* [akpm-mm:mm-unstable 77/240] mm/filemap.c:3283:37-43: WARNING: Consider using vma_pages helper on vma (fwd)
@ 2025-07-07 8:19 Julia Lawall
2025-07-07 9:08 ` Ryan Roberts
0 siblings, 1 reply; 2+ messages in thread
From: Julia Lawall @ 2025-07-07 8:19 UTC (permalink / raw)
To: Ryan Roberts
Cc: Andrew Morton, Linux Memory Management List, Jan Kara, oe-kbuild-all
---------- Forwarded message ----------
Date: Mon, 7 Jul 2025 14:53:04 +0800
From: kernel test robot <lkp@intel.com>
To: oe-kbuild@lists.linux.dev
Cc: lkp@intel.com, Julia Lawall <julia.lawall@inria.fr>
Subject: [akpm-mm:mm-unstable 77/240] mm/filemap.c:3283:37-43: WARNING: Consider
using vma_pages helper on vma
BCC: lkp@intel.com
CC: oe-kbuild-all@lists.linux.dev
TO: Ryan Roberts <ryan.roberts@arm.com>
CC: Andrew Morton <akpm@linux-foundation.org>
CC: Linux Memory Management List <linux-mm@kvack.org>
CC: Jan Kara <jack@suse.cz>
tree: https://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm.git mm-unstable
head: 5aa46327724d1abe18a36e9dec4f26dba8989710
commit: b58187b2ba0af49c48448c3fe1b400cf8fa14232 [77/240] mm/filemap: allow arch to request folio size for exec memory
:::::: branch date: 33 hours ago
:::::: commit date: 33 hours ago
config: i386-randconfig-052-20250707 (https://download.01.org/0day-ci/archive/20250707/202507071459.lLPSj5Bd-lkp@intel.com/config)
compiler: gcc-12 (Debian 12.2.0-14+deb12u1) 12.2.0
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 <lkp@intel.com>
| Reported-by: Julia Lawall <julia.lawall@inria.fr>
| Closes: https://lore.kernel.org/r/202507071459.lLPSj5Bd-lkp@intel.com/
cocci warnings: (new ones prefixed by >>)
>> mm/filemap.c:3283:37-43: WARNING: Consider using vma_pages helper on vma
vim +3283 mm/filemap.c
4687fdbb805a92 Matthew Wilcox (Oracle 2021-07-24 3240)
b58187b2ba0af4 Ryan Roberts 2025-06-09 3241 /*
b58187b2ba0af4 Ryan Roberts 2025-06-09 3242 * If we don't want any read-ahead, don't bother. VM_EXEC case below is
b58187b2ba0af4 Ryan Roberts 2025-06-09 3243 * already intended for random access.
b58187b2ba0af4 Ryan Roberts 2025-06-09 3244 */
b58187b2ba0af4 Ryan Roberts 2025-06-09 3245 if ((vm_flags & (VM_RAND_READ | VM_EXEC)) == VM_RAND_READ)
6b4c9f4469819a Josef Bacik 2019-03-13 3246 return fpin;
275b12bf5486f6 Wu Fengguang 2011-05-24 3247 if (!ra->ra_pages)
6b4c9f4469819a Josef Bacik 2019-03-13 3248 return fpin;
ef00e08e26dd5d Linus Torvalds 2009-06-16 3249
dcfa24ba68991a Matthew Wilcox (Oracle 2022-05-25 3250) if (vm_flags & VM_SEQ_READ) {
6b4c9f4469819a Josef Bacik 2019-03-13 3251 fpin = maybe_unlock_mmap_for_io(vmf, fpin);
fcd9ae4f7f3b5f Matthew Wilcox (Oracle 2021-04-07 3252) page_cache_sync_ra(&ractl, ra->ra_pages);
6b4c9f4469819a Josef Bacik 2019-03-13 3253 return fpin;
ef00e08e26dd5d Linus Torvalds 2009-06-16 3254 }
ef00e08e26dd5d Linus Torvalds 2009-06-16 3255
207d04baa3591a Andi Kleen 2011-05-24 3256 /* Avoid banging the cache line if not needed */
e630bfac79456d Kirill A. Shutemov 2020-08-14 3257 mmap_miss = READ_ONCE(ra->mmap_miss);
e630bfac79456d Kirill A. Shutemov 2020-08-14 3258 if (mmap_miss < MMAP_LOTSAMISS * 10)
e630bfac79456d Kirill A. Shutemov 2020-08-14 3259 WRITE_ONCE(ra->mmap_miss, ++mmap_miss);
ef00e08e26dd5d Linus Torvalds 2009-06-16 3260
ef00e08e26dd5d Linus Torvalds 2009-06-16 3261 /*
ef00e08e26dd5d Linus Torvalds 2009-06-16 3262 * Do we miss much more than hit in this file? If so,
ef00e08e26dd5d Linus Torvalds 2009-06-16 3263 * stop bothering with read-ahead. It will only hurt.
