HWPOISON 2.6.33 pre-merge posting
Get series with b4
b4 am [email protected]Versions newest first
- v1 Latest31 patches 29 repliesIn review
Patches in v1 No applies recorded yet
| # | Subject and files | Size | Replies | Applied |
|---|---|---|---|---|
| 1/31 | HWPOISON: Add Andi Kleen as hwpoison maintainer to MAINTAINERSMAINTAINERS | +9 | 0 replies | |
| 2/31 | HWPOISON: Be more aggressive at freeing non LRU cachesinclude/ mm/ | +23 | 0 replies | |
| 3/31 | page-types: add standard GPL license headerDocumentation/ | +13 | 0 replies | |
| 4/31 | HWPOISON: remove the anonymous entrymm/ | +0 | 0 replies | |
| 5/31 | HWPOISON: return ENXIO on invalid page numbermm/ | +6 | 0 replies | |
| 6/31 | HWPOISON: avoid grabbing the page count multiple times during madvise injectionmm/ mm/ | +4 | 0 replies | |
| 7/31 | HWPOISON: Turn ref argument into flags argumentinclude/ mm/ mm/ | +8 | 0 replies | |
| 8/31 | HWPOISON: abort on failed unmapmm/ | +15 | 0 replies | |
| 9/31 | HWPOISON: comment the possible set_page_dirty() racemm/ | +2 | 0 replies | |
| 10/31 | HWPOISON: comment dirty swapcache pagesmm/ | +4 | 0 replies | |
| 11/31 | HWPOISON: introduce delete_from_lru_cache()mm/ | +37 | 0 replies | |
| 12/31 | HWPOISON: remove the free buddy page handlermm/ | +4 | 0 replies | |
| 13/31 | HWPOISON: detect free buddy pages explicitlymm/ mm/ mm/ | +31 | 0 replies | |
| 14/31 | HWPOISON: Add unpoisoning supportDocumentation/ include/ include/ +2 more | +114 | 0 replies | |
| 15/31 | HWPOISON: make semantics of IGNORED/DELAYED clearmm/ | +7 | 0 replies | |
| 16/31 | HWPOISON: return 0 to indicate success reliablymm/ | +6 | 0 replies | |
| 17/31 | HWPOISON: add fs/device filtersDocumentation/ mm/ mm/ +1 more | +72 | 0 replies | |
| 18/31 | HWPOISON: limit hwpoison injector to known page typesDocumentation/ mm/ mm/ | +43 | 0 replies | |
| 19/31 | mm: export stable page flagsfs/ include/ include/ | +51 | 6 replies | |
| 20/31 | HWPOISON: add page flags filterDocumentation/ mm/ mm/ +1 more | +42 | 0 replies | |
| 21/31 | memcg: rename and export try_get_mem_cgroup_from_page()include/ mm/ | +10 | 0 replies | |
| 22/31 | memcg: add accessor to mem_cgroup.cssinclude/ mm/ | +12 | 0 replies | |
| 23/31 | HWPOISON: add memory cgroup filterDocumentation/ mm/ mm/ +1 more | +66 | 10 replies | |
| 24/31 | HWPOISON: add an interface to switch off/on all the page filtersmm/ mm/ mm/ | +11 | 0 replies | |
| 25/31 | HWPOISON: Don't do early filtering if filter is disabledmm/ | +3 | 0 replies | |
| 26/31 | HWPOISON: mention HWPoison in Kconfig entrymm/ | +1 | 0 replies | |
| 27/31 | HWPOISON: Use correct name for MADV_HWPOISON in documentationDocumentation/ | +1 | 0 replies | |
| 28/31 | HWPOISON: Use new shake_page in memory_failuremm/ | +8 | 0 replies | |
| 29/31 | HWPOISON: Undefine short-hand macros after use to avoid namespace conflictmm/ | +13 | 0 replies | |
| 30/31 | HWPOISON: Add soft page offline supportDocumentation/ drivers/ include/ +2 more | +297 | 0 replies | |
| 31/31 | HWPOISON: Add a madvise() injector for soft page offlininginclude/ mm/ | +13 | 13 replies |
Reviews trailers per patch
3 of 31 patches have a Reviewed-by or Acked-by.
| Patch | KAMEZAWA Hiroyuki | Matt Mackall |
|---|---|---|
| 1/31 | ||
| 2/31 | ||
| 3/31 | ||
| 4/31 | ||
| 5/31 | ||
| 6/31 | ||
| 7/31 | ||
| 8/31 | ||
| 9/31 | ||
| 10/31 | ||
| 11/31 | ||
| 12/31 | ||
| 13/31 | ||
| 14/31 | ||
| 15/31 | ||
| 16/31 | ||
| 17/31 | ||
| 18/31 | ||
| 19/31 | A-b A-b | |
| 20/31 | ||
| 21/31 | A-b | |
| 22/31 | A-b | |
| 23/31 | ||
| 24/31 | ||
| 25/31 | ||
| 26/31 | ||
| 27/31 | ||
| 28/31 | ||
| 29/31 | ||
| 30/31 | ||
| 31/31 |
Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes.
Cover letter [PATCH] [0/31] HWPOISON 2.6.33 pre-merge posting
These are the hwpoison updates for 2.6.33
I plan to send the following patchkit to Linus in a few days. Any additional review would be appreciated.
Major new features:
- Be more aggressive at flushing caches to get access to a page
- Various fixes for the core memory_failure path
- Handle free memory better by detecting higher-order buddy pages reliably too.
- Reliable return value for memory_failure. This allows to implement some other functionality later on.
- New soft offlining feature:
Offline a page without killing a process.
This allows to implement predictive failure analysis for memory, by watching error trends per page and offlining a page that has too many corrected errors. The policy is all in user space; the kernel just offlines the page and reports the errors.
The current git mcelog has support for using this interface.
- Provide a new sysfs interface for both hard and soft offlining. The existing debugfs interface is still there.
- unpoison support
Unpoison a page. This is mainly for testing, it does not do unpoisioning on the hardware level.
- hwpoison filter
Various filters to the hwpoison PFN error injection, including memcg, page type, block device and others.
This is used by the mce-test stress suite to protect the test suite itself and
This touches some code outside hwpoison, mostly for the memcg support and for the page types. All these changes are straight-forward, are in linux-next and have been posted before.
-Andi