From: hch@lst.de (Christoph Hellwig)
Subject: [PATCH 2/5] nvme-rdma: Free the I/O tags when we delete the controller
Date: Mon, 1 Aug 2016 13:03:08 +0200 [thread overview]
Message-ID: <20160801110308.GC16141@lst.de> (raw)
In-Reply-To: <1469822242-3477-3-git-send-email-sagi@grimberg.me>
On Fri, Jul 29, 2016@10:57:19PM +0300, Sagi Grimberg wrote:
> If we wait until we free the controller (free_ctrl) we might
> lose our rdma device without any notification while we still
> have open resources (tags mrs and dma mappings).
>
> Instead, destroy the tags with their rdma resources once we
> delete the device and not when freeing it.
>
> Note that we don't do that in nvme_rdma_shutdown_ctrl because
> controller reset uses it as well and we want to give active I/O
> a chance to complete successfully.
>
> Reported-by: Steve Wise <swise at opengridcomputing.com>
> Signed-off-by: Sagi Grimberg <sagi at grimberg.me>
This looks fine to me, but can we place share the code instead of
duplicating it? E.g.
static void __nvme_rdma_remove_ctrl(struct nvme_rdma_ctrl *ctrl, bool shutdown)
{
nvme_remove_namespaces(&ctrl->ctrl);
if (shutdown)
nvme_rdma_shutdown_ctrl(ctrl);
nvme_uninit_ctrl(&ctrl->ctrl);
if (ctrl->ctrl.tagset) {
blk_cleanup_queue(ctrl->ctrl.connect_q);
blk_mq_free_tag_set(&ctrl->tag_set);
nvme_rdma_dev_put(ctrl->device);
}
nvme_put_ctrl(&ctrl->ctrl);
}
or in a second step we should probably always call shutdown_ctrl
but skip the actual shutdown if the ctrl state doesn't require it.
next prev parent reply other threads:[~2016-08-01 11:03 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-07-29 19:57 [PATCH 0/5] Some fabrics fixes Sagi Grimberg
2016-07-29 19:57 ` [PATCH 1/5] nvme-rdma: Fix device removal handling Sagi Grimberg
2016-08-01 11:00 ` Christoph Hellwig
2016-07-29 19:57 ` [PATCH 2/5] nvme-rdma: Free the I/O tags when we delete the controller Sagi Grimberg
2016-08-01 11:03 ` Christoph Hellwig [this message]
2016-08-01 11:15 ` Sagi Grimberg
2016-08-01 15:47 ` Christoph Hellwig
2016-08-02 6:14 ` Sagi Grimberg
2016-07-29 19:57 ` [PATCH 3/5] nvme-rdma: Make sure to shutdown the controller if we can Sagi Grimberg
2016-08-01 11:04 ` Christoph Hellwig
2016-08-01 11:17 ` Sagi Grimberg
2016-08-01 15:48 ` Christoph Hellwig
2016-08-02 6:19 ` Sagi Grimberg
2016-08-02 12:52 ` Christoph Hellwig
2016-08-02 13:26 ` Sagi Grimberg
2016-07-29 19:57 ` [PATCH 4/5] nvme-rdma: Remove duplicate call to nvme_remove_namespaces Sagi Grimberg
2016-08-01 11:06 ` Christoph Hellwig
2016-08-01 11:21 ` Sagi Grimberg
2016-07-29 19:57 ` [PATCH 5/5] nvme-loop: " 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=20160801110308.GC16141@lst.de \
--to=hch@lst.de \
/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).