{ + void *ret = NULL; + struct page *page = alloc_page(GFP_KERNEL | __GFP_NOTRACK | + __GFP_REPEAT | __GFP_ZERO); + if (!page) + return NULL; + if (!kernel && !pgtable_page_ctor(page)) { + __free_page(page); + return NULL; + } + + ret = page_address(page); + spin_lock(&mm->page_table_lock); + /* + * If we find pgtable_page set, we return + * the allocated page with single fragement + * count. + */ + if (likely(!mm->context.pte_frag)) { + atomic_set(&page->_count, PTE_FRAG_NR); + mm->context.pte_frag = ret + PTE_FRAG_SIZE; + } + spin_unlock(&mm->page_table_lock); + + return (pte_t *)ret; +} + +pte_t *page_table_alloc(struct mm_struct *mm, unsigned long vmaddr, int kernel) +{ + pte_t *pte; + + pte = get_from_cache(mm); + if (pte) + return pte; + + return __alloc_for_cache(mm, kernel); +} + +void page_table_free(struct mm_struct *mm, unsigned long *table, int kernel) +{ + struct page *page = virt_to_page(table); + if (put_page_testzero(page)) { + if (!kernel) + pgtable_page_dtor(page); + free_hot_cold_page(page, 0); + } +} + +#ifdef CONFIG_SMP +static void page_table_free_rcu(void *table) +{ + struct page *page = virt_to_page(table); + if (put_page_testzero(page)) { + pgtable_page_dtor(page); + free_hot_cold_page(page, 0); + } +} + +void pgtable_free_tlb(struct mmu_gather *tlb, void *table, int shift) +{ + unsigned long pgf = (unsigned long)table; + + BUG_ON(shift > MAX_PGTABLE_INDEX_SIZE); + pgf |= shift; + tlb_remove_table(tlb, (void *)pgf); +} + +void __tlb_remove_table(void *_table) +{ + void *table = (void *)((unsigned long)_table & ~MAX_PGTABLE_INDEX_SIZE); + unsigned shift = (unsigned long)_table & MAX_PGTABLE_INDEX_SIZE; + + if (!shift) + /* PTE page needs special handling */ + page_table_free_rcu(table); + else { + BUG_ON(shift > MAX_PGTABLE_INDEX_SIZE); + kmem_cache_free(PGT_CACHE(shift), table); + } +} +#else +void pgtable_free_tlb(struct mmu_gather *tlb, void *table, int shift) +{ + if (!shift) { + /* PTE page needs special handling */ + struct page *page = virt_to_page(table); + if (put_page_testzero(page)) { + pgtable_page_dtor(page); + free_hot_cold_page(page, 0); + } + } else { + BUG_ON(shift > MAX_PGTABLE_INDEX_SIZE); + kmem_cache_free(PGT_CACHE(shift), table); + } +} +#endif +#endif /* CONFIG_PPC_64K_PAGES */ + +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + +/* + * This is called when relaxing access to a hugepage. It's also called in the page + * fault path when we don't hit any of the major fault cases, ie, a minor + * update of _PAGE_ACCESSED, _PAGE_DIRTY, etc... The generic code will have + * handled those two for us, we additionally deal with missing execute + * permission here on some processors + */ +int pmdp_set_access_flags(struct vm_area_struct *vma, unsigned long address, + pmd_t *pmdp, pmd_t entry, int dirty) +{ + int changed; +#ifdef CONFIG_DEBUG_VM + WARN_ON(!pmd_trans_huge(*pmdp)); + assert_spin_locked(&vma->vm_mm->page_table_lock); +#endif + changed = !pmd_same(*(pmdp), entry); + if (changed) { + __ptep_set_access_flags(pmdp_ptep(pmdp), pmd_pte(entry)); + /* + * Since we are not supporting SW TLB systems, we don't + * have any thing similar to flush_tlb_page_nohash() + */ + } + return changed; +} + +unsigned long pmd_hugepage_update(struct mm_struct *mm, unsigned long addr, + pmd_t *pmdp, unsigned long clr, + unsigned long set) +{ + + unsigned long old, tmp; + +#ifdef CONFIG_DEBUG_VM + WARN_ON(!pmd_trans_huge(*pmdp)); + assert_spin_locked(&mm->page_table_lock); +#endif + +#ifdef PTE_ATOMIC_UPDATES + __asm__ __volatile__( + "1: ldarx %0,0,%3\n\ + andi. %1,%0,%6\n\ + bne- 1b \n\ + andc %1,%0,%4 \n\ + or %1,%1,%7\n\ + stdcx. %1,0,%3 \n\ + bne- 1b" + : "=&r" (old), "=&r" (tmp), "=m" (*pmdp) + : "r" (pmdp), "r" (clr), "m" (*pmdp), "i" (_PAGE_BUSY), "r" (set) + : "cc" ); +#else + old = pmd_val(*pmdp); + *pmdp = __pmd((old & ~clr) | set); +#endif + if (old & _PAGE_HASHPTE) + hpte_do_hugepage_flush(mm, addr, pmdp); + return old; +} + +pmd_t pmdp_clear_flush(struct vm_area_struct *vma, unsigned long address, + pmd_t *pmdp) +{ + pmd_t pmd; + + VM_BUG_ON(address & ~HPAGE_PMD_MASK); + if (pmd_trans_huge(*pmdp)) { + pmd = pmdp_get_and_clear(vma->vm_mm, address, pmdp); + } else { + /* + * khugepaged calls this for normal pmd + */ + pmd = *pmdp; + pmd_clear(pmdp); + /* + * Wait for all pending hash_page to finish. This is needed + * in case of subpage collapse. When we collapse normal pages + * to hugepage, we first clear the pmd, then invalidate all + * the PTE entries. The assumption here is that any low level + * page fault will see a none pmd and take the slow path that + * will wait on mmap_sem. But we could very well be in a + * hash_page with local ptep pointer value. Such a hash page + * can result in adding new HPTE entries for normal subpages. + * That means we could be modifying the page content as we + * copy them to a huge page. So wait for parallel hash_page + * to finish before invalidating HPTE entries. We can do this + * by sending an IPI to all the cpus and executing a dummy + * function there. + */ + kick_all_cpus_sync(); + /* + * Now invalidate the hpte entries in the range + * covered by pmd. This make sure we take a + * fault and will find the pmd as none, which will + * result in a major fault which takes mmap_sem and + * hence wait for collapse to complete. Without this + * the __collapse_huge_page_copy can result in copying + * the old content. + */ + flush_tlb_pmd_range(vma->vm_mm, &pmd, address); + } + return pmd; +} + +int pmdp_test_and_clear_young(struct vm_area_struct *vma, + unsigned long address, pmd_t *pmdp) +{ + return __pmdp_test_and_clear_young(vma->vm_mm, address, pmdp); +} + +/* + * We currently remove entries from the hashtable regardless of whether + * the entry was young or dirty. The generic routines only flush if the + * entry was young or dirty which is not good enough. + * + * We should be more intelligent about this but for the moment we override + * these functions and force a tlb flush unconditionally + */ +int pmdp_clear_flush_young(struct vm_area_struct *vma, + unsigned long address, pmd_t *pmdp) +{ + return __pmdp_test_and_clear_young(vma->vm_mm, address, pmdp); +} + +/* + * We mark the pmd splitting and invalidate all the hpte + * entries for this hugepage. + */ +void pmdp_splitting_flush(struct vm_area_struct *vma, + unsigned long address, pmd_t *pmdp) +{ + unsigned long old, tmp; + + VM_BUG_ON(address & ~HPAGE_PMD_MASK); + +#ifdef CONFIG_DEBUG_VM + WARN_ON(!pmd_trans_huge(*pmdp)); + assert_spin_locked(&vma->vm_mm->page_table_lock); +#endif + +#ifdef PTE_ATOMIC_UPDATES + + __asm__ __volatile__( + "1: ldarx %0,0,%3\n\ + andi. %1,%0,%6\n\ + bne- 1b \n\ + ori %1,%0,%4 \n\ + stdcx. %1,0,%3 \n\ + bne- 1b" + : "=&r" (old), "=&r" (tmp), "=m" (*pmdp) + : "r" (pmdp), "i" (_PAGE_SPLITTING), "m" (*pmdp), "i" (_PAGE_BUSY) + : "cc" ); +#else + old = pmd_val(*pmdp); + *pmdp = __pmd(old | _PAGE_SPLITTING); +#endif + /* + * If we didn't had the splitting flag set, go and flush the + * HPTE entries. + */ + if (!(old & _PAGE_SPLITTING)) { + /* We need to flush the hpte */ + if (old & _PAGE_HASHPTE) + hpte_do_hugepage_flush(vma->vm_mm, address, pmdp); + } + /* + * This ensures that generic code that rely on IRQ disabling + * to prevent a parallel THP split work as expected. + */ + kick_all_cpus_sync(); +} + +/* + * We want to put the pgtable in pmd and use pgtable for tracking + * the base page size hptes + */ +void pgtable_trans_huge_deposit(struct mm_struct *mm, pmd_t *pmdp, + pgtable_t pgtable) +{ + pgtable_t *pgtable_slot; + assert_spin_locked(&mm->page_table_lock); + /* + * we store the pgtable in the second half of PMD + */ + pgtable_slot = (pgtable_t *)pmdp + PTRS_PER_PMD; + *pgtable_slot = pgtable; + /* + * expose the deposited pgtable to other cpus. + * before we set the hugepage PTE at pmd level + * hash fault code looks at the deposted pgtable + * to store hash index values. + */ + smp_wmb(); +} + +pgtable_t pgtable_trans_huge_withdraw(struct mm_struct *mm, pmd_t *pmdp) +{ + pgtable_t pgtable; + pgtable_t *pgtable_slot; + + assert_spin_locked(&mm->page_table_lock); + pgtable_slot = (pgtable_t *)pmdp + PTRS_PER_PMD; + pgtable = *pgtable_slot; + /* + * Once we withdraw, mark the entry NULL. + */ + *pgtable_slot = NULL; + /* + * We store HPTE information in the deposited PTE fragment. + * zero out the content on withdraw. + */ + memset(pgtable, 0, PTE_FRAG_SIZE); + return pgtable; +} + +/* + * set a new huge pmd. We should not be called for updating + * an existing pmd entry. That should go via pmd_hugepage_update. + */ +void set_pmd_at(struct mm_struct *mm, unsigned long addr, + pmd_t *pmdp, pmd_t pmd) +{ +#ifdef CONFIG_DEBUG_VM + WARN_ON(pmd_val(*pmdp) & _PAGE_PRESENT); + assert_spin_locked(&mm->page_table_lock); + WARN_ON(!pmd_trans_huge(pmd)); +#endif + return set_pte_at(mm, addr, pmdp_ptep(pmdp), pmd_pte(pmd)); +} + +void pmdp_invalidate(struct vm_area_struct *vma, unsigned long address, + pmd_t *pmdp) +{ + pmd_hugepage_update(vma->vm_mm, address, pmdp, _PAGE_PRESENT, 0); +} + +/* + * A linux hugepage PMD was changed and the corresponding hash table entries + * neesd to be flushed. + */ +void hpte_do_hugepage_flush(struct mm_struct *mm, unsigned long addr, + pmd_t *pmdp) +{ + int ssize, i; + unsigned long s_addr; + int max_hpte_count; + unsigned int psize, valid; + unsigned char *hpte_slot_array; + unsigned long hidx, vpn, vsid, hash, shift, slot; + + /* + * Flush all the hptes mapping this hugepage + */ + s_addr = addr & HPAGE_PMD_MASK; + hpte_slot_array = get_hpte_slot_array(pmdp); + /* + * IF we try to do a HUGE PTE update after a withdraw is done. + * we will find the below NULL. This happens when we do + * split_huge_page_pmd + */ + if (!hpte_slot_array) + return; + + /* get the base page size */ + psize = get_slice_psize(mm, s_addr); + + if (ppc_md.hugepage_invalidate) + return ppc_md.hugepage_invalidate(mm, hpte_slot_array, + s_addr, psize); + /* + * No bluk hpte removal support, invalidate each entry + */ + shift = mmu_psize_defs[psize].shift; + max_hpte_count = HPAGE_PMD_SIZE >> shift; + for (i = 0; i < max_hpte_count; i++) { + /* + * 8 bits per each hpte entries + * 000| [ secondary group (one bit) | hidx (3 bits) | valid bit] + */ + valid = hpte_valid(hpte_slot_array, i); + if (!valid) + continue; + hidx = hpte_hash_index(hpte_slot_array, i); + + /* get the vpn */ + addr = s_addr + (i * (1ul << shift)); + if (!is_kernel_addr(addr)) { + ssize = user_segment_size(addr); + vsid = get_vsid(mm->context.id, addr, ssize); + WARN_ON(vsid == 0); + } else { + vsid = get_kernel_vsid(addr, mmu_kernel_ssize); + ssize = mmu_kernel_ssize; + } + + vpn = hpt_vpn(addr, vsid, ssize); + hash = hpt_hash(vpn, shift, ssize); + if (hidx & _PTEIDX_SECONDARY) + hash = ~hash; + + slot = (hash & htab_hash_mask) * HPTES_PER_GROUP; + slot += hidx & _PTEIDX_GROUP_IX; + ppc_md.hpte_invalidate(slot, vpn, psize, + MMU_PAGE_16M, ssize, 0); + } +} + +static pmd_t pmd_set_protbits(pmd_t pmd, pgprot_t pgprot) +{ + pmd_val(pmd) |= pgprot_val(pgprot); + return pmd; +} + +pmd_t pfn_pmd(unsigned long pfn, pgprot_t pgprot) +{ + pmd_t pmd; + /* + * For a valid pte, we would have _PAGE_PRESENT or _PAGE_FILE always + * set. We use this to check THP page at pmd level. + * leaf pte for huge page, bottom two bits != 00 + */ + pmd_val(pmd) = pfn << PTE_RPN_SHIFT; + pmd_val(pmd) |= _PAGE_THP_HUGE; + pmd = pmd_set_protbits(pmd, pgprot); + return pmd; +} + +pmd_t mk_pmd(struct page *page, pgprot_t pgprot) +{ + return pfn_pmd(page_to_pfn(page), pgprot); +} + +pmd_t pmd_modify(pmd_t pmd, pgprot_t newprot) +{ + + pmd_val(pmd) &= _HPAGE_CHG_MASK; + pmd = pmd_set_protbits(pmd, newprot); + return pmd; +} + +/* + * This is called at the end of handling a user page fault, when the + * fault has been handled by updating a HUGE PMD entry in the linux page tables. + * We use it to preload an HPTE into the hash table corresponding to + * the updated linux HUGE PMD entry. + */ +void update_mmu_cache_pmd(struct vm_area_struct *vma, unsigned long addr, + pmd_t *pmd) +{ + return; +} + +pmd_t pmdp_get_and_clear(struct mm_struct *mm, + unsigned long addr, pmd_t *pmdp) +{ + pmd_t old_pmd; + pgtable_t pgtable; + unsigned long old; + pgtable_t *pgtable_slot; + + old = pmd_hugepage_update(mm, addr, pmdp, ~0UL, 0); + old_pmd = __pmd(old); + /* + * We have pmd == none and we are holding page_table_lock. + * So we can safely go and clear the pgtable hash + * index info. + */ + pgtable_slot = (pgtable_t *)pmdp + PTRS_PER_PMD; + pgtable = *pgtable_slot; + /* + * Let's zero out old valid and hash index details + * hash fault look at them. + */ + memset(pgtable, 0, PTE_FRAG_SIZE); + return old_pmd; +} + +int has_transparent_hugepage(void) +{ + if (!mmu_has_feature(MMU_FTR_16M_PAGE)) + return 0; + /* + * We support THP only if PMD_SIZE is 16MB. + */ + if (mmu_psize_defs[MMU_PAGE_16M].shift != PMD_SHIFT) + return 0; + /* + * We need to make sure that we support 16MB hugepage in a segement + * with base page size 64K or 4K. We only enable THP with a