[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CAOu_J6kb8pQRzhUWD=JOhi1_=mxORUiL-isSaCfrxQ-yxOUepw@mail.gmail.com>
Date: Tue, 30 Sep 2014 20:10:21 +0200
From: Matias Bjørling <m@...rling.me>
To: Keith Busch <keith.busch@...el.com>
Cc: Matthew Wilcox <willy@...ux.intel.com>,
"Sam Bradshaw (sbradshaw)" <sbradshaw@...ron.com>,
Jens Axboe <axboe@...com>, Ming Lei <tom.leiming@...il.com>,
Christoph Hellwig <hch@...radead.org>,
Rob Nelson <rlnelson@...gle.com>,
LKML <linux-kernel@...r.kernel.org>,
linux-nvme <linux-nvme@...ts.infradead.org>
Subject: Re: [PATCH v13] NVMe: Convert to blk-mq
On Tue, Sep 30, 2014 at 7:42 PM, Keith Busch <keith.busch@...el.com> wrote:
> On Tue, 30 Sep 2014, Matias Bjørling wrote:
>>
>> @@ -1967,27 +1801,30 @@ static struct nvme_ns *nvme_alloc_ns(struct
>> nvme_dev *dev, unsigned nsid,
>> {
>
> ...
>>
>> - ns->queue->queue_flags = QUEUE_FLAG_DEFAULT;
>> + queue_flag_set_unlocked(QUEUE_FLAG_DEFAULT, ns->queue);
>
>
> Instead of the above, you want
>
> + ns->queue->queue_flags |= QUEUE_FLAG_DEFAULT;
>
> I only caught it because of the fun dm-mpath attempt with nvme and
> blk-mq. Or maybe we don't want to include "QUEUE_FLAG_STACKABLE" right
> now (part of the default flags) because it will kernel crash if you're
> using dm-multipath with a blk-mq driver today.
>
Great find. We should properly add a bunch of patches that add the
necessary support after the patch has gone in. I have a feeling that
there is a lot of interesting corner cases for getting stackable to
work.
> Otherwise, looks great!
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists