diff options
Diffstat (limited to 'arch/powerpc/mm/pgtable.c')
| -rw-r--r-- | arch/powerpc/mm/pgtable.c | 459 |
1 files changed, 382 insertions, 77 deletions
diff --git a/arch/powerpc/mm/pgtable.c b/arch/powerpc/mm/pgtable.c index edda589795c3..56d7e8960e77 100644 --- a/arch/powerpc/mm/pgtable.c +++ b/arch/powerpc/mm/pgtable.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* * This file contains common routines for dealing with free of page tables * Along with common page table handling code @@ -14,25 +15,26 @@ * * Dave Engebretsen <engebret@us.ibm.com> * Rework for PPC64 port. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. */ #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> -#include <asm/pgalloc.h> #include <asm/tlbflush.h> #include <asm/tlb.h> +#include <asm/hugetlb.h> +#include <asm/pte-walk.h> + +#ifdef CONFIG_PPC64 +#define PGD_ALIGN (sizeof(pgd_t) * MAX_PTRS_PER_PGD) +#else +#define PGD_ALIGN PAGE_SIZE +#endif -#include "mmu_decl.h" +pgd_t swapper_pg_dir[MAX_PTRS_PER_PGD] __section(".bss..page_aligned") __aligned(PGD_ALIGN); static inline int is_exec_fault(void) { @@ -41,17 +43,22 @@ static inline int is_exec_fault(void) /* We only try to do i/d cache coherency on stuff that looks like * reasonably "normal" PTEs. We currently require a PTE to be present - * and we avoid _PAGE_SPECIAL and _PAGE_NO_CACHE. We also only do that + * and we avoid _PAGE_SPECIAL and cache inhibited pte. We also only do that * on userspace PTEs */ -static inline int pte_looks_normal(pte_t pte) +static inline int pte_looks_normal(pte_t pte, unsigned long addr) { - return (pte_val(pte) & - (_PAGE_PRESENT | _PAGE_SPECIAL | _PAGE_NO_CACHE | _PAGE_USER)) == - (_PAGE_PRESENT | _PAGE_USER); + + if (pte_present(pte) && !pte_special(pte)) { + if (pte_ci(pte)) + return 0; + if (!is_kernel_addr(addr)) + return 1; + } + return 0; } -struct page * maybe_pte_to_page(pte_t pte) +static struct folio *maybe_pte_to_folio(pte_t pte) { unsigned long pfn = pte_pfn(pte); struct page *page; @@ -61,10 +68,10 @@ struct page * maybe_pte_to_page(pte_t pte) page = pfn_to_page(pfn); if (PageReserved(page)) return NULL; - return page; + return page_folio(page); } -#if defined(CONFIG_PPC_STD_MMU) || _PAGE_EXEC == 0 +#ifdef CONFIG_PPC_BOOK3S /* Server-style MMU handles coherency when hashing if HW exec permission * is supposed per page (currently 64-bit only). If not, then, we always @@ -72,84 +79,85 @@ 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_hash(pte_t pte, unsigned long addr) { pte = __pte(pte_val(pte) & ~_PAGE_HPTEFLAGS); - if (pte_looks_normal(pte) && !(cpu_has_feature(CPU_FTR_COHERENT_ICACHE) || - cpu_has_feature(CPU_FTR_NOEXECUTE))) { - struct page *pg = maybe_pte_to_page(pte); - if (!pg) + if (pte_looks_normal(pte, addr) && !