PAGE_SIZE + * of 64K. + */ + /* + * If we have 64K HPTE, we will be using that by default + */ + if (mmu_psize_defs[MMU_PAGE_64K].shift && + (mmu_psize_defs[MMU_PAGE_64K].penc[MMU_PAGE_16M] == -1)) + return 0; + /* + * Ok we only have 4K HPTE + */ + if (mmu_psize_defs[MMU_PAGE_4K].penc[MMU_PAGE_16M] == -1) + return 0; + + return 1; +} +#endif /* CONFIG_TRANSPARENT_HUGEPAGE */ diff --git a/arch/powerpc/mm/slb.c b/arch/powerpc/mm/slb.c index a538c80db2d..0399a670295 100644 --- a/arch/powerpc/mm/slb.c +++ b/arch/powerpc/mm/slb.c @@ -66,8 +66,10 @@ static inline void slb_shadow_update(unsigned long ea, int ssize, * we only update the current CPU's SLB shadow buffer. */ get_slb_shadow()->save_area[entry].esid = 0; - get_slb_shadow()->save_area[entry].vsid = mk_vsid_data(ea, ssize, flags); - get_slb_shadow()->save_area[entry].esid = mk_esid_data(ea, ssize, entry); + get_slb_shadow()->save_area[entry].vsid = + cpu_to_be64(mk_vsid_data(ea, ssize, flags)); + get_slb_shadow()->save_area[entry].esid = + cpu_to_be64(mk_esid_data(ea, ssize, entry)); } static inline void slb_shadow_clear(unsigned long entry) @@ -95,7 +97,7 @@ static inline void create_shadowed_slbe(unsigned long ea, int ssize, static void __slb_flush_and_rebolt(void) { /* If you change this make sure you change SLB_NUM_BOLTED - * appropriately too. */ + * and PR KVM appropriately too. */ unsigned long linear_llp, vmalloc_llp, lflags, vflags; unsigned long ksp_esid_data, ksp_vsid_data; @@ -112,7 +114,8 @@ static void __slb_flush_and_rebolt(void) } else { /* Update stack entry; others don't change */ slb_shadow_update(get_paca()->kstack, mmu_kernel_ssize, lflags, 2); - ksp_vsid_data = get_slb_shadow()->save_area[2].vsid; + ksp_vsid_data = + be64_to_cpu(get_slb_shadow()->save_area[2].vsid); } /* We need to do this all in asm, so we're sure we don't touch @@ -253,10 +256,14 @@ static inline void patch_slb_encoding(unsigned int *insn_addr, patch_instruction(insn_addr, insn); } +extern u32 slb_compare_rr_to_size[]; +extern u32 slb_miss_kernel_load_linear[]; +extern u32 slb_miss_kernel_load_io[]; +extern u32 slb_compare_rr_to_size[]; +extern u32 slb_miss_kernel_load_vmemmap[]; + void slb_set_size(u16 size) { - extern unsigned int *slb_compare_rr_to_size; - if (mmu_slb_size == size) return; @@ -269,11 +276,7 @@ void slb_initialize(void) unsigned long linear_llp, vmalloc_llp, io_llp; unsigned long lflags, vflags; static int slb_encoding_inited; - extern unsigned int *slb_miss_kernel_load_linear; - extern unsigned int *slb_miss_kernel_load_io; - extern unsigned int *slb_compare_rr_to_size; #ifdef CONFIG_SPARSEMEM_VMEMMAP - extern unsigned int *slb_miss_kernel_load_vmemmap; unsigned long vmemmap_llp; #endif diff --git a/arch/powerpc/mm/slb_low.S b/arch/powerpc/mm/slb_low.S index 1a16ca22775..736d18b3cef 100644 --- a/arch/powerpc/mm/slb_low.S +++ b/arch/powerpc/mm/slb_low.S @@ -31,10 +31,15 @@ * No other registers are examined or changed. */ _GLOBAL(slb_allocate_realmode) - /* r3 = faulting address */ + /* + * check for bad kernel/user address + * (ea & ~REGION_MASK) >= PGTABLE_RANGE + */ + rldicr. r9,r3,4,(63 - PGTABLE_EADDR_SIZE - 4) + bne- 8f srdi r9,r3,60 /* get region */ - srdi r10,r3,28 /* get esid */ + srdi r10,r3,SID_SHIFT /* get esid */ cmpldi cr7,r9,0xc /* cmp PAGE_OFFSET for later use */ /* r3 = address, r10 = esid, cr7 = <> PAGE_OFFSET */ @@ -54,14 +59,17 @@ _GLOBAL(slb_allocate_realmode) /* Linear mapping encoding bits, the "li" instruction below will * be patched by the kernel at boot */ -_GLOBAL(slb_miss_kernel_load_linear) +.globl slb_miss_kernel_load_linear +slb_miss_kernel_load_linear: li r11,0 - li r9,0x1 /* - * for 1T we shift 12 bits more. slb_finish_load_1T will do - * the necessary adjustment + * context = (MAX_USER_CONTEXT) + ((ea >> 60) - 0xc) + 1 + * r9 = region id. */ - rldimi r10,r9,(CONTEXT_BITS + USER_ESID_BITS),0 + addis r9,r9,(MAX_USER_CONTEXT - 0xc + 1)@ha + addi r9,r9,(MAX_USER_CONTEXT - 0xc + 1)@l + + BEGIN_FTR_SECTION b slb_finish_load END_MMU_FTR_SECTION_IFCLR(MMU_FTR_1T_SEGMENT) @@ -72,7 +80,8 @@ END_MMU_FTR_SECTION_IFCLR(MMU_FTR_1T_SEGMENT) /* Check virtual memmap region. To be patches at kernel boot */ cmpldi cr0,r9,0xf bne 1f -_GLOBAL(slb_miss_kernel_load_vmemmap) +.globl slb_miss_kernel_load_vmemmap +slb_miss_kernel_load_vmemmap: li r11,0 b 6f 1: @@ -88,27 +97,23 @@ _GLOBAL(slb_miss_kernel_load_vmemmap) b 6f 5: /* IO mapping */ - _GLOBAL(slb_miss_kernel_load_io) +.globl slb_miss_kernel_load_io +slb_miss_kernel_load_io: li r11,0 6: - li r9,0x1 /* - * for 1T we shift 12 bits more. slb_finish_load_1T will do - * the necessary adjustment + * context = (MAX_USER_CONTEXT) + ((ea >> 60) - 0xc) + 1 + * r9 = region id. */ - rldimi r10,r9,(CONTEXT_BITS + USER_ESID_BITS),0 + addis r9,r9,(MAX_USER_CONTEXT - 0xc + 1)@ha + addi r9,r9,(MAX_USER_CONTEXT - 0xc + 1)@l + BEGIN_FTR_SECTION b slb_finish_load END_MMU_FTR_SECTION_IFCLR(MMU_FTR_1T_SEGMENT) b slb_finish_load_1T -0: /* user address: proto-VSID = context << 15 | ESID. First check - * if the address is within the boundaries of the user region - */ - srdi. r9,r10,USER_ESID_BITS - bne- 8f /* invalid ea bits set */ - - +0: /* when using slices, we extract the psize off the slice bitmaps * and then we need to get the sllp encoding off the mmu_psize_defs * array. @@ -164,15 +169,13 @@ END_MMU_FTR_SECTION_IFCLR(MMU_FTR_1T_SEGMENT) ld r9,PACACONTEXTID(r13) BEGIN_FTR_SECTION cmpldi r10,0x1000 -END_MMU_FTR_SECTION_IFSET(MMU_FTR_1T_SEGMENT) - rldimi r10,r9,USER_ESID_BITS,0 -BEGIN_FTR_SECTION bge slb_finish_load_1T END_MMU_FTR_SECTION_IFSET(MMU_FTR_1T_SEGMENT) b slb_finish_load 8: /* invalid EA */ li r10,0 /* BAD_VSID */ + li r9,0 /* BAD_VSID */ li r11,SLB_VSID_USER /* flags don't much matter */ b slb_finish_load @@ -221,8 +224,6 @@ _GLOBAL(slb_allocate_user) /* get context to calculate proto-VSID */ ld r9,PACACONTEXTID(r13) - rldimi r10,r9,USER_ESID_BITS,0 - /* fall through slb_finish_load */ #endif /* __DISABLED__ */ @@ -231,9 +232,10 @@ _GLOBAL(slb_allocate_user) /* * Finish loading of an SLB entry and return * - * r3 = EA, r10 = proto-VSID, r11 = flags, clobbers r9, cr7 = <> PAGE_OFFSET + * r3 = EA, r9 = context, r10 = ESID, r11 = flags, clobbers r9, cr7 = <> PAGE_OFFSET */ slb_finish_load: + rldimi r10,r9,ESID_BITS,0 ASM_VSID_SCRAMBLE(r10,r9,256M) /* * bits above VSID_BITS_256M need to be ignored from r10 @@ -251,7 +253,8 @@ slb_finish_load: 7: ld r10,PACASTABRR(r13) addi r10,r10,1 /* This gets soft patched on boot. */ -_GLOBAL(slb_compare_rr_to_size) +.globl