nagai/kernel
3
0
Fork 0
forked from mirrors/linux
Code Pull requests Activity

mm/mmu_notifier: use structure for invalidate_range_start/end calls v2

Browse source 
To avoid having to change many call sites everytime we want to add a
parameter use a structure to group all parameters for the mmu_notifier
invalidate_range_start/end cakks.  No functional changes with this patch.

[akpm@linux-foundation.org: coding style fixes]
Link: http://lkml.kernel.org/r/20181205053628.3210-3-jglisse@redhat.com
Signed-off-by: Jérôme Glisse <jglisse@redhat.com>
Acked-by: Christian König <christian.koenig@amd.com>
Acked-by: Jan Kara <jack@suse.cz>
Cc: Matthew Wilcox <mawilcox@microsoft.com>
Cc: Ross Zwisler <zwisler@kernel.org>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krcmar <rkrcmar@redhat.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Felix Kuehling <felix.kuehling@amd.com>
Cc: Ralph Campbell <rcampbell@nvidia.com>
Cc: John Hubbard <jhubbard@nvidia.com>
From: Jérôme Glisse <jglisse@redhat.com>
Subject: mm/mmu_notifier: use structure for invalidate_range_start/end calls v3

fix build warning in migrate.c when CONFIG_MMU_NOTIFIER=n

Link: http://lkml.kernel.org/r/20181213171330.8489-3-jglisse@redhat.com
Signed-off-by: Jérôme Glisse <jglisse@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This commit is contained in:
Jérôme Glisse 2018-12-28 00:38:09 -08:00 committed by Linus Torvalds
parent 5d6527a784
commit ac46d4f3c4
17 changed files with 262 additions and 250 deletions
repo.diff.show_diff_stats Download patch file Download diff file Expand all files Collapse all files

8
fs/dax.c
View file

@ -779,7 +779,8 @@ static void dax_entry_mkclean(struct address_space *mapping, pgoff_t index,
i_mmap_lock_read(mapping); i_mmap_lock_read(mapping);
vma_interval_tree_foreach(vma, &mapping->i_mmap, index, index) { vma_interval_tree_foreach(vma, &mapping->i_mmap, index, index) {
unsigned long address, start, end; struct mmu_notifier_range range;
unsigned long address;
cond_resched(); cond_resched();
@ -793,7 +794,8 @@ static void dax_entry_mkclean(struct address_space *mapping, pgoff_t index,
* call mmu_notifier_invalidate_range_start() on our behalf * call mmu_notifier_invalidate_range_start() on our behalf
* before taking any lock. * before taking any lock.
*/ */
if (follow_pte_pmd(vma->vm_mm, address, &start, &end, &ptep, &pmdp, &ptl)) if (follow_pte_pmd(vma->vm_mm, address, &range,
&ptep, &pmdp, &ptl))
continue; continue;
/* /*
@ -835,7 +837,7 @@ static void dax_entry_mkclean(struct address_space *mapping, pgoff_t index,
pte_unmap_unlock(ptep, ptl); pte_unmap_unlock(ptep, ptl);
} }
mmu_notifier_invalidate_range_end(vma->vm_mm, start, end); mmu_notifier_invalidate_range_end(&range);
} }
i_mmap_unlock_read(mapping); i_mmap_unlock_read(mapping);
} }

7
fs/proc/task_mmu.c
View file

@ -1096,6 +1096,7 @@ static ssize_t clear_refs_write(struct file *file, const char __user *buf,
return -ESRCH; return -ESRCH;
mm = get_task_mm(task); mm = get_task_mm(task);
if (mm) { if (mm) {
struct mmu_notifier_range range;
struct clear_refs_private cp = { struct clear_refs_private cp = {
.type = type, .type = type,
}; };
@ -1139,11 +1140,13 @@ static ssize_t clear_refs_write(struct file *file, const char __user *buf,
downgrade_write(&mm->mmap_sem); downgrade_write(&mm->mmap_sem);
break; break;
} }
mmu_notifier_invalidate_range_start(mm, 0, -1);
mmu_notifier_range_init(&range, mm, 0, -1UL);
mmu_notifier_invalidate_range_start(&range);
} }
walk_page_range(0, mm->highest_vm_end, &clear_refs_walk); walk_page_range(0, mm->highest_vm_end, &clear_refs_walk);
if (type == CLEAR_REFS_SOFT_DIRTY) if (type == CLEAR_REFS_SOFT_DIRTY)
mmu_notifier_invalidate_range_end(mm, 0, -1); mmu_notifier_invalidate_range_end(&range);
tlb_finish_mmu(&tlb, 0, -1); tlb_finish_mmu(&tlb, 0, -1);
up_read(&mm->mmap_sem); up_read(&mm->mmap_sem);
out_mm: out_mm:

4
include/linux/mm.h
View file

@ -1451,6 +1451,8 @@ struct mm_walk {
void *private; void *private;
}; };
struct mmu_notifier_range;
int walk_page_range(unsigned long addr, unsigned long end, int walk_page_range(unsigned long addr, unsigned long end,
struct mm_walk *walk); struct mm_walk *walk);
int walk_page_vma(struct vm_area_struct *vma, struct mm_walk *walk); int walk_page_vma(struct vm_area_struct *vma, struct mm_walk *walk);
@ -1459,7 +1461,7 @@ void free_pgd_range(struct mmu_gather *tlb, unsigned long addr,
int copy_page_range(struct mm_struct *dst, struct mm_struct *src, int copy_page_range(struct mm_struct *dst, struct mm_struct *src,
struct vm_area_struct *vma); struct vm_area_struct *vma);
int follow_pte_pmd(struct mm_struct *mm, unsigned long address, int follow_pte_pmd(struct mm_struct *mm, unsigned long address,
unsigned long *start, unsigned long *end, struct mmu_notifier_range *range,
pte_t **ptepp, pmd_t **pmdpp, spinlock_t **ptlp); pte_t **ptepp, pmd_t **pmdpp, spinlock_t **ptlp);
int follow_pfn(struct vm_area_struct *vma, unsigned long address, int follow_pfn(struct vm_area_struct *vma, unsigned long address,
unsigned long *pfn); unsigned long *pfn);

