This is the mail archive of the
mailing list for the GCC project.
Re: [PATCH]: Add CSE pass directly after expand / Add register equal notes for divmod4 for AVR
- From: Steven Bosscher <stevenb at suse dot de>
- To: Björn Haase <Bjoern dot M dot Haase at web dot de>
- Cc: gcc-patches at gcc dot gnu dot org
- Date: Thu, 16 Jun 2005 21:12:15 +0200
- Subject: Re: [PATCH]: Add CSE pass directly after expand / Add register equal notes for divmod4 for AVR
- References: <firstname.lastname@example.org>
On Thursday 16 June 2005 21:10, Björn Haase wrote:
> Steven Bosscher <email@example.com> schrieb am 16.06.05 16:43:45:
> > I don't see these testsuite differences on my amd64 box. Did you have
> > other differeces (i.e. other than your patch) between your reference
> > run and test run?
> > Gr.
> > Steven
> I have been working on a fresh checkout. Could it make a difference if you
> are having different processor loads for different testsuite runs? IIRC, I
> had two testsuites running simultaneously when testing with the patch (the
> AVR run on a simulator and the native run).
This definitely makes a difference for the 20001226 failures. Look
at the test case to see why ;-) It is a notorious cycle-sucker.
For the other test, I don't know.