slb_compare_rr_to_size +slb_compare_rr_to_size: cmpldi r10,0 blt+ 4f @@ -298,10 +301,11 @@ _GLOBAL(slb_compare_rr_to_size) /* * Finish loading of a 1T SLB entry (for the kernel linear mapping) and return. * - * r3 = EA, r10 = proto-VSID, r11 = flags, clobbers r9 + * r3 = EA, r9 = context, r10 = ESID(256MB), r11 = flags, clobbers r9 */ slb_finish_load_1T: - srdi r10,r10,40-28 /* get 1T ESID */ + srdi r10,r10,(SID_SHIFT_1T - SID_SHIFT) /* get 1T ESID */ + rldimi r10,r9,ESID_BITS_1T,0 ASM_VSID_SCRAMBLE(r10,r9,1T) /* * bits above VSID_BITS_1T need to be ignored from r10 diff --git a/arch/powerpc/mm/slice.c b/arch/powerpc/mm/slice.c index cf9dada734b..b0c75cc15ef 100644 --- a/arch/powerpc/mm/slice.c +++ b/arch/powerpc/mm/slice.c @@ -237,134 +237,112 @@ static void slice_convert(struct mm_struct *mm, struct slice_mask mask, int psiz #endif } +/* + * Compute which slice addr is part of; + * set *boundary_addr to the start or end boundary of that slice + * (depending on 'end' parameter); + * return boolean indicating if the slice is marked as available in the + * 'available' slice_mark. + */ +static bool slice_scan_available(unsigned long addr, + struct slice_mask available, + int end, + unsigned long *boundary_addr) +{ + unsigned long slice; + if (addr < SLICE_LOW_TOP) { + slice = GET_LOW_SLICE_INDEX(addr); + *boundary_addr = (slice + end) << SLICE_LOW_SHIFT; + return !!(available.low_slices & (1u << slice)); + } else { + slice = GET_HIGH_SLICE_INDEX(addr); + *boundary_addr = (slice + end) ? + ((slice + end) << SLICE_HIGH_SHIFT) : SLICE_LOW_TOP; + return !!(available.high_slices & (1ul << slice)); + } +} + static unsigned long slice_find_area_bottomup(struct mm_struct *mm, unsigned long len, struct slice_mask available, - int psize, int use_cache) + int psize) { - struct vm_area_struct *vma; - unsigned long start_addr, addr; - struct slice_mask mask; int pshift = max_t(int, mmu_psize_defs[psize].shift, PAGE_SHIFT); - - if (use_cache) { - if (len <= mm->cached_hole_size) { - start_addr = addr = TASK_UNMAPPED_BASE; - mm->cached_hole_size = 0; - } else - start_addr = addr = mm->free_area_cache; - } else - start_addr = addr = TASK_UNMAPPED_BASE; - -full_search: - for (;;) { - addr = _ALIGN_UP(addr, 1ul << pshift); - if ((TASK_SIZE - len) < addr) - break; - vma = find_vma(mm, addr); - BUG_ON(vma && (addr >= vma->vm_end)); - - mask = slice_range_to_mask(addr, len); - if (!slice_check_fit(mask, available)) { - if (addr < SLICE_LOW_TOP) - addr = _ALIGN_UP(addr + 1, 1ul << SLICE_LOW_SHIFT); - else - addr = _ALIGN_UP(addr + 1, 1ul << SLICE_HIGH_SHIFT); + unsigned long addr, found, next_end; + struct vm_unmapped_area_info info; + + info.flags = 0; + info.length = len; + info.align_mask = PAGE_MASK & ((1ul << pshift) - 1); + info.align_offset = 0; + + addr = TASK_UNMAPPED_BASE; + while (addr < TASK_SIZE) { + info.low_limit = addr; + if (!slice_scan_available(addr, available, 1, &addr)) continue; + + next_slice: + /* + * At this point [info.low_limit; addr) covers + * available slices only and ends at a slice boundary. + * Check if we need to reduce the range, or if we can + * extend it to cover the next available slice. + */ + if (addr >= TASK_SIZE) + addr = TASK_SIZE; + else if (slice_scan_available(addr, available, 1, &next_end)) { + addr = next_end; + goto next_slice; } - if (!vma || addr + len <= vma->vm_start) { - /* - * Remember the place where we stopped the search: - */ - if (use_cache) - mm->free_area_cache = addr + len; - return addr; - } - if (use_cache && (addr + mm->cached_hole_size) < vma->vm_start) - mm->cached_hole_size = vma->vm_start - addr; - addr = vma->vm_end; - } + info.high_limit = addr; - /* Make sure we didn't miss any holes */ - if (use_cache && start_addr != TASK_UNMAPPED_BASE) { - start_addr = addr = TASK_UNMAPPED_BASE; - mm->cached_hole_size = 0; - goto full_search; + found = vm_unmapped_area(&info); + if (!(found & ~PAGE_MASK)) + return found; } + return -ENOMEM; } static unsigned long slice_find_area_topdown(struct mm_struct *mm, unsigned long len, struct slice_mask available, - int psize, int use_cache) + int psize) { - struct vm_area_struct *vma; - unsigned long addr; - struct slice_mask mask; int pshift = max_t(int, mmu_psize_defs[psize].shift, PAGE_SHIFT); + unsigned long addr, found, prev; + struct vm_unmapped_area_info info; - /* check if free_area_cache is useful for us */ - if (use_cache) { - if (len <= mm->cached_hole_size) { - mm->cached_hole_size = 0; - mm->free_area_cache = mm->mmap_base; - } - - /* either no address requested or can't fit in requested - * address hole - */ - addr = mm->free_area_cache; - - /* make sure it can fit in the remaining address space */ - if (addr > len) { - addr = _ALIGN_DOWN(addr - len, 1ul << pshift); - mask = slice_range_to_mask(addr, len); - if (slice_check_fit(mask, available) && - slice_area_is_free(mm, addr, len)) - /* remember the address as a hint for - * next time - */ - return (mm->free_area_cache = addr); - } - } + info.flags = VM_UNMAPPED_AREA_TOPDOWN; + info.length = len; + info.align_mask = PAGE_MASK & ((1ul << pshift) - 1); + info.align_offset = 0; addr = mm->mmap_base; - while (addr > len) { - /* Go down by chunk size */ - addr = _ALIGN_DOWN(addr - len, 1ul << pshift); - - /* Check for hit with different page size */ - mask = slice_range_to_mask(addr, len); - if (!slice_check_fit(mask, available)) { - if (addr < SLICE_LOW_TOP) - addr = _ALIGN_DOWN(addr, 1ul << SLICE_LOW_SHIFT); - else if (addr < (1ul << SLICE_HIGH_SHIFT)) - addr = SLICE_LOW_TOP; - else - addr = _ALIGN_DOWN(addr, 1ul << SLICE_HIGH_SHIFT); + while (addr > PAGE_SIZE) { + info.high_limit = addr; + if (!slice_scan_available(addr - 1, available, 0, &addr)) continue; - } + prev_slice: /* - * Lookup failure means no vma is above this address, - * else if new region fits below vma->vm_start, - * return with success: + * At this point [addr; info.high_limit) covers + * available slices only and starts at a slice boundary. + * Check if we need to reduce the range, or if we can + * extend it to cover the previous available slice. */ - vma = find_vma(mm, addr); - if (!vma || (addr + len) <= vma->vm_start) { - /* remember the address as a hint for next time */ - if (use_cache) - mm->free_area_cache = addr; - return addr; + if (addr < PAGE_SIZE) + addr = PAGE_SIZE; + else if (slice_scan_available(addr - 1, available, 0, &prev)) { + addr = prev; + goto prev_slice; } + info.low_limit = addr; - /* remember the largest hole we saw so far */ - if (use_cache && (addr + mm->cached_hole_size) < vma->vm_start) - mm->cached_hole_size = vma->vm_start - addr; - - /* try just below the current vma->vm_start */ - addr = vma->vm_start; + found = vm_unmapped_area(&info); + if (!