(cpu_has_feature(CPU_FTR_COHERENT_ICACHE) || + cpu_has_feature(CPU_FTR_NOEXECUTE))) { + struct folio *folio = maybe_pte_to_folio(pte); + if (!folio) 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); + if (!test_bit(PG_dcache_clean, &folio->flags.f)) { + flush_dcache_icache_folio(folio); + set_bit(PG_dcache_clean, &folio->flags.f); } } return pte; } -static pte_t set_access_flags_filter(pte_t pte, struct vm_area_struct *vma, - int dirty) -{ - return pte; -} +#else /* CONFIG_PPC_BOOK3S */ -#else /* defined(CONFIG_PPC_STD_MMU) || _PAGE_EXEC == 0 */ +static pte_t set_pte_filter_hash(pte_t pte, unsigned long addr) { return pte; } + +#endif /* CONFIG_PPC_BOOK3S */ /* Embedded type MMU with HW exec support. This is a bit more complicated * 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. + * + * This is also called once for the folio. So only work with folio->flags here. */ -static pte_t set_pte_filter(pte_t pte, unsigned long addr) +static inline pte_t set_pte_filter(pte_t pte, unsigned long addr) { - struct page *pg; + struct folio *folio; + + if (radix_enabled()) + return pte; + + if (mmu_has_feature(MMU_FTR_HPTE_TABLE)) + return set_pte_filter_hash(pte, addr); /* No exec permission in the first place, move on */ - if (!(pte_val(pte) & _PAGE_EXEC) || !pte_looks_normal(pte)) + if (!pte_exec(pte) || !pte_looks_normal(pte, addr)) return pte; /* If you set _PAGE_EXEC on weird pages you're on your own */ - pg = maybe_pte_to_page(pte); - if (unlikely(!pg)) + folio = maybe_pte_to_folio(pte); + if (unlikely(!folio)) return pte; /* If the page clean, we move on */ - if (test_bit(PG_arch_1, &pg->flags)) + if (test_bit(PG_dcache_clean, &folio->flags.f)) return pte; /* If it's an exec fault, we flush the cache and make it clean */ if (is_exec_fault()) { - flush_dcache_icache_page(pg); - set_bit(PG_arch_1, &pg->flags); + flush_dcache_icache_folio(folio); + set_bit(PG_dcache_clean, &folio->flags.f); return pte; } /* Else, we filter out _PAGE_EXEC */ - return __pte(pte_val(pte) & ~_PAGE_EXEC); + return pte_exprotect(pte); } static pte_t set_access_flags_filter(pte_t pte, struct vm_area_struct *vma, int dirty) { - struct page *pg; + struct folio *folio; + + if (IS_ENABLED(CONFIG_PPC_BOOK3S_64)) + return pte; + + if (mmu_has_feature(MMU_FTR_HPTE_TABLE)) + return pte; /* So here, we only care about exec faults, as we use them * to recover lost _PAGE_EXEC and perform I$/D$ coherency * if necessary. Also if _PAGE_EXEC is already set, same deal, * we just bail out */ - if (dirty || (pte_val(pte) & _PAGE_EXEC) || !is_exec_fault()) + if (dirty || pte_exec(pte) || !is_exec_fault()) return pte; #ifdef CONFIG_DEBUG_VM @@ -162,41 +170,67 @@ static pte_t set_access_flags_filter(pte_t pte, struct vm_area_struct *vma, #endif /* CONFIG_DEBUG_VM */ /* If you set _PAGE_EXEC on weird pages you're on your own */ - pg = maybe_pte_to_page(pte); - if (unlikely(!pg)) + folio = maybe_pte_to_folio(pte); + if (unlikely(!folio)) goto bail; /* If the page is already clean, we move on */ - if (test_bit(PG_arch_1, &pg->flags)) + if (test_bit(PG_dcache_clean, &folio->flags.f)) goto bail; - /* Clean the page and set PG_arch_1 */ - flush_dcache_icache_page(pg); - set_bit(PG_arch_1, &pg->flags); + /* Clean the page and set PG_dcache_clean */ + flush_dcache_icache_folio(folio); + set_bit(PG_dcache_clean, &folio->flags.f); bail: - return __pte(pte_val(pte) | _PAGE_EXEC); + return pte_mkexec(pte); } -#endif /* !