This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [rs6000 0/3] POWER removal
- From: Michael Meissner <meissner at linux dot vnet dot ibm dot com>
- To: Segher Boessenkool <segher at kernel dot crashing dot org>
- Cc: gcc-patches at gcc dot gnu dot org, dje dot gcc at gmail dot com, meissner at gcc dot gnu dot org
- Date: Thu, 26 Jul 2012 15:01:47 -0400
- Subject: Re: [rs6000 0/3] POWER removal
- References: <cover.1343287109.git.segher@kernel.crashing.org>
On Thu, Jul 26, 2012 at 12:38:47AM -0700, Segher Boessenkool wrote:
> This patch series removes support for the old POWER CPUs (the RIOS,
> RSC, and RIOS2).
>
> It does not do any more than that: it leaves common mode alone,
> it does not remove the old assembler mnemonics and the {xx|yy}
> insn template syntax, etc.
>
> Bootstrapped and regtested on powerpc64-linux (c,c++,fortran);
> no new failures.
>
> Comments?
>
> Segher
These look good. It will be good to get rid of them, even with the pain it
causes to my condition code branch (since the power stuff touchs things like
abs, min, max, and some tests).
Note, you will need to update the copyright to include 2012 for the files that
have not been modified in this year. Using grep, etc. the files in question
are:
gcc/config/rs6000/aix53.h
gcc/config/rs6000/rs6000-tables.opt
gcc/config/rs6000/constraints.md
gcc/config/rs6000/predicates.md
gcc/config/rs6000/rs6000-cpus.def
gcc/config/rs6000/aix51.h
gcc/config/rs6000/t-rs6000
gcc/config/rs6000/aix43.h
gcc/config/rs6000/rs6000-c.c
gcc/config/rs6000/rs6000.opt
gcc/config/rs6000/darwin.h
gcc/config/rs6000/rs6000.c
gcc/config/rs6000/aix52.h
gcc/config/rs6000/rs6000.h
gcc/config/rs6000/t-aix43
gcc/config/rs6000/601.md
gcc/config/rs6000/dfp.md
gcc/config/rs6000/aix61.h
gcc/config/rs6000/rs6000.md
gcc/config/rs6000/rs6000-opts.h
gcc/config/rs6000/driver-rs6000.c
gcc/config/rs6000/sysv4.h
gcc/common/config/rs6000/rs6000-common.c
In running parallel tests for both 64-bit and 32-bit, I see 3 differences:
gcc.dg/pr42389.c Trunk fails, Segher passes for 64-bit
gcc.dg/pr46521.c Trunk fails, Segher passes for 64-bit
gcc.dg/tree-prof/pr45354.c Trunk passes, Segher fails for 64-bit
gcc.dg/pr45352-1.c Trunk passes, Segher fails for 32-bit
gcc.dg/pr46522.c Trunk passes, Segher fails for 32-bit
gcc.dg/pr48144.c Trunk fails, Segher passes for 32-bit
gcc.dg/pr50205.c Trunk passes, Segher fails for 32-bit
gcc.dg/tree-prof/pr45354.c Trunk fails, Segher passes for 32-bit
gfortran.dg/pr44691.f Trunk fails, Segher passes for 64-bit
Now, all of these differences are in tests that do selective scheduling, so I'm
not sure they are blockers.
--
Michael Meissner, IBM
5 Technology Place Drive, M/S 2757, Westford, MA 01886-3141, USA
meissner@linux.vnet.ibm.com fax +1 (978) 399-6899