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: [RFC] use symbol_ref flags for rs6000

From: Dale Johannesen <dalej at apple dot com>
To: David Edelsohn <dje at watson dot ibm dot com>
Cc: Dale Johannesen <dalej at apple dot com>, Andrew Pinski <pinskia at physics dot uc dot edu>, Jakub Jelinek <jakub at redhat dot com>, Richard Henderson <rth at redhat dot com>, Janis Johnson <janis187 at us dot ibm dot com>, gcc-patches at gcc dot gnu dot org
Date: Thu, 1 May 2003 15:41:49 -0700
Subject: Re: [RFC] use symbol_ref flags for rs6000

On Thursday, May 1, 2003, at 03:30 PM, David Edelsohn wrote:

Andrew Pinski writes:
Andrew> The only problem with this on darwin, it is okay to call the static int Andrew> so that will become a regression.
	Regession in what sense?  I have heard that calling function
pointer on Darwin actually is faster.

Depends. If you look at the assembly output, a "direct" call actually jumps to a stub of 8 insns and executes those before getting to the real target. However, the linker eliminates stubs in all cases where that's possible, so if the caller and callee are both present at static link time, the direct call winds up really being direct, therefore faster than going through a pointer. Calls to shared libraries do execute the stub code, and in that case going through a pointer is faster.

References:
- Re: [RFC] use symbol_ref flags for rs6000
  - From: David Edelsohn

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