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: [Committed] Fix 54524: Wrong code with some 64bit addition with registers as 32bits

From: Eric Botcazou <ebotcazou at adacore dot com>
To: Richard Sandiford <rdsandiford at googlemail dot com>
Cc: gcc-patches at gcc dot gnu dot org, Andrew Pinski <andrew dot pinski at caviumnetworks dot com>
Date: Mon, 05 Nov 2012 15:13:42 +0100
Subject: Re: [Committed] Fix 54524: Wrong code with some 64bit addition with registers as 32bits
References: <CA+=Sn1k0kJaSERkK8LZF7=Nt-avZ7fwRydKod4t8uYpPmC06HA@mail.gmail.com> <87sj8rf4c6.fsf@talisman.home>

> Hmm, like you said originally in the PR, I think generating the invalid
> PLUS is the bug here.  I agree (as Richard B said) that an invalid PLUS
> shouldn't cause us to generate wrong code, but an assert seems better
> than a check.  Eric, what do you think?

I think that this boils down to deciding whether (PLUS X 0) is invalid or 
merely non-canonical.  If the former, then I agree that an assertion is 
appropriate; if the latter, then I don't think that we should stop the 
compiler because of it.

No strong opinion, although I'd lean towards the latter.  In any case, the 
piece of code which generates this PLUS needs also to be fixed (at least on 
the mainline).

-- 
Eric Botcazou

References:
- [Committed] Fix 54524: Wrong code with some 64bit addition with registers as 32bits
  - From: Andrew Pinski
- Re: [Committed] Fix 54524: Wrong code with some 64bit addition with registers as 32bits
  - From: Richard Sandiford

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