mm/swap: Fix release_pages() when releasing devmap pages
PATCH v4 1/1 mm/swap: Fix release_pages() when releasing devmap pages
From: Ira Weiny <[email protected]>
release_pages() is an optimized version of a loop around put_page(). Unfortunately for devmap pages the logic is not entirely correct in release_pages(). This is because device pages can be more than type MEMORY_DEVICE_PUBLIC. There are in fact 4 types, private, public, FS DAX, and PCI P2PDMA. Some of these have specific needs to "put" the page while others do not.
This logic to handle any special needs is contained in put_devmap_managed_page(). Therefore all devmap pages should be processed by this function where we can contain the correct logic for a page put.
Handle all device type pages within release_pages() by calling put_devmap_managed_page() on all devmap pages. If put_devmap_managed_page() returns true the page has been put and we continue with the next page. A false return of put_devmap_managed_page() means the page did not require special processing and should fall to "normal" processing.
This was found via code inspection while determining if release_pages() and the new put_user_pages() could be interchangeable.[1]
[1] https://lore.kernel.org/lkml/[email protected]/
---
Changes from V3:
Update comment to the one provided by John
Changes from V2:
Update changelog for more clarity as requested by Michal Update comment WRT "failing" of put_devmap_managed_page()
Changes from V1:
Add comment clarifying that put_devmap_managed_page() can still fail.
Add Reviewed-by tags.
mm/swap.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-)
mm/swap.c +9 -4
@@ -740,15 +740,20 @@ void release_pages(struct page **pages, int nr)if (is_huge_zero_page (page))continue ; -/* Device public page can not be huge page */ -if (is_device_public_page (page)) { +if (is_zone_device_page (page)) {if (locked_pgdat) {spin_unlock_irqrestore (&locked_pgdat->lru_lock, flags); locked_pgdat =NULL ; } -put_devmap_managed_page (page); -continue ; +/* +* ZONE_DEVICE pages that return 'false' from +* put_devmap_managed_page() do not require special +* processing, and instead, expect a call to +* put_page_testzero(). +*/ +if (put_devmap_managed_page (page)) +continue ; } page =compound_head (page);
Signature · 1 line
2.20.1
Sorry for a late reply.
Quoted 21 lines · On Wed 05-06-19 14:49:22, [email protected] wrote:
From: Ira Weiny <[email protected]>
release_pages()is an optimized version of a loop aroundput_page(). Unfortunately for devmap pages the logic is not entirely correct inrelease_pages(). This is because device pages can be more than typeMEMORY_DEVICE_PUBLIC. There are in fact 4 types, private, public, FS DAX, and PCI P2PDMA. Some of these have specific needs to "put" the page while others do not.This logic to handle any special needs is contained in
put_devmap_managed_page(). Therefore all devmap pages should be processed by this function where we can contain the correct logic for a page put.Handle all device type pages within
release_pages()by callingput_devmap_managed_page()on all devmap pages. Ifput_devmap_managed_page()returns true the page has been put and we continue with the next page. A false return ofput_devmap_managed_page()means the page did not require special processing and should fall to "normal" processing.
Quoted 2 lines
This was found via code inspection while determining if
release_pages()and the newput_user_pages()could be interchangeable.[1]
This is much more clear than the previous version I've looked at. Thanks a lot!
Quoted 8 lines
[1] https://lore.kernel.org/lkml/[email protected]/
Cc Jérôme Glisse <[email protected]>Cc Michal Hocko <[email protected]>Reviewed-by Dan Williams <[email protected]>Reviewed-by John Hubbard <[email protected]>Signed-off-by Ira Weiny <[email protected]>
Quoted 49 lines
---
Changes from V3:
Update comment to the one provided by JohnChanges from V2:
Update changelog for more clarity as requested by Michal Update comment WRT "failing" ofput_devmap_managed_page()Changes from V1:
Add comment clarifying thatput_devmap_managed_page()can still fail.
Add Reviewed-by tags.mm/swap.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-)mm/swap.c +9
-4@@ -740,15 +740,20 @@ void release_pages(struct page **pages, int nr)if (is_huge_zero_page (page))continue ; -/* Device public page can not be huge page */ -if (is_device_public_page (page)) { +if (is_zone_device_page (page)) {if (locked_pgdat) {spin_unlock_irqrestore (&locked_pgdat->lru_lock, flags); locked_pgdat =NULL ; } -put_devmap_managed_page (page); -continue ; +/* +* ZONE_DEVICE pages that return 'false' from +* put_devmap_managed_page() do not require special +* processing, and instead, expect a call to +* put_page_testzero(). +*/ +if (put_devmap_managed_page (page)) +continue ; } page =compound_head (page);Signature · 1 line
2.20.1
Signature · 2 lines
Michal Hocko SUSE Labs
Quoted 40 lines · On Wed, Jun 05, 2019 at 02:49:22PM -0700, [email protected] wrote:
From: Ira Weiny <[email protected]>
release_pages()is an optimized version of a loop aroundput_page(). Unfortunately for devmap pages the logic is not entirely correct inrelease_pages(). This is because device pages can be more than typeMEMORY_DEVICE_PUBLIC. There are in fact 4 types, private, public, FS DAX, and PCI P2PDMA. Some of these have specific needs to "put" the page while others do not.This logic to handle any special needs is contained in
put_devmap_managed_page(). Therefore all devmap pages should be processed by this function where we can contain the correct logic for a page put.Handle all device type pages within
release_pages()by callingput_devmap_managed_page()on all devmap pages. Ifput_devmap_managed_page()returns true the page has been put and we continue with the next page. A false return ofput_devmap_managed_page()means the page did not require special processing and should fall to "normal" processing.This was found via code inspection while determining if
release_pages()and the newput_user_pages()could be interchangeable.[1][1] https://lore.kernel.org/lkml/[email protected]/
Cc Jérôme Glisse <[email protected]>Cc Michal Hocko <[email protected]>Reviewed-by Dan Williams <[email protected]>Reviewed-by John Hubbard <[email protected]>Signed-off-by Ira Weiny <[email protected]>---
Changes from V3:
Update comment to the one provided by JohnChanges from V2:
Update changelog for more clarity as requested by Michal Update comment WRT "failing" ofput_devmap_managed_page()
Quoted 7 lines
Changes from V1:
Add comment clarifying thatput_devmap_managed_page()can still fail.
Add Reviewed-by tags.mm/swap.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-)
Andrew,
As per the discussion on the hmm thread I took this patch into the hmm.git as the conflict that was created with CH's rework was tricky - the resolution is simple, but keeping Ira's hunk instead of the delete is, IMHO, subtle.
Regards,
Jason