Hey all -
I've got a couple of simple questions that haven't been able to figure
out by looking in the source or in any of the kernel docs. (Please clue
me in if the answers to these are in some document somewhere, and I
promise to RTFM. ;-)
While doing some kernel profiling w/ the SGI patches, quite a lot of the
CPU time (typically 7% in network-heavy tests I've been doing) seems to
get sucked up by do_softirq(). This surprised me a bit, b/c AFAIK both
the network stuff and the profiling should generate hardware
interrupts. Does anyone have any suggestions as to what might be
triggering so many soft IRQs? Any suggestions on how I can examine this
more thorougly?
My other question is w/ regard to coding style in schedule.c. I noticed
that schedule() has a large number of constructs like this:
if(condition)
goto label;
label_back:
....
label: do_stuff()
goto label_back;
I'm probably showing my naivette as a C hacker, but why the gotos
instead of just inline functions?
Best regards,
-Rob von Behren
time spent in do_softirq()
strange goto statements in schedule() from schedule.c
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Thu Jun 15 2000 - 21:00:28 EST