On 4/19/23 17:14, Bernhard Reutner-Fischer via Gcc-patches wrote:
On Wed, 19 Apr 2023 at 03:03, Jerry D via Fortran <fortran@gcc.gnu.org> wrote:
On 4/18/23 12:39 PM, Harald Anlauf via Fortran wrote:
Dear all,
the attached patch adjusts the scan-tree-dump patterns of the
reported testcases which likely were run in a location such
that a path in an error message showing in the tree-dump might
have accidentally matched "free" or "data", respectively.
For the testcase gfortran.dg/reshape_8.f90 I checked with a
failing gfortran-11 that the pattern is appropriate.
OK for mainline?
Thanks,
Harald
Yes, OK
I'm certainly not opposed to this specific incarnation of such a fix.
These failures are really unpleasant :)
As proposed in
https://inbox.sourceware.org/gcc-patches/20220426010029.2b476337@nbbrfq/
we could add a -fno-file to suppress the assembler .file output
(whatever the prefix looks like depends on the assembler dialect). Or
we could nuke the .file directives by a sed(1), but that would
probably be cumbersome for remote targets. I don't have a better idea
than -fno-file or -ffile=foo.c .
Fixing them case-by-case does not scale all that well IMHO.
Thoughts?
?
It wasn't the tree-dumps being at fault, it was the scan patterns.