ef00e08e26dd5d Linus Torvalds 2009-06-16 3264 */
e630bfac79456d Kirill A. Shutemov 2020-08-14 3265 if (mmap_miss > MMAP_LOTSAMISS)
6b4c9f4469819a Josef Bacik 2019-03-13 3266 return fpin;
ef00e08e26dd5d Linus Torvalds 2009-06-16 3267
b58187b2ba0af4 Ryan Roberts 2025-06-09 3268 fpin = maybe_unlock_mmap_for_io(vmf, fpin);
b58187b2ba0af4 Ryan Roberts 2025-06-09 3269 if (vm_flags & VM_EXEC) {
b58187b2ba0af4 Ryan Roberts 2025-06-09 3270 /*
b58187b2ba0af4 Ryan Roberts 2025-06-09 3271 * Allow arch to request a preferred minimum folio order for
b58187b2ba0af4 Ryan Roberts 2025-06-09 3272 * executable memory. This can often be beneficial to
b58187b2ba0af4 Ryan Roberts 2025-06-09 3273 * performance if (e.g.) arm64 can contpte-map the folio.
b58187b2ba0af4 Ryan Roberts 2025-06-09 3274 * Executable memory rarely benefits from readahead, due to its
b58187b2ba0af4 Ryan Roberts 2025-06-09 3275 * random access nature, so set async_size to 0.
b58187b2ba0af4 Ryan Roberts 2025-06-09 3276 *
b58187b2ba0af4 Ryan Roberts 2025-06-09 3277 * Limit to the boundaries of the VMA to avoid reading in any
b58187b2ba0af4 Ryan Roberts 2025-06-09 3278 * pad that might exist between sections, which would be a waste
b58187b2ba0af4 Ryan Roberts 2025-06-09 3279 * of memory.
b58187b2ba0af4 Ryan Roberts 2025-06-09 3280 */
b58187b2ba0af4 Ryan Roberts 2025-06-09 3281 struct vm_area_struct *vma = vmf->vma;
b58187b2ba0af4 Ryan Roberts 2025-06-09 3282 unsigned long start = vma->vm_pgoff;
b58187b2ba0af4 Ryan Roberts 2025-06-09 @3283 unsigned long end = start + ((vma->vm_end - vma->vm_start) >> PAGE_SHIFT);
b58187b2ba0af4 Ryan Roberts 2025-06-09 3284 unsigned long ra_end;
b58187b2ba0af4 Ryan Roberts 2025-06-09 3285
b58187b2ba0af4 Ryan Roberts 2025-06-09 3286 ra->order = exec_folio_order();
b58187b2ba0af4 Ryan Roberts 2025-06-09 3287 ra->start = round_down(vmf->pgoff, 1UL << ra->order);
b58187b2ba0af4 Ryan Roberts 2025-06-09 3288 ra->start = max(ra->start, start);
b58187b2ba0af4 Ryan Roberts 2025-06-09 3289 ra_end = round_up(ra->start + ra->ra_pages, 1UL << ra->order);
b58187b2ba0af4 Ryan Roberts 2025-06-09 3290 ra_end = min(ra_end, end);
b58187b2ba0af4 Ryan Roberts 2025-06-09 3291 ra->size = ra_end - ra->start;
b58187b2ba0af4 Ryan Roberts 2025-06-09 3292 ra->async_size = 0;
b58187b2ba0af4 Ryan Roberts 2025-06-09 3293 } else {
d30a11004e3411 Wu Fengguang 2009-06-16 3294 /*
d30a11004e3411 Wu Fengguang 2009-06-16 3295 * mmap read-around
d30a11004e3411 Wu Fengguang 2009-06-16 3296 */
db660d462525c4 David Howells 2020-10-15 3297 ra->start = max_t(long, 0, vmf->pgoff - ra->ra_pages / 2);
600e19afc5f8a6 Roman Gushchin 2015-11-05 3298 ra->size = ra->ra_pages;
600e19afc5f8a6 Roman Gushchin 2015-11-05 3299 ra->async_size = ra->ra_pages / 4;
28b31a2b2dbfc0 Ryan Roberts 2025-06-09 3300 ra->order = 0;
b58187b2ba0af4 Ryan Roberts 2025-06-09 3301 }
db660d462525c4 David Howells 2020-10-15 3302 ractl._index = ra->start;
28b31a2b2dbfc0 Ryan Roberts 2025-06-09 3303 page_cache_ra_order(&ractl, ra);
6b4c9f4469819a Josef Bacik 2019-03-13 3304 return fpin;
ef00e08e26dd5d Linus Torvalds 2009-06-16 3305 }
ef00e08e26dd5d Linus Torvalds 2009-06-16 3306
--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
^ permalink raw reply [flat|nested] 2+ messages in thread* Re: [akpm-mm:mm-unstable 77/240] mm/filemap.c:3283:37-43: WARNING: Consider using vma_pages helper on vma (fwd)
2025-07-07 8:19 [akpm-mm:mm-unstable 77/240] mm/filemap.c:3283:37-43: WARNING: Consider using vma_pages helper on vma (fwd) Julia Lawall
@ 2025-07-07 9:08 ` Ryan Roberts
0 siblings, 0 replies; 2+ messages in thread
From: Ryan Roberts @ 2025-07-07 9:08 UTC (permalink / raw)
To: Julia Lawall
Cc: Andrew Morton, Linux Memory Management List, Jan Kara, oe-kbuild-all
Thanks for the report.
