Re: [tree-ssa] performance regression after yesterdays mainline merge

[Janis Johnson <janis187 at us dot ibm dot com>]

On Fri, Apr 02, 2004 at 11:09:57AM +0200, Richard Guenther wrote:
> Hi!
> 
> The automatic tester at
> http://www.tat.physik.uni-tuebingen.de/~rguenth/gcc/monitor-summary.html
> 
> caught a performance regression of the compiled tramp3d-v3 code. It's a
> 7.43s to 7.64s (per iteration) aka 3% regression.
> 
> As this is between a mainline merge I suspect something from there causing
> this regression (I don't monitor mainline, maybe I should).  Usually
> regressins in this test are caused by FP loop pessimizations (as nearly
> all time is spent there), and as this is ia64 I suspected
> 
> 2004-03-29  Fariborz Jahanian <fjahanian@apple.com>
> 
>         * fold-const.c (fold): Reassociate multiply expression
>         with an adjacent non-multiply expression to use
>         architecture's multiply-add instruction.
> 
> to be the cause of the regression, but just reverting the above patch
> causes another regression to 7.69s - so I'm lost.  Diffs of assembly are
> not helpful, as all non-label differences seem to be complete different
> compiled routines.
> 
> But you may notice the raise in bootstrap time which might (didnt check)
> be caused by a regression in gengtype which consumes quite a lot of
> bootstrap time on ia64.

There's an example in http://gcc.gnu.org/bugs/reghunt.html of getting
sources from a branch for a particular date.  If you suspect a particular
patch on a branch of causing some sort of regression, get the UTC time of
its checkin from the cvs log and then get sources from just before and
just after it was added.  This is much more reliable than reverting the
patch from a later version.

Janis