This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: patch: disable string insns on ppc e500
- From: Aldy Hernandez <aldy at quesejoda dot com>
- To: Geoff Keating <geoffk at geoffk dot org>
- Cc: gcc-patches at gcc dot gnu dot org, dje at watson dot ibm dot com
- Date: Tue, 11 Mar 2003 13:48:12 -0800
- Subject: Re: patch: disable string insns on ppc e500
On Tuesday, March 11, 2003, at 01:32 PM, Geoff Keating wrote:
Date: Tue, 11 Mar 2003 13:23:04 -0800
Cc: Geoff Keating <geoffk at geoffk dot org>, David Edelsohn
<dje at watson dot ibm dot com>
From: Aldy Hernandez <aldy at quesejoda dot com>
X-OriginalArrivalTime: 11 Mar 2003 21:23:09.0890 (UTC)
FILETIME=[6A38EA20:01C2E814]
As discussed. This fixes a gazillion -Os regressions on the e500.
Applied to mainline.
No, no, no. TARGET_SPE != CPU_E500. Please use the right test.
Actually it does, but that's purely coincidental:
eabi.h:#define TARGET_SPE (rs6000_cpu == PROCESSOR_PPC8540)
Now that you mention it, sometime I should fix 90% of TARGET_SPE macros
to TARGET_E500 or something like it.
Do you approve of this version?
2003-03-11 Aldy Hernandez <aldyh at redhat dot com>
* config/rs6000/rs6000.c (rs6000_override_options): Use cpu type
instead of TARGET_SPE.
Index: config/rs6000/rs6000.c
===================================================================
RCS file: /cvs/gcc/gcc/gcc/config/rs6000/rs6000.c,v
retrieving revision 1.437
diff -c -p -r1.437 rs6000.c
*** config/rs6000/rs6000.c 11 Mar 2003 21:26:12 -0000 1.437
--- config/rs6000/rs6000.c 11 Mar 2003 21:46:16 -0000
*************** rs6000_override_options (default_cpu)
*** 700,706 ****
/* The e500 does not have string instructions, and we set
MASK_STRING above when optimizing for size. */
! if (TARGET_SPE && (target_flags & MASK_STRING) != 0)
target_flags = target_flags & ~MASK_STRING;
/* Handle -m(no-)longcall option. This is a bit of a cheap hack,
--- 700,706 ----
/* The e500 does not have string instructions, and we set
MASK_STRING above when optimizing for size. */
! if (rs6000_cpu == PROCESSOR_PPC8540 && (target_flags & MASK_STRING)
!= 0)
target_flags = target_flags & ~MASK_STRING;
/* Handle -m(no-)longcall option. This is a bit of a cheap hack,