87
include/linux/mmu_notifier.h
View file

@ -220,11 +220,8 @@ extern int __mmu_notifier_test_young(struct mm_struct *mm,
unsigned long address); unsigned long address);
extern void __mmu_notifier_change_pte(struct mm_struct *mm, extern void __mmu_notifier_change_pte(struct mm_struct *mm,
unsigned long address, pte_t pte); unsigned long address, pte_t pte);
extern int __mmu_notifier_invalidate_range_start(struct mm_struct *mm, extern int __mmu_notifier_invalidate_range_start(struct mmu_notifier_range *r);
unsigned long start, unsigned long end, extern void __mmu_notifier_invalidate_range_end(struct mmu_notifier_range *r,
bool blockable);
extern void __mmu_notifier_invalidate_range_end(struct mm_struct *mm,
unsigned long start, unsigned long end,
bool only_end); bool only_end);
extern void __mmu_notifier_invalidate_range(struct mm_struct *mm, extern void __mmu_notifier_invalidate_range(struct mm_struct *mm,
unsigned long start, unsigned long end); unsigned long start, unsigned long end);
@ -268,33 +265,37 @@ static inline void mmu_notifier_change_pte(struct mm_struct *mm,
__mmu_notifier_change_pte(mm, address, pte); __mmu_notifier_change_pte(mm, address, pte);
} }
static inline void mmu_notifier_invalidate_range_start(struct mm_struct *mm, static inline void
unsigned long start, unsigned long end) mmu_notifier_invalidate_range_start(struct mmu_notifier_range *range)
{ {
if (mm_has_notifiers(mm)) if (mm_has_notifiers(range->mm)) {
__mmu_notifier_invalidate_range_start(mm, start, end, true); range->blockable = true;
__mmu_notifier_invalidate_range_start(range);
}
} }
static inline int mmu_notifier_invalidate_range_start_nonblock(struct mm_struct *mm, static inline int
unsigned long start, unsigned long end) mmu_notifier_invalidate_range_start_nonblock(struct mmu_notifier_range *range)
{ {
if (mm_has_notifiers(mm)) if (mm_has_notifiers(range->mm)) {
return __mmu_notifier_invalidate_range_start(mm, start, end, false); range->blockable = false;
return __mmu_notifier_invalidate_range_start(range);
}
return 0; return 0;
} }
static inline void mmu_notifier_invalidate_range_end(struct mm_struct *mm, static inline void
unsigned long start, unsigned long end) mmu_notifier_invalidate_range_end(struct mmu_notifier_range *range)
{ {
if (mm_has_notifiers(mm)) if (mm_has_notifiers(range->mm))
__mmu_notifier_invalidate_range_end(mm, start, end, false); __mmu_notifier_invalidate_range_end(range, false);
} }
static inline void mmu_notifier_invalidate_range_only_end(struct mm_struct *mm, static inline void
unsigned long start, unsigned long end) mmu_notifier_invalidate_range_only_end(struct mmu_notifier_range *range)
{ {
if (mm_has_notifiers(mm)) if (mm_has_notifiers(range->mm))
__mmu_notifier_invalidate_range_end(mm, start, end, true); __mmu_notifier_invalidate_range_end(range, true);
} }
static inline void mmu_notifier_invalidate_range(struct mm_struct *mm, static inline void mmu_notifier_invalidate_range(struct mm_struct *mm,
@ -315,6 +316,17 @@ static inline void mmu_notifier_mm_destroy(struct mm_struct *mm)
__mmu_notifier_mm_destroy(mm); __mmu_notifier_mm_destroy(mm);
} }
static inline void mmu_notifier_range_init(struct mmu_notifier_range *range,
struct mm_struct *mm,
unsigned long start,
unsigned long end)
{
range->mm = mm;
range->start = start;
range->end = end;
}
#define ptep_clear_flush_young_notify(__vma, __address, __ptep) \ #define ptep_clear_flush_young_notify(__vma, __address, __ptep) \
({ \ ({ \
int __young; \ int __young; \
@ -427,6 +439,23 @@ extern void mmu_notifier_call_srcu(struct rcu_head *rcu,
#else /* CONFIG_MMU_NOTIFIER */ #else /* CONFIG_MMU_NOTIFIER */
struct mmu_notifier_range {
unsigned long start;
unsigned long end;
};
static inline void _mmu_notifier_range_init(struct mmu_notifier_range *range,
unsigned long start,
unsigned long end)
{
range->start = start;
range->end = end;
}
#define mmu_notifier_range_init(range, mm, start, end) \
_mmu_notifier_range_init(range, start, end)
static inline int mm_has_notifiers(struct mm_struct *mm) static inline int mm_has_notifiers(struct mm_struct *mm)
{ {
return 0; return 0;
@ -454,24 +483,24 @@ static inline void mmu_notifier_change_pte(struct mm_struct *mm,
{ {
} }
static inline void mmu_notifier_invalidate_range_start(struct mm_struct *mm, static inline void
unsigned long start, unsigned long end) mmu_notifier_invalidate_range_start(struct mmu_notifier_range *range)
{ {
} }
static inline int mmu_notifier_invalidate_range_start_nonblock(struct mm_struct *mm, static inline int
unsigned long start, unsigned long end) mmu_notifier_invalidate_range_start_nonblock(struct mmu_notifier_range *range)
{ {
return 0; return 0;
} }
static inline void mmu_notifier_invalidate_range_end(struct mm_struct *mm, static inline
unsigned long start, unsigned long end) void mmu_notifier_invalidate_range_end(struct mmu_notifier_range *range)
{ {
} }
static inline void mmu_notifier_invalidate_range_only_end(struct mm_struct *mm, static inline void
unsigned long start, unsigned long end) mmu_notifier_invalidate_range_only_end(struct mmu_notifier_range *range)
{ {
} }

10
kernel/events/uprobes.c
View file

@ -171,11 +171,11 @@ static int __replace_page(struct vm_area_struct *vma, unsigned long addr,
.address = addr, .address = addr,
}; };
int err; int err;
/* For mmu_notifiers */ struct mmu_notifier_range range;
const unsigned long mmun_start = addr;
const unsigned long mmun_end = addr + PAGE_SIZE;
struct mem_cgroup *memcg; struct mem_cgroup *memcg;
mmu_notifier_range_init(&range, mm, addr, addr + PAGE_SIZE);
VM_BUG_ON_PAGE(PageTransHuge(old_page), old_page); VM_BUG_ON_PAGE(PageTransHuge(old_page), old_page);
err = mem_cgroup_try_charge(new_page, vma->vm_mm, GFP_KERNEL, &memcg, err = mem_cgroup_try_charge(new_page, vma->vm_mm, GFP_KERNEL, &memcg,
@ -186,7 +186,7 @@ static int __replace_page(struct vm_area_struct *vma, unsigned long addr,
/* For try_to_free_swap() and munlock_vma_page() below */ /* For try_to_free_swap() and munlock_vma_page() below */
lock_page(old_page); lock_page(old_page);
mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end); mmu_notifier_invalidate_range_start(&range);
err = -EAGAIN; err = -EAGAIN;
if (!page_vma_mapped_walk(&pvmw)) { if (!page_vma_mapped_walk(&pvmw)) {
mem_cgroup_cancel_charge(new_page, memcg, false); mem_cgroup_cancel_charge(new_page, memcg, false);
@ -220,7 +220,7 @@ static int __replace_page(struct vm_area_struct *vma, unsigned long addr,
err = 0; err = 0;
unlock: unlock:
mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end); mmu_notifier_invalidate_range_end(&range);
unlock_page(old_page); unlock_page(old_page);
return err; return err;
} }

