mm/memory: fix folio_set_dirty() vs. folio_mark_dirty() in zap_pte_range()
Subsystems (2)
PATCH 1/1 mm/memory: fix folio_set_dirty() vs. folio_mark_dirty() in zap_pte_range()
The correct folio replacement for "set_page_dirty()" is "folio_mark_dirty()", not "folio_set_dirty()". Using the latter won't properly inform the FS using the dirty_folio() callback.
This has been found by code inspection, but likely this can result in some real trouble when zapping dirty PTEs that point at clean pagecache folios.
--- mm/memory.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
mm/memory.c +1 -1
@@ -1464,7 +1464,7 @@ static unsigned long zap_pte_range(struct mmu_gather *tlb, delay_rmap =0 ;if (!folio_test_anon (folio)) {if (pte_dirty (ptent)) { -folio_set_dirty (folio); +folio_mark_dirty (folio);if (tlb_delay_rmap (tlb)) { delay_rmap =1 ; force_flush =1 ;
Signature · 1 line
2.43.0
Quoted 3 lines · David Hildenbrand
The correct folio replacement for "
set_page_dirty()" is "folio_mark_dirty()", not "folio_set_dirty()". Using the latter won't properly inform the FS using thedirty_folio()callback.
That set_page_dirty() naming is pretty nasty, hey.
Quoted 11 lines
This has been found by code inspection, but likely this can result in some real trouble when zapping dirty PTEs that point at clean pagecache folios.
Reported-by Ryan Roberts <[email protected]>Fixes c46265030b0f ("mm/memory: page_remove_rmap() -> folio_remove_rmap_pte()")Cc Matthew Wilcox (Oracle) <[email protected]>Cc Andrew Morton <[email protected]>Signed-off-by David Hildenbrand <[email protected]>
Quoted 17 lines
--- mm/memory.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)mm/memory.c +1
-1@@ -1464,7 +1464,7 @@ static unsigned long zap_pte_range(struct mmu_gather *tlb, delay_rmap =0 ;if (!folio_test_anon (folio)) {if (pte_dirty (ptent)) { -folio_set_dirty (folio); +folio_mark_dirty (folio);if (tlb_delay_rmap (tlb)) { delay_rmap =1 ; force_flush =1 ;
Without this fix, testing the latest exfat with xfstests, test cases generic/029 and generic/030 will fail.
Quoted 46 lines
-----Original Message-----
From: [email protected] <[email protected]> On Behalf Of David Hildenbrand
Sent: Tuesday, January 23, 2024 1:18 AM
To: [email protected]
Cc: [email protected]; David Hildenbrand <[email protected]>; Ryan Roberts <[email protected]>; Matthew Wilcox <[email protected]>; Andrew Morton <[email protected]>
Subject: [PATCH v1] mm/memory: fixfolio_set_dirty()vs.folio_mark_dirty()inzap_pte_range()The correct folio replacement for "
set_page_dirty()" is "folio_mark_dirty()", not "folio_set_dirty()". Using the latter won't properly inform the FS using thedirty_folio()callback.This has been found by code inspection, but likely this can result in some real trouble when zapping dirty PTEs that point at clean pagecache folios.
Reported-by Ryan Roberts <[email protected]>Closes:
https://lkml.kernel.org/r/[email protected]Fixes c46265030b0f ("mm/memory: page_remove_rmap() ->
folio_remove_rmap_pte()")Cc Matthew Wilcox (Oracle) <[email protected]>Cc Andrew Morton <[email protected]>Signed-off-by David Hildenbrand <[email protected]>Reviewed-by Ryan Roberts <[email protected]>--- mm/memory.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)mm/memory.c +1
-1@@ -1464,7 +1464,7 @@ static unsigned long zap_pte_range(structmmu_gather *tlb, delay_rmap =0 ;if (!folio_test_anon (folio)) {if (pte_dirty (ptent)) { -folio_set_dirty (folio); +folio_mark_dirty (folio);if (tlb_delay_rmap (tlb)) { delay_rmap =1 ;
force_flush= 1;
Quoted 4 lines · On 23.01.24 09:49, [email protected] wrote:
Reviewed-by Yuezhang Mo <[email protected]>Without this fix, testing the latest exfat with xfstests, test cases generic/029 and generic/030 will fail.
Great, thanks for testing and for providing actual reproducers!
Signature · 3 lines
Cheers, David / dhildenb