This is the mail archive of the newlib@sourceware.org mailing list for the newlib 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: Optimized arm string routines

From: Jeff Johnston <jjohnstn at redhat dot com>
To: Richard Earnshaw <rearnsha at arm dot com>
Cc: "Schwarz, Konrad" <konrad dot schwarz at siemens dot com>, newlib at sourceware dot org
Date: Thu, 22 Jan 2009 13:05:01 -0500
Subject: Re: Optimized arm string routines
References: <1232446912.1485.12.camel@pc960.cambridge.arm.com> <4977B998.4080805@redhat.com> <A0725A2C3F6E9A4699A9A416DE9265EF02EADCDA@MCHP7RCA.ww002.siemens.net> <1232617584.6156.0.camel@pc960.cambridge.arm.com>

Richard Earnshaw wrote:

On Thu, 2009-01-22 at 08:59 +0100, Schwarz, Konrad wrote:
Subject: Re: Optimized arm string routines
How do these fit in with the optimized strcmp() provided by Eric Blake?

I've not seen that. Do you have a link?

R.

I believe Konrad is referring to the work Eric did for handling unaligned data to avoid performance penalties. A similar optimization might make sense in the ASM version.

Perhaps if you were to publish some performance numbers (aligned, unaliged data) of the new code vs the generic, that would probably help. Do you happen to have any of these numbers handy?

-- Jeff J.

Follow-Ups:
- Re: Optimized arm string routines
  - From: Richard Earnshaw

References:
- Optimized arm string routines
  - From: Richard Earnshaw
- Re: Optimized arm string routines
  - From: Jeff Johnston
- RE: Optimized arm string routines
  - From: Schwarz, Konrad
- RE: Optimized arm string routines
  - From: Richard Earnshaw

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