[lkml]   [2001]   [Aug]   [18]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
Subject2.4.9aa2 and please test mmap-rb-4
Diff between 2.4.9aa1 and 2.4.9aa2:

Only in 2.4.9aa2: 00_megaraid-paranoid-1

Possibly it's a bug in the anti-bounce buffer patch but make sure the
megaraid gets not confused by the fact I'm really allowing it to get
the 512k large requests with the sd-max_sectors patch (in mainline the
line I #ifdeffed out are meaningless so it never got tested).

Only in 2.4.9aa1: 00_o_direct-13
Only in 2.4.9aa2: 00_o_direct-14

Janet Morgan promptly noticed I needed to s/>=/</ (it was an editing
error while preparing the -13 patch).

Only in 2.4.9aa1: 00_poll-nfds-1
Only in 2.4.9aa2: 00_poll-nfds-2

Fixed off by one check.

Only in 2.4.9aa1: 40_blkdev-pagecache-13
Only in 2.4.9aa2: 40_blkdev-pagecache-14

Rediffed against o_direct-14.

Only in 2.4.9aa1: 70_mmap-rb-3
Only in 2.4.9aa2: 70_mmap-rb-4

Fixed severe bug that caused mm corruption and crashes. (splitted a
__vma_link_file off __vma_link_rb, build_mmap_rb must not mess with
anything but the rb itself) Improved other bits to cover all the
merging cases and some minor optimization.

Since I couldn't find other issues in my workloads I'd now ask
everybody who complained about the lack of vma merging in 2.4 to test
this patch on their systems and to report the performance difference
while running their vma intensive applications. Thanks!

You don't need to use the -aa tree to test this patch, I ported it
on top of 2.4.9 vanilla too, you can find it here:

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
More majordomo info at
Please read the FAQ at

 \ /
  Last update: 2005-03-22 12:57    [W:0.031 / U:1.456 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site