mps(4) driver (LSI 6Gb SAS) commited to stable/8

[Jeremy Chadwick]

On Sat, Feb 19, 2011 at 01:08:41AM +0300, Dmitry Morozovsky wrote:
> On Fri, 18 Feb 2011, Kenneth D. Merry wrote:
> 
> KDM> I just merged the mps(4) driver to stable/8, for those of you with LSI 6Gb
> KDM> SAS hardware.
> 
> [snip]
> 
> Again, thank you very much Ken.  I'm planning to stress test this on 846 case 
> filled with 12 (yet) WD RE4 disks organized as raidz2, and will post the 
> results.
> 
> Any hints to particularly I/O stressing patterns?  Out of my mind, I'm planning 
> multiple parallel -j'ed builds, parallel tars, *SQL benchmarks -- what else 
> could you suppose?

Please be aware of a performance issue affecting certain models of WD
RE4 disks.  Specifics are still sketchy, but you should read the thread
"immense delayed write to file system (ZFS and UFS2), performance
issues" in full to get an idea of the problem:

http://lists.freebsd.org/pipermail/freebsd-stable/2010-January/thread.html#54185

There's confirmation of the problem here, with the statement that WD
gave at least one person a firmware image that fixed the problem:

http://lists.freebsd.org/pipermail/freebsd-stable/2010-January/054489.html

Buyer beware.  :-)  And remember, it's not an issue with the
brand/vendor, just certain models.  Caviar Black drives don't appear
affected (and don't let the TLER stuff make you lose focus), which is
one reason why I advocate them.

-- 
| Jeremy Chadwick                                   jdc at parodius.com |
| Parodius Networking                       http://www.parodius.com/ |
| UNIX Systems Administrator                  Mountain View, CA, USA |
| Making life hard for others since 1977.               PGP 4BD6C0CB |