@@ -860,7 +860,7 @@ static int show_smap(struct seq_file *m, void *v)
__show_smap(m, &mss, false);
seq_printf(m, "THPeligible: %d\n",
- hugepage_vma_check(vma, vma->vm_flags, true));
+ hugepage_vma_check(vma, vma->vm_flags, true, false));
if (arch_pkeys_enabled())
seq_printf(m, "ProtectionKey: %8u\n", vma_pkey(vma));
@@ -153,48 +153,6 @@ static inline bool transhuge_vma_suitable(struct vm_area_struct *vma,
return true;
}
-static inline bool transhuge_vma_enabled(struct vm_area_struct *vma,
- unsigned long vm_flags)
-{
- /* Explicitly disabled through madvise. */
- if ((vm_flags & VM_NOHUGEPAGE) ||
- test_bit(MMF_DISABLE_THP, &vma->vm_mm->flags))
- return false;
- return true;
-}
-
-/*
- * to be used on vmas which are known to support THP.
- * Use transparent_hugepage_active otherwise
- */
-static inline bool __transparent_hugepage_enabled(struct vm_area_struct *vma)
-{
-
- /*
- * If the hardware/firmware marked hugepage support disabled.
- */
- if (transparent_hugepage_flags & (1 << TRANSPARENT_HUGEPAGE_NEVER_DAX))
- return false;
-
- if (!transhuge_vma_enabled(vma, vma->vm_flags))
- return false;
-
- if (vma_is_temporary_stack(vma))
- return false;
-
- if (transparent_hugepage_flags & (1 << TRANSPARENT_HUGEPAGE_FLAG))
- return true;
-
- if (vma_is_dax(vma))
- return true;
-
- if (transparent_hugepage_flags &
- (1 << TRANSPARENT_HUGEPAGE_REQ_MADV_FLAG))
- return !!(vma->vm_flags & VM_HUGEPAGE);
-
- return false;
-}
-
static inline bool file_thp_enabled(struct vm_area_struct *vma)
{
struct inode *inode;
@@ -211,7 +169,7 @@ static inline bool file_thp_enabled(struct vm_area_struct *vma)
extern bool hugepage_vma_check(struct vm_area_struct *vma,
unsigned long vm_flags,
- bool smaps);
+ bool smaps, bool in_pf);
#define transparent_hugepage_use_zero_page() \
(transparent_hugepage_flags & \
@@ -355,11 +313,6 @@ static inline bool folio_test_pmd_mappable(struct folio *folio)
return false;
}
-static inline bool __transparent_hugepage_enabled(struct vm_area_struct *vma)
-{
- return false;
-}
-
static inline bool transhuge_vma_size_ok(struct vm_area_struct *vma)
{
return false;
@@ -371,15 +324,9 @@ static inline bool transhuge_vma_suitable(struct vm_area_struct *vma,
return false;
}
-static inline bool transhuge_vma_enabled(struct vm_area_struct *vma,
- unsigned long vm_flags)
-{
- return false;
-}
-
static inline bool hugepage_vma_check(struct vm_area_struct *vma,
unsigned long vm_flags,
- bool smaps)
+ bool smaps, bool in_pf)
{
return false;
}
@@ -55,7 +55,7 @@ static inline void khugepaged_enter(struct vm_area_struct *vma,
{
if (!test_bit(MMF_VM_HUGEPAGE, &vma->vm_mm->flags) &&
khugepaged_enabled()) {
- if (hugepage_vma_check(vma, vm_flags, false))
+ if (hugepage_vma_check(vma, vm_flags, false, false))
__khugepaged_enter(vma->vm_mm);
}
}
@@ -71,17 +71,25 @@ unsigned long huge_zero_pfn __read_mostly = ~0UL;
bool hugepage_vma_check(struct vm_area_struct *vma,
unsigned long vm_flags,
- bool smaps)
+ bool smaps, bool in_pf)
{
- if (!transhuge_vma_enabled(vma, vm_flags))
+ /* Explicitly disabled through madvise or prctl. */
+ if ((vm_flags & VM_NOHUGEPAGE) ||
+ test_bit(MMF_DISABLE_THP, &vma->vm_mm->flags))
+ return false;
+ /*
+ * If the hardware/firmware marked hugepage support disabled.
+ */
+ if (transparent_hugepage_flags & (1 << TRANSPARENT_HUGEPAGE_NEVER_DAX))
return false;
+ /* Special VMA and hugetlb VMA */
if (vm_flags & VM_NO_KHUGEPAGED)
return false;
- /* Don't run khugepaged against DAX vma */
+ /* khugepaged doesn't collapse DAX vma, but page fault is fine. */
if (vma_is_dax(vma))
- return false;
+ return in_pf;
if (vma->vm_file && !IS_ALIGNED((vma->vm_start >> PAGE_SHIFT) -
vma->vm_pgoff, HPAGE_PMD_NR))
@@ -91,7 +99,7 @@ bool hugepage_vma_check(struct vm_area_struct *vma,
return false;
/* Enabled via shmem mount options or sysfs settings. */
- if (shmem_file(vma->vm_file))
+ if (!in_pf && shmem_file(vma->vm_file))
return shmem_huge_enabled(vma);
if (!khugepaged_enabled())
@@ -102,7 +110,7 @@ bool hugepage_vma_check(struct vm_area_struct *vma,
return false;
/* Only regular file is valid */
- if (file_thp_enabled(vma))
+ if (!in_pf && file_thp_enabled(vma))
return true;
if (!vma_is_anonymous(vma))
@@ -114,9 +122,12 @@ bool hugepage_vma_check(struct vm_area_struct *vma,
/*
* THPeligible bit of smaps should show 1 for proper VMAs even
* though anon_vma is not initialized yet.
