Re: [perf-discuss] [osol-code] extremly bad performance of Solaris regex

2007-10-31 Thread Fintan Ryan
Hi, >>> Linux i6860.704u 0.004s 0:00.70 100.0%5587 ms 2079.593 >> 1. Which locale do you use ? > > LC_CTYPE is set to en_US and LC_LANG/LC_ALL are unset. So I assume > 'en_US'. > Can you try using the C locale, en_US is now a utf8 locale afaik, the I18N folks would be abl

Re: [perf-discuss] RAM throughput - how can it be measured ?

2007-08-03 Thread Fintan Ryan
Hi, >> What size are you setting N in the streams code to? We pull this out of >> smbios info where its available. >> > > Yes I had the Env-Var set. > > It was the default N= 2 setting. > ah, what was the L2 cache size? See the "Adjust the Problem Size" section of the stream refer

Re: [perf-discuss] RAM throughput - how can it be measured ?

2007-08-02 Thread Fintan Ryan
Hi, >compling with -m64 -xopenmp gives one or two more hundred MB/s > but the variation of the values between different runs can also vary almost > the same amount. So one probably have to run a series of 10 tests or more > and compute > the mean value to see a consistent diff

Re: [perf-discuss] RAM throughput - how can it be measured ?

2007-07-31 Thread Fintan Ryan
Hi, > On 7/31/07, Lars Tunkrans <[EMAIL PROTECTED]> wrote: > >> So yes , Thanks to the Stream Benchmark I have now sort of proved >> that the new >> DDR2 memory controller on the AM2 socket Athlon's very nearly >> doubles memory throughput. >> > > BTW, if you are really into getting the

Re: [perf-discuss] Tools used to test performance between builds.

2007-06-27 Thread Fintan Ryan
Sean McGrath - Sun Microsystems Ireland wrote: Brian Gupta stated:

Re: [perf-discuss] PDU error in solaris

Hi, I am getting the following error in my messages file. Could you please provide any workaround for the same. Apr 12 00:46:33 sr1-ebiz1 /usr/lib/snmp/snmpdx: [ID 702911 daemon.error] error while receiving a pdu from localhost.32809: Decode the header of the variables failed: asn length to

[perf-discuss] Performance Benchmark List

Hi, We just updated the list of performance benchmarks for possible use with OpenSolaris. These benchmarks will also be available for the community to use in the upcoming Self Test facility. Futher details are available at http://www.opensolaris.org/os/community/testing/perfbench/ Feedback appr

Re: [perf-discuss] kcopy() implementation

Hi, This is being tracked in bugid 6291299 [1]. - Fintan On Mon, Sep 12, 2005 at 08:30:42PM +0400, [EMAIL PROTECTED] wrote: < < < Hi, < < I have noticed that the amd64 version of kcopy() uses the same "rep smov" < the i386 version use. < < The AMD's "Opteron Optimization Guide" states that