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: RFC - COST of const_double for x86 prevents constant copy propagation in cse

From: Andrew Pinski <pinskia at physics dot uc dot edu>
To: Fariborz Jahanian <fjahanian at apple dot com>
Cc: Ian Lance Taylor <ian at airs dot com>, gcc mailing list <gcc at gcc dot gnu dot org>
Date: Thu, 25 Aug 2005 14:15:57 -0400
Subject: Re: RFC - COST of const_double for x86 prevents constant copy propagation in cse
References: <CCC52908-3E57-4F28-ABC9-93BF45D82DC3@apple.com>

On Aug 25, 2005, at 2:09 PM, Fariborz Jahanian wrote:

Compiled with -O1 -mdynamic-no-pic -march=pentium4 produces:
        pxor    %xmm0, %xmm0
        movsd   %xmm0, 16(%eax)
        movsd   %xmm0, 8(%eax)
        movsd   %xmm0, (%eax)
But following code results in 7% performance gain in eon as reported by one of Apple's performance people:
        movl    $0, 16(%eax)
        movl    $0, 20(%eax)
        movl    $0, 8(%eax)
        movl    $0, 12(%eax)
        movl    $0, (%eax)
        movl    $0, 4(%eax)

Actually that does not make sense for a real processor but then again this is x86 we are talking about. Because we have less complex instructions and less instructions at that matter.

-- Pinski

References:
- RFC - COST of const_double for x86 prevents constant copy propagation in cse
  - From: Fariborz Jahanian

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