(found & ~PAGE_MASK)) + return found; } /* @@ -373,28 +351,18 @@ static unsigned long slice_find_area_topdown(struct mm_struct *mm, * can happen with large stack limits and large mmap() * allocations. */ - addr = slice_find_area_bottomup(mm, len, available, psize, 0); - - /* - * Restore the topdown base: - */ - if (use_cache) { - mm->free_area_cache = mm->mmap_base; - mm->cached_hole_size = ~0UL; - } - - return addr; + return slice_find_area_bottomup(mm, len, available, psize); } static unsigned long slice_find_area(struct mm_struct *mm, unsigned long len, struct slice_mask mask, int psize, - int topdown, int use_cache) + int topdown) { if (topdown) - return slice_find_area_topdown(mm, len, mask, psize, use_cache); + return slice_find_area_topdown(mm, len, mask, psize); else - return slice_find_area_bottomup(mm, len, mask, psize, use_cache); + return slice_find_area_bottomup(mm, len, mask, psize); } #define or_mask(dst, src) do { \ @@ -415,7 +383,7 @@ static unsigned long slice_find_area(struct mm_struct *mm, unsigned long len, unsigned long slice_get_unmapped_area(unsigned long addr, unsigned long len, unsigned long flags, unsigned int psize, - int topdown, int use_cache) + int topdown) { struct slice_mask mask = {0, 0}; struct slice_mask good_mask; @@ -430,8 +398,8 @@ unsigned long slice_get_unmapped_area(unsigned long addr, unsigned long len, BUG_ON(mm->task_size == 0); slice_dbg("slice_get_unmapped_area(mm=%p, psize=%d...\n", mm, psize); - slice_dbg(" addr=%lx, len=%lx, flags=%lx, topdown=%d, use_cache=%d\n", - addr, len, flags, topdown, use_cache); + slice_dbg(" addr=%lx, len=%lx, flags=%lx, topdown=%d\n", + addr, len, flags, topdown); if (len > mm->task_size) return -ENOMEM; @@ -440,7 +408,7 @@ unsigned long slice_get_unmapped_area(unsigned long addr, unsigned long len, if (fixed && (addr & ((1ul << pshift) - 1))) return -EINVAL; if (fixed && addr > (mm->task_size - len)) - return -EINVAL; + return -ENOMEM; /* If hint, make sure it matches our alignment restrictions */ if (!fixed && addr) { @@ -503,8 +471,7 @@ unsigned long slice_get_unmapped_area(unsigned long addr, unsigned long len, /* Now let's see if we can find something in the existing * slices for that size */ - newaddr = slice_find_area(mm, len, good_mask, psize, topdown, - use_cache); + newaddr = slice_find_area(mm, len, good_mask, psize, topdown); if (newaddr != -ENOMEM) { /* Found within the good mask, we don't have to setup, * we thus return directly @@ -536,8 +503,7 @@ unsigned long slice_get_unmapped_area(unsigned long addr, unsigned long len, * anywhere in the good area. */ if (addr) { - addr = slice_find_area(mm, len, good_mask, psize, topdown, - use_cache); + addr = slice_find_area(mm, len, good_mask, psize, topdown); if (addr != -ENOMEM) { slice_dbg(" found area at 0x%lx\n", addr); return addr; @@ -547,15 +513,14 @@ unsigned long slice_get_unmapped_area(unsigned long addr, unsigned long len, /* Now let's see if we can find something in the existing slices * for that size plus free slices */ - addr = slice_find_area(mm, len, potential_mask, psize, topdown, - use_cache); + addr = slice_find_area(mm, len, potential_mask, psize, topdown); #ifdef CONFIG_PPC_64K_PAGES if (addr == -ENOMEM && psize == MMU_PAGE_64K) { /* retry the search with 4k-page slices included */ or_mask(potential_mask, compat_mask); addr = slice_find_area(mm, len, potential_mask, psize, - topdown, use_cache); + topdown); } #endif @@ -586,8 +551,7 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long flags) { return slice_get_unmapped_area(addr, len, flags, - current->mm->context.user_psize, - 0, 1); + current->mm->context.user_psize, 0); } unsigned long arch_get_unmapped_area_topdown(struct file *filp, @@ -597,8 +561,7 @@ unsigned long arch_get_unmapped_area_topdown(struct file *filp, const unsigned long flags) { return slice_get_unmapped_area(addr0, len, flags, - current->mm->context.user_psize, - 1, 1); + current->mm->context.user_psize, 1); } unsigned int get_slice_psize(struct mm_struct *mm, unsigned long addr) diff --git a/arch/powerpc/mm/subpage-prot.c b/arch/powerpc/mm/subpage-prot.c index 7c415ddde94..6c0b1f5f8d2 100644 --- a/arch/powerpc/mm/subpage-prot.c +++ b/arch/powerpc/mm/subpage-prot.c @@ -78,7 +78,7 @@ static void hpte_flush_range(struct mm_struct *mm, unsigned long addr, pte = pte_offset_map_lock(mm, pmd, addr, &ptl); arch_enter_lazy_mmu_mode(); for (; npages > 0; --npages) { - pte_update(mm, addr, pte, 0, 0); + pte_update(mm, addr, pte, 0, 0, 0); addr += PAGE_SIZE; ++pte; } @@ -105,7 +105,7 @@ static void subpage_prot_clear(unsigned long addr, unsigned long len) limit = spt->maxaddr; for (; addr < limit; addr = next) { next = pmd_addr_end(addr, limit); - if (addr < 0x100000000) { + if (addr < 0x100000000UL) { spm = spt->low_prot; } else { spm = spt->protptrs[addr >> SBP_L3_SHIFT]; @@ -130,6 +130,53 @@ static void subpage_prot_clear(unsigned long addr, unsigned long len) up_write(&mm->mmap_sem); } +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +static int subpage_walk_pmd_entry(pmd_t *pmd, unsigned long addr, + unsigned long end, struct mm_walk *walk) +{ + struct vm_area_struct *vma = walk->private; + split_huge_page_pmd(vma, addr, pmd); + return 0; +} + +static void subpage_mark_vma_nohuge(struct mm_struct *mm, unsigned long addr, + unsigned long len) +{ + struct vm_area_struct *vma; + struct mm_walk subpage_proto_walk = { + .mm = mm, + .pmd_entry = subpage_walk_pmd_entry, + }; + + /* + * We don't try too hard, we just mark all the vma in that range + * VM_NOHUGEPAGE and split them. + */ + vma = find_vma(mm, addr); + /* + * If the range is in unmapped range, just return + */ + if (vma && ((addr + len) <= vma->vm_start)) + return; + + while (vma) { + if (vma->vm_start >= (addr + len)) + break; + vma->vm_flags |= VM_NOHUGEPAGE; + subpage_proto_walk.private = vma; + walk_page_range(vma->vm_start, vma->vm_end, + &subpage_proto_walk); + vma = vma->vm_next; + } +} +#else +static void subpage_mark_vma_nohuge(struct mm_struct *mm, unsigned long addr, + unsigned long len) +{ + return; +} +#endif + /* * Copy in a subpage protection map for an address range. * The map has 2 bits per 4k subpage, so 32 bits per 64k page. @@ -168,10 +215,11 @@ long sys_subpage_prot(unsigned long addr, unsigned long len, u32 __user *map) return -EFAULT; down_write(&mm->mmap_sem); + subpage_mark_vma_nohuge(mm, addr, len); for (limit = addr + len; addr < limit; addr = next) { next = pmd_addr_end(addr, limit); err = -ENOMEM; - if (addr < 0x100000000) { + if (addr < 0x100000000UL) { spm = spt->low_prot; } else { spm = spt->protptrs[addr >> SBP_L3_SHIFT]; diff --git a/arch/powerpc/mm/tlb_hash64.c b/arch/powerpc/mm/tlb_hash64.c index 0d82ef50dc3..c99f6510a0b 100644 --- a/arch/powerpc/mm/tlb_hash64.c +++ b/arch/powerpc/mm/tlb_hash64.c @@ -23,7 +23,6 @@ #include <linux/kernel.h> #include <linux/mm.h> -#include <linux/init.h> #include <linux/percpu.h> #include <linux/hardirq.h> #include <asm/pgalloc.h> @@ -82,11 +81,11 @@ void hpte_need_flush(struct mm_struct *mm, unsigned long addr, if (!is_kernel_addr(addr)) { ssize = user_segment_size(addr); vsid = get_vsid(mm->context.id, addr, ssize); - WARN_ON(vsid == 0); } else { vsid = get_kernel_vsid(addr, mmu_kernel_ssize); ssize = mmu_kernel_ssize; } + WARN_ON(vsid == 0); vpn = hpt_vpn(addr, vsid, ssize); rpte = __real_pte(__pte(pte), ptep); @@ -183,12 +182,13 @@ void tlb_flush(struct mmu_gather *tlb) * since 64K pages may overlap with other bridges when using 64K pages * with 4K HW pages on IO space. * - * Because of that usage pattern, it's only available with CONFIG_HOTPLUG - * and is implemented for small size rather than speed. + * Because of that usage pattern, it is implemented for small size rather + * than speed. */ void __flush_hash_table_range(struct mm_struct *mm, unsigned long start, unsigned long end) { + int hugepage_shift; unsigned long flags; start = _ALIGN_DOWN(start, PAGE_SIZE); @@ -206,7 +206,8 @@ void __flush_hash_table_range(struct mm_struct *mm, unsigned long start, local_irq_save(flags); arch_enter_lazy_mmu_mode(); for (; start < end; start += PAGE_SIZE) { - pte_t *ptep = find_linux_pte(mm->pgd, start); + pte_t *ptep = find_linux_pte_or_hugepte(mm->pgd, start, + &hugepage_shift); unsigned long pte; if (ptep == NULL) @@ -214,7 +215,37 @@ void __flush_hash_table_range(struct mm_struct *mm, unsigned long start, pte = pte_val(*ptep); if (!(pte & _PAGE_HASHPTE)) continue; - hpte_need_flush(mm, start, ptep, pte, 0); + if (unlikely(hugepage_shift && pmd_trans_huge(*(pmd_t *)pte))) + hpte_do_hugepage_flush(mm, start, (pmd_t *)pte); + else + hpte_need_flush(mm, start, ptep, pte, 0); + } + arch_leave_lazy_mmu_mode(); + local_irq_restore(flags); +} + +void flush_tlb_pmd_range(struct mm_struct *mm, pmd_t *pmd, unsigned long addr) +{ + pte_t *pte; + pte_t *start_pte; + unsigned long flags; + + addr = _ALIGN_DOWN(addr, PMD_SIZE); + /* Note: Normally, we should only ever use a batch within a + * PTE locked section. This violates the rule, but will work + * since we don't actually modify the PTEs, we just flush the + * hash while leaving the PTEs intact (including their reference + * to being hashed). This is not the most performance oriented + * way to do things but is fine for our needs here. + */ + local_irq_save(flags); + arch_enter_lazy_mmu_mode(); + start_pte = pte_offset_map(pmd, addr); + for (pte = start_pte; pte < start_pte + PTRS_PER_PTE; pte++) { + unsigned long pteval = pte_val(*pte); + if (pteval & _PAGE_HASHPTE) + hpte_need_flush(mm, addr, pte, pteval, 0); + addr += PAGE_SIZE; } arch_leave_lazy_mmu_mode(); local_irq_restore(flags); diff --git a/arch/powerpc/mm/tlb_low_64e.S b/arch/powerpc/mm/tlb_low_64e.S index b4113bf8635..356e8b41fb0 100644 --- a/arch/powerpc/mm/tlb_low_64e.S +++ b/arch/powerpc/mm/tlb_low_64e.S @@ -39,37 +39,49 @@ * * **********************************************************************/ +/* + * Note that, unlike non-bolted handlers, TLB_EXFRAME is not + * modified by the TLB miss handlers themselves, since the TLB miss + * handler code will not itself cause a recursive TLB miss. + * + * TLB_EXFRAME will be modified when crit/mc/debug exceptions are + * entered/exited. + */ .macro tlb_prolog_bolted intnum addr - mtspr SPRN_SPRG_GEN_SCRATCH,r13 + mtspr SPRN_SPRG_GEN_SCRATCH,r12 + mfspr r12,SPRN_SPRG_TLB_EXFRAME + std r13,EX_TLB_R13(r12) + std r10,EX_TLB_R10(r12) mfspr r13,SPRN_SPRG_PACA - std r10,PACA_EXTLB+EX_TLB_R10(r13) + mfcr r10 - std r11,PACA_EXTLB+EX_TLB_R11(r13) + std r11,EX_TLB_R11(r12) #ifdef CONFIG_KVM_BOOKE_HV BEGIN_FTR_SECTION mfspr r11, SPRN_SRR1 END_FTR_SECTION_IFSET(CPU_FTR_EMB_HV) #endif DO_KVM \intnum, SPRN_SRR1 - std r16,PACA_EXTLB+EX_TLB_R16(r13) + std r16,EX_TLB_R16(r12) mfspr r16,\addr /* get faulting address */ - std r14,PACA_EXTLB+EX_TLB_R14(r13) + std r14,EX_TLB_R14(r12) ld r14,PACAPGD(r13) - std r15,PACA_EXTLB+EX_TLB_R15(r13) - std r10,PACA_EXTLB+EX_TLB_CR(r13) - TLB_MISS_PROLOG_STATS_BOLTED + std r15,EX_TLB_R15(r12) + std r10,EX_TLB_CR(r12) + TLB_MISS_PROLOG_STATS .endm .macro tlb_epilog_bolted - ld r14,PACA_EXTLB+EX_TLB_CR(r13) - ld r10,PACA_EXTLB+EX_TLB_R10(r13) - ld r11,PACA_EXTLB+EX_TLB_R11(r13) + ld r14,EX_TLB_CR(r12) + ld r10,EX_TLB_R10(r12) + ld r11,EX_TLB_R11(r12) + ld r13,EX_TLB_R13(r12) mtcr r14 - ld r14,PACA_EXTLB+EX_TLB_R14(r13) - ld r15,PACA_EXTLB+EX_TLB_R15(r13) - TLB_MISS_RESTORE_STATS_BOLTED - ld r16,PACA_EXTLB+EX_TLB_R16(r13) - mfspr r13,SPRN_SPRG_GEN_SCRATCH + ld r14,EX_TLB_R14(r12) + ld r15,EX_TLB_R15(r12) + TLB_MISS_RESTORE_STATS + ld r16,EX_TLB_R16(r12) + mfspr r12,SPRN_SPRG_GEN_SCRATCH .endm /* Data TLB miss */ @@ -136,7 +148,7 @@ BEGIN_MMU_FTR_SECTION */ PPC_TLBSRX_DOT(0,R16) ldx r14,r14,r15 /* grab pgd entry */ - beq normal_tlb_miss_done /* tlb exists already, bail */ + beq tlb_miss_done_bolted /* tlb exists already, bail */ MMU_FTR_SECTION_ELSE ldx r14,r14,r15 /* grab pgd entry */ ALT_MMU_FTR_SECTION_END_IFSET(MMU_FTR_USE_TLBRSRV) @@ -192,6 +204,7 @@ ALT_MMU_FTR_SECTION_END_IFSET(MMU_FTR_USE_TLBRSRV) mtspr SPRN_MAS7_MAS3,r15 tlbwe +tlb_miss_done_bolted: TLB_MISS_STATS_X(MMSTAT_TLB_MISS_NORM_OK) tlb_epilog_bolted rfi @@ -239,6 +252,183 @@ itlb_miss_fault_bolted: beq tlb_miss_common_bolted b itlb_miss_kernel_bolted +#ifdef CONFIG_PPC_FSL_BOOK3E +/* + * TLB miss handling for e6500 and derivatives, using hardware tablewalk. + * + * Linear mapping is bolted: no virtual page table or nested TLB misses + * Indirect entries in TLB1, hardware loads resulting direct entries + * into TLB0 + * No HES or NV hint on TLB1, so we need to do software round-robin + * No tlbsrx. so we need a spinlock, and we have to deal + * with MAS-damage caused by tlbsx + * 4K pages only + */ + + START_EXCEPTION(instruction_tlb_miss_e6500) + tlb_prolog_bolted BOOKE_INTERRUPT_ITLB_MISS SPRN_SRR0 + + ld