[perf-discuss] Problem Recording external stats

2009-04-24 Thread Ted Battershill
I wanted to make use of the external stats collection described in the assorted .func files. All the .prof files refer to function=generic, which doesn't collect any external stats. I tried to keep it simple and just collecting iostat and vmstat with the randomread2k. Has anyone seen or solved

Re: [perf-discuss] ZFS performance issue - READ is slow as hell...

2009-04-24 Thread roland
i have created bugticket nr 6833814 (http://bugs.opensolaris.org/view_bug.do?bug_id=6833814 ) for this issue -- This message posted from opensolaris.org ___ perf-discuss mailing list perf-discuss@opensolaris.org

Re: [perf-discuss] Seeking feedback on LatencyTOP

2009-04-24 Thread Krishnendu Sadhukhan
Hi Lejun, Yes, I think we can display the data from the "adaptive-block" probe as a separate count so that users will know if a thread was blocked on an adaptive lock. I will open CRs for this and other enhancements that I mentioned in the mail. Thanks, Krish -- This message posted from ope

Re: [perf-discuss] New release of FIlebench on SourceForge

2009-04-24 Thread Andrew Wilson
On 04/24/09 05:52 AM, Dearl Neal wrote: Hey Drew.. Thanks for the update on filebench. I have dowloaded and compiled it for use on a sun sparc system. One thing I have noticed about using this tool is that it does not clean up the ipc shared memory segements, active semaphores, or active message

Re: [perf-discuss] New release of FIlebench on SourceForge

2009-04-24 Thread Andrew Wilson
There is code to clean those up, though it doesn't seem to always get run, especially if you quit early. But normally FB does clean all those things up. I suppose it is possible that my new shutdown code broke the cleanup. I'll investigate to see if I have the same cleanup failures you are

Re: [perf-discuss] New release of FIlebench on SourceForge

2009-04-24 Thread Dearl Neal
Hey Drew.. Thanks for the update on filebench. I have dowloaded and compiled it for use on a sun sparc system. One thing I have noticed about using this tool is that it does not clean up the ipc shared memory segements, active semaphores, or active message queues. When I run the db workloads,