Hi,
On 04/06/06 00:51, Alexander Kolbasov wrote:
Hello,
here is you chance to bite a piece of the scheduler code.
The purpose of the fix is to make idle thread less less aggressive in stealin
work from other CPUs when the work was only placed on the run-queue recently.
The diffs are available
On 02/02/06 23:22, Jonathan Adams wrote:
On Thu, Feb 02, 2006 at 03:01:12PM -0800, Roman wrote:
I thought an OS kernel was supposed to synchronise hardware counters
on multiple CPUs when it was loading??
It does when it can. On x86 hardware, depending on how the BIOS sets things
up, it ca
On 12/18/05 09:50, Josip Gracin wrote:
This happens on my Thinkpad R50 with ATI Radeon Mobility 9000. However,
the exact same behavior is present on my desktop Athlon64 3000+ with
NVidia GeForce 5200 card, first running S10GA and now running SXCE b28,
the very latest. Two completely differen
On 11/04/05 08:11, Andrei Dorofeev wrote:
BEGIN
{
ncpu = 2; /* XXX Can DTrace give us ncpus? */
}
Yes `ncpus will give you access to the kernel variable ncpus.
Qualify with a module name if necessary, as in unix`ncpus.
Cheers
Gavin
___
perf
On 08/18/05 08:17, Jonathan Adams wrote:
This is very cool; one additional point of interest might be
"non-DEBUG w/ kmem_flags = f", so that the kmem debugging overhead can
be factored out.
I put a TRAPTRACE-only build (a small part of DEBUG) through Perfpit
a while back to judge the impact o
On 07/02/05 07:30, Michael Schulte wrote:
I am of the opinion, that context switch time has nothing to do
with multithreaded applications (at least on Solaris) because a
multithreaded application is always in the same "context"
(because of shared global address space of such a program).
There
