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: recent troubles with float vectors & bitwise ops

From: Tim Prince <tprince at computer dot org>
To: bonzini at gnu dot org
Cc: Mark Mitchell <mark at codesourcery dot com>, Ross Ridge <rridge at csclub dot uwaterloo dot ca>, gcc at gcc dot gnu dot org
Date: Fri, 24 Aug 2007 05:14:59 -0700
Subject: Re: recent troubles with float vectors & bitwise ops
References: <20070823013503.426FE74B95@caffeine.csclub.uwaterloo.ca> <46CD273E.6060009@gnu.org> <46CDF9E2.1050505@codesourcery.com> <46CE7A09.3010702@lu.unisi.ch>

Paolo Bonzini wrote:


2) selection operations on vectors, kind of (v1 <= v2 ? v3 : v4).  These
can be written for example like this:

     cmpleps xmm1, xmm2   ; xmm1 = xmm1 <= xmm2 ? all-ones : 0
     andnps  xmm4, xmm1   ; xmm4 = xmm1 <= xmm2 ? 0 : xmm4
     andps   xmm1, xmm3   ; xmm1 = xmm1 <= xmm2 ? xmm3 : 0
     orps    xmm1, xmm4   ; xmm1 = xmm1 <= xmm2 ? xmm3 : xmm4

SSE4 introduces specific instruction support, with a shorter sequence for this purpose. It seems to be quite difficult to persuade gcc to use it.

References:
- Re: recent troubles with float vectors & bitwise ops
  - From: Ross Ridge
- Re: recent troubles with float vectors & bitwise ops
  - From: Paolo Bonzini
- Re: recent troubles with float vectors & bitwise ops
  - From: Mark Mitchell
- Re: recent troubles with float vectors & bitwise ops
  - From: Paolo Bonzini

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