This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] Perform frame_pointer_needed hfp -> cfa replacements even if fp setter is not in the first bb (PR debug/54402)
- From: Jakub Jelinek <jakub at redhat dot com>
- To: Steven Bosscher <stevenb dot gcc at gmail dot com>
- Cc: gcc-patches at gcc dot gnu dot org, Alexandre Oliva <aoliva at redhat dot com>
- Date: Mon, 29 Oct 2012 21:52:12 +0100
- Subject: Re: [PATCH] Perform frame_pointer_needed hfp -> cfa replacements even if fp setter is not in the first bb (PR debug/54402)
- References: <20121029204328.GM1752@tucnak.redhat.com> <CABu31nPyRVgOke=LGXYgi=qndOwkJpPmEXxFsv1X=hJ5BBRhSA@mail.gmail.com>
- Reply-to: Jakub Jelinek <jakub at redhat dot com>
On Mon, Oct 29, 2012 at 09:46:24PM +0100, Steven Bosscher wrote:
> Will you also add the test case?
There already is, libgo make reflect/check. I don't see what other kind of
testcase should I provide for that, unless you mean scanning -dA assembly
for DW_OP_fbreg (which would be inherently target specific).
BTW, only the go testcase has been worked on, I haven't been able to
reproduce any significant slowness on the #c0 testcase (perhaps I've been
trying wrong target or options), and the clobber_overlapping_mems
routine, while certainly desirable for correct debug info (without it
debug info might say something is available in a mem where it isn't
available), is still quite expensive and it would be nice to find some
ways to speed it up.
Jakub