Re: [PATCH] Fix ACATS failures in GCC 3.4

[Laurent GUERBY <laurent at guerby dot net>]

On Thu, 2004-02-26 at 20:06, Zack Weinberg wrote:
> This is why the c-torture testsuite cycles through a bunch of
> different optimization options.
...
> Failures exclusively at -O0 and at -O1 do occur - they usually come
> from backend bugs handling constructs that are normally optimized
> out.  The most recent example I can think of was 

Yes I understand this, but since ACATS is quite expansive
to run and my year-long observation is that I never saw
an ACATS failure at -On and success at -On+1, a good trade-off
would be to just run once at higher optimisation level by default.
Careful people can always run more if they want, Ada is not
required these days anyway :). 

I did a run at -O2 on HEAD/x86-linux, it gets 8 failures, so
2 failures are missing from -O3:

c94008c	cc70a01

Run time is virtually identical at -O2 and -O3, however I agree
with Arnaud that -O2 is much more likely to be used in the Ada
case.

Should I submit a patch to HEAD to set a new default to "-O2"?

Laurent