r11,PACA_TCD_PTR(r13) + srdi. r15,r16,60 /* get region */ + ori r16,r16,1 + + TLB_MISS_STATS_SAVE_INFO_BOLTED + bne tlb_miss_kernel_e6500 /* user/kernel test */ + + b tlb_miss_common_e6500 + + START_EXCEPTION(data_tlb_miss_e6500) + tlb_prolog_bolted BOOKE_INTERRUPT_DTLB_MISS SPRN_DEAR + + ld r11,PACA_TCD_PTR(r13) + srdi. r15,r16,60 /* get region */ + rldicr r16,r16,0,62 + + TLB_MISS_STATS_SAVE_INFO_BOLTED + bne tlb_miss_kernel_e6500 /* user vs kernel check */ + +/* + * This is the guts of the TLB miss handler for e6500 and derivatives. + * We are entered with: + * + * r16 = page of faulting address (low bit 0 if data, 1 if instruction) + * r15 = crap (free to use) + * r14 = page table base + * r13 = PACA + * r11 = tlb_per_core ptr + * r10 = cpu number + */ +tlb_miss_common_e6500: + /* + * Search if we already have an indirect entry for that virtual + * address, and if we do, bail out. + * + * MAS6:IND should be already set based on MAS4 + */ +1: lbarx r15,0,r11 + lhz r10,PACAPACAINDEX(r13) + cmpdi r15,0 + cmpdi cr1,r15,1 /* set cr1.eq = 0 for non-recursive */ + bne 2f + stbcx. r10,0,r11 + bne 1b +3: + .subsection 1 +2: cmpd cr1,r15,r10 /* recursive lock due to mcheck/crit/etc? */ + beq cr1,3b /* unlock will happen if cr1.eq = 0 */ + lbz r15,0(r11) + cmpdi r15,0 + bne 2b + b 1b + .previous + + mfspr r15,SPRN_MAS2 + + tlbsx 0,r16 + mfspr r10,SPRN_MAS1 + andis. r10,r10,MAS1_VALID@h + bne tlb_miss_done_e6500 + + /* Undo MAS-damage from the tlbsx */ + mfspr r10,SPRN_MAS1 + oris r10,r10,MAS1_VALID@h + mtspr SPRN_MAS1,r10 + mtspr SPRN_MAS2,r15 + + /* Now, we need to walk the page tables. First check if we are in + * range. + */ + rldicl. r10,r16,64-PGTABLE_EADDR_SIZE,PGTABLE_EADDR_SIZE+4 + bne- tlb_miss_fault_e6500 + + rldicl r15,r16,64-PGDIR_SHIFT+3,64-PGD_INDEX_SIZE-3 + cmpldi cr0,r14,0 + clrrdi r15,r15,3 + beq- tlb_miss_fault_e6500 /* No PGDIR, bail */ + ldx r14,r14,r15 /* grab pgd entry */ + + rldicl r15,r16,64-PUD_SHIFT+3,64-PUD_INDEX_SIZE-3 + clrrdi r15,r15,3 + cmpdi cr0,r14,0 + bge tlb_miss_fault_e6500 /* Bad pgd entry or hugepage; bail */ + ldx r14,r14,r15 /* grab pud entry */ + + rldicl r15,r16,64-PMD_SHIFT+3,64-PMD_INDEX_SIZE-3 + clrrdi r15,r15,3 + cmpdi cr0,r14,0 + bge tlb_miss_fault_e6500 + ldx r14,r14,r15 /* Grab pmd entry */ + + mfspr r10,SPRN_MAS0 + cmpdi cr0,r14,0 + bge tlb_miss_fault_e6500 + + /* Now we build the MAS for a 2M indirect page: + * + * MAS 0 : ESEL needs to be filled by software round-robin + * MAS 1 : Fully set up + * - PID already updated by caller if necessary + * - TSIZE for now is base ind page size always + * - TID already cleared if necessary + * MAS 2 : Default not 2M-aligned, need to be redone + * MAS 3+7 : Needs to be done + */ + + ori r14,r14,(BOOK3E_PAGESZ_4K << MAS3_SPSIZE_SHIFT) + mtspr SPRN_MAS7_MAS3,r14 + + clrrdi r15,r16,21 /* make EA 2M-aligned */ + mtspr SPRN_MAS2,r15 + + lbz r15,TCD_ESEL_NEXT(r11) + lbz r16,TCD_ESEL_MAX(r11) + lbz r14,TCD_ESEL_FIRST(r11) + rlwimi r10,r15,16,0x00ff0000 /* insert esel_next into MAS0 */ + addi r15,r15,1 /* increment esel_next */ + mtspr SPRN_MAS0,r10 + cmpw r15,r16 + iseleq r15,r14,r15 /* if next == last use first */ + stb r15,TCD_ESEL_NEXT(r11) + + tlbwe + +tlb_miss_done_e6500: + .macro tlb_unlock_e6500 + beq cr1,1f /* no unlock if lock was recursively grabbed */ + li r15,0 + isync + stb r15,0(r11) +1: + .endm + + tlb_unlock_e6500 + TLB_MISS_STATS_X(MMSTAT_TLB_MISS_NORM_OK) + tlb_epilog_bolted + rfi + +tlb_miss_kernel_e6500: + mfspr r10,SPRN_MAS1 + ld r14,PACA_KERNELPGD(r13) + cmpldi cr0,r15,8 /* Check for vmalloc region */ + rlwinm r10,r10,0,16,1 /* Clear TID */ + mtspr SPRN_MAS1,r10 + beq+ tlb_miss_common_e6500 + +tlb_miss_fault_e6500: + tlb_unlock_e6500 + /* We need to check if it was an instruction miss */ + andi. r16,r16,1 + bne itlb_miss_fault_e6500 +dtlb_miss_fault_e6500: + TLB_MISS_STATS_D(MMSTAT_TLB_MISS_NORM_FAULT) + tlb_epilog_bolted + b exc_data_storage_book3e +itlb_miss_fault_e6500: + TLB_MISS_STATS_I(MMSTAT_TLB_MISS_NORM_FAULT) + tlb_epilog_bolted + b exc_instruction_storage_book3e +#endif /* CONFIG_PPC_FSL_BOOK3E */ + /********************************************************************** * * * TLB miss handling for Book3E with TLB reservation and HES support * @@ -918,7 +1108,8 @@ tlb_load_linear: ld r11,PACATOC(r13) ld r11,linear_map_top@got(r11) ld r10,0(r11) - cmpld cr0,r10,r16 + tovirt(10,10) + cmpld cr0,r16,r10 bge tlb_load_linear_fault /* MAS1 need whole new setup. */ diff --git a/arch/powerpc/mm/tlb_nohash.c b/arch/powerpc/mm/tlb_nohash.c index df32a838dcf..92cb18d52ea 100644 --- a/arch/powerpc/mm/tlb_nohash.c +++ b/arch/powerpc/mm/tlb_nohash.c @@ -43,6 +43,7 @@ #include <asm/tlb.h> #include <asm/code-patching.h> #include <asm/hugetlb.h> +#include <asm/paca.h> #include "mmu_decl.h" @@ -58,6 +59,10 @@ struct mmu_psize_def mmu_psize_defs[MMU_PAGE_COUNT] = { .shift = 12, .enc = BOOK3E_PAGESZ_4K, }, + [MMU_PAGE_2M] = { + .shift = 21, + .enc = BOOK3E_PAGESZ_2M, + }, [MMU_PAGE_4M] = { .shift = 22, .enc = BOOK3E_PAGESZ_4M, @@ -136,9 +141,18 @@ static inline int mmu_get_tsize(int psize) int mmu_linear_psize; /* Page size used for the linear mapping */ int mmu_pte_psize; /* Page size used for PTE pages */ int mmu_vmemmap_psize; /* Page size used for the virtual mem map */ -int book3e_htw_enabled; /* Is HW tablewalk enabled ? */ +int book3e_htw_mode; /* HW tablewalk? Value is PPC_HTW_* */ unsigned long linear_map_top; /* Top of linear mapping */ + +/* + * Number of bytes to add to SPRN_SPRG_TLB_EXFRAME on crit/mcheck/debug + * exceptions. This is used for bolted and e6500 TLB miss handlers which + * do not modify this SPRG in the TLB miss code; for other TLB miss handlers, + * this is set to zero. + */ +int extlb_level_exc; + #endif /* CONFIG_PPC64 */ #ifdef CONFIG_PPC_FSL_BOOK3E @@ -305,7 +319,7 @@ void __flush_tlb_page(struct mm_struct *mm, unsigned long vmaddr, void flush_tlb_page(struct vm_area_struct *vma, unsigned long vmaddr) { #ifdef CONFIG_HUGETLB_PAGE - if (is_vm_hugetlb_page(vma)) + if (vma && is_vm_hugetlb_page(vma)) flush_hugetlb_page(vma, vmaddr); #endif @@ -377,7 +391,7 @@ void tlb_flush_pgtable(struct mmu_gather *tlb, unsigned long address) { int tsize = mmu_psize_defs[mmu_pte_psize].enc; - if (book3e_htw_enabled) { + if (book3e_htw_mode != PPC_HTW_NONE) { unsigned long start = address & PMD_MASK; unsigned long end = address + PMD_SIZE; unsigned long size = 1UL << mmu_psize_defs[mmu_pte_psize].shift; @@ -414,9 +428,9 @@ static void setup_page_sizes(void) #ifdef CONFIG_PPC_FSL_BOOK3E unsigned