On Tue, 2011-09-27 at 18:01 -0700, Roland Dreier wrote: > > (v2: Make access of qla_tgt_sess->sess_cmd_lock use normal spin_lock_irq() > > access as qla_hw_data->hardware_lock is already disabling interrupts) > > Do you have a tree where the end result of this is applied? > http://www.risingtidesystems.com/git/?p=lio-core-2.6.git;a=summary Unfortuately i've had to disable public git access for the moment as the number of full clones have been causing OOMs on the machine, so you'll need to use gitweb for now.. Sorry. > I'm just curious if there's anywhere that sess_cmd_lock is taken outside > of hardware_lock. If not then we don't even need that extra lock. > Correct.. > Assuming I merged everything correctly, in my tree sess_cmd_lock looks > like it is never taken outside of hardware_lock. So we can just get > rid of the superfluous lock. > I left that in for the moment as it looks kinda strange to have a qla_tgt_sess->sess_cmd_list, but no qla_tgt_sess context lock to protect it.. --nab -- To unsubscribe from this list: send the line "unsubscribe target-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
