cciss: Do not automatically rescan on UNIT ATTENTION/LUN DATA CHANGED
authorStephen M. Cameron <scameron@beardog.cce.hp.com>
Thu, 12 Nov 2009 18:49:55 +0000 (12:49 -0600)
committerJens Axboe <jens.axboe@oracle.com>
Fri, 13 Nov 2009 07:45:53 +0000 (08:45 +0100)
commitda0021841c3ea6a82588efae3260015847ea5d33
treeef789cce615bfef67721fe34d70deb5b0b25e4a3
parentd06dfbd236795acbb67e22e51bb8af12e953ced3
cciss: Do not automatically rescan on UNIT ATTENTION/LUN DATA CHANGED

cciss: Do not automatically rescan on UNIT ATTENTION/LUN DATA CHANGED
There are problems with doing this.  If, say, several logical drives
are deleted at once, several such UNIT ATTENTIONS will be encountered,
often during the rescan triggered by the first such UNIT ATTENTION.
The block layer may be in the midst of trying to add logical drives
which were just deleted (resulting in the subsequent UNIT ATTENTION(s).)
Making the rescan code robust enough to tolerate this kind of thing
is too complicated for the moment.  So, for now, we just don't do it.
Note: This UNIT ATTENTION/LUN DATA CHANGED situation only occurs on
the MSA2012.

Signed-off-by: Stephen M. Cameron <scameron@beardog.cce.hp.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
drivers/block/cciss.c