This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PARCH 2/2, x86, PR63534] Fix darwin bootstrap
- From: Jakub Jelinek <jakub at redhat dot com>
- To: Evgeny Stupachenko <evstupac at gmail dot com>
- Cc: GCC Patches <gcc-patches at gcc dot gnu dot org>, Jeff Law <law at redhat dot com>, Uros Bizjak <ubizjak at gmail dot com>, iains at gcc dot gnu dot org
- Date: Fri, 17 Oct 2014 16:20:54 +0200
- Subject: Re: [PARCH 2/2, x86, PR63534] Fix darwin bootstrap
- Authentication-results: sourceware.org; auth=none
- References: <CAOvf_xxtWrrAm_z3s3-xGMqkPPA7SppL-woFG-4docMrmZ1xGw at mail dot gmail dot com>
- Reply-to: Jakub Jelinek <jakub at redhat dot com>
On Fri, Oct 17, 2014 at 06:16:41PM +0400, Evgeny Stupachenko wrote:
> Hi,
>
> Some instructions (like one in PR63534) could have hidden use of PIC register.
> Therefore we need to leave SET_GOT not deleted till reload completed.
> The patch prevents SET_GOT from deleting while PIC register is pseudo.
Just curious, do you emit the init_pic_reg unconditionally at the start of
the function in -fpic mode? What does IRA do in that case, if it sees
a dead setter of something that doesn't seem to be used at that point?
Doesn't it penalize generated code, even if we don't end up with any PIC
references during/after reload?
Jakub