54
mm/huge_memory.c
View file

@ -1134,8 +1134,7 @@ static vm_fault_t do_huge_pmd_wp_page_fallback(struct vm_fault *vmf,
int i; int i;
vm_fault_t ret = 0; vm_fault_t ret = 0;
struct page **pages; struct page **pages;
unsigned long mmun_start; /* For mmu_notifiers */ struct mmu_notifier_range range;
unsigned long mmun_end; /* For mmu_notifiers */
pages = kmalloc_array(HPAGE_PMD_NR, sizeof(struct page *), pages = kmalloc_array(HPAGE_PMD_NR, sizeof(struct page *),
GFP_KERNEL); GFP_KERNEL);
@ -1173,9 +1172,9 @@ static vm_fault_t do_huge_pmd_wp_page_fallback(struct vm_fault *vmf,
cond_resched(); cond_resched();
} }
mmun_start = haddr; mmu_notifier_range_init(&range, vma->vm_mm, haddr,
mmun_end = haddr + HPAGE_PMD_SIZE; haddr + HPAGE_PMD_SIZE);
mmu_notifier_invalidate_range_start(vma->vm_mm, mmun_start, mmun_end); mmu_notifier_invalidate_range_start(&range);
vmf->ptl = pmd_lock(vma->vm_mm, vmf->pmd); vmf->ptl = pmd_lock(vma->vm_mm, vmf->pmd);
if (unlikely(!pmd_same(*vmf->pmd, orig_pmd))) if (unlikely(!pmd_same(*vmf->pmd, orig_pmd)))
@ -1220,8 +1219,7 @@ static vm_fault_t do_huge_pmd_wp_page_fallback(struct vm_fault *vmf,
* No need to double call mmu_notifier->invalidate_range() callback as * No need to double call mmu_notifier->invalidate_range() callback as
* the above pmdp_huge_clear_flush_notify() did already call it. * the above pmdp_huge_clear_flush_notify() did already call it.
*/ */
mmu_notifier_invalidate_range_only_end(vma->vm_mm, mmun_start, mmu_notifier_invalidate_range_only_end(&range);
mmun_end);
ret |= VM_FAULT_WRITE; ret |= VM_FAULT_WRITE;
put_page(page); put_page(page);
@ -1231,7 +1229,7 @@ static vm_fault_t do_huge_pmd_wp_page_fallback(struct vm_fault *vmf,
out_free_pages: out_free_pages:
spin_unlock(vmf->ptl); spin_unlock(vmf->ptl);
mmu_notifier_invalidate_range_end(vma->vm_mm, mmun_start, mmun_end); mmu_notifier_invalidate_range_end(&range);
for (i = 0; i < HPAGE_PMD_NR; i++) { for (i = 0; i < HPAGE_PMD_NR; i++) {
memcg = (void *)page_private(pages[i]); memcg = (void *)page_private(pages[i]);
set_page_private(pages[i], 0); set_page_private(pages[i], 0);
@ -1248,8 +1246,7 @@ vm_fault_t do_huge_pmd_wp_page(struct vm_fault *vmf, pmd_t orig_pmd)
struct page *page = NULL, *new_page; struct page *page = NULL, *new_page;
struct mem_cgroup *memcg; struct mem_cgroup *memcg;
unsigned long haddr = vmf->address & HPAGE_PMD_MASK; unsigned long haddr = vmf->address & HPAGE_PMD_MASK;
unsigned long mmun_start; /* For mmu_notifiers */ struct mmu_notifier_range range;
unsigned long mmun_end; /* For mmu_notifiers */
gfp_t huge_gfp; /* for allocation and charge */ gfp_t huge_gfp; /* for allocation and charge */
vm_fault_t ret = 0; vm_fault_t ret = 0;
@ -1338,9 +1335,9 @@ vm_fault_t do_huge_pmd_wp_page(struct vm_fault *vmf, pmd_t orig_pmd)
vma, HPAGE_PMD_NR); vma, HPAGE_PMD_NR);
__SetPageUptodate(new_page); __SetPageUptodate(new_page);
mmun_start = haddr; mmu_notifier_range_init(&range, vma->vm_mm, haddr,
mmun_end = haddr + HPAGE_PMD_SIZE; haddr + HPAGE_PMD_SIZE);
mmu_notifier_invalidate_range_start(vma->vm_mm, mmun_start, mmun_end); mmu_notifier_invalidate_range_start(&range);
spin_lock(vmf->ptl); spin_lock(vmf->ptl);
if (page) if (page)
@ -1375,8 +1372,7 @@ vm_fault_t do_huge_pmd_wp_page(struct vm_fault *vmf, pmd_t orig_pmd)
* No need to double call mmu_notifier->invalidate_range() callback as * No need to double call mmu_notifier->invalidate_range() callback as
* the above pmdp_huge_clear_flush_notify() did already call it. * the above pmdp_huge_clear_flush_notify() did already call it.
*/ */
mmu_notifier_invalidate_range_only_end(vma->vm_mm, mmun_start, mmu_notifier_invalidate_range_only_end(&range);
mmun_end);
out: out:
return ret; return ret;
out_unlock: out_unlock:
@ -2015,14 +2011,15 @@ void __split_huge_pud(struct vm_area_struct *vma, pud_t *pud,
unsigned long address) unsigned long address)
{ {
spinlock_t *ptl; spinlock_t *ptl;
struct mm_struct *mm = vma->vm_mm; struct mmu_notifier_range range;
unsigned long haddr = address & HPAGE_PUD_MASK;
mmu_notifier_invalidate_range_start(mm, haddr, haddr + HPAGE_PUD_SIZE); mmu_notifier_range_init(&range, vma->vm_mm, address & HPAGE_PUD_MASK,
ptl = pud_lock(mm, pud); (address & HPAGE_PUD_MASK) + HPAGE_PUD_SIZE);
mmu_notifier_invalidate_range_start(&range);
ptl = pud_lock(vma->vm_mm, pud);
if (unlikely(!pud_trans_huge(*pud) && !pud_devmap(*pud))) if (unlikely(!pud_trans_huge(*pud) && !pud_devmap(*pud)))
goto out; goto out;
__split_huge_pud_locked(vma, pud, haddr); __split_huge_pud_locked(vma, pud, range.start);
out: out:
spin_unlock(ptl); spin_unlock(ptl);
@ -2030,8 +2027,7 @@ void __split_huge_pud(struct vm_area_struct *vma, pud_t *pud,
* No need to double call mmu_notifier->invalidate_range() callback as * No need to double call mmu_notifier->invalidate_range() callback as
* the above pudp_huge_clear_flush_notify() did already call it. * the above pudp_huge_clear_flush_notify() did already call it.
*/ */
mmu_notifier_invalidate_range_only_end(mm, haddr, haddr + mmu_notifier_invalidate_range_only_end(&range);
HPAGE_PUD_SIZE);
} }
#endif /* CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD */ #endif /* CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD */
@ -2233,11 +2229,12 @@ void __split_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
unsigned long address, bool freeze, struct page *page) unsigned long address, bool freeze, struct page *page)
{ {
spinlock_t *ptl; spinlock_t *ptl;
struct mm_struct *mm = vma->vm_mm; struct mmu_notifier_range range;
unsigned long haddr = address & HPAGE_PMD_MASK;
mmu_notifier_invalidate_range_start(mm, haddr, haddr + HPAGE_PMD_SIZE); mmu_notifier_range_init(&range, vma->vm_mm, address & HPAGE_PMD_MASK,
ptl = pmd_lock(mm, pmd); (address & HPAGE_PMD_MASK) + HPAGE_PMD_SIZE);
mmu_notifier_invalidate_range_start(&range);
ptl = pmd_lock(vma->vm_mm, pmd);
/* /*
* If caller asks to setup a migration entries, we need a page to check * If caller asks to setup a migration entries, we need a page to check
@ -2253,7 +2250,7 @@ void __split_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
clear_page_mlock(page); clear_page_mlock(page);
} else if (!(pmd_devmap(*pmd) || is_pmd_migration_entry(*pmd))) } else if (!(pmd_devmap(*pmd) || is_pmd_migration_entry(*pmd)))
goto out; goto out;
__split_huge_pmd_locked(vma, pmd, haddr, freeze); __split_huge_pmd_locked(vma, pmd, range.start, freeze);
out: out:
spin_unlock(ptl); spin_unlock(ptl);
/* /*
@ -2269,8 +2266,7 @@ void __split_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
* any further changes to individual pte will notify. So no need * any further changes to individual pte will notify. So no need
* to call mmu_notifier->invalidate_range() * to call mmu_notifier->invalidate_range()
*/ */
mmu_notifier_invalidate_range_only_end(mm, haddr, haddr + mmu_notifier_invalidate_range_only_end(&range);
HPAGE_PMD_SIZE);
} }
void split_huge_pmd_address(struct vm_area_struct *vma, unsigned long address, void split_huge_pmd_address(struct vm_area_struct *vma, unsigned long address,

