6.12-stable review patch. If anyone has any objections, please let me know. ------------------ From: Lorenzo Stoakes (ARM) [ Upstream commit e384abeb559d10d6505aec053ede9368d81d4c71 ] The sysfs THP tuneables at /sys/kernel/mm/transparent_huge_pages/ rather confusingly only control the behaviour of THP in some instances. They are not applicable to MADV_COLLAPSE operations, nor to DAX mappings. Long-term, THP is predicated upon compaction being able to obtain large folios to populate THP ranges. However, vm_normal_folio() returns NULL for PFN map mappings, thus their reference count is maintained by the driver, not core mm. As a consequence, the folios are not subject to reclaim nor compaction, so are not truly part of the THP mechanism at all. However, since commit 5dd40721f147 ("mm: allow THP orders for PFNMAPs") introduced the ability to establish huge PFN maps, they have been subject to THP tuneables. This is incorrect - if a huge PFN map is available (defined by vma->vm_ops->huge_fault being non-NULL for a VM_PFNMAP VMA), then it should be mapped huge upon fault-in. Correct this by explicitly checking for this while ensuring that smaps continues to accurately report THPeligible statistics. While here, abstract the entire file-backed THP check in vma_can_map_huge_file(), with sensible separation of logic into helper functions. Failing to map huge PFN maps has resulted in significant real-world performance degradation, see links for details. [ziy@nvidia.com: rename some functions] Link: https://lore.kernel.org/DL1HIHWYJ7TB.1CY76SJS0V03L@nvidia.com Fixes: 5dd40721f147 ("mm: allow THP orders for PFNMAPs") Signed-off-by: Lorenzo Stoakes (ARM) Signed-off-by: Zi Yan Reported-by: Cedric Le Goater Closes: https://lore.kernel.org/linux-mm/20260805055544.1568534-1-clg@redhat.com/ Reported-by: Saravanan D Closes: https://lore.kernel.org/linux-mm/20260821070520.25759-1-saravanand@crusoe.ai/ Reviewed-by: Zi Yan Tested-by: Saravanan D Tested-by: Lance Yang Reviewed-by: SJ Park Reviewed-by: Baolin Wang Cc: Signed-off-by: Andrew Morton [ ljs - Converted to tva_flags, made file_thp_enabled() const in huge_mm.h + other smaller changes. Updated commit msg to reflect. ] Signed-off-by: Lorenzo Stoakes (ARM) Signed-off-by: Sasha Levin --- include/linux/huge_mm.h | 2 +- mm/huge_memory.c | 79 ++++++++++++++++++++++++++++++----------- 2 files changed, 59 insertions(+), 22 deletions(-) diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h index bd34b70dd0cf9..10e41e8b16349 100644 --- a/include/linux/huge_mm.h +++ b/include/linux/huge_mm.h @@ -254,7 +254,7 @@ static inline unsigned long thp_vma_suitable_orders(struct vm_area_struct *vma, return orders; } -static inline bool file_thp_enabled(struct vm_area_struct *vma) +static inline bool file_thp_enabled(const struct vm_area_struct *vma) { struct inode *inode; diff --git a/mm/huge_memory.c b/mm/huge_memory.c index f6ba19b57265b..d175e93988a92 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -86,6 +86,62 @@ unsigned long huge_anon_orders_madvise __read_mostly; unsigned long huge_anon_orders_inherit __read_mostly; static bool anon_orders_configured __initdata; +static bool vma_file_bypass_thp_tuneables(const struct vm_area_struct *vma, + unsigned long tva_flags) +{ + const bool has_huge_fault = vma->vm_ops->huge_fault; + + /* MADV_COLLAPSE ignores tuneables. */ + if (!(tva_flags & TVA_ENFORCE_SYSFS)) + return true; + /* Huge PFN mappings are uncompactable so the policy doesn't apply. */ + if ((vma->vm_flags & VM_PFNMAP) && has_huge_fault) + return true; + return false; +} + +static bool vma_file_allow_thp_tuneables(vm_flags_t vm_flags) +{ + /* THP=always? */ + if (hugepage_global_always()) + return true; + /* THP=madvise and marked MADV_HUGEPAGE? */ + if (hugepage_global_enabled() && (vm_flags & VM_HUGEPAGE)) + return true; + return false; +} + +static bool vma_file_check_thp_tuneables(const struct vm_area_struct *vma, + vm_flags_t vm_flags, unsigned long tva_flags) +{ + return vma_file_bypass_thp_tuneables(vma, tva_flags) || + vma_file_allow_thp_tuneables(vm_flags); +} + +static bool vma_can_map_huge_file(const struct vm_area_struct *vma, + vm_flags_t vm_flags, unsigned long tva_flags) +{ + const bool has_huge_fault = vma->vm_ops->huge_fault; + + /* + * Enforce THP collapse requirements as necessary. Anonymous vmas + * were already handled in thp_vma_allowable_orders(). + */ + if (!vma_file_check_thp_tuneables(vma, vm_flags, tva_flags)) + return false; + + /* + * Trust that ->huge_fault() handlers know what they are doing + * in fault path. + */ + if (tva_flags & TVA_IN_PF) + return has_huge_fault; + if ((tva_flags & TVA_SMAPS) && has_huge_fault) + return true; + /* Only regular file is valid in collapse path. */ + return file_thp_enabled(vma); +} + unsigned long __thp_vma_allowable_orders(struct vm_area_struct *vma, unsigned long vm_flags, unsigned long tva_flags, @@ -158,27 +214,8 @@ unsigned long __thp_vma_allowable_orders(struct vm_area_struct *vma, vma, vma->vm_pgoff, 0, !enforce_sysfs); - if (!vma_is_anonymous(vma)) { - /* - * Enforce sysfs THP requirements as necessary. Anonymous vmas - * were already handled in thp_vma_allowable_orders(). - */ - if (enforce_sysfs && - (!hugepage_global_enabled() || (!(vm_flags & VM_HUGEPAGE) && - !hugepage_global_always()))) - return 0; - - /* - * Trust that ->huge_fault() handlers know what they are doing - * in fault path. - */ - if (((in_pf || smaps)) && vma->vm_ops->huge_fault) - return orders; - /* Only regular file is valid in collapse path */ - if (((!in_pf || smaps)) && file_thp_enabled(vma)) - return orders; - return 0; - } + if (!vma_is_anonymous(vma)) + return vma_can_map_huge_file(vma, vm_flags, tva_flags) ? orders : 0; if (vma_is_temporary_stack(vma)) return 0; -- 2.53.0