Linux-NVME Archive on lore.kernel.org
 help / color / Atom feed
From: keith.busch@intel.com (Keith Busch)
Subject: [PATCH 04/47] block: provide a new BLK_EH_QUIESCED timeout return value
Date: Tue, 24 Nov 2015 16:47:50 +0000
Message-ID: <20151124164750.GB5193@localhost.localdomain> (raw)
In-Reply-To: <x49wpt71hm9.fsf@segfault.boston.devel.redhat.com>

On Tue, Nov 24, 2015@11:34:22AM -0500, Jeff Moyer wrote:
> CPU 0 executes the following:
> 
> blk_rq_check_expired():
>         if (!blk_mark_rq_complete(rq))  // this succeeds, so we call blk_rq_timed_out
>                 blk_rq_timed_out(rq);
> blk_rq_timed_out():
>         if (q->rq_timed_out_fn)
>                 ret = q->rq_timed_out_fn(req);
>         switch (ret) {  // QUIESCED is returned
> ...
>         case BLK_EH_QUIESCED:
>                 set_bit(REQ_ATOM_QUIESCED, &req->atomic_flags);
>                 break;
> 
> CPU 1 takes an interrupt for the completion of the same request:
> 
> blk_complete_request():
>         if (!blk_mark_rq_complete(req) ||  // this fails, as it's already marked complete
>             test_and_clear_bit(REQ_ATOM_QUIESCED, &req->atomic_flags))  // this succeeds
>                 __blk_complete_request(req); // so we complete the request
> 
> Later, after the adapter reset is finished, you mentioned that all
> requests will be completed.  My question is: will this result in a
> double completion for this particular request?

The reset action clears the queue's busy tags only after the adapter
is disabled. At that point, no interrupts can happen and h/w will never
complete more active requests. If an interrupt completes the request
while the reset is in progress, the request tag won't be "busy" when
the final clean-up occurs.

