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] Prevent noisey warnings on arm-elf

From: Mark Mitchell <mark at codesourcery dot com>
To: micros at valvers dot com, Julian Brown <julian at codesourcery dot com>
Cc: Andrew Pinski <pinskia at gmail dot com>, GCC Patches <gcc-patches at gcc dot gnu dot org>
Date: Thu, 30 Aug 2007 20:04:36 -0700
Subject: Re: [PATCH] Prevent noisey warnings on arm-elf
References: <46D567DA.4040500@valvers.com> <de8d50360708290926k2293fc47l15a5cb5f48dea6ec@mail.gmail.com> <46D6C127.80206@valvers.com>

Brian Sidebotham wrote:
> Andrew Pinski wrote:
> 
>>> -  HOST_WIDE_INT elem = ffs (operands[2]) - 1;
>>> +  HOST_WIDE_INT elem = ffs ((int) operands[2]) - 1;
>>
>> I think this should be:
>> HOST_WIDE_INT elem = ffs (INTVAL (operands[2])) - 1;
> 
> Hi Andrew,
> 
> I have changed this as requested, though I can't find the definition of
> INTVAL, or what difference there is from a plain (int) cast. Attached is
> an updated patch, regression tested, with no new regressions on
> arm-none-elf.

The revised form is correct.  The original code looks like it's just
buggy; the elements of operands[] are not integers but RTX nodes.
INTVAL is the way you get the "int" value of an RTX for a "CONST_INT";
see gcc/rtl.h.

Julian, would you please review and check in Brian's patch, unless you
disagree?

Thanks,

-- 
Mark Mitchell
CodeSourcery
mark@codesourcery.com
(650) 331-3385 x713

References:
- [PATCH] Prevent noisey warnings on arm-elf
  - From: Brian Sidebotham
- Re: [PATCH] Prevent noisey warnings on arm-elf
  - From: Andrew Pinski
- Re: [PATCH] Prevent noisey warnings on arm-elf
  - From: Brian Sidebotham

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