fixup: mm/huge_memory.c: introduce folio_split_unmapped
Subsystems (3)
PATCH v2 1/1 fixup: mm/huge_memory.c: introduce folio_split_unmapped
Code refactoring of __folio_split() via helper __folio_freeze_and_split_unmapped() caused a regression with clang-20 with CONFIG_SHMEM=n, the compiler was not able to optimize away the call to shmem_uncharge() due to changes in nr_shmem_dropped. Fix this by adding a stub function for shmem_uncharge when CONFIG_SHMEM is not defined.
smatch also complained about parameter end being used without initialization, which is a false positive, but keep the tool happy by sending in initialized parameters. end is initialized to 0. smatch still complains about mapping being NULL and nr_shmem_dropped may not be 0, but that is not true prior to or after the changes.
Add detailed documentation comments for folio_split_unmapped()
---
This fixup should be squashed into the patch "mm/huge_memory.c:
introduce folio_split_unmapped" in mm/mm-unstable
include/linux/shmem_fs.h | 6 +++++- mm/huge_memory.c | 30 +++++++++++++++++++++--------- 2 files changed, 26 insertions(+), 10 deletions(-)
include/linux/shmem_fs.h +5 -1
@@ -136,11 +136,16 @@ static inline bool shmem_hpage_pmd_enabled(void)#ifdef CONFIG_SHMEM extern unsigned long shmem_swap_usage (struct vm_area_struct *vma); +extern void shmem_uncharge (struct inode *inode,long pages);#else static inline unsigned long shmem_swap_usage (struct vm_area_struct *vma) {return 0 ; } + +static void shmem_uncharge (struct inode *inode,long pages) +{ +}#endif extern unsigned long shmem_partial_swap_usage (struct address_space *mapping,pgoff_t start,pgoff_t end);
@@ -194,7 +199,6 @@ static inline pgoff_t shmem_fallocend(struct inode *inode, pgoff_t eof) }extern bool shmem_charge (struct inode *inode,long pages); -extern void shmem_uncharge (struct inode *inode,long pages);#ifdef CONFIG_USERFAULTFD #ifdef CONFIG_SHMEM
mm/huge_memory.c +21 -9
@@ -3751,6 +3751,7 @@ static int __folio_freeze_and_split_unmapped(struct folio *folio, unsigned int nint ret =0 ;struct deferred_split *ds_queue; +VM_WARN_ON_ONCE (!mapping && end);/* Prevent deferred_split_scan() touching ->_refcount */ ds_queue =folio_split_queue_lock (folio);if (folio_ref_freeze (folio,1 + extra_pins)) {
@@ -3919,7 +3920,7 @@ static int __folio_split(struct folio *folio, unsigned int new_order,int nr_shmem_dropped =0 ;int remap_flags =0 ;int extra_pins, ret; -pgoff_t end; +pgoff_t end =0 ;bool is_hzp;VM_WARN_ON_ONCE_FOLIO (!folio_test_locked (folio), folio);
@@ -4092,16 +4093,27 @@ static int __folio_split(struct folio *folio, unsigned int new_order,return ret; } -/* -* This function is a helper for splitting folios that have already been unmapped. -* The use case is that the device or the CPU can refuse to migrate THP pages in -* the middle of migration, due to allocation issues on either side +/** +* folio_split_unmapped() - split a large anon folio that is already unmapped +* @folio: folio to split +* @new_order: the order of folios after split +* +* This function is a helper for splitting folios that have already been +* unmapped. The use case is that the device or the CPU can refuse to migrate +* THP pages in the middle of migration, due to allocation issues on either +* side. +* +* anon_vma_lock is not required to be held, mmap_read_lock() or +* mmap_write_lock() should be held. @folio is expected to be locked by the +* caller. device-private and non device-private folios are supported along +* with folios that are in the swapcache. @folio should also be unmapped and +* isolated from LRU (if applicable) * -* The high level code is copied from __folio_split, since the pages are anonymous -* and are already isolated from the LRU, the code has been simplified to not -* burden __folio_split with unmapped sprinkled into the code. +* Upon return, the folio is not remapped, split folios are not added to LRU, +* free_folio_and_swap_cache() is not called, and new folios remain locked. * -* None of the split folios are unlocked +* Return: 0 on success, -EAGAIN if the folio cannot be split (e.g., due to +* insufficient reference count or extra pins). */ int folio_split_unmapped (struct folio *folio,unsigned int new_order) {