  reply index

Thread overview: 63+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-11-20 16:34 NVMe mega patchbomb for Linux 4.5-rc Christoph Hellwig
2015-11-20 16:34 ` [PATCH 01/47] nvme: add missing unmaps in nvme_queue_rq Christoph Hellwig
2015-11-20 16:34 ` [PATCH 02/47] block: fix blk_abort_request for blk-mq drivers Christoph Hellwig
2015-11-20 21:43   ` Jeff Moyer
2015-11-20 21:47     ` Jens Axboe
2015-11-20 21:54       ` Jeff Moyer
2015-11-20 22:20   ` Jeff Moyer
2015-11-21  7:34     ` Christoph Hellwig
2015-11-20 16:34 ` [PATCH 03/47] block: defer timeouts to a workqueue Christoph Hellwig
2015-11-23 20:31   ` Jeff Moyer
2015-11-23 20:48     ` Christoph Hellwig
2015-11-23 20:59       ` Jeff Moyer
2015-11-20 16:34 ` [PATCH 04/47] block: provide a new BLK_EH_QUIESCED timeout return value Christoph Hellwig
2015-11-24 15:16   ` Jeff Moyer
2015-11-24 15:40     ` Christoph Hellwig
2015-11-24 15:51       ` Jeff Moyer
2015-11-24 15:56         ` Christoph Hellwig
2015-11-24 16:34           ` Jeff Moyer
2015-11-24 16:47             ` Keith Busch [this message]
2015-11-24 17:56             ` Christoph Hellwig
2015-11-24 18:12               ` Jeff Moyer
2015-11-24 19:40                 ` Christoph Hellwig
2015-11-20 16:35 ` [PATCH 05/47] block: remoe REQ_ATOM_COMPLETE wrappers Christoph Hellwig
2015-11-23 21:23   ` Jeff Moyer
2015-11-24 21:22   ` Jens Axboe
2015-11-20 16:35 ` [PATCH 06/47] blk-mq: add a flags parameter to blk_mq_alloc_request Christoph Hellwig
2015-11-24 15:19   ` Jeff Moyer
2015-11-24 15:32     ` Christoph Hellwig
2015-11-24 21:21   ` Jens Axboe
2015-11-24 22:22     ` Christoph Hellwig
2015-11-24 22:25       ` Jens Axboe
2015-11-20 16:35 ` [PATCH 07/47] nvme: move struct nvme_iod to pci.c Christoph Hellwig
2015-11-20 16:35 ` [PATCH 08/47] nvme: split command submission helpers out of pci.c Christoph Hellwig
2015-11-20 16:35 ` [PATCH 09/47] nvme: add a vendor field to struct nvme_dev Christoph Hellwig
2015-11-20 16:35 ` [PATCH 10/47] nvme: use offset instead of a struct for registers Christoph Hellwig
2015-11-20 16:35 ` [PATCH 11/47] nvme: split a new struct nvme_ctrl out of struct nvme_dev Christoph Hellwig
2015-11-20 16:35 ` [PATCH 12/47] nvme: simplify nvme_setup_prps calling convention Christoph Hellwig
2015-11-20 16:35 ` [PATCH 13/47] nvme: refactor nvme_queue_rq Christoph Hellwig
2015-11-20 16:35 ` [PATCH 14/47] nvme: move nvme_error_status to common code Christoph Hellwig
2015-11-20 16:35 ` [PATCH 15/47] nvme: move nvme_setup_flush and nvme_setup_rw " Christoph Hellwig
2015-11-20 16:35 ` [PATCH 16/47] nvme: split __nvme_submit_sync_cmd Christoph Hellwig
2015-11-20 16:35 ` [PATCH 17/47] nvme: use the block layer for userspace passthrough metadata Christoph Hellwig
2015-11-20 16:35 ` [PATCH 18/47] nvme: move block_device_operations and ns/ctrl freeing to common code Christoph Hellwig
2015-11-20 16:35 ` [PATCH 19/47] nvme: add explicit quirk handling Christoph Hellwig
2015-11-20 16:35 ` [PATCH 20/47] nvme: add a common helper to read Identify Controller data Christoph Hellwig
2015-11-20 16:35 ` [PATCH 21/47] nvme: move the call to nvme_init_identify earlier Christoph Hellwig
2015-11-20 16:35 ` [PATCH 22/47] nvme: move namespace scanning to common code Christoph Hellwig
2015-11-20 16:35 ` [PATCH 23/47] nvme: move chardev and sysfs interface " Christoph Hellwig
2015-11-20 16:35 ` [PATCH 24/47] nvme: only add a controller to dev_list after it's been fully initialized Christoph Hellwig
2015-11-20 16:35 ` [PATCH 25/47] nvme: don't take the I/O queue q_lock in nvme_timeout Christoph Hellwig
2017-03-10 12:51   ` David Woodhouse
2017-03-10 14:24     ` Christoph Hellwig
2015-11-20 16:35 ` [PATCH 26/47] nvme: merge nvme_abort_req and nvme_timeout Christoph Hellwig
2015-11-20 16:35 ` [PATCH 27/47] nvme: do not restart the request timeout if we're resetting the controller Christoph Hellwig
2015-11-20 16:35 ` [PATCH 28/47] nvme: simplify resets Christoph Hellwig
2015-11-20 16:35 ` [PATCH 29/47] nvme: merge probe_work and reset_work Christoph Hellwig
2015-11-20 16:35 ` [PATCH 30/47] nvme: remove dead controllers from a work item Christoph Hellwig
2015-11-20 16:35 ` [PATCH 31/47] nvme: switch abort_limit to an atomic_t Christoph Hellwig
2015-11-20 16:35 ` [PATCH 32/47] NVMe: Implement namespace list scanning Christoph Hellwig
2015-11-20 16:35 ` [PATCH 33/47] NVMe: Use unbounded work queue for all work Christoph Hellwig
2015-11-20 16:35 ` [PATCH 34/47] NVMe: Remove device management handles on remove Christoph Hellwig
2015-11-20 16:50 ` NVMe mega patchbomb for Linux 4.5-rc Christoph Hellwig
2015-11-21  7:19 NVMe mega patchbomb for Linux 4.5-rc (resend) Christoph Hellwig
2015-11-21  7:19 ` [PATCH 04/47] block: provide a new BLK_EH_QUIESCED timeout return value Christoph Hellwig

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=20151124164750.GB5193@localhost.localdomain \
    --to=keith.busch@intel.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

Linux-NVME Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/linux-nvme/0 linux-nvme/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-nvme linux-nvme/ https://lore.kernel.org/linux-nvme \
		linux-nvme@lists.infradead.org
	public-inbox-index linux-nvme

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.infradead.lists.linux-nvme


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git