On 01/27/2011 09:48 AM, Ken McDonell wrote:
Thanks Arthur ... I'll commit this patch.
But, can some who's familiar with the code in this area please cast an
eye over the proc_interrupts->syscall[cpu] and is_syscall code in this
same source file.
a) the is_syscall logic does not seem right compared to
the /proc/interrupts contents on my local system, and
b) proc_interrupts->syscall[cpu] is set but never returned in a
fetch ... which is why the apparent error in a) has not been noticed?
These oddities are independent of the other interrupts metrics that
Arthur's patch accommodates.
I'm not 100% sure, but I think the "SYS" entries in /proc/interrupts
might have been something I added to SGI only kernels many years
ago. This was an SGI-only patch for per-cpu system call counters.
(IIRC, the patch was never posted because it needed changes on every
arch).
Nathan and I just checked a few kernels, including as far back as
the RHEL4 2.6.9 kernel and there's no sign of any SYS entries in
/proc/interrupts, so unless someone else pipes up, I think this is
probably dead code and we can nuke it.
Cheers
-- Mark
|