52
mm/hugetlb.c
View file

@ -3240,16 +3240,16 @@ int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src,
int cow; int cow;
struct hstate *h = hstate_vma(vma); struct hstate *h = hstate_vma(vma);
unsigned long sz = huge_page_size(h); unsigned long sz = huge_page_size(h);
unsigned long mmun_start; /* For mmu_notifiers */ struct mmu_notifier_range range;
unsigned long mmun_end; /* For mmu_notifiers */
int ret = 0; int ret = 0;
cow = (vma->vm_flags & (VM_SHARED | VM_MAYWRITE)) == VM_MAYWRITE; cow = (vma->vm_flags & (VM_SHARED | VM_MAYWRITE)) == VM_MAYWRITE;
mmun_start = vma->vm_start; if (cow) {
mmun_end = vma->vm_end; mmu_notifier_range_init(&range, src, vma->vm_start,
if (cow) vma->vm_end);
mmu_notifier_invalidate_range_start(src, mmun_start, mmun_end); mmu_notifier_invalidate_range_start(&range);
}
for (addr = vma->vm_start; addr < vma->vm_end; addr += sz) { for (addr = vma->vm_start; addr < vma->vm_end; addr += sz) {
spinlock_t *src_ptl, *dst_ptl; spinlock_t *src_ptl, *dst_ptl;
@ -3325,7 +3325,7 @@ int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src,
} }
if (cow) if (cow)
mmu_notifier_invalidate_range_end(src, mmun_start, mmun_end); mmu_notifier_invalidate_range_end(&range);
return ret; return ret;
} }
@ -3342,8 +3342,7 @@ void __unmap_hugepage_range(struct mmu_gather *tlb, struct vm_area_struct *vma,
struct page *page; struct page *page;
struct hstate *h = hstate_vma(vma); struct hstate *h = hstate_vma(vma);
unsigned long sz = huge_page_size(h); unsigned long sz = huge_page_size(h);
unsigned long mmun_start = start; /* For mmu_notifiers */ struct mmu_notifier_range range;
unsigned long mmun_end = end; /* For mmu_notifiers */
WARN_ON(!is_vm_hugetlb_page(vma)); WARN_ON(!is_vm_hugetlb_page(vma));
BUG_ON(start & ~huge_page_mask(h)); BUG_ON(start & ~huge_page_mask(h));
@ -3359,8 +3358,9 @@ void __unmap_hugepage_range(struct mmu_gather *tlb, struct vm_area_struct *vma,
/* /*
* If sharing possible, alert mmu notifiers of worst case. * If sharing possible, alert mmu notifiers of worst case.
*/ */
adjust_range_if_pmd_sharing_possible(vma, &mmun_start, &mmun_end); mmu_notifier_range_init(&range, mm, start, end);
mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end); adjust_range_if_pmd_sharing_possible(vma, &range.start, &range.end);
mmu_notifier_invalidate_range_start(&range);
address = start; address = start;
for (; address < end; address += sz) { for (; address < end; address += sz) {
ptep = huge_pte_offset(mm, address, sz); ptep = huge_pte_offset(mm, address, sz);
@ -3428,7 +3428,7 @@ void __unmap_hugepage_range(struct mmu_gather *tlb, struct vm_area_struct *vma,
if (ref_page) if (ref_page)
break; break;
} }
mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end); mmu_notifier_invalidate_range_end(&range);
tlb_end_vma(tlb, vma); tlb_end_vma(tlb, vma);
} }
@ -3546,9 +3546,8 @@ static vm_fault_t hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma,
struct page *old_page, *new_page; struct page *old_page, *new_page;
int outside_reserve = 0; int outside_reserve = 0;
vm_fault_t ret = 0; vm_fault_t ret = 0;
unsigned long mmun_start; /* For mmu_notifiers */
unsigned long mmun_end; /* For mmu_notifiers */
unsigned long haddr = address & huge_page_mask(h); unsigned long haddr = address & huge_page_mask(h);
struct mmu_notifier_range range;
pte = huge_ptep_get(ptep); pte = huge_ptep_get(ptep);
old_page = pte_page(pte); old_page = pte_page(pte);
@ -3627,9 +3626,8 @@ static vm_fault_t hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma,
__SetPageUptodate(new_page); __SetPageUptodate(new_page);
set_page_huge_active(new_page); set_page_huge_active(new_page);
mmun_start = haddr; mmu_notifier_range_init(&range, mm, haddr, haddr + huge_page_size(h));
mmun_end = mmun_start + huge_page_size(h); mmu_notifier_invalidate_range_start(&range);
mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);
/* /*
* Retake the page table lock to check for racing updates * Retake the page table lock to check for racing updates
@ -3642,7 +3640,7 @@ static vm_fault_t hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma,
/* Break COW */ /* Break COW */
huge_ptep_clear_flush(vma, haddr, ptep); huge_ptep_clear_flush(vma, haddr, ptep);
mmu_notifier_invalidate_range(mm, mmun_start, mmun_end); mmu_notifier_invalidate_range(mm, range.start, range.end);
set_huge_pte_at(mm, haddr, ptep, set_huge_pte_at(mm, haddr, ptep,
make_huge_pte(vma, new_page, 1)); make_huge_pte(vma, new_page, 1));
page_remove_rmap(old_page, true); page_remove_rmap(old_page, true);
@ -3651,7 +3649,7 @@ static vm_fault_t hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma,
new_page = old_page; new_page = old_page;
} }
spin_unlock(ptl); spin_unlock(ptl);
mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end); mmu_notifier_invalidate_range_end(&range);
out_release_all: out_release_all:
restore_reserve_on_error(h, vma, haddr, new_page); restore_reserve_on_error(h, vma, haddr, new_page);
put_page(new_page); put_page(new_page);
@ -4340,21 +4338,21 @@ unsigned long hugetlb_change_protection(struct vm_area_struct *vma,
pte_t pte; pte_t pte;
struct hstate *h = hstate_vma(vma); struct hstate *h = hstate_vma(vma);
unsigned long pages = 0; unsigned long pages = 0;
unsigned long f_start = start;
unsigned long f_end = end;
bool shared_pmd = false; bool shared_pmd = false;
struct mmu_notifier_range range;
/* /*
* In the case of shared PMDs, the area to flush could be beyond * In the case of shared PMDs, the area to flush could be beyond
* start/end. Set f_start/f_end to cover the maximum possible * start/end. Set range.start/range.end to cover the maximum possible
* range if PMD sharing is possible. * range if PMD sharing is possible.
*/ */
adjust_range_if_pmd_sharing_possible(vma, &f_start, &f_end); mmu_notifier_range_init(&range, mm, start, end);
adjust_range_if_pmd_sharing_possible(vma, &range.start, &range.end);
BUG_ON(address >= end); BUG_ON(address >= end);
flush_cache_range(vma, f_start, f_end); flush_cache_range(vma, range.start, range.end);
mmu_notifier_invalidate_range_start(mm, f_start, f_end); mmu_notifier_invalidate_range_start(&range);
i_mmap_lock_write(vma->vm_file->f_mapping); i_mmap_lock_write(vma->vm_file->f_mapping);
for (; address < end; address += huge_page_size(h)) { for (; address < end; address += huge_page_size(h)) {
spinlock_t *ptl; spinlock_t *ptl;
@ -4405,7 +4403,7 @@ unsigned long hugetlb_change_protection(struct vm_area_struct *vma,
* did unshare a page of pmds, flush the range corresponding to the pud. * did unshare a page of pmds, flush the range corresponding to the pud.
*/ */
if (shared_pmd) if (shared_pmd)
flush_hugetlb_tlb_range(vma, f_start, f_end); flush_hugetlb_tlb_range(vma, range.start, range.end);
else else
flush_hugetlb_tlb_range(vma, start, end); flush_hugetlb_tlb_range(vma, start, end);
/* /*
@ -4415,7 +4413,7 @@ unsigned long hugetlb_change_protection(struct vm_area_struct *vma,
* See Documentation/vm/mmu_notifier.rst * See Documentation/vm/mmu_notifier.rst
*/ */
i_mmap_unlock_write(vma->vm_file->f_mapping); i_mmap_unlock_write(vma->vm_file->f_mapping);
mmu_notifier_invalidate_range_end(mm, f_start, f_end); mmu_notifier_invalidate_range_end(&range);
return pages << h->order; return pages << h->order;
} }

10
mm/khugepaged.c
View file

@ -944,8 +944,7 @@ static void collapse_huge_page(struct mm_struct *mm,
int isolated = 0, result = 0; int isolated = 0, result = 0;
struct mem_cgroup *memcg; struct mem_cgroup *memcg;
struct vm_area_struct *vma; struct vm_area_struct *vma;
unsigned long mmun_start; /* For mmu_notifiers */ struct mmu_notifier_range range;
unsigned long mmun_end; /* For mmu_notifiers */
gfp_t gfp; gfp_t gfp;
VM_BUG_ON(address & ~HPAGE_PMD_MASK); VM_BUG_ON(address & ~HPAGE_PMD_MASK);
@ -1017,9 +1016,8 @@ static void collapse_huge_page(struct mm_struct *mm,
pte = pte_offset_map(pmd, address); pte = pte_offset_map(pmd, address);
pte_ptl = pte_lockptr(mm, pmd); pte_ptl = pte_lockptr(mm, pmd);
mmun_start = address; mmu_notifier_range_init(&range, mm, address, address + HPAGE_PMD_SIZE);
mmun_end = address + HPAGE_PMD_SIZE; mmu_notifier_invalidate_range_start(&range);
mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);
pmd_ptl = pmd_lock(mm, pmd); /* probably unnecessary */ pmd_ptl = pmd_lock(mm, pmd); /* probably unnecessary */
/* /*
* After this gup_fast can't run anymore. This also removes * After this gup_fast can't run anymore. This also removes
@ -1029,7 +1027,7 @@ static void collapse_huge_page(struct mm_struct *mm,
*/ */
_pmd = pmdp_collapse_flush(vma, address, pmd); _pmd = pmdp_collapse_flush(vma, address, pmd);
spin_unlock(pmd_ptl); spin_unlock(pmd_ptl);
mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end); mmu_notifier_invalidate_range_end(&range);
spin_lock(pte_ptl); spin_lock(pte_ptl);
isolated = __collapse_huge_page_isolate(vma, address, pte); isolated = __collapse_huge_page_isolate(vma, address, pte);