(defined(CONFIG_PPC_STD_MMU) || _PAGE_EXEC == 0) */ - /* * set_pte stores a linux PTE into the linux page table. */ -void set_pte_at(struct mm_struct *mm, unsigned long addr, pte_t *ptep, - pte_t pte) +void set_ptes(struct mm_struct *mm, unsigned long addr, pte_t *ptep, + pte_t pte, unsigned int nr) { -#ifdef CONFIG_DEBUG_VM - WARN_ON(pte_present(*ptep)); -#endif + /* Note: mm->context.id might not yet have been assigned as * this context might not have been activated yet when this - * is called. + * is called. Filter the pte value and use the filtered value + * to setup all the ptes in the range. */ pte = set_pte_filter(pte, addr); - /* Perform the setting of the PTE */ - __set_pte_at(mm, addr, ptep, pte, 0); + /* + * We don't need to call arch_enter/leave_lazy_mmu_mode() + * because we expect set_ptes to be only be used on not present + * and not hw_valid ptes. Hence there is no translation cache flush + * involved that need to be batched. + */ + for (;;) { + + /* + * Make sure hardware valid bit is not set. We don't do + * tlb flush for this update. + */ + VM_WARN_ON(pte_hw_valid(*ptep) && !pte_protnone(*ptep)); + + /* Perform the setting of the PTE */ + __set_pte_at(mm, addr, ptep, pte, 0); + if (--nr == 0) + break; + ptep++; + addr += PAGE_SIZE; + pte = pte_next_pfn(pte); + } +} + +void unmap_kernel_page(unsigned long va) +{ + pmd_t *pmdp = pmd_off_k(va); + pte_t *ptep = pte_offset_kernel(pmdp, va); + + pte_clear(&init_mm, va, ptep); + flush_tlb_kernel_range(va, va + PAGE_SIZE); } /* @@ -213,38 +247,309 @@ int ptep_set_access_flags(struct vm_area_struct *vma, unsigned long address, entry = set_access_flags_filter(entry, vma, dirty); changed = !pte_same(*(ptep), entry); if (changed) { - if (!is_vm_hugetlb_page(vma)) - assert_pte_locked(vma->vm_mm, address); - __ptep_set_access_flags(ptep, entry); - flush_tlb_page_nohash(vma, address); + assert_pte_locked(vma->vm_mm, address); + __ptep_set_access_flags(vma, ptep, entry, + address, mmu_virtual_psize); + } + return changed; +} + +#ifdef CONFIG_HUGETLB_PAGE +int huge_ptep_set_access_flags(struct vm_area_struct *vma, + unsigned long addr, pte_t *ptep, + pte_t pte, int dirty) +{ +#ifdef HUGETLB_NEED_PRELOAD + /* + * The "return 1" forces a call of update_mmu_cache, which will write a + * TLB entry. Without this, platforms that don't do a write of the TLB + * entry in the TLB miss handler asm will fault ad infinitum. + */ + ptep_set_access_flags(vma, addr, ptep, pte, dirty); + return 1; +#else + int changed, psize; + + pte = set_access_flags_filter(pte, vma, dirty); + changed = !pte_same(*(ptep), pte); + if (changed) { + +#ifdef CONFIG_PPC_BOOK3S_64 + struct hstate *h = hstate_vma(vma); + + psize = hstate_get_psize(h); +#ifdef CONFIG_DEBUG_VM + assert_spin_locked(huge_pte_lockptr(h, vma->vm_mm, ptep)); +#endif + +#else + /* + * Not used on non book3s64 platforms. + * 8xx compares it with mmu_virtual_psize to + * know if it is a huge page or not. + */ + psize = MMU_PAGE_COUNT; +#endif + __ptep_set_access_flags(vma, ptep, pte, addr, psize); } return changed; +#endif +} + +#if defined(CONFIG_PPC_8xx) + +#if defined(CONFIG_SPLIT_PTE_PTLOCKS) || defined(CONFIG_SPLIT_PMD_PTLOCKS) +/* We need the same lock to protect the PMD table and the two PTE tables. */ +#error "8M hugetlb folios are incompatible with split page table locks" +#endif + +static void __set_huge_pte_at(pmd_t *pmd, pte_t *ptep, pte_basic_t val) +{ + pte_basic_t *entry = (pte_basic_t *)ptep; + int num, i; + + /* + * Make sure hardware valid bit is not set. We don't do + * tlb flush for this update. + */ + VM_WARN_ON(pte_hw_valid(*ptep) && !pte_protnone(*ptep)); + + num = number_of_cells_per_pte(pmd, val, 1); + + for (i = 0; i < num; i++, entry++, val += SZ_4K) + *entry = val; +} + +void set_huge_pte_at(struct mm_struct *mm, unsigned long addr, pte_t *ptep, + pte_t pte, unsigned long sz) +{ + pmd_t *pmdp = pmd_off(mm, addr); + + pte = set_pte_filter(pte, addr); + + if (sz == SZ_8M) { /* Flag both PMD entries as 8M and fill both page tables */ + *pmdp = __pmd(pmd_val(*pmdp) | _PMD_PAGE_8M); + *(pmdp + 1) = __pmd(pmd_val(*(pmdp + 1)) | _PMD_PAGE_8M); + + __set_huge_pte_at(pmdp, pte_offset_kernel(pmdp, 0), pte_val(pte)); + __set_huge_pte_at(pmdp, pte_offset_kernel(pmdp + 1, 0), pte_val(pte) + SZ_4M); + } else { + __set_huge_pte_at(pmdp, ptep, pte_val(pte)); + } +} +#else +void set_huge_pte_at(struct mm_struct *mm, unsigned long addr, pte_t *ptep, + pte_t pte, unsigned long sz) +{ + unsigned long pdsize; + int i; + + pte = set_pte_filter(pte, addr); + + /* + * Make sure hardware valid bit is not set. We don't do + * tlb flush for this update. + */ + VM_WARN_ON(pte_hw_valid(*ptep) && !pte_protnone(*ptep)); + + if (sz < PMD_SIZE) + pdsize = PAGE_SIZE; + else if (sz < PUD_SIZE) + pdsize = PMD_SIZE; + else if (sz < P4D_SIZE) + pdsize = PUD_SIZE; + else if (sz < PGDIR_SIZE) + pdsize = P4D_SIZE; + else + pdsize = PGDIR_SIZE; + + for (i = 0; i < sz / pdsize; i++, ptep++, addr += pdsize) { + __set_pte_at(mm, addr, ptep, pte, 0); + pte = __pte(pte_val(pte) + ((unsigned long long)pdsize / PAGE_SIZE << PFN_PTE_SHIFT)); + } } +#endif +#endif /* CONFIG_HUGETLB_PAGE */ #ifdef CONFIG_DEBUG_VM void assert_pte_locked(struct mm_struct *mm, unsigned long addr) { pgd_t *pgd; + p4d_t *p4d; pud_t *pud; pmd_t *pmd; + pte_t *pte; + spinlock_t *ptl; if (mm == &init_mm) return; pgd = mm->pgd + pgd_index(addr); BUG_ON(pgd_none(*pgd)); - pud = pud_offset(pgd, addr); + p4d = p4d_offset(pgd, addr); + BUG_ON(p4d_none(*p4d)); + pud = pud_offset(p4d, 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 to none to force page fault/gup to take mmap_lock. 