page_mkwrite fixes
Get series with b4
b4 am [email protected]Versions newest first
- v1 Latest6 patches 1 replyIn review
Patches in v1 No applies recorded yet
| # | Subject and files | Size | Replies | Applied |
|---|---|---|---|---|
| 1/6 | mm: page_mkwrite change prototype to match faultDocumentation/ drivers/ fs/ +11 more | +60 | 0 replies | |
| 2/6 | fs: fix page_mkwrite error cases in core code and btrfsfs/ | +8 | 0 replies | |
| 3/6 | mm: close page_mkwrite racesDocumentation/ fs/ mm/ | +98 | 0 replies | |
| 4/6 | GFS2: Fix page_mkwrite() return codefs/ | +3 | 0 replies | |
| 5/6 | NFS: Fix the return value in nfs_page_mkwrite()fs/ | +0 | 0 replies | |
| 6/6 | NFS: Close page_mkwrite() racesfs/ | +3 | 0 replies |
Reviews trailers per patch
1 of 6 patches has a Reviewed-by or Acked-by.
| Patch | Chris Mason |
|---|---|
| 1/6 | |
| 2/6 | A-b |
| 3/6 | |
| 4/6 | |
| 5/6 | |
| 6/6 |
Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes.
Cover letter [2.6.27.stable][patch 0/6] page_mkwrite fixes
Hi stable team,
This is the sequence of applicable fixes for 2.6.27.stable to solve the page_mkwrite issues. The series solves 2 problems in the core APIs:
- the problem of incorrect return values (eg. causing SIGBUS when there is an OOM or page invalidation).
- the problem of insufficient synchronisation between mm and filesystem, which results in at least NFS being quite easily broken with mmap()ed writes.
And it also updates several filesystems to fix the issues on their side.
I think it has been agreed to get these fixes into .stable kernels (they are all upstream).
One issue is that there are several filesystems (basically, any which supply ->page_mkwrite) which could be broken but maybe have not had anybody look at them yet (OCFS2, UBIFS maybe). Anyway, if fs maintainers eventually do add any fixes here, it would be appreciated if you keep .stable in mind as well.
Thanks,
Nick