lkml.org 
[lkml]   [1997]   [Jul]   [3]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
Subjectrefill_freelist futility
In the discussion of the changes to refill_freelist, one factor that has
been overlooked is that there are a number of places where the code will
block, allowing other tasks to steal away the buffers freed up by the
task running refill_freelist. This is especially true of the blocking
call to wakeup bdflush. Thus even when the refill_freelist finally gets
the needed amount of memory, little of it may be available after
returning.

Since the "needed" amount is more-or-less arbitrary, wouldn't it be
better to return if _some_ buffers have been freed? That is, wake up
bdflush if there are dirty buffers, but return to the caller instead of
waiting.

\
 
 \ /
  Last update: 2005-03-22 13:39    [W:0.025 / U:21.680 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site