| To: | Greg Lindahl <lindahl@keyresearch.com> |
|---|---|
| Subject: | Re: [RFC] FPU context switch |
| From: | Daniel Jacobowitz <dan@debian.org> |
| Date: | Tue, 17 Sep 2002 15:38:20 -0400 |
| Cc: | Jun Sun <jsun@mvista.com>, linux-mips@linux-mips.org |
| In-reply-to: | <20020917120310.A2043@wumpus.internal.keyresearch.com> |
| Original-recipient: | rfc822;linux-mips@linux-mips.org |
| References: | <20020917110423.E17321@mvista.com> <1032288140.28433.165.camel@gs256.sp.cs.cmu.edu> <20020917114831.G17321@mvista.com> <20020917120310.A2043@wumpus.internal.keyresearch.com> |
| Sender: | linux-mips-bounce@linux-mips.org |
| User-agent: | Mutt/1.5.1i |
On Tue, Sep 17, 2002 at 12:03:10PM -0700, Greg Lindahl wrote: > On Tue, Sep 17, 2002 at 11:48:31AM -0700, Jun Sun wrote: > > > I think this gives a big performance improvement because most processes > > don't use FPU during their runs but they all have used_math flag set! > > Jun, > > You really ought to prove that first. Many people spend a lot of time > optimizing things that aren't important. If it isn't important, than > the simplest scheme is the best choice. Oh, he's quite correct. There's a setjmp() early in the execution path, and it saves FP registers on machines with FP support configured on. So tasks are marked as FPU users. I've never thought of a terribly good way around this. -- Daniel Jacobowitz MontaVista Software Debian GNU/Linux Developer |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: [PATCH] ip22 console selection fixes, Florian Lohoff |
|---|---|
| Next by Date: | Re: cannot debug multi-threaded programs with gdb/gdbserver, Steven J. Hill |
| Previous by Thread: | Re: [RFC] FPU context switch, Greg Lindahl |
| Next by Thread: | Re: [RFC] FPU context switch, Dominic Sweetman |
| Indexes: | [Date] [Thread] [Top] [All Lists] |