[lkml]   [1997]   [Jul]   [1]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: Page cache patch - less aggressive
Dr. Werner Fink wrote:
> During testing the new buffer code I found that find_candidate()
> called very often in refill_freelist() runs try_to_free_buffer()
> ... this seems to free the previous buffer pages allocated by
> grow_buffers() in refill_freelist() and leads to heavy swap I/O

find_candidate() only calls try_to_free_buffer() if the buffer size
doesn't match what it's looking for. Therefore it won't release the
buffer pages allocated by the previous grow_buffers().

However, if two tasks were simultaneously using refill_freelist to get
different sizes of buffers, they could end up releasing each other's
newly added buffers.

> Does any body know a simple fix for find_candidate() in the
> new buffer code to avoid this behaviour?

Glad you asked :-) I've found some bugs in refill_freelist() that could
explain the oopses people have been reporting. I'm working on a patch
and will post it later (hopefully today.)


 \ /
  Last update: 2005-03-22 13:39    [W:0.036 / U:4.188 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site