> > honor to read your message. I believe that commit mentioned was from you, if not, my bad.
> > Would you please > [1] introduce a few words about the environment in which it is noisy, It's pretty obvious, particularly if you are in rport removing that resets the exch pool, which is what my test has seen. > [2] the version of this patch, and Don't expect this one would go more than one round, if so, this is v1. > [3] the git cmdline since this patch looks cool? Not sure what you are asking, this patch was sent by 'stg mail'. Thanks, yi > > thanks//Hillf > > On Wed, Nov 10, 2010 at 8:03 AM, Yi Zou <[email protected]> wrote: > > commit 137c34f8 adds pool alloc/release counter, which should be reset > to > > 0 when the pool is being reset, otherwise it gets too noisy for > previously > > warned mismatch of the two counters. > > > > Signed-off-by: Yi Zou <[email protected]> > > --- > > > > drivers/scsi/libfc/fc_exch.c | 5 ++++- > > 1 files changed, 4 insertions(+), 1 deletions(-) > > > > diff --git a/drivers/scsi/libfc/fc_exch.c b/drivers/scsi/libfc/fc_exch.c > > index b7935e7..b50eb61 100644 > > --- a/drivers/scsi/libfc/fc_exch.c > > +++ b/drivers/scsi/libfc/fc_exch.c > > @@ -1736,12 +1736,15 @@ restart: > > } > > } > > > > - if (atomic_read(&pool->allocated) != atomic_read(&pool- > >released)) > > + if (atomic_read(&pool->allocated) != atomic_read(&pool- > >released)) { > > printk(KERN_WARNING "libfc: host %u exch pool %p " > > "allocated %d released %d\n", > > lport->host->host_no, pool, > > atomic_read(&pool->allocated), > > atomic_read(&pool->released)); > > + atomic_set(&pool->allocated, 0); > > + atomic_set(&pool->released, 0); > > + } > > > > spin_unlock_bh(&pool->lock); > > } > > > > _______________________________________________ devel mailing list [email protected] http://www.open-fcoe.org/mailman/listinfo/devel
