This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] Check for binutils .cfi_* advance bug with code alignment ?factor > 1 (PR target/37610)
> I guess better would be to improve the search for readelf. If you use gas
> (I don't think Sun as supports .cfi_* directives), then you should have
> corresponding readelf somewhere around.
Sure, but the canonical default is 'no' for objdump-based tests so I see no
reason why this should be different for readelf-based tests. See just above
in the series of CFI tests:
else
# no objdump, err on the side of caution
gcc_cv_as_cfi_directive=no
fi
That's rather critical for any non-Linux/BSD systems.
> Is it because it is installed as greadelf instead of readelf, or under
> which name and where it is usually installed?
On Solaris you usually configure --with-as --with-ld explicitly when you want
to use GNU binutils so readelf can be anywhere.
--
Eric Botcazou