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: PR target/17101: question about powerpc s<cond> expanders

From: David Edelsohn <dje at watson dot ibm dot com>
To: Nathan Sidwell <nathan at codesourcery dot com>
Cc: gcc <gcc at gcc dot gnu dot org>, Geoffrey Keating <geoffk at apple dot com>
Date: Fri, 12 Nov 2004 14:45:40 -0500
Subject: Re: PR target/17101: question about powerpc s<cond> expanders
References: <4194EC32.9080308@codesourcery.com> <200411121824.iACIOrD26520@makai.watson.ibm.com> <419502F3.90700@codesourcery.com>

>>>>> Nathan Sidwell writes:

Nathan> that is not what I observe, when I remove the TARGET_POWER check.
Nathan> For instance, V > 55

.sgt:
         cmpwi 7,3,55
         mfcr 3
         rlwinm 3,3,30,1
is that worse than
	li  9,0
	cmpwe 7,3,55
	ble- 7,.skip
	li  9,1
.skip:

	The branch sequence may be better, depending on how well predicted
it is. 

David

References:
- PR target/17101: question about powerpc s<cond> expanders
  - From: Nathan Sidwell
- Re: PR target/17101: question about powerpc s<cond> expanders
  - From: David Edelsohn
- Re: PR target/17101: question about powerpc s<cond> expanders
  - From: Nathan Sidwell

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