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 ICE in gen_rtx_SUBREG, gen_lowpart_if_possible trying to create an invalid subreg

From: Paolo Bonzini <bonzini at gnu dot org>
To: Andrew_Pinski at PlayStation dot Sony dot Com
Cc: gcc-patches at gcc dot gnu dot org
Date: Tue, 29 May 2007 10:23:25 +0200
Subject: Re: [PATCH] Fix ICE in gen_rtx_SUBREG, gen_lowpart_if_possible trying to create an invalid subreg
References: <OFE5E1DB15.E62B5E2A-ON882572E9.006D7BC7-882572E9.006E7C36@playstation.sony.com>

Andrew_Pinski@PlayStation.Sony.Com wrote:

Hi, This patch fixes the ICE in gen_rtx_SUBREG when CSE calls gen_lowpart_if_possible but the lowpart would cause an invalid subreg (on spu-elf where a lot of subregs are invalid because of the way the register is laid out). gen_lowpart_if_possible does not check if gen_lowpart_SUBREG would try to create an invalid subreg which causes a crash in gen_rtx_SUBREG. This patch fixes the problem by checking about the invalid subreg in gen_lowpart_if_possible before calling gen_lowpart_SUBREG.

I would use an "&& validate_subreg (...)" (this would make a difference if another "else" was added after the one you are modifying).

Paolo

References:
- [PATCH] Fix ICE in gen_rtx_SUBREG, gen_lowpart_if_possible trying to create an invalid subreg
  - From: Andrew_Pinski

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