21
mm/ksm.c
View file

@ -1042,8 +1042,7 @@ static int write_protect_page(struct vm_area_struct *vma, struct page *page,
}; };
int swapped; int swapped;
int err = -EFAULT; int err = -EFAULT;
unsigned long mmun_start; /* For mmu_notifiers */ struct mmu_notifier_range range;
unsigned long mmun_end; /* For mmu_notifiers */
pvmw.address = page_address_in_vma(page, vma); pvmw.address = page_address_in_vma(page, vma);
if (pvmw.address == -EFAULT) if (pvmw.address == -EFAULT)
@ -1051,9 +1050,9 @@ static int write_protect_page(struct vm_area_struct *vma, struct page *page,
BUG_ON(PageTransCompound(page)); BUG_ON(PageTransCompound(page));
mmun_start = pvmw.address; mmu_notifier_range_init(&range, mm, pvmw.address,
mmun_end = pvmw.address + PAGE_SIZE; pvmw.address + PAGE_SIZE);
mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end); mmu_notifier_invalidate_range_start(&range);
if (!page_vma_mapped_walk(&pvmw)) if (!page_vma_mapped_walk(&pvmw))
goto out_mn; goto out_mn;
@ -1105,7 +1104,7 @@ static int write_protect_page(struct vm_area_struct *vma, struct page *page,
out_unlock: out_unlock:
page_vma_mapped_walk_done(&pvmw); page_vma_mapped_walk_done(&pvmw);
out_mn: out_mn:
mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end); mmu_notifier_invalidate_range_end(&range);
out: out:
return err; return err;
} }
@ -1129,8 +1128,7 @@ static int replace_page(struct vm_area_struct *vma, struct page *page,
spinlock_t *ptl; spinlock_t *ptl;
unsigned long addr; unsigned long addr;
int err = -EFAULT; int err = -EFAULT;
unsigned long mmun_start; /* For mmu_notifiers */ struct mmu_notifier_range range;
unsigned long mmun_end; /* For mmu_notifiers */
addr = page_address_in_vma(page, vma); addr = page_address_in_vma(page, vma);
if (addr == -EFAULT) if (addr == -EFAULT)
@ -1140,9 +1138,8 @@ static int replace_page(struct vm_area_struct *vma, struct page *page,
if (!pmd) if (!pmd)
goto out; goto out;
mmun_start = addr; mmu_notifier_range_init(&range, mm, addr, addr + PAGE_SIZE);
mmun_end = addr + PAGE_SIZE; mmu_notifier_invalidate_range_start(&range);
mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);
ptep = pte_offset_map_lock(mm, pmd, addr, &ptl); ptep = pte_offset_map_lock(mm, pmd, addr, &ptl);
if (!pte_same(*ptep, orig_pte)) { if (!pte_same(*ptep, orig_pte)) {
@ -1188,7 +1185,7 @@ static int replace_page(struct vm_area_struct *vma, struct page *page,
pte_unmap_unlock(ptep, ptl); pte_unmap_unlock(ptep, ptl);
err = 0; err = 0;
out_mn: out_mn:
mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end); mmu_notifier_invalidate_range_end(&range);
out: out:
return err; return err;
} }

21
mm/madvise.c
View file

@ -458,29 +458,30 @@ static void madvise_free_page_range(struct mmu_gather *tlb,
static int madvise_free_single_vma(struct vm_area_struct *vma, static int madvise_free_single_vma(struct vm_area_struct *vma,
unsigned long start_addr, unsigned long end_addr) unsigned long start_addr, unsigned long end_addr)
{ {
unsigned long start, end;
struct mm_struct *mm = vma->vm_mm; struct mm_struct *mm = vma->vm_mm;
struct mmu_notifier_range range;
struct mmu_gather tlb; struct mmu_gather tlb;
/* MADV_FREE works for only anon vma at the moment */ /* MADV_FREE works for only anon vma at the moment */
if (!vma_is_anonymous(vma)) if (!vma_is_anonymous(vma))
return -EINVAL; return -EINVAL;
start = max(vma->vm_start, start_addr); range.start = max(vma->vm_start, start_addr);
if (start >= vma->vm_end) if (range.start >= vma->vm_end)
return -EINVAL; return -EINVAL;
end = min(vma->vm_end, end_addr); range.end = min(vma->vm_end, end_addr);
if (end <= vma->vm_start) if (range.end <= vma->vm_start)
return -EINVAL; return -EINVAL;
mmu_notifier_range_init(&range, mm, range.start, range.end);
lru_add_drain(); lru_add_drain();
tlb_gather_mmu(&tlb, mm, start, end); tlb_gather_mmu(&tlb, mm, range.start, range.end);
update_hiwater_rss(mm); update_hiwater_rss(mm);
mmu_notifier_invalidate_range_start(mm, start, end); mmu_notifier_invalidate_range_start(&range);
madvise_free_page_range(&tlb, vma, start, end); madvise_free_page_range(&tlb, vma, range.start, range.end);
mmu_notifier_invalidate_range_end(mm, start, end); mmu_notifier_invalidate_range_end(&range);
tlb_finish_mmu(&tlb, start, end); tlb_finish_mmu(&tlb, range.start, range.end);
return 0; return 0;
} }

