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: Patch (rs6000): Prefer CTR to LR in indirect calls

From: Dale Johannesen <dalej at apple dot com>
To: David Edelsohn <dje at watson dot ibm dot com>
Cc: gcc-patches at gcc dot gnu dot org, Dale Johannesen <dalej at apple dot com>
Date: Sun, 5 Jun 2005 12:21:19 -0700
Subject: Re: Patch (rs6000): Prefer CTR to LR in indirect calls
References: <7a08fc025dd1283cc143c59102596b14@apple.com> <200506050405.j5545O031274@makai.watson.ibm.com>

On Jun 4, 2005, at 9:05 PM, David Edelsohn wrote:

I just realized that the patch is more intrusive than it needed to be. The `*' modifier modifies the next character and does not need to be a separate alternative. In other words, one should be able to achieve the same effect with

[(call (mem:SI (match_operand:SI 0 "register_operand" "c*l,c*l"))

Would you please revert the patterns back to fewer alternatives using the solution above, after bootstrapping and testing?

I guess. Are you sure this works? I was following rth's suggestion here: http://gcc.gnu.org/ml/gcc-patches/2001-12/msg00636.html

References:
- Patch (rs6000): Prefer CTR to LR in indirect calls
  - From: Dale Johannesen
- Re: Patch (rs6000): Prefer CTR to LR in indirect calls
  - From: David Edelsohn

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