Hi Alex,
> On Feb 2, 2026, Rainer Orth <[email protected]> wrote:
>
>> PR other/123841
>> PR target/123926
>> * acinclude.m4 (gcc_GAS_FLAGS): Handle x86 and SPARC/gas cases
>> separately.
>> * configure: Regenerate.
>
> LGTM, thanks.
there's one thing I have begun to wonder about at this point: as is,
this patch restricts the use of gas options (--32/--x32/--64) to
assemblers that identify themselves as gas. However, this isn't really
about gas or non-gas at all but accepting those options. This is
similar to the Darwin situation where older assemblers identify as GNU
assembler (an ancient port of gas 1.38 actually), but still accept the
same options as the later clang-based assembler, both different from
what gas usually uses.
Maybe it would be better to use gas options irrespective of whether the
assembler self-identifies as gas at this point? We've already handled
the sparc and x86 cases that are known to differ: Solaris as and Darwin
as. I'm not really certain how to handle this. Or just commit as is
and wait if there's yet more fallout so we know for certain?
Thanks.
Rainer
--
-----------------------------------------------------------------------------
Rainer Orth, Center for Biotechnology, Bielefeld University