97
mm/memory.c
View file

@ -973,8 +973,7 @@ int copy_page_range(struct mm_struct *dst_mm, struct mm_struct *src_mm,
unsigned long next; unsigned long next;
unsigned long addr = vma->vm_start; unsigned long addr = vma->vm_start;
unsigned long end = vma->vm_end; unsigned long end = vma->vm_end;
unsigned long mmun_start; /* For mmu_notifiers */ struct mmu_notifier_range range;
unsigned long mmun_end; /* For mmu_notifiers */
bool is_cow; bool is_cow;
int ret; int ret;
@ -1008,11 +1007,11 @@ int copy_page_range(struct mm_struct *dst_mm, struct mm_struct *src_mm,
* is_cow_mapping() returns true. * is_cow_mapping() returns true.
*/ */
is_cow = is_cow_mapping(vma->vm_flags); is_cow = is_cow_mapping(vma->vm_flags);
mmun_start = addr;
mmun_end = end; if (is_cow) {
if (is_cow) mmu_notifier_range_init(&range, src_mm, addr, end);
mmu_notifier_invalidate_range_start(src_mm, mmun_start, mmu_notifier_invalidate_range_start(&range);
mmun_end); }
ret = 0; ret = 0;
dst_pgd = pgd_offset(dst_mm, addr); dst_pgd = pgd_offset(dst_mm, addr);
@ -1029,7 +1028,7 @@ int copy_page_range(struct mm_struct *dst_mm, struct mm_struct *src_mm,
} while (dst_pgd++, src_pgd++, addr = next, addr != end); } while (dst_pgd++, src_pgd++, addr = next, addr != end);
if (is_cow) if (is_cow)
mmu_notifier_invalidate_range_end(src_mm, mmun_start, mmun_end); mmu_notifier_invalidate_range_end(&range);
return ret; return ret;
} }
@ -1332,12 +1331,13 @@ void unmap_vmas(struct mmu_gather *tlb,
struct vm_area_struct *vma, unsigned long start_addr, struct vm_area_struct *vma, unsigned long start_addr,
unsigned long end_addr) unsigned long end_addr)
{ {
struct mm_struct *mm = vma->vm_mm; struct mmu_notifier_range range;
mmu_notifier_invalidate_range_start(mm, start_addr, end_addr); mmu_notifier_range_init(&range, vma->vm_mm, start_addr, end_addr);
mmu_notifier_invalidate_range_start(&range);
for ( ; vma && vma->vm_start < end_addr; vma = vma->vm_next) for ( ; vma && vma->vm_start < end_addr; vma = vma->vm_next)
unmap_single_vma(tlb, vma, start_addr, end_addr, NULL); unmap_single_vma(tlb, vma, start_addr, end_addr, NULL);
mmu_notifier_invalidate_range_end(mm, start_addr, end_addr); mmu_notifier_invalidate_range_end(&range);
} }
/** /**
@ -1351,18 +1351,18 @@ void unmap_vmas(struct mmu_gather *tlb,
void zap_page_range(struct vm_area_struct *vma, unsigned long start, void zap_page_range(struct vm_area_struct *vma, unsigned long start,
unsigned long size) unsigned long size)
{ {
struct mm_struct *mm = vma->vm_mm; struct mmu_notifier_range range;
struct mmu_gather tlb; struct mmu_gather tlb;
unsigned long end = start + size;
lru_add_drain(); lru_add_drain();
tlb_gather_mmu(&tlb, mm, start, end); mmu_notifier_range_init(&range, vma->vm_mm, start, start + size);
update_hiwater_rss(mm); tlb_gather_mmu(&tlb, vma->vm_mm, start, range.end);
mmu_notifier_invalidate_range_start(mm, start, end); update_hiwater_rss(vma->vm_mm);
for ( ; vma && vma->vm_start < end; vma = vma->vm_next) mmu_notifier_invalidate_range_start(&range);
unmap_single_vma(&tlb, vma, start, end, NULL); for ( ; vma && vma->vm_start < range.end; vma = vma->vm_next)
mmu_notifier_invalidate_range_end(mm, start, end); unmap_single_vma(&tlb, vma, start, range.end, NULL);
tlb_finish_mmu(&tlb, start, end); mmu_notifier_invalidate_range_end(&range);
tlb_finish_mmu(&tlb, start, range.end);
} }
/** /**
@ -1377,17 +1377,17 @@ void zap_page_range(struct vm_area_struct *vma, unsigned long start,
static void zap_page_range_single(struct vm_area_struct *vma, unsigned long address, static void zap_page_range_single(struct vm_area_struct *vma, unsigned long address,
unsigned long size, struct zap_details *details) unsigned long size, struct zap_details *details)
{ {
struct mm_struct *mm = vma->vm_mm; struct mmu_notifier_range range;
struct mmu_gather tlb; struct mmu_gather tlb;
unsigned long end = address + size;
lru_add_drain(); lru_add_drain();
tlb_gather_mmu(&tlb, mm, address, end); mmu_notifier_range_init(&range, vma->vm_mm, address, address + size);
update_hiwater_rss(mm); tlb_gather_mmu(&tlb, vma->vm_mm, address, range.end);
mmu_notifier_invalidate_range_start(mm, address, end); update_hiwater_rss(vma->vm_mm);
unmap_single_vma(&tlb, vma, address, end, details); mmu_notifier_invalidate_range_start(&range);
mmu_notifier_invalidate_range_end(mm, address, end); unmap_single_vma(&tlb, vma, address, range.end, details);
tlb_finish_mmu(&tlb, address, end); mmu_notifier_invalidate_range_end(&range);
tlb_finish_mmu(&tlb, address, range.end);
} }
/** /**
@ -2247,9 +2247,8 @@ static vm_fault_t wp_page_copy(struct vm_fault *vmf)
struct page *new_page = NULL; struct page *new_page = NULL;
pte_t entry; pte_t entry;
int page_copied = 0; int page_copied = 0;
const unsigned long mmun_start = vmf->address & PAGE_MASK;
const unsigned long mmun_end = mmun_start + PAGE_SIZE;
struct mem_cgroup *memcg; struct mem_cgroup *memcg;
struct mmu_notifier_range range;
if (unlikely(anon_vma_prepare(vma))) if (unlikely(anon_vma_prepare(vma)))
goto oom; goto oom;
@ -2272,7 +2271,9 @@ static vm_fault_t wp_page_copy(struct vm_fault *vmf)
__SetPageUptodate(new_page); __SetPageUptodate(new_page);
mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end); mmu_notifier_range_init(&range, mm, vmf->address & PAGE_MASK,
(vmf->address & PAGE_MASK) + PAGE_SIZE);
mmu_notifier_invalidate_range_start(&range);
/* /*
* Re-check the pte - we dropped the lock * Re-check the pte - we dropped the lock
@ -2349,7 +2350,7 @@ static vm_fault_t wp_page_copy(struct vm_fault *vmf)
* No need to double call mmu_notifier->invalidate_range() callback as * No need to double call mmu_notifier->invalidate_range() callback as
* the above ptep_clear_flush_notify() did already call it. * the above ptep_clear_flush_notify() did already call it.
*/ */
mmu_notifier_invalidate_range_only_end(mm, mmun_start, mmun_end); mmu_notifier_invalidate_range_only_end(&range);
if (old_page) { if (old_page) {
/* /*
* Don't let another task, with possibly unlocked vma, * Don't let another task, with possibly unlocked vma,
@ -4030,7 +4031,7 @@ int __pmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long address)
#endif /* __PAGETABLE_PMD_FOLDED */ #endif /* __PAGETABLE_PMD_FOLDED */
static int __follow_pte_pmd(struct mm_struct *mm, unsigned long address, static int __follow_pte_pmd(struct mm_struct *mm, unsigned long address,
unsigned long *start, unsigned long *end, struct mmu_notifier_range *range,
pte_t **ptepp, pmd_t **pmdpp, spinlock_t **ptlp) pte_t **ptepp, pmd_t **pmdpp, spinlock_t **ptlp)
{ {
pgd_t *pgd; pgd_t *pgd;
@ -4058,10 +4059,10 @@ static int __follow_pte_pmd(struct mm_struct *mm, unsigned long address,
if (!pmdpp) if (!pmdpp)
goto out; goto out;
if (start && end) { if (range) {
*start = address & PMD_MASK; mmu_notifier_range_init(range, mm, address & PMD_MASK,
*end = *start + PMD_SIZE; (address & PMD_MASK) + PMD_SIZE);
mmu_notifier_invalidate_range_start(mm, *start, *end); mmu_notifier_invalidate_range_start(range);
} }
*ptlp = pmd_lock(mm, pmd); *ptlp = pmd_lock(mm, pmd);
if (pmd_huge(*pmd)) { if (pmd_huge(*pmd)) {
@ -4069,17 +4070,17 @@ static int __follow_pte_pmd(struct mm_struct *mm, unsigned long address,
return 0; return 0;
} }
spin_unlock(*ptlp); spin_unlock(*ptlp);
if (start && end) if (range)
mmu_notifier_invalidate_range_end(mm, *start, *end); mmu_notifier_invalidate_range_end(range);
} }
if (pmd_none(*pmd) || unlikely(pmd_bad(*pmd))) if (pmd_none(*pmd) || unlikely(pmd_bad(*pmd)))
goto out; goto out;
if (start && end) { if (range) {
*start = address & PAGE_MASK; range->start = address & PAGE_MASK;
*end = *start + PAGE_SIZE; range->end = range->start + PAGE_SIZE;
mmu_notifier_invalidate_range_start(mm, *start, *end); mmu_notifier_invalidate_range_start(range);
} }
ptep = pte_offset_map_lock(mm, pmd, address, ptlp); ptep = pte_offset_map_lock(mm, pmd, address, ptlp);
if (!pte_present(*ptep)) if (!pte_present(*ptep))
@ -4088,8 +4089,8 @@ static int __follow_pte_pmd(struct mm_struct *mm, unsigned long address,
return 0; return 0;
unlock: unlock:
pte_unmap_unlock(ptep, *ptlp); pte_unmap_unlock(ptep, *ptlp);
if (start && end) if (range)
mmu_notifier_invalidate_range_end(mm, *start, *end); mmu_notifier_invalidate_range_end(range);
out: out:
return -EINVAL; return -EINVAL;
} }
@ -4101,20 +4102,20 @@ static inline int follow_pte(struct mm_struct *mm, unsigned long address,
/* (void) is needed to make gcc happy */ /* (void) is needed to make gcc happy */
(void) __cond_lock(*ptlp, (void) __cond_lock(*ptlp,
!(res = __follow_pte_pmd(mm, address, NULL, NULL, !(res = __follow_pte_pmd(mm, address, NULL,
ptepp, NULL, ptlp))); ptepp, NULL, ptlp)));
return res; return res;
} }
int follow_pte_pmd(struct mm_struct *mm, unsigned long address, int follow_pte_pmd(struct mm_struct *mm, unsigned long address,
unsigned long *start, unsigned long *end, struct mmu_notifier_range *range,
pte_t **ptepp, pmd_t **pmdpp, spinlock_t **ptlp) pte_t **ptepp, pmd_t **pmdpp, spinlock_t **ptlp)
{ {
int res; int res;
/* (void) is needed to make gcc happy */ /* (void) is needed to make gcc happy */
(void) __cond_lock(*ptlp, (void) __cond_lock(*ptlp,
!(res = __follow_pte_pmd(mm, address, start, end, !(res = __follow_pte_pmd(mm, address, range,
ptepp, pmdpp, ptlp))); ptepp, pmdpp, ptlp)));
return res; return res;
} }