+ *
+ * Allow page fault since anon_vma may be not initialized until
+ * the first page fault.
*/
if (!vma->anon_vma)
- return smaps;
+ return (smaps || in_pf);
return true;
}
@@ -473,7 +473,7 @@ void khugepaged_enter_vma(struct vm_area_struct *vma,
{
if (!test_bit(MMF_VM_HUGEPAGE, &vma->vm_mm->flags) &&
khugepaged_enabled()) {
- if (hugepage_vma_check(vma, vm_flags, false))
+ if (hugepage_vma_check(vma, vm_flags, false, false))
__khugepaged_enter(vma->vm_mm);
}
}
@@ -918,7 +918,7 @@ static int hugepage_vma_revalidate(struct mm_struct *mm, unsigned long address,
if (!transhuge_vma_suitable(vma, address))
return SCAN_ADDRESS_RANGE;
- if (!hugepage_vma_check(vma, vma->vm_flags, false))
+ if (!hugepage_vma_check(vma, vma->vm_flags, false, false))
return SCAN_VMA_CHECK;
return 0;
}
@@ -1399,7 +1399,7 @@ void collapse_pte_mapped_thp(struct mm_struct *mm, unsigned long addr)
* the valid THP. Add extra VM_HUGEPAGE so hugepage_vma_check()
* will not fail the vma for missing VM_HUGEPAGE
*/
- if (!hugepage_vma_check(vma, vma->vm_flags | VM_HUGEPAGE, false))
+ if (!hugepage_vma_check(vma, vma->vm_flags | VM_HUGEPAGE, false, false))
return;
/* Keep pmd pgtable for uffd-wp; see comment in retract_page_tables() */
@@ -2089,7 +2089,7 @@ static unsigned int khugepaged_scan_mm_slot(unsigned int pages,
progress++;
break;
}
- if (!hugepage_vma_check(vma, vma->vm_flags, false)) {
+ if (!hugepage_vma_check(vma, vma->vm_flags, false, false)) {
skip:
progress++;
continue;
@@ -4962,6 +4962,7 @@ static vm_fault_t __handle_mm_fault(struct vm_area_struct *vma,
.gfp_mask = __get_fault_gfp_mask(vma),
};
struct mm_struct *mm = vma->vm_mm;
+ unsigned long vm_flags = vma->vm_flags;
pgd_t *pgd;
p4d_t *p4d;
vm_fault_t ret;
@@ -4975,7 +4976,8 @@ static vm_fault_t __handle_mm_fault(struct vm_area_struct *vma,
if (!vmf.pud)
return VM_FAULT_OOM;
retry_pud:
- if (pud_none(*vmf.pud) && __transparent_hugepage_enabled(vma)) {
+ if (pud_none(*vmf.pud) &&
+ hugepage_vma_check(vma, vm_flags, false, true)) {
ret = create_huge_pud(&vmf);
if (!(ret & VM_FAULT_FALLBACK))
return ret;
@@ -5008,7 +5010,8 @@ static vm_fault_t __handle_mm_fault(struct vm_area_struct *vma,
if (pud_trans_unstable(vmf.pud))
goto retry_pud;
- if (pmd_none(*vmf.pmd) && __transparent_hugepage_enabled(vma)) {
+ if (pmd_none(*vmf.pmd) &&
+ hugepage_vma_check(vma, vm_flags, false, true)) {
ret = create_huge_pmd(&vmf);
if (!(ret & VM_FAULT_FALLBACK))
return ret;
The page fault path checks THP eligibility with __transhuge_page_enabled() which does the similar thing as hugepage_vma_check(), so use hugepage_vma_check() instead. However page fault allows DAX and !anon_vma cases, so added a new flag, in_pf, to hugepage_vma_check() to make page fault work correctly. The in_pf flag is also used to skip shmem and file THP for page fault since shmem handles THP in its own shmem_fault() and file THP allocation on fault is not supported yet. Also remove hugepage_vma_enabled() since hugepage_vma_check() is the only caller now, it is not necessary to have a helper function. Signed-off-by: Yang Shi <shy828301@gmail.com> --- fs/proc/task_mmu.c | 2 +- include/linux/huge_mm.h | 57 ++------------------------------------ include/linux/khugepaged.h | 2 +- mm/huge_memory.c | 25 ++++++++++++----- mm/khugepaged.c | 8 +++--- mm/memory.c | 7 +++-- 6 files changed, 31 insertions(+), 70 deletions(-)