Re: [PATCH] enhance strlen to understand MEM_REF and partial overlaps (PR 86042, 86043)

[Martin Sebor <msebor at gmail dot com>]

On 07/27/2018 03:05 AM, Eric Botcazou wrote:
> > I missed your approval and didn't get to committing the patch
> > until today.  While retesting it on top of fresh trunk I noticed
> > a few test failures due to other recent strlen changes.  I made
> > adjustments to the patch to avoid most of them and opened bug
> > 86688 for one that I think needs a separate code change and
> > xfailed the test cases until the bug gets resolved.
>
> But it has introduced a couple of regressions in the ACATS testsuite:
>
> FAIL:   c52103c
> FAIL:   cde0001
>
>                 === acats Summary ===
> # of expected passes            2318
> # of unexpected failures        2
> Native configuration is x86_64-suse-linux-gnu

Yes, I didn't see these failures, sorry.

FWIW, there are 128 failures in the GCC test suite on x86_64.
Many of these have been there for weeks (e.g., the lto failures
due to PR86004), even years (guality).  My script to compare
the results against a baseline uses the following regular
expression to extract the names of failing (and other
"interesting") tests.  As it happens, the Ada logs have
extra spaces before the test names that makes the script
exclude them from the list:

	    tlist=$(sed -n -e "s/^FAIL: \([^ ]*\) .*/\1/p" \
 			   -e "s/^XFAIL: \([^ ]*\) .*/\1/p" \
 			   -e "s/^UNRESOLVED: \([^ ]*\) .*/\1/p" \
			   -e "s/^XPASS: \([^ ]*\) .*/\1/p" \
			   test_summary.log \
			| sort -u)

Martin