This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [tuples][patch] Disable generation of SFTs
- From: "Richard Guenther" <richard dot guenther at gmail dot com>
- To: "Rafael Espindola" <espindola at google dot com>
- Cc: GCC <gcc-patches at gcc dot gnu dot org>, "Diego Novillo" <dnovillo at google dot com>
- Date: Mon, 21 Apr 2008 11:00:23 +0200
- Subject: Re: [tuples][patch] Disable generation of SFTs
- References: <38a0d8450804210038k241e5af4ie2519351d27c6d77@mail.gmail.com>
On Mon, Apr 21, 2008 at 9:38 AM, Rafael Espindola <espindola@google.com> wrote:
> This patch disables SFTs on the tuples branch. It is disabled on trunk
> (http://gcc.gnu.org/ml/gcc-patches/2008-04/msg01213.html).
>
> The patch fixes 8083 fortran tests. It also causes some missed
interesting... ;)
> optimization regressions and one missing warning (see below). OK for
> tuples if it bootstraps C?
>
> Tests that now fail, but worked before:
>
> gcc.dg/tree-ssa/alias-15.c scan-tree-dump salias "SFT.5 created for
> var m offset 128"
Are you sure you also merged the testsuite changes? The above test
doesn't any longer scan for SFT.5 ...
> gcc.dg/tree-ssa/alias-15.c scan-tree-dump-times salias "VUSE <SFT.5_"
> 2
> gcc.dg/tree-ssa/salias-1.c scan-tree-dump-times salias "structure
> field tag SFT" 2
> gcc.dg/tree-ssa/ssa-lim-3.c scan-tree-dump lim "Executing store motion
> of global.y"
> gcc.dg/tree-ssa/ssa-lim-3.c scan-tree-dump lim "Moving
> statement.*global.x.*out of loop 1"
> gcc.dg/tree-ssa/structopt-2.c scan-tree-dump-times optimized "a.e" 0
> gcc.dg/tree-ssa/structopt-2.c scan-tree-dump-times optimized "a.f" 0
> gcc.dg/tree-ssa/structopt-2.c scan-tree-dump-times optimized "a.g" 0
> gcc.dg/tree-ssa/structopt-3.c scan-tree-dump-times optimized "return
> 11" 1
Some of them are likely due to missing FRE/PRE on tuples.
>
> Tests that now fail, but worked before:
>
> g++.dg/torture/pr34850.C -O0 (test for warnings, line 14)
> g++.dg/torture/pr34850.C -O1 (test for warnings, line 14)
> g++.dg/torture/pr34850.C -O2 (test for warnings, line 14)
> g++.dg/torture/pr34850.C -O3 -fomit-frame-pointer (test for
> warnings, line 14)
> g++.dg/torture/pr34850.C -O3 -g (test for warnings, line 14)
> g++.dg/torture/pr34850.C -Os (test for warnings, line 14)
I guess this is another missing FRE fallout and we don't recognize
__len as constant.
If you make sure the testsuite parts are properly merged this is ok.
Thanks,
Richard.