26
mm/migrate.c
View file

@ -2299,6 +2299,7 @@ static int migrate_vma_collect_pmd(pmd_t *pmdp,
*/ */
static void migrate_vma_collect(struct migrate_vma *migrate) static void migrate_vma_collect(struct migrate_vma *migrate)
{ {
struct mmu_notifier_range range;
struct mm_walk mm_walk; struct mm_walk mm_walk;
mm_walk.pmd_entry = migrate_vma_collect_pmd; mm_walk.pmd_entry = migrate_vma_collect_pmd;
@ -2310,13 +2311,11 @@ static void migrate_vma_collect(struct migrate_vma *migrate)
mm_walk.mm = migrate->vma->vm_mm; mm_walk.mm = migrate->vma->vm_mm;
mm_walk.private = migrate; mm_walk.private = migrate;
mmu_notifier_invalidate_range_start(mm_walk.mm, mmu_notifier_range_init(&range, mm_walk.mm, migrate->start,
migrate->start,
migrate->end); migrate->end);
mmu_notifier_invalidate_range_start(&range);
walk_page_range(migrate->start, migrate->end, &mm_walk); walk_page_range(migrate->start, migrate->end, &mm_walk);
mmu_notifier_invalidate_range_end(mm_walk.mm, mmu_notifier_invalidate_range_end(&range);
migrate->start,
migrate->end);
migrate->end = migrate->start + (migrate->npages << PAGE_SHIFT); migrate->end = migrate->start + (migrate->npages << PAGE_SHIFT);
} }
@ -2697,9 +2696,8 @@ static void migrate_vma_pages(struct migrate_vma *migrate)
{ {
const unsigned long npages = migrate->npages; const unsigned long npages = migrate->npages;
const unsigned long start = migrate->start; const unsigned long start = migrate->start;
struct vm_area_struct *vma = migrate->vma; struct mmu_notifier_range range;
struct mm_struct *mm = vma->vm_mm; unsigned long addr, i;
unsigned long addr, i, mmu_start;
bool notified = false; bool notified = false;
for (i = 0, addr = start; i < npages; addr += PAGE_SIZE, i++) { for (i = 0, addr = start; i < npages; addr += PAGE_SIZE, i++) {
@ -2718,11 +2716,12 @@ static void migrate_vma_pages(struct migrate_vma *migrate)
continue; continue;
} }
if (!notified) { if (!notified) {
mmu_start = addr;
notified = true; notified = true;
mmu_notifier_invalidate_range_start(mm,
mmu_start, mmu_notifier_range_init(&range,
migrate->end); migrate->vma->vm_mm,
addr, migrate->end);
mmu_notifier_invalidate_range_start(&range);
} }
migrate_vma_insert_page(migrate, addr, newpage, migrate_vma_insert_page(migrate, addr, newpage,
&migrate->src[i], &migrate->src[i],
@ -2763,8 +2762,7 @@ static void migrate_vma_pages(struct migrate_vma *migrate)
* did already call it. * did already call it.
*/ */
if (notified) if (notified)
mmu_notifier_invalidate_range_only_end(mm, mmu_start, mmu_notifier_invalidate_range_only_end(&range);
migrate->end);
} }
/* /*

35
mm/mmu_notifier.c
View file

@ -167,28 +167,20 @@ void __mmu_notifier_change_pte(struct mm_struct *mm, unsigned long address,
srcu_read_unlock(&srcu, id); srcu_read_unlock(&srcu, id);
} }
int __mmu_notifier_invalidate_range_start(struct mm_struct *mm, int __mmu_notifier_invalidate_range_start(struct mmu_notifier_range *range)
unsigned long start, unsigned long end,
bool blockable)
{ {
struct mmu_notifier_range _range, *range = &_range;
struct mmu_notifier *mn; struct mmu_notifier *mn;
int ret = 0; int ret = 0;
int id; int id;
range->blockable = blockable;
range->start = start;
range->end = end;
range->mm = mm;
id = srcu_read_lock(&srcu); id = srcu_read_lock(&srcu);
hlist_for_each_entry_rcu(mn, &mm->mmu_notifier_mm->list, hlist) { hlist_for_each_entry_rcu(mn, &range->mm->mmu_notifier_mm->list, hlist) {
if (mn->ops->invalidate_range_start) { if (mn->ops->invalidate_range_start) {
int _ret = mn->ops->invalidate_range_start(mn, range); int _ret = mn->ops->invalidate_range_start(mn, range);
if (_ret) { if (_ret) {
pr_info("%pS callback failed with %d in %sblockable context.\n", pr_info("%pS callback failed with %d in %sblockable context.\n",
mn->ops->invalidate_range_start, _ret, mn->ops->invalidate_range_start, _ret,
!blockable ? "non-" : ""); !range->blockable ? "non-" : "");
ret = _ret; ret = _ret;
} }
} }
@ -199,27 +191,14 @@ int __mmu_notifier_invalidate_range_start(struct mm_struct *mm,
} }
EXPORT_SYMBOL_GPL(__mmu_notifier_invalidate_range_start); EXPORT_SYMBOL_GPL(__mmu_notifier_invalidate_range_start);
void __mmu_notifier_invalidate_range_end(struct mm_struct *mm, void __mmu_notifier_invalidate_range_end(struct mmu_notifier_range *range,
unsigned long start,
unsigned long end,
bool only_end) bool only_end)
{ {
struct mmu_notifier_range _range, *range = &_range;
struct mmu_notifier *mn; struct mmu_notifier *mn;
int id; int id;
/*
* The end call back will never be call if the start refused to go
* through because of blockable was false so here assume that we
* can block.
*/
range->blockable = true;
range->start = start;
range->end = end;
range->mm = mm;
id = srcu_read_lock(&srcu); id = srcu_read_lock(&srcu);
hlist_for_each_entry_rcu(mn, &mm->mmu_notifier_mm->list, hlist) { hlist_for_each_entry_rcu(mn, &range->mm->mmu_notifier_mm->list, hlist) {
/* /*
* Call invalidate_range here too to avoid the need for the * Call invalidate_range here too to avoid the need for the
* subsystem of having to register an invalidate_range_end * subsystem of having to register an invalidate_range_end
@ -234,7 +213,9 @@ void __mmu_notifier_invalidate_range_end(struct mm_struct *mm,
* already happen under page table lock. * already happen under page table lock.
*/ */
if (!only_end && mn->ops->invalidate_range) if (!only_end && mn->ops->invalidate_range)
mn->ops->invalidate_range(mn, mm, start, end); mn->ops->invalidate_range(mn, range->mm,
range->start,
range->end);
if (mn->ops->invalidate_range_end) if (mn->ops->invalidate_range_end)
mn->ops->invalidate_range_end(mn, range); mn->ops->invalidate_range_end(mn, range);
} }