Andrew, would you mind squashing in the below?
On 07/07/2025 09:19, Julia Lawall wrote:
>
>
> ---------- Forwarded message ----------
> Date: Mon, 7 Jul 2025 14:53:04 +0800
> From: kernel test robot <lkp@intel.com>
> To: oe-kbuild@lists.linux.dev
> Cc: lkp@intel.com, Julia Lawall <julia.lawall@inria.fr>
> Subject: [akpm-mm:mm-unstable 77/240] mm/filemap.c:3283:37-43: WARNING: Consider
> using vma_pages helper on vma
>
> BCC: lkp@intel.com
> CC: oe-kbuild-all@lists.linux.dev
> TO: Ryan Roberts <ryan.roberts@arm.com>
> CC: Andrew Morton <akpm@linux-foundation.org>
> CC: Linux Memory Management List <linux-mm@kvack.org>
> CC: Jan Kara <jack@suse.cz>
>
> tree: https://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm.git mm-unstable
> head: 5aa46327724d1abe18a36e9dec4f26dba8989710
> commit: b58187b2ba0af49c48448c3fe1b400cf8fa14232 [77/240] mm/filemap: allow arch to request folio size for exec memory
> :::::: branch date: 33 hours ago
> :::::: commit date: 33 hours ago
> config: i386-randconfig-052-20250707 (https://download.01.org/0day-ci/archive/20250707/202507071459.lLPSj5Bd-lkp@intel.com/config)
> compiler: gcc-12 (Debian 12.2.0-14+deb12u1) 12.2.0
>
> 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 <lkp@intel.com>
> | Reported-by: Julia Lawall <julia.lawall@inria.fr>
> | Closes: https://lore.kernel.org/r/202507071459.lLPSj5Bd-lkp@intel.com/
>
> cocci warnings: (new ones prefixed by >>)
>>> mm/filemap.c:3283:37-43: WARNING: Consider using vma_pages helper on vma
>
> vim +3283 mm/filemap.c
>
> 4687fdbb805a92 Matthew Wilcox (Oracle 2021-07-24 3240)
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3241 /*
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3242 * If we don't want any read-ahead, don't bother. VM_EXEC case below is
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3243 * already intended for random access.
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3244 */
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3245 if ((vm_flags & (VM_RAND_READ | VM_EXEC)) == VM_RAND_READ)
> 6b4c9f4469819a Josef Bacik 2019-03-13 3246 return fpin;
> 275b12bf5486f6 Wu Fengguang 2011-05-24 3247 if (!ra->ra_pages)
> 6b4c9f4469819a Josef Bacik 2019-03-13 3248 return fpin;
> ef00e08e26dd5d Linus Torvalds 2009-06-16 3249
> dcfa24ba68991a Matthew Wilcox (Oracle 2022-05-25 3250) if (vm_flags & VM_SEQ_READ) {
> 6b4c9f4469819a Josef Bacik 2019-03-13 3251 fpin = maybe_unlock_mmap_for_io(vmf, fpin);
> fcd9ae4f7f3b5f Matthew Wilcox (Oracle 2021-04-07 3252) page_cache_sync_ra(&ractl, ra->ra_pages);
> 6b4c9f4469819a Josef Bacik 2019-03-13 3253 return fpin;
> ef00e08e26dd5d Linus Torvalds 2009-06-16 3254 }
> ef00e08e26dd5d Linus Torvalds 2009-06-16 3255
> 207d04baa3591a Andi Kleen 2011-05-24 3256 /* Avoid banging the cache line if not needed */
> e630bfac79456d Kirill A. Shutemov 2020-08-14 3257 mmap_miss = READ_ONCE(ra->mmap_miss);
> e630bfac79456d Kirill A. Shutemov 2020-08-14 3258 if (mmap_miss < MMAP_LOTSAMISS * 10)
> e630bfac79456d Kirill A. Shutemov 2020-08-14 3259 WRITE_ONCE(ra->mmap_miss, ++mmap_miss);
> ef00e08e26dd5d Linus Torvalds 2009-06-16 3260
> ef00e08e26dd5d Linus Torvalds 2009-06-16 3261 /*
> ef00e08e26dd5d Linus Torvalds 2009-06-16 3262 * Do we miss much more than hit in this file? If so,
> ef00e08e26dd5d Linus Torvalds 2009-06-16 3263 * stop bothering with read-ahead. It will only hurt.