Signature · 1 line
2.51.1
Quoted 60 lines · Balbir Singh
Code refactoring of
__folio_split()via helper__folio_freeze_and_split_unmapped()caused a regression with clang-20 withCONFIG_SHMEM=n, the compiler was not able to optimize away the call toshmem_uncharge()due to changes innr_shmem_dropped. Fix this by adding a stub function forshmem_unchargewhenCONFIG_SHMEMis not defined.smatch also complained about parameter end being used without initialization, which is a false positive, but keep the tool happy by sending in initialized parameters. end is initialized to 0. smatch still complains about mapping being NULL and
nr_shmem_droppedmay not be 0, but that is not true prior to or after the changes.Add detailed documentation comments for
folio_split_unmapped()Cc Andrew Morton <[email protected]>Cc David Hildenbrand <[email protected]>Cc Zi Yan <[email protected]>Cc Joshua Hahn <[email protected]>Cc Rakie Kim <[email protected]>Cc Byungchul Park <[email protected]>Cc Gregory Price <[email protected]>Cc Ying Huang <[email protected]>Cc Alistair Popple <[email protected]>Cc Oscar Salvador <[email protected]>Cc Lorenzo Stoakes <[email protected]>Cc Baolin Wang <[email protected]>Cc "Liam R. Howlett" <[email protected]>Cc Nico Pache <[email protected]>Cc Ryan Roberts <[email protected]>Cc Dev Jain <[email protected]>Cc Barry Song <[email protected]>Cc Lyude Paul <[email protected]>Cc Danilo Krummrich <[email protected]>Cc David Airlie <[email protected]>Cc Simona Vetter <[email protected]>Cc Ralph Campbell <[email protected]>Cc Mika Penttilä <[email protected]>Cc Matthew Brost <[email protected]>Cc Francois Dugast <[email protected]>Suggested-by David Hildenbrand <[email protected]>Signed-off-by Balbir Singh <[email protected]>---
This fixup should be squashed into the patch "mm/huge_memory.c:
introducefolio_split_unmapped" inmm/mm-unstableinclude/linux/shmem_fs.h | 6 +++++- mm/huge_memory.c | 30 +++++++++++++++++++++--------- 2 files changed, 26 insertions(+), 10 deletions(-)include/linux/shmem_fs.h +1
-0@@ -136,11 +136,16 @@ static inline bool shmem_hpage_pmd_enabled(void)#ifdef CONFIG_SHMEM extern unsigned long shmem_swap_usage (struct vm_area_struct *vma); +extern void shmem_uncharge (struct inode *inode,long pages);
include/linux/shmem_fs.h static inline bool shmem_hpage_pmd_enabled(void) · line 149
Quoted 10 lines
include/linux/shmem_fs.h +4
-0#else static inline unsigned long shmem_swap_usage (struct vm_area_struct *vma) {return 0 ; } + +static void shmem_uncharge (struct inode *inode,long pages) +{ +}#endif
This is missing an inline attribute, my compilation tests did not catch it, but I'll do a new version tomorrow
Balbir
Balbir
kernel test robot
build botkernel test robot noticed the following build warnings:
Hi Balbir,
kernel test robot noticed the following build warnings:
[auto build test WARNING on akpm-mm/mm-everything]
[also build test WARNING on next-20251120]
[cannot apply to drm-misc/drm-misc-next drm-tip/drm-tip linus/master v6.18-rc6] [If your patch is applied to the wrong git tree, kindly drop us a note. And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]
url: https://github.com/intel-lab-lkp/linux/commits/Balbir-Singh/fixup-mm-huge_memory-c-introduce-folio_split_unmapped/20251120-214322 base: https://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm.git mm-everything patch link: https://lore.kernel.org/r/20251120134232.3588203-1-balbirs%40nvidia.com patch subject: [PATCH v2] fixup: mm/huge_memory.c: introduce folio_split_unmapped config: m68k-allnoconfig (https://download.01.org/0day-ci/archive/20251120/[email protected]/config) compiler: m68k-linux-gcc (GCC) 15.1.0 reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20251120/[email protected]/reproduce)
If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <[email protected]>
| Closes: https://lore.kernel.org/oe-kbuild-all/[email protected]/
All warnings (new ones prefixed by >>):
In file included from mm/filemap.c:39:
>> include/linux/shmem_fs.h:160:13: warning: 'shmem_uncharge' defined but not used [-Wunused-function]
160 | static void shmem_uncharge(struct inode *inode, long pages)
| ^~~~~~~~~~~~~~
--
mm/shmem.c: In function '__shmem_file_setup':
mm/shmem.c:5838:23: warning: unused variable 'flags' [-Wunused-variable]
5838 | unsigned long flags = (vm_flags & VM_NORESERVE) ? SHMEM_F_NORESERVE : 0;
| ^~~~~
In file included from mm/shmem.c:36:
include/linux/shmem_fs.h: At top level:
>> include/linux/shmem_fs.h:160:13: warning: 'shmem_uncharge' defined but not used [-Wunused-function] 160 | static void shmem_uncharge(struct inode *inode, long pages)
| ^~~~~~~~~~~~~~
vim +/shmem_uncharge +160 include/linux/shmem_fs.h
159
> 160 static void shmem_uncharge(struct inode *inode, long pages)
161 {
162 }
163 #endif
164 extern unsigned long shmem_partial_swap_usage(struct address_space *mapping,
165 pgoff_t start, pgoff_t end);
166
Signature · 2 lines
0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki
David Hildenbrand (Red Hat)
maintainerQuoted 56 lines · Balbir Singh
Code refactoring of
__folio_split()via helper__folio_freeze_and_split_unmapped()caused a regression with clang-20 withCONFIG_SHMEM=n, the compiler was not able to optimize away the call toshmem_uncharge()due to changes innr_shmem_dropped. Fix this by adding a stub function forshmem_unchargewhenCONFIG_SHMEMis not defined.smatch also complained about parameter end being used without initialization, which is a false positive, but keep the tool happy by sending in initialized parameters. end is initialized to 0. smatch still complains about mapping being NULL and
nr_shmem_droppedmay not be 0, but that is not true prior to or after the changes.Add detailed documentation comments for
folio_split_unmapped()Cc Andrew Morton <[email protected]>Cc David Hildenbrand <[email protected]>Cc Zi Yan <[email protected]>Cc Joshua Hahn <[email protected]>Cc Rakie Kim <[email protected]>Cc Byungchul Park <[email protected]>Cc Gregory Price <[email protected]>Cc Ying Huang <[email protected]>Cc Alistair Popple <[email protected]>Cc Oscar Salvador <[email protected]>Cc Lorenzo Stoakes <[email protected]>Cc Baolin Wang <[email protected]>Cc "Liam R. Howlett" <[email protected]>Cc Nico Pache <[email protected]>Cc Ryan Roberts <[email protected]>Cc Dev Jain <[email protected]>Cc Barry Song <[email protected]>Cc Lyude Paul <[email protected]>Cc Danilo Krummrich <[email protected]>Cc David Airlie <[email protected]>Cc Simona Vetter <[email protected]>Cc Ralph Campbell <[email protected]>Cc Mika Penttilä <[email protected]>Cc Matthew Brost <[email protected]>Cc Francois Dugast <[email protected]>Suggested-by David Hildenbrand <[email protected]>Signed-off-by Balbir Singh <[email protected]>---
This fixup should be squashed into the patch "mm/huge_memory.c:
introducefolio_split_unmapped" inmm/mm-unstableinclude/linux/shmem_fs.h | 6 +++++- mm/huge_memory.c | 30 +++++++++++++++++++++--------- 2 files changed, 26 insertions(+), 10 deletions(-)include/linux/shmem_fs.h +0
-0
include/linux/shmem_fs.h static inline bool shmem_hpage_pmd_enabled(void) · line 139
Quoted 4 lines
include/linux/shmem_fs.h +1
-0@@ -136,11 +136,16 @@ static inline bool shmem_hpage_pmd_enabled(void)#ifdef CONFIG_SHMEM extern unsigned long shmem_swap_usage (struct vm_area_struct *vma); +extern void shmem_uncharge (struct inode *inode,long pages);
I forgot to add, we drop the "extern" on functions whenever we move them.
include/linux/shmem_fs.h static inline bool shmem_hpage_pmd_enabled(void) · line 146
Quoted 7 lines
include/linux/shmem_fs.h +2
-0#else static inline unsigned long shmem_swap_usage (struct vm_area_struct *vma) {return 0 ; } + +static void shmem_uncharge (struct inode *inode,long pages)
"static inline void" should resolve the compilation issue.