15
mm/mprotect.c
View file

@ -167,11 +167,12 @@ static inline unsigned long change_pmd_range(struct vm_area_struct *vma,
pgprot_t newprot, int dirty_accountable, int prot_numa) pgprot_t newprot, int dirty_accountable, int prot_numa)
{ {
pmd_t *pmd; pmd_t *pmd;
struct mm_struct *mm = vma->vm_mm;
unsigned long next; unsigned long next;
unsigned long pages = 0; unsigned long pages = 0;
unsigned long nr_huge_updates = 0; unsigned long nr_huge_updates = 0;
unsigned long mni_start = 0; struct mmu_notifier_range range;
range.start = 0;
pmd = pmd_offset(pud, addr); pmd = pmd_offset(pud, addr);
do { do {
@ -183,9 +184,9 @@ static inline unsigned long change_pmd_range(struct vm_area_struct *vma,
goto next; goto next;
/* invoke the mmu notifier if the pmd is populated */ /* invoke the mmu notifier if the pmd is populated */
if (!mni_start) { if (!range.start) {
mni_start = addr; mmu_notifier_range_init(&range, vma->vm_mm, addr, end);
mmu_notifier_invalidate_range_start(mm, mni_start, end); mmu_notifier_invalidate_range_start(&range);
} }
if (is_swap_pmd(*pmd) || pmd_trans_huge(*pmd) || pmd_devmap(*pmd)) { if (is_swap_pmd(*pmd) || pmd_trans_huge(*pmd) || pmd_devmap(*pmd)) {
@ -214,8 +215,8 @@ static inline unsigned long change_pmd_range(struct vm_area_struct *vma,
cond_resched(); cond_resched();
} while (pmd++, addr = next, addr != end); } while (pmd++, addr = next, addr != end);
if (mni_start) if (range.start)
mmu_notifier_invalidate_range_end(mm, mni_start, end); mmu_notifier_invalidate_range_end(&range);
if (nr_huge_updates) if (nr_huge_updates)
count_vm_numa_events(NUMA_HUGE_PTE_UPDATES, nr_huge_updates); count_vm_numa_events(NUMA_HUGE_PTE_UPDATES, nr_huge_updates);

10
mm/mremap.c
View file

@ -197,16 +197,14 @@ unsigned long move_page_tables(struct vm_area_struct *vma,
bool need_rmap_locks) bool need_rmap_locks)
{ {
unsigned long extent, next, old_end; unsigned long extent, next, old_end;
struct mmu_notifier_range range;
pmd_t *old_pmd, *new_pmd; pmd_t *old_pmd, *new_pmd;
unsigned long mmun_start; /* For mmu_notifiers */
unsigned long mmun_end; /* For mmu_notifiers */
old_end = old_addr + len; old_end = old_addr + len;
flush_cache_range(vma, old_addr, old_end); flush_cache_range(vma, old_addr, old_end);
mmun_start = old_addr; mmu_notifier_range_init(&range, vma->vm_mm, old_addr, old_end);
mmun_end = old_end; mmu_notifier_invalidate_range_start(&range);
mmu_notifier_invalidate_range_start(vma->vm_mm, mmun_start, mmun_end);
for (; old_addr < old_end; old_addr += extent, new_addr += extent) { for (; old_addr < old_end; old_addr += extent, new_addr += extent) {
cond_resched(); cond_resched();
@ -247,7 +245,7 @@ unsigned long move_page_tables(struct vm_area_struct *vma,
new_pmd, new_addr, need_rmap_locks); new_pmd, new_addr, need_rmap_locks);
} }
mmu_notifier_invalidate_range_end(vma->vm_mm, mmun_start, mmun_end); mmu_notifier_invalidate_range_end(&range);
return len + old_addr - old_end; /* how much done */ return len + old_addr - old_end; /* how much done */
} }

17
mm/oom_kill.c
View file

@ -528,19 +528,20 @@ bool __oom_reap_task_mm(struct mm_struct *mm)
* count elevated without a good reason. * count elevated without a good reason.
*/ */
if (vma_is_anonymous(vma) || !(vma->vm_flags & VM_SHARED)) { if (vma_is_anonymous(vma) || !(vma->vm_flags & VM_SHARED)) {
const unsigned long start = vma->vm_start; struct mmu_notifier_range range;
const unsigned long end = vma->vm_end;
struct mmu_gather tlb; struct mmu_gather tlb;
tlb_gather_mmu(&tlb, mm, start, end); mmu_notifier_range_init(&range, mm, vma->vm_start,
if (mmu_notifier_invalidate_range_start_nonblock(mm, start, end)) { vma->vm_end);
tlb_finish_mmu(&tlb, start, end); tlb_gather_mmu(&tlb, mm, range.start, range.end);
if (mmu_notifier_invalidate_range_start_nonblock(&range)) {
tlb_finish_mmu(&tlb, range.start, range.end);
ret = false; ret = false;
continue; continue;
} }
unmap_page_range(&tlb, vma, start, end, NULL); unmap_page_range(&tlb, vma, range.start, range.end, NULL);
mmu_notifier_invalidate_range_end(mm, start, end); mmu_notifier_invalidate_range_end(&range);
tlb_finish_mmu(&tlb, start, end); tlb_finish_mmu(&tlb, range.start, range.end);
} }
} }

30
mm/rmap.c
View file

@ -889,15 +889,17 @@ static bool page_mkclean_one(struct page *page, struct vm_area_struct *vma,
.address = address, .address = address,
.flags = PVMW_SYNC, .flags = PVMW_SYNC,
}; };
unsigned long start = address, end; struct mmu_notifier_range range;
int *cleaned = arg; int *cleaned = arg;
/* /*
* We have to assume the worse case ie pmd for invalidation. Note that * We have to assume the worse case ie pmd for invalidation. Note that
* the page can not be free from this function. * the page can not be free from this function.
*/ */
end = min(vma->vm_end, start + (PAGE_SIZE << compound_order(page))); mmu_notifier_range_init(&range, vma->vm_mm, address,
mmu_notifier_invalidate_range_start(vma->vm_mm, start, end); min(vma->vm_end, address +
(PAGE_SIZE << compound_order(page))));
mmu_notifier_invalidate_range_start(&range);
while (page_vma_mapped_walk(&pvmw)) { while (page_vma_mapped_walk(&pvmw)) {
unsigned long cstart; unsigned long cstart;
@ -949,7 +951,7 @@ static bool page_mkclean_one(struct page *page, struct vm_area_struct *vma,
(*cleaned)++; (*cleaned)++;
} }
mmu_notifier_invalidate_range_end(vma->vm_mm, start, end); mmu_notifier_invalidate_range_end(&range);
return true; return true;
} }
@ -1345,7 +1347,7 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma,
pte_t pteval; pte_t pteval;
struct page *subpage; struct page *subpage;
bool ret = true; bool ret = true;
unsigned long start = address, end; struct mmu_notifier_range range;
enum ttu_flags flags = (enum ttu_flags)arg; enum ttu_flags flags = (enum ttu_flags)arg;
/* munlock has nothing to gain from examining un-locked vmas */ /* munlock has nothing to gain from examining un-locked vmas */
@ -1369,15 +1371,18 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma,
* Note that the page can not be free in this function as call of * Note that the page can not be free in this function as call of
* try_to_unmap() must hold a reference on the page. * try_to_unmap() must hold a reference on the page.
*/ */
end = min(vma->vm_end, start + (PAGE_SIZE << compound_order(page))); mmu_notifier_range_init(&range, vma->vm_mm, vma->vm_start,
min(vma->vm_end, vma->vm_start +
(PAGE_SIZE << compound_order(page))));
if (PageHuge(page)) { if (PageHuge(page)) {
/* /*
* If sharing is possible, start and end will be adjusted * If sharing is possible, start and end will be adjusted
* accordingly. * accordingly.
*/ */
adjust_range_if_pmd_sharing_possible(vma, &start, &end); adjust_range_if_pmd_sharing_possible(vma, &range.start,
&range.end);
} }
mmu_notifier_invalidate_range_start(vma->vm_mm, start, end); mmu_notifier_invalidate_range_start(&range);
while (page_vma_mapped_walk(&pvmw)) { while (page_vma_mapped_walk(&pvmw)) {
#ifdef CONFIG_ARCH_ENABLE_THP_MIGRATION #ifdef CONFIG_ARCH_ENABLE_THP_MIGRATION
@ -1428,9 +1433,10 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma,
* we must flush them all. start/end were * we must flush them all. start/end were
* already adjusted above to cover this range. * already adjusted above to cover this range.
*/ */
flush_cache_range(vma, start, end); flush_cache_range(vma, range.start, range.end);
flush_tlb_range(vma, start, end); flush_tlb_range(vma, range.start, range.end);
mmu_notifier_invalidate_range(mm, start, end); mmu_notifier_invalidate_range(mm, range.start,
range.end);
/* /*
* The ref count of the PMD page was dropped * The ref count of the PMD page was dropped
@ -1650,7 +1656,7 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma,
put_page(page); put_page(page);
} }
mmu_notifier_invalidate_range_end(vma->vm_mm, start, end); mmu_notifier_invalidate_range_end(&range);
return ret; return ret;
} }