mm/filemap: set folio->mapping to NULL before xas_store()
Subsystems (3)
PATCH 1/1 mm/filemap: set folio->mapping to NULL before xas_store()
Functions such as __filemap_get_folio() check the truncation of folios based on the mapping field. Therefore setting this field to NULL earlier prevents unnecessary operations on already removed folios.
--- mm/filemap.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
mm/filemap.c +3 -2
@@ -139,11 +139,12 @@ static void page_cache_delete(struct address_space *mapping,VM_BUG_ON_FOLIO (!folio_test_locked (folio), folio); + folio->mapping =NULL ; +/* Leave page->index set: truncation lookup relies upon it */ +xas_store (&xas, shadow);xas_init_marks (&xas); - folio->mapping =NULL ; -/* Leave page->index set: truncation lookup relies upon it */ mapping->nrpages -= nr; }
Signature · 1 line
2.25.1
Andrew Morton
maintainerQuoted 12 lines · On Sat, 23 Mar 2024 06:04:54 +0900 Soma Nakata <[email protected]> wrote:
Functions such as
__filemap_get_folio()check the truncation of folios based on the mapping field. Therefore setting this field to NULL earlier prevents unnecessary operations on already removed folios....
mm/filemap.c +0
-0@@ -139,11 +139,12 @@ static void page_cache_delete(struct address_space *mapping,VM_BUG_ON_FOLIO (!folio_test_locked (folio), folio);
mm/filemap.c static void page_cache_delete(struct address_space *mapping, · line 149
Quoted 10 lines
mm/filemap.c +3
-2+ folio->mapping =NULL ; +/* Leave page->index set: truncation lookup relies upon it */ +xas_store (&xas, shadow);xas_init_marks (&xas); - folio->mapping =NULL ; -/* Leave page->index set: truncation lookup relies upon it */ mapping->nrpages -= nr; }
Seems at least harmless, but I wonder if it can really make any difference. Don't readers of folio->mapping lock the folio first?
Matthew Wilcox
maintainerQuoted 19 lines · Andrew Morton
Quoted 18 lines · On Sat, 23 Mar 2024 06:04:54 +0900 Soma Nakata <[email protected]> wrote:
Functions such as
__filemap_get_folio()check the truncation of folios based on the mapping field. Therefore setting this field to NULL earlier prevents unnecessary operations on already removed folios....
mm/filemap.c +3
-0@@ -139,11 +139,12 @@ static void page_cache_delete(struct address_space *mapping,VM_BUG_ON_FOLIO (!folio_test_locked (folio), folio); + folio->mapping =NULL ; +/* Leave page->index set: truncation lookup relies upon it */ +xas_store (&xas, shadow);xas_init_marks (&xas);
Quoted 7 lines
mm/filemap.c static void page_cache_delete(struct address_space *mapping, · line 149
Quoted 4 lines
mm/filemap.c +0
-2- folio->mapping =NULL ; -/* Leave page->index set: truncation lookup relies upon it */ mapping->nrpages -= nr; }Seems at least harmless, but I wonder if it can really make any difference. Don't readers of
folio->mappinglock the folio first?
I can't think of anywhere that doesn't ... most of the places that check folio->mapping have "goto unlock" as the very next line. I don't think this patch accomplishes anything.
Quoted 21 lines · Andrew Morton
Quoted 18 lines · On Sat, 23 Mar 2024 06:04:54 +0900 Soma Nakata <[email protected]> wrote:
Functions such as
__filemap_get_folio()check the truncation of folios based on the mapping field. Therefore setting this field to NULL earlier prevents unnecessary operations on already removed folios....
mm/filemap.c +3
-0@@ -139,11 +139,12 @@ static void page_cache_delete(struct address_space *mapping,VM_BUG_ON_FOLIO (!folio_test_locked (folio), folio); + folio->mapping =NULL ; +/* Leave page->index set: truncation lookup relies upon it */ +xas_store (&xas, shadow);xas_init_marks (&xas);
Quoted 7 lines
mm/filemap.c static void page_cache_delete(struct address_space *mapping, · line 149
Quoted 4 lines
mm/filemap.c +0
-2- folio->mapping =NULL ; -/* Leave page->index set: truncation lookup relies upon it */ mapping->nrpages -= nr; }Seems at least harmless, but I wonder if it can really make any difference. Don't readers of
folio->mappinglock the folio first?
Yes, the reader locks the folio.
Only __filemap_remove_folio() calls page_cache_delete(), and it says the caller has to lock the folio or make sure that usage is safe. In the latter case, this patch improves efficiency a little bit.
However, I found that there is not any latter case actually, so discard it or apply, also to make the order of operations in page_cache_delete() and page_cache_delete_batch() the same for a cleanup.
Thanks,