Ingo Molnar wrote:
>
> On Tue, 14 Mar 2000, Manfred Spraul wrote:
>
> > I've seen such code from egcs-1.1.2, IIRC in do_softirq().
>
> oh! does this explain the do_softirq() weirdness you have noticed?
>
No, the asm output looks fine:
egcs loads local_bh_count into a register for the in_interrupt() test,
then it increases the register and stores the register back into
local_bh_count.
I have 2 explanations for the high do_softirq numbers:
* problem with readprofile (version 2.00, May 1996): it also reported
that sys_shutdown() got called, and I doubt that this is true ;)
[prof_shift=1]
* do_softirq reenables the interrupts after a hardware interrupt. If a
timer interrupt was pending, then do_softirq gets the profiling hit.
-- Manfred- 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 : Wed Mar 15 2000 - 21:00:29 EST