> ef00e08e26dd5d Linus Torvalds 2009-06-16 3264 */
> e630bfac79456d Kirill A. Shutemov 2020-08-14 3265 if (mmap_miss > MMAP_LOTSAMISS)
> 6b4c9f4469819a Josef Bacik 2019-03-13 3266 return fpin;
> ef00e08e26dd5d Linus Torvalds 2009-06-16 3267
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3268 fpin = maybe_unlock_mmap_for_io(vmf, fpin);
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3269 if (vm_flags & VM_EXEC) {
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3270 /*
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3271 * Allow arch to request a preferred minimum folio order for
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3272 * executable memory. This can often be beneficial to
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3273 * performance if (e.g.) arm64 can contpte-map the folio.
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3274 * Executable memory rarely benefits from readahead, due to its
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3275 * random access nature, so set async_size to 0.
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3276 *
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3277 * Limit to the boundaries of the VMA to avoid reading in any
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3278 * pad that might exist between sections, which would be a waste
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3279 * of memory.
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3280 */
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3281 struct vm_area_struct *vma = vmf->vma;
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3282 unsigned long start = vma->vm_pgoff;
> b58187b2ba0af4 Ryan Roberts 2025-06-09 @3283 unsigned long end = start + ((vma->vm_end - vma->vm_start) >> PAGE_SHIFT);
Please squash the following into Commit 9fc12248e06c ("mm/filemap: allow arch to
request folio size for exec memory") (that's the SHA it has in today's mm-
unstable), to use the vma_pages() helper instead of open-coding:
---8<---
diff --git a/mm/filemap.c b/mm/filemap.c
index 957189a12968..0d0369fb5fa1 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -3279,7 +3279,7 @@ static struct file *do_sync_mmap_readahead(struct vm_fault *vmf)
*/
struct vm_area_struct *vma = vmf->vma;
unsigned long start = vma->vm_pgoff;
- unsigned long end = start + ((vma->vm_end - vma->vm_start) >> PAGE_SHIFT);
+ unsigned long end = start + vma_pages(vma);
unsigned long ra_end;
ra->order = exec_folio_order();
---8<---
Thanks,
Ryan
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3284 unsigned long ra_end;
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3285
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3286 ra->order = exec_folio_order();
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3287 ra->start = round_down(vmf->pgoff, 1UL << ra->order);
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3288 ra->start = max(ra->start, start);
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3289 ra_end = round_up(ra->start + ra->ra_pages, 1UL << ra->order);
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3290 ra_end = min(ra_end, end);
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3291 ra->size = ra_end - ra->start;
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3292 ra->async_size = 0;
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3293 } else {
> d30a11004e3411 Wu Fengguang 2009-06-16 3294 /*
> d30a11004e3411 Wu Fengguang 2009-06-16 3295 * mmap read-around
> d30a11004e3411 Wu Fengguang 2009-06-16 3296 */
> db660d462525c4 David Howells 2020-10-15 3297 ra->start = max_t(long, 0, vmf->pgoff - ra->ra_pages / 2);
> 600e19afc5f8a6 Roman Gushchin 2015-11-05 3298 ra->size = ra->ra_pages;
> 600e19afc5f8a6 Roman Gushchin 2015-11-05 3299 ra->async_size = ra->ra_pages / 4;
> 28b31a2b2dbfc0 Ryan Roberts 2025-06-09 3300 ra->order = 0;
> b58187b2ba0af4 Ryan Roberts 2025-06-09 3301 }
> db660d462525c4 David Howells 2020-10-15 3302 ractl._index = ra->start;
> 28b31a2b2dbfc0 Ryan Roberts 2025-06-09 3303 page_cache_ra_order(&ractl, ra);
> 6b4c9f4469819a Josef Bacik 2019-03-13 3304 return fpin;
> ef00e08e26dd5d Linus Torvalds 2009-06-16 3305 }
> ef00e08e26dd5d Linus Torvalds 2009-06-16 3306
>
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2025-07-07 9:08 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2025-07-07 8:19 [akpm-mm:mm-unstable 77/240] mm/filemap.c:3283:37-43: WARNING: Consider using vma_pages helper on vma (fwd) Julia Lawall
2025-07-07 9:08 ` Ryan Roberts
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox