This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [Patch MIPS/libgcc] fix crts.
- From: Richard Sandiford <rdsandiford at googlemail dot com>
- To: Iain Sandoe <Iain_Sandoe at mentor dot com>
- Cc: "\<gcc-patches\ at gcc dot gnu dot org\>" <gcc-patches at gcc dot gnu dot org>
- Date: Fri, 22 Jun 2012 22:06:13 +0100
- Subject: Re: [Patch MIPS/libgcc] fix crts.
- References: <BC889A6F-A26F-4568-8B0F-55209B436482@mentor.com>
Iain Sandoe <Iain_Sandoe@mentor.com> writes:
> Hello,
>
> While working on mips recently, I noticed that all the execute tests
> fail for simulator.
>
> This appears to be caused by an oversight in the move from gcc/config
> => libgcc, where t-elf defined extra parts including crt{begin,end}.o
> but these have been omitted from the re-built libgcc/config.host.
>
> The proposed patch adds the two crts to mips targets in
> libgcc/config.host (where they include t-elf) and coalesces those that
> have the same config.
Which target were you using? Was it something like mipsisa64-elfoabi?
crtbegin.o and crtend.o are supposed to be added via:
*-*-elf)
extra_parts="crtbegin.o crtend.o"
;;
but I suspect that that "elf)" ought to be "elf*)". Would that fix
your case? If so, it sounds like the right fix, although I can't
approve it. It probably ought to go on 4.7 too.
I still like the idea of combining the stanzas though. That part
of the patch -- i.e. without the changes to extra_parts -- is OK.
I think the tx39 differences are purely accidental or historical.
Your new generic mips*-*-elf*) stanza ought to be OK for that too,
so would you mind removing the tx39 case altogether while you're there?
No need to test mipstx39-elf explicitly; it's very much a legacy target.
Thanks,
Richard