lkml.org 
[lkml]   [2008]   [May]   [9]   [last100]   RSS-feed
Views: [more markup]  [less markup]  [headers]  [forward] 
 
Messages in this thread
Complete thread (EXPERIMENTAL)
/
SubjectRe: [patch] speed up / fix the new generic semaphore code (fix AIM7 40% regression with 2.6.26-rc1)
FromAndi Kleen <>
DateFri, 09 May 2008 10:29:14 +0200
Digg This
Linus Torvalds <torvalds@linux-foundation.org> writes:

Well, it turns out that Ingo's fixed statistics actually put the real cost in fcntl/ioctl/open/release:

310 down <= lock_kernel <= sys_fcntl <= system_call_after_swapgs <

That must be ->fasync? If it was file locks the lock_kernel would not be inlined into sys_fcntl. Or is that an truncated back trace?
-Andi (wondering if he should plug ->fasync_unlocked again ...)


\
ISP Services
Valid XHTML 1.0!\ /
Valid CSS! Last update: 2008-05-09 10:33    [W:0.927 / U:0.130 seconds]
©2003-2005 Jasper Spaans