This is the mail archive of the
glibc-linux@ricardo.ecn.wfu.edu
mailing list for the glibc project.
Re: Slow pthread_create() under high load
- To: David Wragg <dpw at doc dot ic dot ac dot uk>
- Subject: Re: Slow pthread_create() under high load
- From: Kaz Kylheku <kaz at ashi dot footprints dot net>
- Date: Sun, 26 Mar 2000 15:21:38 -0800 (PST)
- cc: glibc-linux at ricardo dot ecn dot wfu dot edu, linux-kernel at vger dot rutgers dot edu
- Reply-To: glibc-linux at ricardo dot ecn dot wfu dot edu
On 26 Mar 2000, David Wragg wrote:
> Date: 26 Mar 2000 17:36:18 +0000
> From: David Wragg <dpw@doc.ic.ac.uk>
> To: Kaz Kylheku <kaz@ashi.footprints.net>
> Cc: glibc-linux@ricardo.ecn.wfu.edu, linux-kernel@vger.rutgers.edu
> Subject: Re: Slow pthread_create() under high load
>
> Kaz Kylheku <kaz@ashi.footprints.net> writes:
> > Even if you make it a high priority thread, it does not start that
> > way; it becomes high priority after it executes the code which
> > performs the adjustment.
>
> This isn't necessarily so. sched_setscheduler() takes a pid parameter,
> so the parent thread and the child thread can both perform the
> adjustment, ensuring that on return from pthread_create() the new
> thread has the appropriate priority.
LinuxThreads doesn't do it this way, however. Perhaps it should!