Improve hugetlbfs read on HWPOISON hugepages
Status from this message: [merged mm-stable] selftests-mm-add-tests-for-hwpoison-hugetlbfs-read.patch removed from -mm tree
Get series with b4
b4 am [email protected]Versions newest first · linked by author and title
- v4 Latest4 patches 14 repliesApplied 4/4
- v34 patches 12 repliesSuperseded
- v24 patches 14 repliesSuperseded
Earlier versions · 1
| Version | Posted | Patches | Replies | Status |
|---|---|---|---|---|
| v1 | 3 patches | 25 replies | Superseded |
Changes in v4 from the cover letter
* incorporate feedbacks from Matthew Wilcox <[email protected]>, Miaohe Lin <[email protected]>, Mike Kravetz <[email protected]> and Naoya Horiguchi <[email protected]>.
* rename is_raw_hwp_subpage => is_raw_hwpoison_page_in_hugepage.
* instead of taking hugetlb_lock, is_raw_hwpoison_page_in_hugepage needs to hold mf_mutex.
* use llist_for_each_entry, instead of llist_for_each_entry_safe in is_raw_hwpoison_page_in_hugepage.
* is_raw_hwpoison_page_in_hugepage doesn't need the folio argument.
* no need to export struct raw_hwp_page to header file.
* v4 is based on commit fd3006d2d0e7 ("Sync mm-stable with v6.5-rc1.").
Patches in v4 All 4 applied
| # | Subject and files | Size | Replies | Applied |
|---|---|---|---|---|
| 1/4 | mm/hwpoison: delete all entries before traversal in __folio_free_raw_hwpmm/ | +3 | 0 replies | Applied |
| 2/4 | mm/hwpoison: check if a raw page in a hugetlb folio is raw HWPOISONinclude/ mm/ | +43 | 0 replies | Applied |
| 3/4 | hugetlbfs: improve read HWPOISON hugepagefs/ | +51 | 0 replies | Applied |
| 4/4 | selftests/mm: add tests for HWPOISON hugetlbfs readtools/ tools/ tools/ | +324 | 14 replies | Applied |
Reviews trailers per patch
All 4 patches have a Reviewed-by or Acked-by.
| Patch | Mike Kravetz | Naoya Horiguchi | Miaohe Lin | Muhammad Usama Anjum |
|---|---|---|---|---|
| 1/4 | A-b | A-b | ||
| 2/4 | R-b | R-b | R-b | |
| 3/4 | R-b | R-b | ||
| 4/4 | A-b | R-b | T-b |
Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes. Scroll the table sideways for all 4 reviewers.
Cover letter [PATCH v4 0/4] Improve hugetlbfs read on HWPOISON hugepages
Today when hardware memory is corrupted in a hugetlb hugepage, kernel leaves the hugepage in pagecache [1]; otherwise future mmap or read will suject to silent data corruption. This is implemented by returning -EIO from hugetlb_read_iter immediately if the hugepage has HWPOISON flag set.
Since memory_failure already tracks the raw HWPOISON subpages in a hugepage, a natural improvement is possible: if userspace only asks for healthy subpages in the pagecache, kernel can return these data.
This patchset implements this improvement. It consist of three parts. The 1st commit exports the functionality to tell if a subpage inside a hugetlb hugepage is a raw HWPOISON page. The 2nd commit teaches hugetlbfs_read_iter to return as many healthy bytes as possible. The 3rd commit properly tests this new feature.
[1] commit 8625147cafaa ("hugetlbfs: don't delete error page from pagecache")
Changelog
v3 => v4
* incorporate feedbacks from Matthew Wilcox <[email protected]>, Miaohe Lin <[email protected]>, Mike Kravetz <[email protected]> and Naoya Horiguchi <[email protected]>.
* rename is_raw_hwp_subpage => is_raw_hwpoison_page_in_hugepage.
* instead of taking hugetlb_lock, is_raw_hwpoison_page_in_hugepage needs to hold mf_mutex.
* use llist_for_each_entry, instead of llist_for_each_entry_safe in is_raw_hwpoison_page_in_hugepage.
* is_raw_hwpoison_page_in_hugepage doesn't need the folio argument.
* no need to export struct raw_hwp_page to header file.
* v4 is based on commit fd3006d2d0e7 ("Sync mm-stable with v6.5-rc1.").
v2 => v3
* Update commit messages for future reader to know background and code details.
* v3 is based on commit 5bb367dca2b9 ("Merge branch 'master' into mm-stable").
v1 => v2
* incorporate feedbacks from both Mike Kravetz <[email protected]> and Naoya Horiguchi <[email protected]>.
* __folio_free_raw_hwp deletes all entries in raw_hwp_list before it traverses and frees raw_hwp_page.
* find_raw_hwp_page => __is_raw_hwp_subpage and __is_raw_hwp_subpage only returns bool instead of a raw_hwp_page entry.
* is_raw_hwp_subpage holds hugetlb_lock while checking __is_raw_hwp_subpage.
* No need to do folio_lock in adjust_range_hwpoison.
* v2 is based on commit a6e79df92e4a ("mm/gup: disallow FOLL_LONGTERM GUP-fast writing to file-backed mappings").
Jiaqi Yan (4):
mm/hwpoison: delete all entries before traversal in
__folio_free_raw_hwp
mm/hwpoison: check if a raw page in a hugetlb folio is raw HWPOISON
hugetlbfs: improve read HWPOISON hugepage
selftests/mm: add tests for HWPOISON hugetlbfs read
fs/hugetlbfs/inode.c | 57 +++- include/linux/hugetlb.h | 5 + mm/memory-failure.c | 48 ++- tools/testing/selftests/mm/.gitignore | 1 + tools/testing/selftests/mm/Makefile | 1 + .../selftests/mm/hugetlb-read-hwpoison.c | 322 ++++++++++++++++++ 6 files changed, 421 insertions(+), 13 deletions(-) create mode 100644 tools/testing/selftests/mm/hugetlb-read-hwpoison.c
Signature · 1 line
2.41.0.255.g8b1d071c50-goog