diff options
Diffstat (limited to 'arch/powerpc/mm/pgtable_64.c')
| -rw-r--r-- | arch/powerpc/mm/pgtable_64.c | 877 |
1 files changed, 717 insertions, 160 deletions
diff --git a/arch/powerpc/mm/pgtable_64.c b/arch/powerpc/mm/pgtable_64.c index 51b78694097..f6ce1f111f5 100644 --- a/arch/powerpc/mm/pgtable_64.c +++ b/arch/powerpc/mm/pgtable_64.c @@ -7,7 +7,6 @@ * Modifications by Paul Mackerras (PowerMac) (paulus@samba.org) * and Cort Dougan (PReP) (cort@cs.nmt.edu) * Copyright (C) 1996 Paul Mackerras - * Amiga/APUS changes by Jesper Skov (jskov@cygnus.co.uk). * * Derived from "arch/i386/mm/init.c" * Copyright (C) 1991, 1992, 1993, 1994 Linus Torvalds @@ -22,87 +21,81 @@ * */ -#include <linux/config.h> #include <linux/signal.h> #include <linux/sched.h> #include <linux/kernel.h> #include <linux/errno.h> #include <linux/string.h> +#include <linux/export.h> #include <linux/types.h> #include <linux/mman.h> #include <linux/mm.h> #include <linux/swap.h> #include <linux/stddef.h> #include <linux/vmalloc.h> -#include <linux/init.h> -#include <linux/delay.h> #include <linux/bootmem.h> -#include <linux/highmem.h> -#include <linux/idr.h> -#include <linux/nodemask.h> -#include <linux/module.h> +#include <linux/memblock.h> +#include <linux/slab.h> #include <asm/pgalloc.h> #include <asm/page.h> #include <asm/prom.h> -#include <asm/lmb.h> -#include <asm/rtas.h> #include <asm/io.h> #include <asm/mmu_context.h> #include <asm/pgtable.h> #include <asm/mmu.h> -#include <asm/uaccess.h> #include <asm/smp.h> #include <asm/machdep.h> #include <asm/tlb.h> -#include <asm/eeh.h> #include <asm/processor.h> -#include <asm/mmzone.h> #include <asm/cputable.h> -#include <asm/ppcdebug.h> #include <asm/sections.h> -#include <asm/system.h> -#include <asm/iommu.h> -#include <asm/abs_addr.h> -#include <asm/vdso.h> -#include <asm/imalloc.h> +#include <asm/firmware.h> -unsigned long ioremap_bot = IMALLOC_BASE; -static unsigned long phbs_io_bot = PHBS_IO_BASE; +#include "mmu_decl.h" -#ifdef CONFIG_PPC_ISERIES +/* Some sanity checking */ +#if TASK_SIZE_USER64 > PGTABLE_RANGE +#error TASK_SIZE_USER64 exceeds pagetable range +#endif -void __iomem *ioremap(unsigned long addr, unsigned long size) -{ - return (void __iomem *)addr; -} +#ifdef CONFIG_PPC_STD_MMU_64 +#if TASK_SIZE_USER64 > (1UL << (ESID_BITS + SID_SHIFT)) +#error TASK_SIZE_USER64 exceeds user VSID range +#endif +#endif -extern void __iomem *__ioremap(unsigned long addr, unsigned long size, - unsigned long flags) -{ - return (void __iomem *)addr; -} +unsigned long ioremap_bot = IOREMAP_BASE; -void iounmap(volatile void __iomem *addr) +#ifdef CONFIG_PPC_MMU_NOHASH +static void *early_alloc_pgtable(unsigned long size) { - return; -} + void *pt; -#else + if (init_bootmem_done) + pt = __alloc_bootmem(size, size, __pa(MAX_DMA_ADDRESS)); + else + pt = __va(memblock_alloc_base(size, size, + __pa(MAX_DMA_ADDRESS))); + memset(pt, 0, size); + + return pt; +} +#endif /* CONFIG_PPC_MMU_NOHASH */ /* - * map_io_page currently only called by __ioremap - * map_io_page adds an entry to the ioremap page table + * map_kernel_page currently only called by __ioremap + * map_kernel_page adds an entry to the ioremap page table * and adds an entry to the HPT, possibly bolting it */ -static int map_io_page(unsigned long ea, unsigned long pa, int flags) +int map_kernel_page(unsigned long ea, unsigned long pa, int flags) { pgd_t *pgdp; pud_t *pudp; pmd_t *pmdp; pte_t *ptep; - if (mem_init_done) { + if (slab_is_available()) { pgdp = pgd_offset_k(ea); pudp = pud_alloc(&init_mm, pgdp, ea); if (!pudp) @@ -116,6 +109,35 @@ static int map_io_page(unsigned long ea, unsigned long pa, int flags) set_pte_at(&init_mm, ea, ptep, pfn_pte(pa >> PAGE_SHIFT, __pgprot(flags))); } else { +#ifdef CONFIG_PPC_MMU_NOHASH + /* Warning ! This will blow up if bootmem is not initialized + * which our ppc64 code is keen to do that, we'll need to + * fix it and/or be more careful + */ + pgdp = pgd_offset_k(ea); +#ifdef PUD_TABLE_SIZE + if (pgd_none(*pgdp)) { + pudp = early_alloc_pgtable(PUD_TABLE_SIZE); + BUG_ON(pudp == NULL); + pgd_populate(&init_mm, pgdp, pudp); + } +#endif /* PUD_TABLE_SIZE */ + pudp = pud_offset(pgdp, ea); + if (pud_none(*pudp)) { + pmdp = early_alloc_pgtable(PMD_TABLE_SIZE); + BUG_ON(pmdp == NULL); + pud_populate(&init_mm, pudp, pmdp); + } + pmdp = pmd_offset(pudp, ea); + if (!pmd_present(*pmdp)) { + ptep = early_alloc_pgtable(PAGE_SIZE); + BUG_ON(ptep == NULL); + pmd_populate_kernel(&init_mm, pmdp, ptep); + } + ptep = pte_offset_kernel(pmdp, ea); + set_pte_at(&init_mm, ea, ptep, pfn_pte(pa >> PAGE_SHIFT, + __pgprot(flags))); +#else /* CONFIG_PPC_MMU_NOHASH */ /* * If the mm subsystem is not fully up, we cannot create a * linux page table entry for this mapping. Simply bolt an @@ -123,40 +145,79 @@ static int map_io_page(unsigned long ea, unsigned long pa, int flags) * */ if (htab_bolt_mapping(ea, ea + PAGE_SIZE, pa, flags, - mmu_virtual_psize)) - panic("Can't map bolted IO mapping"); + mmu_io_psize, mmu_kernel_ssize)) { + printk(KERN_ERR "Failed to do bolted mapping IO " + "memory at %016lx !\n", pa); + return -ENOMEM; + } +#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; } -static void __iomem * __ioremap_com(unsigned long addr, unsigned long pa, - unsigned long ea, unsigned long size, +/** + * __ioremap_at - Low level function to establish the page tables + * for an IO mapping + */ +void __iomem * __ioremap_at(phys_addr_t pa, void *ea, unsigned long size, unsigned long flags) { unsigned long i; + /* Make sure we have the base flags */ if ((flags & _PAGE_PRESENT) == 0) flags |= pgprot_val(PAGE_KERNEL); + /* Non-cacheable page cannot be coherent */ + if (flags & _PAGE_NO_CACHE) + flags &= ~_PAGE_COHERENT; + + /* We don't support the 4K PFN hack with ioremap */ + if (flags & _PAGE_4K_PFN) + return NULL; + + WARN_ON(pa & ~PAGE_MASK); + WARN_ON(((unsigned long)ea) & ~PAGE_MASK); + WARN_ON(size & ~PAGE_MASK); + for (i = 0; i < size; i += PAGE_SIZE) - if (map_io_page(ea+i, pa+i, flags)) + if (map_kernel_page((unsigned long)ea+i, pa+i, flags)) return NULL; - return (void __iomem *) (ea + (addr & ~PAGE_MASK)); + return (void __iomem *)ea; } - -void __iomem * -ioremap(unsigned long addr, unsigned long size) +/** + * __iounmap_from - Low level function to tear down the page tables + * for an IO mapping. This is used for mappings that + * are manipulated manually, like partial unmapping of + * PCI IOs or ISA space. + */ +void __iounmap_at(void *ea, unsigned long size) { - return __ioremap(addr, size, _PAGE_NO_CACHE | _PAGE_GUARDED); + WARN_ON(((unsigned long)ea) & ~PAGE_MASK); + WARN_ON(size & ~PAGE_MASK); + + unmap_kernel_range((unsigned long)ea, size); } -void __iomem * __ioremap(unsigned long addr, unsigned long size, - unsigned long flags) +void __iomem * __ioremap_caller(phys_addr_t addr, unsigned long size, + unsigned long flags, void *caller) { - unsigned long pa, ea; + phys_addr_t paligned; void __iomem *ret; /* @@ -168,166 +229,662 @@ void __iomem * __ioremap(unsigned long addr, unsigned long size, * IMALLOC_END * */ - pa = addr & PAGE_MASK; - size = PAGE_ALIGN(addr + size) - pa; + paligned = addr & PAGE_MASK; + size = PAGE_ALIGN(addr + size) - paligned; - if (size == 0) + if ((size == 0) || (paligned == 0)) return NULL; if (mem_init_done) { struct vm_struct *area; - area = im_get_free_area(size); + + area = __get_vm_area_caller(size, VM_IOREMAP, + ioremap_bot, IOREMAP_END, + caller); if (area == NULL) return NULL; - ea = (unsigned long)(area->addr); - ret = __ioremap_com(addr, pa, ea, size, flags); + + area->phys_addr = paligned; + ret = __ioremap_at(paligned, area->addr, size, flags); if (!ret) - im_free(area->addr); + vunmap(area->addr); } else { - ea = ioremap_bot; - ret = __ioremap_com(addr, pa, ea, size, flags); + ret = __ioremap_at(paligned, (void *)ioremap_bot, size, flags); if (ret) ioremap_bot += size; } + + if (ret) + ret += addr & ~PAGE_MASK; return ret; } -#define IS_PAGE_ALIGNED(_val) ((_val) == ((_val) & PAGE_MASK)) +void __iomem * __ioremap(phys_addr_t addr, unsigned long size, + unsigned long flags) +{ + return __ioremap_caller(addr, size, flags, __builtin_return_address(0)); +} -int __ioremap_explicit(unsigned long pa, unsigned long ea, - unsigned long size, unsigned long flags) +void __iomem * ioremap(phys_addr_t addr, unsigned long size) { - struct vm_struct *area; - void __iomem *ret; - - /* For now, require page-aligned values for pa, ea, and size */ - if (!IS_PAGE_ALIGNED(pa) || !IS_PAGE_ALIGNED(ea) || - !IS_PAGE_ALIGNED(size)) { - printk(KERN_ERR "unaligned value in %s\n", __FUNCTION__); - return 1; - } - - if (!mem_init_done) { - /* Two things to consider in this case: - * 1) No records will be kept (imalloc, etc) that the region - * has been remapped - * 2) It won't be easy to iounmap() the region later (because - * of 1) - */ - ; - } else { - area = im_get_area(ea, size, - IM_REGION_UNUSED|IM_REGION_SUBSET|IM_REGION_EXISTS); - if (area == NULL) { - /* Expected when PHB-dlpar is in play */ - return 1; - } - if (ea != (unsigned long) area->addr) { - printk(KERN_ERR "unexpected addr return from " - "im_get_area\n"); - return 1; - } - } - - ret = __ioremap_com(pa, pa, ea, size, flags); - if (ret == NULL) { - printk(KERN_ERR "ioremap_explicit() allocation failure !\n"); - return 1; - } - if (ret != (void *) ea) { - printk(KERN_ERR "__ioremap_com() returned unexpected addr\n"); - return 1; - } + unsigned long flags = _PAGE_NO_CACHE | _PAGE_GUARDED; + void *caller = __builtin_return_address(0); - return 0; + if (ppc_md.ioremap) + return ppc_md.ioremap(addr, size, flags, caller); + return __ioremap_caller(addr, size, flags, caller); +} + +void __iomem * ioremap_wc(phys_addr_t addr, unsigned long size) +{ + unsigned long flags = _PAGE_NO_CACHE; + void *caller = __builtin_return_address(0); + + if (ppc_md.ioremap) + return ppc_md.ioremap(addr, size, flags, caller); + return __ioremap_caller(addr, size, flags, caller); +} + +void __iomem * ioremap_prot(phys_addr_t addr, unsigned long size, + unsigned long flags) +{ + void *caller = __builtin_return_address(0); + + /* writeable implies dirty for kernel addresses */ + if (flags & _PAGE_RW) + flags |= _PAGE_DIRTY; + + /* we don't want to let _PAGE_USER and _PAGE_EXEC leak out */ + flags &= ~(_PAGE_USER | _PAGE_EXEC); + +#ifdef _PAGE_BAP_SR + /* _PAGE_USER contains _PAGE_BAP_SR on BookE using the new PTE format + * which means that we just cleared supervisor access... oops ;-) This + * restores it + */ + flags |= _PAGE_BAP_SR; +#endif + + if (ppc_md.ioremap) + return ppc_md.ioremap(addr, size, flags, caller); + return __ioremap_caller(addr, size, flags, caller); } + /* * Unmap an IO region and remove it from imalloc'd list. * Access to IO memory should be serialized by driver. - * This code is modeled after vmalloc code - unmap_vm_area() - * - * XXX what about calls before mem_init_done (ie python_countermeasures()) */ -void iounmap(volatile void __iomem *token) +void __iounmap(volatile void __iomem *token) { void *addr; if (!mem_init_done) return; - addr = (void *) ((unsigned long __force) token & PAGE_MASK); + addr = (void *) ((unsigned long __force) + PCI_FIX_ADDR(token) & PAGE_MASK); + if ((unsigned long)addr < ioremap_bot) { + printk(KERN_WARNING "Attempt to iounmap early bolted mapping" + " at 0x%p\n", addr); + return; + } + vunmap(addr); +} - im_free(addr); +void iounmap(volatile void __iomem *token) +{ + if (ppc_md.iounmap) + ppc_md.iounmap(token); + else + __iounmap(token); } -static int iounmap_subset_regions(unsigned long addr, unsigned long size) +EXPORT_SYMBOL(ioremap); +EXPORT_SYMBOL(ioremap_wc); +EXPORT_SYMBOL(ioremap_prot); +EXPORT_SYMBOL(__ioremap); +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) { - struct vm_struct *area; +#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)); +} - /* Check whether subsets of this region exist */ - area = im_get_area(addr, size, IM_REGION_SUPERSET); - if (area == NULL) - return 1; +#ifdef CONFIG_PPC_64K_PAGES +static pte_t *get_from_cache(struct mm_struct *mm) +{ + void *pte_frag, *ret; - while (area) { - iounmap((void __iomem *) area->addr); - area = im_get_area(addr, size, - IM_REGION_SUPERSET); + 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; } - - return 0; + spin_unlock(&mm->page_table_lock); + return (pte_t *)ret; } -int iounmap_explicit(volatile void __iomem *start, unsigned long size) +static pte_t *__alloc_for_cache(struct mm_struct *mm, int kernel) { - struct vm_struct *area; - unsigned long addr; - int rc; - - addr = (unsigned long __force) start & PAGE_MASK; + 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; + } - /* Verify that the region either exists or is a subset of an existing - * region. In the latter case, split the parent region to create - * the exact region + 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. */ - area = im_get_area(addr, size, - IM_REGION_EXISTS | IM_REGION_SUBSET); - if (area == NULL) { - /* Determine whether subset regions exist. If so, unmap */ - rc = iounmap_subset_regions(addr, size); - if (rc) { - printk(KERN_ERR - "%s() cannot unmap nonexistent range 0x%lx\n", - __FUNCTION__, addr); - return 1; + 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 { - iounmap((void __iomem *) area->addr); + 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 /* - * FIXME! This can't be right: - iounmap(area->addr); - * Maybe it should be "iounmap(area);" + * If we didn't had the splitting flag set, go and flush the + * HPTE entries. */ - return 0; + 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)); +} -EXPORT_SYMBOL(ioremap); -EXPORT_SYMBOL(__ioremap); -EXPORT_SYMBOL(iounmap); +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); +} -void __iomem * reserve_phb_iospace(unsigned long size) +/* + * 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) { - void __iomem *virt_addr; - - if (phbs_io_bot >= IMALLOC_BASE) - panic("reserve_phb_iospace(): phb io space overflow\n"); - - virt_addr = (void __iomem *) phbs_io_bot; - phbs_io_bot += size; + 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 virt_addr; + return 1; } +#endif /* CONFIG_TRANSPARENT_HUGEPAGE */ |