Signature · 3 lines
Cheers David
Quoted 111 lines · Balbir Singh
Code refactoring of
__folio_split()via helper__folio_freeze_and_split_unmapped()caused a regression with clang-20 withCONFIG_SHMEM=n, the compiler was not able to optimize away the call toshmem_uncharge()due to changes innr_shmem_dropped. Fix this by adding a stub function forshmem_unchargewhenCONFIG_SHMEMis not defined.smatch also complained about parameter end being used without initialization, which is a false positive, but keep the tool happy by sending in initialized parameters. end is initialized to 0. smatch still complains about mapping being NULL and
nr_shmem_droppedmay not be 0, but that is not true prior to or after the changes.Add detailed documentation comments for
folio_split_unmapped()Cc Andrew Morton <[email protected]>Cc David Hildenbrand <[email protected]>Cc Zi Yan <[email protected]>Cc Joshua Hahn <[email protected]>Cc Rakie Kim <[email protected]>Cc Byungchul Park <[email protected]>Cc Gregory Price <[email protected]>Cc Ying Huang <[email protected]>Cc Alistair Popple <[email protected]>Cc Oscar Salvador <[email protected]>Cc Lorenzo Stoakes <[email protected]>Cc Baolin Wang <[email protected]>Cc "Liam R. Howlett" <[email protected]>Cc Nico Pache <[email protected]>Cc Ryan Roberts <[email protected]>Cc Dev Jain <[email protected]>Cc Barry Song <[email protected]>Cc Lyude Paul <[email protected]>Cc Danilo Krummrich <[email protected]>Cc David Airlie <[email protected]>Cc Simona Vetter <[email protected]>Cc Ralph Campbell <[email protected]>Cc Mika Penttilä <[email protected]>Cc Matthew Brost <[email protected]>Cc Francois Dugast <[email protected]>Suggested-by David Hildenbrand <[email protected]>Signed-off-by Balbir Singh <[email protected]>---
This fixup should be squashed into the patch "mm/huge_memory.c:
introducefolio_split_unmapped" inmm/mm-unstable
include/linux/shmem_fs.h| 6 +++++-mm/huge_memory.c| 30 +++++++++++++++++++++---------2 files changed, 26 insertions(+), 10 deletions(-)include/linux/shmem_fs.h +5
-1@@ -136,11 +136,16 @@ static inline bool shmem_hpage_pmd_enabled(void)#ifdef CONFIG_SHMEM extern unsigned long shmem_swap_usage (struct vm_area_struct *vma); +extern void shmem_uncharge (struct inode *inode,long pages);#else static inline unsigned long shmem_swap_usage (struct vm_area_struct *vma) {return 0 ; } + +static void shmem_uncharge (struct inode *inode,long pages) +{ +}#endif extern unsigned long shmem_partial_swap_usage (struct address_space *mapping,pgoff_t start,pgoff_t end);@@ -194,7 +199,6 @@ static inline pgoff_t shmem_fallocend(struct inode *inode, pgoff_t eof) }extern bool shmem_charge (struct inode *inode,long pages); -extern void shmem_uncharge (struct inode *inode,long pages);#ifdef CONFIG_USERFAULTFD #ifdef CONFIG_SHMEM mm/huge_memory.c +4
-5@@ -3751,6 +3751,7 @@ static int __folio_freeze_and_split_unmapped(struct folio *folio, unsigned int nint ret =0 ;struct deferred_split *ds_queue; +VM_WARN_ON_ONCE (!mapping && end);/* Prevent deferred_split_scan() touching ->_refcount */ ds_queue =folio_split_queue_lock (folio);if (folio_ref_freeze (folio,1 + extra_pins)) {@@ -3919,7 +3920,7 @@ static int __folio_split(struct folio *folio, unsigned int new_order,int nr_shmem_dropped =0 ;int remap_flags =0 ;int extra_pins, ret; -pgoff_t end; +pgoff_t end =0 ;bool is_hzp;VM_WARN_ON_ONCE_FOLIO (!folio_test_locked (folio), folio);@@ -4092,16 +4093,27 @@ static int __folio_split(struct folio *folio, unsigned int new_order,return ret; } -/* -* This function is a helper for splitting folios that have already been unmapped. -* The use case is that the device or the CPU can refuse to migrate THP pages in -* the middle of migration, due to allocation issues on either side +/** +* folio_split_unmapped() - split a large anon folio that is already unmapped
mm/huge_memory.c static int __folio_split(struct folio *folio, unsigned int new_order, · line 4107
Quoted 10 lines
mm/huge_memory.c +10
-0+* @folio: folio to split +* @new_order: the order of folios after split +* +* This function is a helper for splitting folios that have already been +* unmapped. The use case is that the device or the CPU can refuse to migrate +* THP pages in the middle of migration, due to allocation issues on either +* side. +* +* anon_vma_lock is not required to be held, mmap_read_lock() or +* mmap_write_lock() should be held. @folio is expected to be locked by the
Took a look into its caller:
__migrate_device_pages()
migrate_vma_split_unmapped_folio()
folio_split_unmapped()
I don't see where get the folio lock.
Would you mind giving me a hint where we toke the lock? Seems I missed that.
Quoted 19 lines
mm/huge_memory.c +7
-4+* caller. device-private and non device-private folios are supported along +* with folios that are in the swapcache. @folio should also be unmapped and +* isolated from LRU (if applicable) * -* The high level code is copied from __folio_split, since the pages are anonymous -* and are already isolated from the LRU, the code has been simplified to not -* burden __folio_split with unmapped sprinkled into the code. +* Upon return, the folio is not remapped, split folios are not added to LRU, +* free_folio_and_swap_cache() is not called, and new folios remain locked. * -* None of the split folios are unlocked +* Return: 0 on success, -EAGAIN if the folio cannot be split (e.g., due to +* insufficient reference count or extra pins). */ int folio_split_unmapped (struct folio *folio,unsigned int new_order) {Signature · 1 line
2.51.1
Signature · 2 lines
Wei Yang Help you, Help me