Re: [perf-discuss] Kernel Asynchronous I/O

Hi Matty, Kernel async I/O just works on RAW devices at the moment. Kernel async I/O give a small incremental improvement, but only in the order of a few percent in optimal cases. The user-land thread emulation works quite well when direct I/O is enabled (without it, the single writer lock will

Re: [perf-discuss] LibMicro: Portable Microbenchmarks won't compile on FreeBSD

Hi Petri, libMicro only supports Solaris, Linux and Aix out of the box. The Makefile checks for a OS specific makefile named Makefile., so you'll need to create that and modify it as needed to run on BSD. Check out my diffs for Darwin at this thread: http:// www.opensolaris.org/jive/thread

Re: [perf-discuss] Re: FileBench Discuss

Brendan Gregg wrote: G'Day Folks, FileBench looks great. I've been wanting to write such a tool for a while, but this is far more comprehensive than I was planning. :) Yes, it is definetly cool. Imagine using this for a practical Solaris interview. Create a well defined testload with fi

Re: [perf-discuss] Re: Re: libMicro page added

On Thu, Aug 11, 2005 at 01:55:40PM -0700, Dan Price wrote: > On Thu 11 Aug 2005 at 01:52PM, Devon H. O'Dell wrote: > > b) There's actually currently no limit on the amount of byte-range > > locks you can hold. > > (Except that the problem we're discussing here is libMicro running on > Darwin). >

Re: [perf-discuss] Re: FileBench Discuss

Sorry mate :^) A world-class chef would never serve the souffle' until it's perfect, would they? ;^) /jim James Dickens wrote: On 8/11/05, *James Mauro* <[EMAIL PROTECTED] > wrote: Hey Kevin - In addition to all the good stuff Richard provided, be aw

Re: [perf-discuss] Re: Re: libMicro page added

> This might sound dumb, but are we sure that processes are the resource > which is temporarily unavailable? I think what you are saying, Dan, is that "this might be so obvious that it has escaped you", and you very well might be right. :) -Eric Eric Saxe Solaris Kernel Development 510.550.4866

Re: [perf-discuss] Re: Re: libMicro page added

Phil Harman wrote: Bob Palowoda wrote: There is probably nothing wrong with connection it just toke 123sec on a 3.2ghz machine. I just didn't wait long enough. Phew! 123sec is a long time. Maybe your machine was busy doing other stuff in the background? Before 0.3.0 we used to run eac

[perf-discuss] LibMicro: Portable Microbenchmarks won't compile on FreeBSD

Hi! [EMAIL PROTECTED]:~/libMicro-0.3.0$ make make: cannot open ../Makefile.FreeBSD. *** Error code 2 Stop in /home/users/t/th/thoron/libMicro-0.3.0. [EMAIL PROTECTED]:~/libMicro-0.3.0$ gmake gmake[1]: Entering directory `/home/users/t/th/thoron/libMicro-0.3.0/bin-i386' gmake[1]: ../Makefile.FreeB

[perf-discuss] Kernel Asynchronous I/O

Howdy, Does anyone know why kernel asynchronous I/O is limited to raw devices? Is it possible to extend kaio to the file system layer? Would there be any benefit? Is there a noticeable performance difference between the userland and kernel asynchronous I/O implemenation when UFS is used with l

Re: [perf-discuss] Re: Re: libMicro page added

On Thu, Aug 11, 2005 at 01:28:08PM -0700, Dan Price wrote: > On Thu 11 Aug 2005 at 11:56AM, Eric Saxe wrote: > > > This might sound dumb, but are we sure that processes are the resource > > > which is temporarily unavailable? > > > > I think what you are saying, Dan, is that "this might be so obvio