Patchset for fixing poison failure of THP.
Get series with b4
b4 am [email protected]Versions newest first
- v1 Latest3 patches 1 replyIn review
Patches in v1 No applies recorded yet
| # | Subject and files | Size | Replies | Applied |
|---|---|---|---|---|
| 1/3 | Fix splitting poisoned THP.mm/ | +28 | 1 reply | |
| 2/3 | Fix the wrong reported address of poisoned THP.mm/ mm/ | +28 | 0 replies | |
| 3/3 | Fix unsuitable behavior for poisoned tail pages of THP.mm/ | +15 | 0 replies |
Reviews trailers per patch
All 3 patches have a Reviewed-by or Acked-by.
| Patch | Hidetoshi Seto |
|---|---|
| 1/3 | R-b |
| 2/3 | R-b |
| 3/3 | R-b |
Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes.
Cover letter [PATCH -v2 0/3] Patchset for fixing poison failure of THP.
I found the following problems while testing the poison of THP:
1. If the poisoned THP is failed to be split, the process could not be killed.
2. The reported address with sigbus is the address of the head page. This will make that the unexpected process on Guest OS(running on KVM) to be killed.
3. The test of poisoned tail page always fails. This is because the lru check of THP is done always, though the tail page is not set with PG_lru flag.
To resolve the above problems, I made the following patches.
Jin Dongming (3):
[PATCH 1/3] Fix splitting poisoned THP.
[PATCH 2/3] Fix the wrong reported address of poisoned THP. [PATCH 3/3] Fix unsuitable behavior for poisoned tail pages of THP.
mm/huge_memory.c | 7 ++++- mm/memory-failure.c | 84 +++++++++++++++++++++++++++++++++++++++----------- 2 files changed, 71 insertions(+), 20 deletions(-)