This is the mail archive of the gcc@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: reload question

From: Paolo Bonzini <bonzini at gnu dot org>
To: Pat Haugen <pthaugen at us dot ibm dot com>
Cc: gcc at gcc dot gnu dot org, Ulrich Weigand <Ulrich dot Weigand at de dot ibm dot com>
Date: Sat, 11 Aug 2007 13:45:25 +0200
Subject: Re: reload question
References: <OF4CB9C8F6.34DA5C1C-ON86257333.0067A2B6-86257333.006E343A@us.ibm.com>

Pat Haugen wrote:

Is this being done on purpose (going on assumption that operands are
commutative), such as to allow more opportunities for a successful
allocation with reduced spill?

I think so, but reordering the "else if"s should improve the results.

I've also seen the same situation come up during register renaming
(regrename.c), but not too surprising since the code there says it's based
off find_reloads_address_1() and is coded similarly.

... and likewise here.

Paolo

References:
- reload question
  - From: Pat Haugen

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