From: Sagi Grimberg <sagi@grimberg.me>
To: Logan Gunthorpe <logang@deltatee.com>,
linux-kernel@vger.kernel.org, linux-nvme@lists.infradead.org
Cc: Keith Busch <kbusch@kernel.org>, Max Gurtovoy <maxg@mellanox.com>,
Christoph Hellwig <hch@lst.de>,
Chaitanya Kulkarni <Chaitanya.Kulkarni@wdc.com>,
Stephen Bates <sbates@raithlin.com>
Subject: Re: [RFC PATCH 3/3] nvme: Introduce nvme_execute_passthru_rq_nowait()
Date: Fri, 25 Oct 2019 14:40:20 -0700 [thread overview]
Message-ID: <952c93fa-ef69-e113-a285-b1e9a0ddcafc@grimberg.me> (raw)
In-Reply-To: <11006dd2-718f-b569-4ef3-c01232354d5f@deltatee.com>
>>> +#ifdef CONFIG_NVME_TARGET_PASSTHRU
>>> +static void nvme_execute_passthru_rq_work(struct work_struct *w)
>>> +{
>>> + struct nvme_request *req = container_of(w, struct nvme_request,
>>> work);
>>> + struct request *rq = blk_mq_rq_from_pdu(req);
>>> + rq_end_io_fn *done = rq->end_io;
>>> + void *end_io_data = rq->end_io_data;
>>
>> Why is end_io_data stored to a local variable here? where is it set?
>
> blk_execute_rq() (which is called by nvme_execute_rq()) will overwrite
> rq->endio and rq->end_io_data. We store them here so we can call the
> callback appropriately after the request completes. It would be set by
> the caller in the same way they set it if they were calling
> blk_execute_rq_nowait().
I see..
>>> +
>>> + nvme_execute_passthru_rq(rq);
>>> +
>>> + if (done) {
>>> + rq->end_io_data = end_io_data;
>>> + done(rq, 0);
>>> + }
>>> +}
>>> +
>>> +void nvme_execute_passthru_rq_nowait(struct request *rq, rq_end_io_fn
>>> *done)
>>> +{
>>> + struct nvme_command *cmd = nvme_req(rq)->cmd;
>>> + struct nvme_ctrl *ctrl = nvme_req(rq)->ctrl;
>>> + struct nvme_ns *ns = rq->q->queuedata;
>>> + struct gendisk *disk = ns ? ns->disk : NULL;
>>> + u32 effects;
>>> +
>>> + /*
>>> + * This function may be called in interrupt context, so we cannot
>>> sleep
>>> + * but nvme_passthru_[start|end]() may sleep so we need to execute
>>> + * the command in a work queue.
>>> + */
>>> + effects = nvme_command_effects(ctrl, ns, cmd->common.opcode);
>>> + if (effects) {
>>> + rq->end_io = done;
>>> + INIT_WORK(&nvme_req(rq)->work, nvme_execute_passthru_rq_work);
>>> + queue_work(nvme_wq, &nvme_req(rq)->work);
>>
>> This work will need to be flushed when in nvme_stop_ctrl. That is
>> assuming that it will fail-fast and not hang (which it should given
>> that its a passthru command that is allocated via nvme_alloc_request).
>
> Hmm, that's going to be a bit tricky. Seeing the work_struct belongs
> potentially to a number of different requests, we can't just flush the
> individual work items. I think we'd have to create a work-queue per ctrl
> and flush that. Any objections to that?
I'd object to that overhead...
How about marking the request if the workqueue path is taken and
in nvme_stop_ctrl you add a blk_mq_tagset_busy_iter and cancel
it in the callback?
Something like:
--
diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c
index fa7ba09dca77..13dbbec5497d 100644
--- a/drivers/nvme/host/core.c
+++ b/drivers/nvme/host/core.c
@@ -3955,12 +3955,33 @@ void nvme_complete_async_event(struct nvme_ctrl
*ctrl, __le16 status,
}
EXPORT_SYMBOL_GPL(nvme_complete_async_event);
+static bool nvme_flush_async_passthru_request(struct request *rq,
+ void *data, bool reserved)
+{
+ if (!(nvme_req(rq)->flags & NVME_REQ_ASYNC_PASSTHRU))
+ return true;
+
+ dev_dbg_ratelimited(((struct nvme_ctrl *) data)->device,
+ "Cancelling passthru I/O %d", req->tag);
+ flush_work(&nvme_req(rq)->work);
+ return true;
+}
+
+static void nvme_flush_async_passthru_requests(struct nvme_ctrl *ctrl)
+{
+ blk_mq_tagset_busy_iter(ctrl->tagset,
+ nvme_flush_async_passthru_request, ctrl);
+ blk_mq_tagset_busy_iter(ctrl->admin_tagset,
+ nvme_flush_async_passthru_request, ctrl);
+}
+
void nvme_stop_ctrl(struct nvme_ctrl *ctrl)
{
nvme_mpath_stop(ctrl);
nvme_stop_keep_alive(ctrl);
flush_work(&ctrl->async_event_work);
cancel_work_sync(&ctrl->fw_act_work);
+ nvme_flush_async_passthru_requests(ctrl);
}
EXPORT_SYMBOL_GPL(nvme_stop_ctrl);
--
_______________________________________________
Linux-nvme mailing list
Linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme
next prev parent reply other threads:[~2019-10-25 21:41 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-10-25 20:25 [RFC PATCH 0/3] Passthru Execute Request Interface Logan Gunthorpe
2019-10-25 20:25 ` [RFC PATCH 1/3] nvme: Move nvme_passthru_[start|end]() calls to common code Logan Gunthorpe
2019-10-25 20:25 ` [RFC PATCH 2/3] nvme: Create helper function to obtain command effects Logan Gunthorpe
2019-10-27 15:05 ` Christoph Hellwig
2019-10-25 20:25 ` [RFC PATCH 3/3] nvme: Introduce nvme_execute_passthru_rq_nowait() Logan Gunthorpe
2019-10-25 20:41 ` Sagi Grimberg
2019-10-25 21:12 ` Logan Gunthorpe
2019-10-25 21:40 ` Sagi Grimberg [this message]
2019-10-25 21:55 ` Logan Gunthorpe
2019-10-27 15:09 ` Christoph Hellwig
2019-10-28 16:58 ` Logan Gunthorpe
2019-10-28 21:04 ` Sagi Grimberg
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=952c93fa-ef69-e113-a285-b1e9a0ddcafc@grimberg.me \
--to=sagi@grimberg.me \
--cc=Chaitanya.Kulkarni@wdc.com \
--cc=hch@lst.de \
--cc=kbusch@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-nvme@lists.infradead.org \
--cc=logang@deltatee.com \
--cc=maxg@mellanox.com \
--cc=sbates@raithlin.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).