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)); + pte = pte_offset_map_ro_nolock(mm, pmd, addr, &ptl); + BUG_ON(!pte); + assert_spin_locked(ptl); + pte_unmap(pte); } #endif /* CONFIG_DEBUG_VM */ +unsigned long vmalloc_to_phys(void *va) +{ + unsigned long pfn = vmalloc_to_pfn(va); + + BUG_ON(!pfn); + return __pa(pfn_to_kaddr(pfn)) + offset_in_page(va); +} +EXPORT_SYMBOL_GPL(vmalloc_to_phys); + +/* + * We have 3 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 _PAGE_PTE set + * + * So long as we atomically load page table pointers we are safe against teardown, + * we can follow the address down to the page and take a ref on it. + * This function need to be called with interrupts disabled. We use this variant + * when we have MSR[EE] = 0 but the paca->irq_soft_mask = IRQS_ENABLED + */ +pte_t *__find_linux_pte(pgd_t *pgdir, unsigned long ea, + bool *is_thp, unsigned *hpage_shift) +{ + pgd_t *pgdp; +#ifdef CONFIG_PPC64 + p4d_t p4d, *p4dp; + pud_t pud, *pudp; +#endif + pmd_t pmd, *pmdp; + pte_t *ret_pte; + unsigned pdshift; + + if (hpage_shift) + *hpage_shift = 0; + + if (is_thp) + *is_thp = false; + + /* + * 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. + * Top level is an exception because it is folded into p4d. + * + * On PPC32, P4D/PUD/PMD are folded into PGD so go straight to + * PMD level. + */ + pgdp = pgdir + pgd_index(ea); +#ifdef CONFIG_PPC64 + p4dp = p4d_offset(pgdp, ea); + p4d = READ_ONCE(*p4dp); + pdshift = P4D_SHIFT; + + if (p4d_none(p4d)) + return NULL; + + if (p4d_leaf(p4d)) { + ret_pte = (pte_t *)p4dp; + goto out; + } + + /* + * 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(&p4d, ea); + pud = READ_ONCE(*pudp); + + if (pud_none(pud)) + return NULL; + + if (pud_leaf(pud)) { + ret_pte = (pte_t *)pudp; + goto out; + } + + pmdp = pmd_offset(&pud, ea); +#else + pmdp = pmd_offset(pud_offset(p4d_offset(pgdp, ea), ea), ea); +#endif + pdshift = PMD_SHIFT; + pmd = READ_ONCE(*pmdp); + + /* + * A hugepage collapse is captured by this condition, see + * pmdp_collapse_flush. + */ + if (pmd_none(pmd)) + return NULL; + +#ifdef CONFIG_PPC_BOOK3S_64 + /* + * A hugepage split is captured by this condition, see + * pmdp_invalidate. + * + * Huge page modification can be caught here too. + */ + if (pmd_is_serializing(pmd)) + return NULL; +#endif + + if (pmd_trans_huge(pmd)) { + if (is_thp) + *is_thp = true; + ret_pte = (pte_t *)pmdp; + goto out; + } + + if (pmd_leaf(pmd)) { + ret_pte = (pte_t *)pmdp; + goto out; + } + + return pte_offset_kernel(&pmd, ea); + +out: + if (hpage_shift) + *hpage_shift = pdshift; + return ret_pte; +} +EXPORT_SYMBOL_GPL(__find_linux_pte); + +/* Note due to the way vm flags are laid out, the bits are XWR */ +const pgprot_t protection_map[16] = { + [VM_NONE] = PAGE_NONE, + [VM_READ] = PAGE_READONLY, + [VM_WRITE] = PAGE_COPY, + [VM_WRITE | VM_READ] = PAGE_COPY, + [VM_EXEC] = PAGE_EXECONLY_X, + [VM_EXEC | VM_READ] = PAGE_READONLY_X, + [VM_EXEC | VM_WRITE] = PAGE_COPY_X, + [VM_EXEC | VM_WRITE | VM_READ] = PAGE_COPY_X, + [VM_SHARED] = PAGE_NONE, + [VM_SHARED | VM_READ] = PAGE_READONLY, + [VM_SHARED | VM_WRITE] = PAGE_SHARED, + [VM_SHARED | VM_WRITE | VM_READ] = PAGE_SHARED, + [VM_SHARED | VM_EXEC] = PAGE_EXECONLY_X, + [VM_SHARED | VM_EXEC | VM_READ] = PAGE_READONLY_X, + [VM_SHARED | VM_EXEC | VM_WRITE] = PAGE_SHARED_X, + [VM_SHARED | VM_EXEC | VM_WRITE | VM_READ] = PAGE_SHARED_X +}; + +#ifndef CONFIG_PPC_BOOK3S_64 +DECLARE_VM_GET_PAGE_PROT +#endif |