| Message ID | 20220616174840.1202070-5-shy828301@gmail.com (mailing list archive) |
|---|---|
| State | New |
| Headers | show |
| Series | Cleanup transhuge_xxx helpers | expand |
On 16 Jun 10:48, Yang Shi wrote: > The transparent_hugepage_active() was introduced to show THP eligibility > bit in smaps in proc, smaps is the only user. But it actually does the > similar check as hugepage_vma_check() which is used by khugepaged. We > definitely don't have to maintain two similar checks, so kill > transparent_hugepage_active(). > > This patch also fixed the wrong behavior for VM_NO_KHUGEPAGED vmas. > > Also move hugepage_vma_check() to huge_memory.c and huge_mm.h since it > is not only for khugepaged anymore. > > Reviewed-by: Zach O'Keefe <zokeefe@google.com> > Signed-off-by: Yang Shi <shy828301@gmail.com> > --- > fs/proc/task_mmu.c | 2 +- > include/linux/huge_mm.h | 16 +++++++----- > include/linux/khugepaged.h | 2 -- > mm/huge_memory.c | 50 +++++++++++++++++++++++++++++++------- > mm/khugepaged.c | 48 +++--------------------------------- > 5 files changed, 56 insertions(+), 62 deletions(-) > > diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c > index 37ccb5c9f4f8..39a40ec181e7 100644 > --- a/fs/proc/task_mmu.c > +++ b/fs/proc/task_mmu.c > @@ -863,7 +863,7 @@ static int show_smap(struct seq_file *m, void *v) > __show_smap(m, &mss, false); > > seq_printf(m, "THPeligible: %d\n", > - transparent_hugepage_active(vma)); > + hugepage_vma_check(vma, vma->vm_flags, true)); > > if (arch_pkeys_enabled()) > seq_printf(m, "ProtectionKey: %8u\n", vma_pkey(vma)); > diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h > index 8a5a8bfce0f5..64487bcd0c7b 100644 > --- a/include/linux/huge_mm.h > +++ b/include/linux/huge_mm.h > @@ -202,7 +202,9 @@ static inline bool file_thp_enabled(struct vm_area_struct *vma) > !inode_is_open_for_write(inode) && S_ISREG(inode->i_mode); > } > > -bool transparent_hugepage_active(struct vm_area_struct *vma); > +bool hugepage_vma_check(struct vm_area_struct *vma, > + unsigned long vm_flags, > + bool smaps); > > #define transparent_hugepage_use_zero_page() \ > (transparent_hugepage_flags & \ > @@ -351,11 +353,6 @@ static inline bool __transparent_hugepage_enabled(struct vm_area_struct *vma) > return false; > } > > -static inline bool transparent_hugepage_active(struct vm_area_struct *vma) > -{ > - return false; > -} > - > static inline bool transhuge_vma_suitable(struct vm_area_struct *vma, > unsigned long addr) > { > @@ -368,6 +365,13 @@ static inline bool transhuge_vma_enabled(struct vm_area_struct *vma, > return false; > } > > +static inline bool hugepage_vma_check(struct vm_area_struct *vma, > + unsigned long vm_flags, > + bool smaps) > +{ > + return false; > +} > + > static inline void prep_transhuge_page(struct page *page) {} > > #define transparent_hugepage_flags 0UL > diff --git a/include/linux/khugepaged.h b/include/linux/khugepaged.h > index 31ca8a7f78f4..ea5fd4c398f7 100644 > --- a/include/linux/khugepaged.h > +++ b/include/linux/khugepaged.h > @@ -10,8 +10,6 @@ extern struct attribute_group khugepaged_attr_group; > extern int khugepaged_init(void); > extern void khugepaged_destroy(void); > extern int start_stop_khugepaged(void); > -extern bool hugepage_vma_check(struct vm_area_struct *vma, > - unsigned long vm_flags); > extern void __khugepaged_enter(struct mm_struct *mm); > extern void __khugepaged_exit(struct mm_struct *mm); > extern void khugepaged_enter_vma(struct vm_area_struct *vma, > diff --git a/mm/huge_memory.c b/mm/huge_memory.c > index b530462c4493..a28c6100b491 100644 > --- a/mm/huge_memory.c > +++ b/mm/huge_memory.c > @@ -69,21 +69,53 @@ static atomic_t huge_zero_refcount; > struct page *huge_zero_page __read_mostly; > unsigned long huge_zero_pfn __read_mostly = ~0UL; > > -bool transparent_hugepage_active(struct vm_area_struct *vma) > +bool hugepage_vma_check(struct vm_area_struct *vma, > + unsigned long vm_flags, > + bool smaps) > { > - /* The addr is used to check if the vma size fits */ > - unsigned long addr = (vma->vm_end & HPAGE_PMD_MASK) - HPAGE_PMD_SIZE; > + if (!transhuge_vma_enabled(vma, vm_flags)) > + return false; > + During testing my work on top this patch, I found a small bug here. Namely, transhuge_vma_enabled() will check vma->vm_mm->flags (to see if MMF_DISABLE_THP is set); however, for vDSO vmas, vma->vm_mm is NULL. Previously, transparent_hugepage_active() in smaps path would check transhuge_vma_suitable() before checking these flags, which would fail for vDSO vma since we'd take the !vma_is_anonymous() branch and find the vma (most likely) wasn't suitably aligned (by chance ?). Anyways, I think we need to check vma->vm_mm. > + if (vm_flags & VM_NO_KHUGEPAGED) > + return false; > + > + /* Don't run khugepaged against DAX vma */ > + if (vma_is_dax(vma)) > + return false; > > - if (!transhuge_vma_suitable(vma, addr)) > + /* Check alignment for file vma and size for both file and anon vma */ > + if (!transhuge_vma_suitable(vma, (vma->vm_end - HPAGE_PMD_SIZE))) > return false; > - if (vma_is_anonymous(vma)) > - return __transparent_hugepage_enabled(vma); > - if (vma_is_shmem(vma)) > + > + /* Enabled via shmem mount options or sysfs settings. */ > + if (shmem_file(vma->vm_file)) > return shmem_huge_enabled(vma); > - if (transhuge_vma_enabled(vma, vma->vm_flags) && file_thp_enabled(vma)) > + > + if (!khugepaged_enabled()) > + return false; > + > + /* THP settings require madvise. */ > + if (!(vm_flags & VM_HUGEPAGE) && !khugepaged_always()) > + return false; > + > + /* Only regular file is valid */ > + if (file_thp_enabled(vma)) > return true; > > - return false; > + if (!vma_is_anonymous(vma)) > + return false; > + > + if (vma_is_temporary_stack(vma)) > + return false; > + > + /* > + * THPeligible bit of smaps should show 1 for proper VMAs even > + * though anon_vma is not initialized yet. > + */ > + if (!vma->anon_vma) > + return smaps; > + > + return true; > } > > static bool get_huge_zero_page(void) > diff --git a/mm/khugepaged.c b/mm/khugepaged.c > index 5baa394e34c8..3afd87f8c0b1 100644 > --- a/mm/khugepaged.c > +++ b/mm/khugepaged.c > @@ -437,46 +437,6 @@ static inline int khugepaged_test_exit(struct mm_struct *mm) > return atomic_read(&mm->mm_users) == 0; > } > > -bool hugepage_vma_check(struct vm_area_struct *vma, > - unsigned long vm_flags) > -{ > - if (!transhuge_vma_enabled(vma, vm_flags)) > - return false; > - > - if (vm_flags & VM_NO_KHUGEPAGED) > - return false; > - > - /* Don't run khugepaged against DAX vma */ > - if (vma_is_dax(vma)) > - return false; > - > - /* Check alignment for file vma and size for both file and anon vma */ > - if (!transhuge_vma_suitable(vma, (vma->vm_end - HPAGE_PMD_SIZE))) > - return false; > - > - /* Enabled via shmem mount options or sysfs settings. */ > - if (shmem_file(vma->vm_file)) > - return shmem_huge_enabled(vma); > - > - if (!khugepaged_enabled()) > - return false; > - > - /* THP settings require madvise. */ > - if (!