Messages in this thread | | | Date | Fri, 22 May 2015 00:44:53 +0530 | Subject | Re: [PATCH] NVMe: Avoid interrupt disable during queue init. | From | Parav Pandit <> |
| |
On Fri, May 22, 2015 at 12:09 AM, Jens Axboe <axboe@kernel.dk> wrote: > On 05/21/2015 06:12 PM, Parav Pandit wrote: >> >> Avoid diabling interrupt and holding q_lock for the queue >> which is just getting initialized. >> >> With this change, online_queues is also incremented without >> lock during queue setup stage. >> if Power management nvme_suspend() kicks in during queue setup time, >> per nvmeq based q_lock spinlock cannot protect device wide >> online_queues variable anyway. > > > Seems fairly pointless, it's not like it's a hot path... > I didn't follow your comments. Do you mean we should still hold the lock, even if its not needed?
I meant to say in above patch/comment that holding q_lock is not necessary in this path. So I removed it.
> -- > Jens Axboe >
| |