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][PING]: Fix PR target/29682 - [4.2/4.3 Regression] ICE on ia64 in scheduler

From: Maxim Kuvyrkov <mkuvyrkov at ispras dot ru>
To: Jim Wilson <wilson at specifix dot com>
Cc: Vladimir Makarov <vmakarov at redhat dot com>, Martin Michlmayr <tbm at cyrius dot com>, gcc-patches at gcc dot gnu dot org
Date: Tue, 30 Jan 2007 10:09:15 +0300
Subject: Re: [PATCH][PING]: Fix PR target/29682 - [4.2/4.3 Regression] ICE on ia64 in scheduler
References: <455C4123.6090602@ispras.ru> <4573CD6D.7070808@ispras.ru> <1170109001.2534.31.camel@localhost.localdomain>

Jim Wilson wrote:

On Mon, 2006-12-04 at 10:25 +0300, Maxim Kuvyrkov wrote:
This patch fixes PR target/29682 on ia64. The bug is that we generate a speculative load into ar.lc register and fail when try to recognize it. The patch adds necessary check to ia64.c: ia64_speculate_insn () function so that only loads to general or fp registers will be considered for speculation.
OK.
While looking at this patch, I noticed that the existing code isn't
handling subregs.  There are two places where it checks REG_P but
doesn't handle or check for subregs.  Presumably subregs can occur here,
which means we could get better optimization if we allowed them.

Thanks. I'll commit the patch with the flags added to the testcase and will post one for handling subregs in about a week.

Thanks,

Maxim

References:
- Re: [PATCH][PING]: Fix PR target/29682 - [4.2/4.3 Regression] ICE on ia64 in scheduler
  - From: Jim Wilson

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