(vm_flags & VM_HUGEPAGE) && !khugepaged_always()) > - return false; > - > - /* Only regular file is valid */ > - if (file_thp_enabled(vma)) > - return true; > - > - if (!vma->anon_vma || !vma_is_anonymous(vma)) > - return false; > - if (vma_is_temporary_stack(vma)) > - return false; > - > - return true; > -} > - > void __khugepaged_enter(struct mm_struct *mm) > { > struct mm_slot *mm_slot; > @@ -513,7 +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)) > + if (hugepage_vma_check(vma, vm_flags, false)) > __khugepaged_enter(vma->vm_mm); > } > } > @@ -958,7 +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)) > + if (!hugepage_vma_check(vma, vma->vm_flags, false)) > return SCAN_VMA_CHECK; > /* > * Anon VMA expected, the address may be unmapped then > @@ -1448,7 +1408,7 @@ 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)) > + if (!hugepage_vma_check(vma, vma->vm_flags | VM_HUGEPAGE, false)) > return; > > /* Keep pmd pgtable for uffd-wp; see comment in retract_page_tables() */ > @@ -2143,7 +2103,7 @@ static unsigned int khugepaged_scan_mm_slot(unsigned int pages, > progress++; > break; > } > - if (!hugepage_vma_check(vma, vma->vm_flags)) { > + if (!hugepage_vma_check(vma, vma->vm_flags, false)) { > skip: > progress++; > continue; > -- > 2.26.3 >
On Tue, 21 Jun 2022 11:58:34 -0700 "Zach O'Keefe" <zokeefe@google.com> wrote: > > -bool transparent_hugepage_active(struct vm_area_struct *vma) > > +bool hugepage_vma_check(struct vm_area_struct *vma, > > + unsigned long vm_flags, > > + bool smaps) > > { > > - /* The addr is used to check if the vma size fits */ > > - unsigned long addr = (vma->vm_end & HPAGE_PMD_MASK) - HPAGE_PMD_SIZE; > > + if (!transhuge_vma_enabled(vma, vm_flags)) > > + return false; > > + > > During testing my work on top this patch, I found a small bug here. > > Namely, transhuge_vma_enabled() will check vma->vm_mm->flags (to see if > MMF_DISABLE_THP is set); however, for vDSO vmas, vma->vm_mm is NULL. > > Previously, transparent_hugepage_active() in smaps path would check > transhuge_vma_suitable() before checking these flags, which would fail for vDSO > vma since we'd take the !vma_is_anonymous() branch and find the vma (most > likely) wasn't suitably aligned (by chance ?). > > Anyways, I think we need to check vma->vm_mm. Like this? --- a/mm/huge_memory.c~mm-thp-kill-transparent_hugepage_active-fix +++ a/mm/huge_memory.c @@ -73,6 +73,9 @@ bool hugepage_vma_check(struct vm_area_s unsigned long vm_flags, bool smaps) { + if (!vma->vm_mm) + return false; + if (!transhuge_vma_enabled(vma, vm_flags)) return false;
On 21 Jun 17:54, Andrew Morton wrote: > On Tue, 21 Jun 2022 11:58:34 -0700 "Zach O'Keefe" <zokeefe@google.com> wrote: > > > > -bool transparent_hugepage_active(struct vm_area_struct *vma) > > > +bool hugepage_vma_check(struct vm_area_struct *vma, > > > + unsigned long vm_flags, > > > + bool smaps) > > > { > > > - /* The addr is used to check if the vma size fits */ > > > - unsigned long addr = (vma->vm_end & HPAGE_PMD_MASK) - HPAGE_PMD_SIZE; > > > + if (!transhuge_vma_enabled(vma, vm_flags)) > > > + return false; > > > + > > > > During testing my work on top this patch, I found a small bug here. > > > > Namely, transhuge_vma_enabled() will check vma->vm_mm->flags (to see if > > MMF_DISABLE_THP is set); however, for vDSO vmas, vma->vm_mm is NULL. > > > > Previously, transparent_hugepage_active() in smaps path would check > > transhuge_vma_suitable() before checking these flags, which would fail for vDSO > > vma since we'd take the !vma_is_anonymous() branch and find the vma (most > > likely) wasn't suitably aligned (by chance ?). > > > > Anyways, I