Messages in this thread | | | Date | Mon, 15 Apr 2002 01:47:39 -0700 | From | Andrew Morton <> | Subject | Re: [prepatch] address_space-based writeback |
| |
An update to these patches is at
http://www.zip.com.au/~akpm/linux/patches/2.5/2.5.8/
All work at this time is on
dallocbase-10-page_alloc_fail.patch dallocbase-35-pageprivate_fixes.patch dallocbase-55-ext2_dir.patch dallocbase-60-page_accounting.patch ratcache-pf_memalloc.patch mempool-node.patch dallocbase-70-writeback.patch ttd (my current 2.5 things-to-do-list, just for fun)
none of the patches beyond these have even been tested in a week..
The new buffer<->page relationship rules seem to be working out OK. In a six-hour stress test on a quad Xeon with 1k blocksize ext3 in ordered-data mode there was one failure: a block in a file which came up with wrong data. There appears to be a race in ext3 or the patch or 2.5 or something somewhere. Still hunting this one.
It is all relatively stable now. ramdisk, loop, reiserfs, ext2, ext3-ordered, ext3-journalled, JFS and vfat have been tested.
minixfs and sysvfs are broken with these patches. They rely on preservation of directory data outside i_size. Will fix.
- - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |