This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: [PATCH] Fix -fsched-stalled-insns on ia64 (PR middle-end/33670)

From: Vladimir Makarov <vmakarov at redhat dot com>
To: Jakub Jelinek <jakub at redhat dot com>
Cc: Maxim Kuvyrkov <maxim at codesourcery dot com>, gcc-patches at gcc dot gnu dot org
Date: Tue, 06 Nov 2007 18:06:16 -0500
Subject: Re: [PATCH] Fix -fsched-stalled-insns on ia64 (PR middle-end/33670)
References: <20071105190632.GU5451@devserv.devel.redhat.com>

Jakub Jelinek wrote:

Hi!

The newly added pr33670.c which the last haifa-sched.c patch
fixed on ppc64 still ICEs on ia64.  The problem is that check_sched_flags
asserts that -fsched-stalled-insns{,=N} isn't used on ia64 unless
-mno-sched-ar-data-spec is also used (which turns off DO_SPECULATION).
Maxim said in bugzilla he doesn't remember why he enforced it and that he
doesn't think it is needed, so the following patch just removes those
assertions.

As I remember, -fsched-stalled-insns was added for a power processor (probably for power4). It was not supposed to be used for Itanium. Probably therefore Maxim added the check.

Bootstrapped/regtested on ia64-linux, additionally I have also run
make check with RUNTESTFLAGS=--target_board=unix/-fsched-stalled-insns=0
and there were no regressions against normal make check.

Ok for trunk?

Yes, the patch is ok of the trunk. Thanks for the patch, Jakub.

References:
- [PATCH] Fix -fsched-stalled-insns on ia64 (PR middle-end/33670)
  - From: Jakub Jelinek

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]