think we need to check vma->vm_mm. > > Like this? > > --- a/mm/huge_memory.c~mm-thp-kill-transparent_hugepage_active-fix > +++ a/mm/huge_memory.c > @@ -73,6 +73,9 @@ bool hugepage_vma_check(struct vm_area_s > unsigned long vm_flags, > bool smaps) > { > + if (!vma->vm_mm) > + return false; > + > if (!transhuge_vma_enabled(vma, vm_flags)) > return false; > > _ > Hey Andrew, In principle, yes that would fix this. I don't know precisely how this fix will be applied, but note that the subsequent patch "mm: thp: kill __transhuge_page_enabled()" won't apply on top of this automatically. Also, I wonder if we should add a comment for future travellers who wonder what kind of vmas don't have an associated mm (it was news to me); though, I'm not sure if vDSO is the only such case (though show_map_vma() seems to think so), or if this just asking for stale comments down the road. Maybe it's fine as is. Thanks, Zach
diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 37ccb5c9f4f8..39a40ec181e7 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -863,7 +863,7 @@ static int show_smap(struct seq_file *m, void *v) __show_smap(m, &mss, false); seq_printf(m, "THPeligible: %d\n", - transparent_hugepage_active(vma)); + hugepage_vma_check(vma, vma->vm_flags, true)); if (arch_pkeys_enabled()) seq_printf(m, "ProtectionKey: %8u\n", vma_pkey(vma)); diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h index 8a5a8bfce0f5..64487bcd0c7b 100644 --- a/include/linux/huge_mm.h +++ b/include/linux/huge_mm.h @@ -202,7 +202,9 @@ static inline bool file_thp_enabled(struct vm_area_struct *vma) !inode_is_open_for_write(inode) && S_ISREG(inode->i_mode); } -bool transparent_hugepage_active(struct vm_area_struct *vma); +bool hugepage_vma_check(struct vm_area_struct *vma, + unsigned long vm_flags, + bool smaps); #define transparent_hugepage_use_zero_page() \ (transparent_hugepage_flags & \ @@ -351,11 +353,6 @@ static inline bool __transparent_hugepage_enabled(struct vm_area_struct *vma) return false; } -static inline bool transparent_hugepage_active(struct vm_area_struct *vma) -{ - return false; -} - static inline bool transhuge_vma_suitable(struct vm_area_struct *vma, unsigned long addr) { @@ -368,6 +365,13 @@ static inline bool transhuge_vma_enabled(struct vm_area_struct *vma, return false; } +static inline bool hugepage_vma_check(struct vm_area_struct *vma, + unsigned long vm_flags, + bool smaps) +{ + return false; +} + static inline void prep_transhuge_page(struct page *page) {} #define transparent_hugepage_flags 0UL diff --git a/include/linux/khugepaged.h b/include/linux/khugepaged.h index 31ca8a7f78f4..ea5fd4c398f7 100644 --- a/include/linux/khugepaged.h +++ b/include/linux/khugepaged.h @@ -10,8 +10,6 @@ extern struct attribute_group khugepaged_attr_group; extern int khugepaged_init(void); extern void khugepaged_destroy(void); extern int start_stop_khugepaged(void); -extern bool hugepage_vma_check(struct vm_area_struct *vma, - unsigned long vm_flags); extern void __khugepaged_enter(struct mm_struct *mm); extern void __khugepaged_exit(struct mm_struct *mm); extern void khugepaged_enter_vma(struct vm_area_struct *vma, diff --git a/mm/huge_memory.c b/mm/huge_memory.c index b530462c4493..a28c6100b491 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -69,21 +69,53 @@ static atomic_t huge_zero_refcount; struct page *huge_zero_page __read_mostly; unsigned long huge_zero_pfn __read_mostly = ~0UL; -bool transparent_hugepage_active(struct vm_area_struct *vma) +bool hugepage_vma_check(struct vm_area_struct *vma, + unsigned long vm_flags, + bool smaps) { - /* The addr is used to check if the vma size fits */ - unsigned long addr = (vma->vm_end & HPAGE_PMD_MASK) - HPAGE_PMD_SIZE; + if (!transhuge_vma_enabled(vma, vm_flags)) + return false; + + if (vm_flags & VM_NO_KHUGEPAGED) + return false; + + /* Don't run khugepaged against DAX vma */ + if (vma_is_dax(vma)) + return false; - if (!transhuge_vma_suitable(vma, addr)) + /* Check alignment for file vma and size for both file and anon vma */ + if (!transhuge_vma_suitable(vma, (vma->vm_end - HPAGE_PMD_SIZE))) return false; - if (vma_is_anonymous(vma)) - return __transparent_hugepage_enabled(vma); - if (vma_is_shmem(vma)) + + /* Enabled via shmem mount options or sysfs settings. */ + if (shmem_file(vma->vm_file)) return shmem_huge_enabled(vma); - if (transhuge_vma_enabled(vma, vma->vm_flags) && file_thp_enabled(vma)) + + if (!khugepaged_enabled()) + return false; + + /* THP settings require madvise. */ + if (!(vm_flags & VM_HUGEPAGE) && !khugepaged_always()) + return false; + + /* Only regular file is valid */ + if (file_thp_enabled(vma)) return true; - return false; + if (!vma_is_anonymous(vma)) + return false; + + if (vma_is_temporary_stack(vma)) + return false; + + /* + * THPeligible bit of smaps should show 1 for proper VMAs even + * though anon_vma is not initialized yet. + */ + if (!vma->anon_vma) + return smaps; + + return true; } static bool get_huge_zero_page(void) diff --git a/mm/khugepaged.c b/mm/khugepaged.c index 5baa394e34c8..3afd87f8c0b1 100644 --- a/mm/khugepaged.c +++ b/mm/khugepaged.c @@ -437,46 +437,6 @@ static inline int khugepaged_test_exit(struct mm_struct *mm) return atomic_read(&mm->mm_users) == 0; } -bool hugepage_vma_check(struct vm_area_struct *vma, - unsigned long vm_flags) -{ - if (!transhuge_vma_enabled(vma, vm_flags)) - return false; - - if (vm_flags & VM_NO_KHUGEPAGED) - return false; - - /* Don't run khugepaged against DAX vma */ - if (vma_is_dax(vma)) - return false; - - /* Check alignment for file vma and size for both file and anon vma */ - if (!transhuge_vma_suitable(vma, (vma->vm_end - HPAGE_PMD_SIZE))) - return false; - - /* Enabled via shmem mount options or sysfs settings. */ - if (shmem_file(vma->vm_file)) - return shmem_huge_enabled(vma); - - if (!khugepaged_enabled()) - return false; - - /* THP settings require madvise. */ - if (!(vm_flags & VM_HUGEPAGE) && !khugepaged_always()) - return false; - - /* Only regular file is valid */ - if (file_thp_enabled(vma)) - return true; - - if (!vma->anon_vma || !vma_is_anonymous(vma)) - return false; - if (vma_is_temporary_stack(vma)) - return false; - - return true; -} - void __khugepaged_enter(struct mm_struct *mm) { struct mm_slot *mm_slot; @@ -513,7 +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)) + if (hugepage_vma_check(vma, vm_flags, false)) __khugepaged_enter(vma->vm_mm); } } @@ -958,7 +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)) + if (!hugepage_vma_check(vma, vma->vm_flags, false)) return SCAN_VMA_CHECK; /* * Anon VMA expected, the address may be unmapped then @@ -1448,7 +1408,7 @@ 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)) + if (!hugepage_vma_check(vma, vma->vm_flags | VM_HUGEPAGE, false)) return; /* Keep pmd pgtable for uffd-wp; see comment in retract_page_tables() */ @@ -2143,7 +2103,7 @@ static unsigned int khugepaged_scan_mm_slot(unsigned int pages, progress++; break; } - if (!hugepage_vma_check(vma, vma->vm_flags)) { + if (!hugepage_vma_check(vma, vma->vm_flags, false)) { skip: progress++; continue;