int mmucfg = mfspr(SPRN_MMUCFG); + int fsl_mmu = mmu_has_feature(MMU_FTR_TYPE_FSL_E); - if (((mmucfg & MMUCFG_MAVN) == MMUCFG_MAVN_V1) && - (mmu_has_feature(MMU_FTR_TYPE_FSL_E))) { + if (fsl_mmu && (mmucfg & MMUCFG_MAVN) == MMUCFG_MAVN_V1) { unsigned int tlb1cfg = mfspr(SPRN_TLB1CFG); unsigned int min_pg, max_pg; @@ -430,7 +444,7 @@ static void setup_page_sizes(void) def = &mmu_psize_defs[psize]; shift = def->shift; - if (shift == 0) + if (shift == 0 || shift & 1) continue; /* adjust to be in terms of 4^shift Kb */ @@ -440,7 +454,40 @@ static void setup_page_sizes(void) def->flags |= MMU_PAGE_SIZE_DIRECT; } - goto no_indirect; + goto out; + } + + if (fsl_mmu && (mmucfg & MMUCFG_MAVN) == MMUCFG_MAVN_V2) { + u32 tlb1cfg, tlb1ps; + + tlb0cfg = mfspr(SPRN_TLB0CFG); + tlb1cfg = mfspr(SPRN_TLB1CFG); + tlb1ps = mfspr(SPRN_TLB1PS); + eptcfg = mfspr(SPRN_EPTCFG); + + if ((tlb1cfg & TLBnCFG_IND) && (tlb0cfg & TLBnCFG_PT)) + book3e_htw_mode = PPC_HTW_E6500; + + /* + * We expect 4K subpage size and unrestricted indirect size. + * The lack of a restriction on indirect size is a Freescale + * extension, indicated by PSn = 0 but SPSn != 0. + */ + if (eptcfg != 2) + book3e_htw_mode = PPC_HTW_NONE; + + for (psize = 0; psize < MMU_PAGE_COUNT; ++psize) { + struct mmu_psize_def *def = &mmu_psize_defs[psize]; + + if (tlb1ps & (1U << (def->shift - 10))) { + def->flags |= MMU_PAGE_SIZE_DIRECT; + + if (book3e_htw_mode && psize == MMU_PAGE_2M) + def->flags |= MMU_PAGE_SIZE_INDIRECT; + } + } + + goto out; } #endif @@ -457,8 +504,11 @@ static void setup_page_sizes(void) } /* Indirect page sizes supported ? */ - if ((tlb0cfg & TLBnCFG_IND) == 0) - goto no_indirect; + if ((tlb0cfg & TLBnCFG_IND) == 0 || + (tlb0cfg & TLBnCFG_PT) == 0) + goto out; + + book3e_htw_mode = PPC_HTW_IBM; /* Now, we only deal with one IND page size for each * direct size. Hopefully all implementations today are @@ -483,8 +533,8 @@ static void setup_page_sizes(void) def->ind = ps + 10; } } - no_indirect: +out: /* Cleanup array and print summary */ pr_info("MMU: Supported page sizes\n"); for (psize = 0; psize < MMU_PAGE_COUNT; ++psize) { @@ -504,44 +554,28 @@ static void setup_page_sizes(void) } } -static void __patch_exception(int exc, unsigned long addr) -{ - extern unsigned int interrupt_base_book3e; - unsigned int *ibase = &interrupt_base_book3e; - - /* Our exceptions vectors start with a NOP and -then- a branch - * to deal with single stepping from userspace which stops on - * the second instruction. Thus we need to patch the second - * instruction of the exception, not the first one - */ - - patch_branch(ibase + (exc / 4) + 1, addr, 0); -} - -#define patch_exception(exc, name) do { \ - extern unsigned int name; \ - __patch_exception((exc), (unsigned long)&name); \ -} while (0) - static void setup_mmu_htw(void) { - /* Check if HW tablewalk is present, and if yes, enable it by: - * - * - patching the TLB miss handlers to branch to the - * one dedicates to it - * - * - setting the global book3e_htw_enabled - */ - unsigned int tlb0cfg = mfspr(SPRN_TLB0CFG); + /* + * If we want to use HW tablewalk, enable it by patching the TLB miss + * handlers to branch to the one dedicated to it. + */ - if ((tlb0cfg & TLBnCFG_IND) && - (tlb0cfg & TLBnCFG_PT)) { + switch (book3e_htw_mode) { + case PPC_HTW_IBM: patch_exception(0x1c0, exc_data_tlb_miss_htw_book3e); patch_exception(0x1e0, exc_instruction_tlb_miss_htw_book3e); - book3e_htw_enabled = 1; + break; +#ifdef CONFIG_PPC_FSL_BOOK3E + case PPC_HTW_E6500: + extlb_level_exc = EX_TLB_SIZE; + patch_exception(0x1c0, exc_data_tlb_miss_e6500_book3e); + patch_exception(0x1e0, exc_instruction_tlb_miss_e6500_book3e); + break; +#endif } pr_info("MMU: Book3E HW tablewalk %s\n", - book3e_htw_enabled ? "enabled" : "not supported"); + book3e_htw_mode != PPC_HTW_NONE ? "enabled" : "not supported"); } /* @@ -562,8 +596,13 @@ static void __early_init_mmu(int boot_cpu) /* XXX This should be decided at runtime based on supported * page sizes in the TLB, but for now let's assume 16M is * always there and a good fit (which it probably is) + * + * Freescale booke only supports 4K pages in TLB0, so use that. */ - mmu_vmemmap_psize = MMU_PAGE_16M; + if (mmu_has_feature(MMU_FTR_TYPE_FSL_E)) + mmu_vmemmap_psize = MMU_PAGE_4K; + else + mmu_vmemmap_psize = MMU_PAGE_16M; /* XXX This code only checks for TLB 0 capabilities and doesn't * check what page size combos are supported by the HW. It @@ -581,8 +620,16 @@ static void __early_init_mmu(int boot_cpu) /* Set MAS4 based on page table setting */ mas4 = 0x4 << MAS4_WIMGED_SHIFT; - if (book3e_htw_enabled) { - mas4 |= mas4 | MAS4_INDD; + switch (book3e_htw_mode) { + case PPC_HTW_E6500: + mas4 |= MAS4_INDD; + mas4 |= BOOK3E_PAGESZ_2M << MAS4_TSIZED_SHIFT; + mas4 |= MAS4_TLBSELD(1); + mmu_pte_psize = MMU_PAGE_2M; + break; + + case PPC_HTW_IBM: + mas4 |= MAS4_INDD; #ifdef CONFIG_PPC_64K_PAGES mas4 |= BOOK3E_PAGESZ_256M << MAS4_TSIZED_SHIFT; mmu_pte_psize = MMU_PAGE_256M; @@ -590,13 +637,16 @@ static void __early_init_mmu(int boot_cpu) mas4 |= BOOK3E_PAGESZ_1M << MAS4_TSIZED_SHIFT; mmu_pte_psize = MMU_PAGE_1M; #endif - } else { + break; + + case PPC_HTW_NONE: #ifdef CONFIG_PPC_64K_PAGES mas4 |= BOOK3E_PAGESZ_64K << MAS4_TSIZED_SHIFT; #else mas4 |= BOOK3E_PAGESZ_4K << MAS4_TSIZED_SHIFT; #endif mmu_pte_psize = mmu_virtual_psize; + break; } mtspr(SPRN_MAS4, mas4); @@ -616,8 +666,12 @@ static void __early_init_mmu(int boot_cpu) /* limit memory so we dont have linear faults */ memblock_enforce_memory_limit(linear_map_top); - patch_exception(0x1c0, exc_data_tlb_miss_bolted_book3e); - patch_exception(0x1e0, exc_instruction_tlb_miss_bolted_book3e); + if (book3e_htw_mode == PPC_HTW_NONE) { + extlb_level_exc = EX_TLB_SIZE; + patch_exception(0x1c0, exc_data_tlb_miss_bolted_book3e); + patch_exception(0x1e0, + exc_instruction_tlb_miss_bolted_book3e); + } } #endif @@ -634,7 +688,7 @@ void __init early_init_mmu(void) __early_init_mmu(1); } -void __cpuinit early_init_mmu_secondary(void) +void early_init_mmu_secondary(void) { __early_init_mmu(0); } diff --git a/arch/powerpc/mm/tlb_nohash_low.S b/arch/powerpc/mm/tlb_nohash_low.S index 626ad081639..43ff3c797fb 100644 --- a/arch/powerpc/mm/tlb_nohash_low.S +++ b/arch/powerpc/mm/tlb_nohash_low.S @@ -402,7 +402,9 @@ _GLOBAL(set_context) * Load TLBCAM[index] entry in to the L2 CAM MMU */ _GLOBAL(loadcam_entry) - LOAD_REG_ADDR(r4, TLBCAM) + mflr r5 + LOAD_REG_ADDR_PIC(r4, TLBCAM) + mtlr r5 mulli r5,r3,TLBCAM_SIZE add r3,r5,r4 lwz r4,TLBCAM_MAS0(r3) |