From: hch@lst.de (Christoph Hellwig)
Subject: [PATCH 7/7] nvme: remove failed controllers directly from probe/reset context
Date: Fri, 2 Oct 2015 19:58:19 +0200 [thread overview]
Message-ID: <1443808699-4738-8-git-send-email-hch@lst.de> (raw)
In-Reply-To: <1443808699-4738-1-git-send-email-hch@lst.de>
Now that all probing and resets happen from workqueue context we can remove
the controller directly without any risk of deadlocking.
Signed-off-by: Christoph Hellwig <hch at lst.de>
---
drivers/block/nvme-core.c | 37 +++++++++++--------------------------
1 file changed, 11 insertions(+), 26 deletions(-)
diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c
index 48688d6..39c40bb 100644
--- a/drivers/block/nvme-core.c
+++ b/drivers/block/nvme-core.c
@@ -87,7 +87,6 @@ static struct class *nvme_class;
static int __nvme_reset(struct nvme_dev *dev);
static int nvme_reset(struct nvme_dev *dev);
static int nvme_process_cq(struct nvme_queue *nvmeq);
-static void nvme_dead_ctrl(struct nvme_dev *dev);
struct async_cmd_info {
struct kthread_work work;
@@ -2960,6 +2959,15 @@ static void nvme_set_irq_hints(struct nvme_dev *dev)
}
}
+static void nvme_remove_dead_ctrl(struct nvme_dev *dev)
+{
+ struct pci_dev *pdev = to_pci_dev(dev->dev);
+
+ dev_warn(dev->dev, "Device failed to start.\n");
+
+ if (pci_get_drvdata(pdev))
+ pci_stop_and_remove_bus_device_locked(pdev);
+}
static void nvme_probe_work(struct work_struct *work)
{
@@ -3033,30 +3041,7 @@ static void nvme_probe_work(struct work_struct *work)
nvme_dev_unmap(dev);
out:
if (!work_busy(&dev->reset_work))
- nvme_dead_ctrl(dev);
-}
-
-static int nvme_remove_dead_ctrl(void *arg)
-{
- struct nvme_dev *dev = (struct nvme_dev *)arg;
- struct pci_dev *pdev = to_pci_dev(dev->dev);
-
- if (pci_get_drvdata(pdev))
- pci_stop_and_remove_bus_device_locked(pdev);
- kref_put(&dev->kref, nvme_free_dev);
- return 0;
-}
-
-static void nvme_dead_ctrl(struct nvme_dev *dev)
-{
- dev_warn(dev->dev, "Device failed to resume\n");
- kref_get(&dev->kref);
- if (IS_ERR(kthread_run(nvme_remove_dead_ctrl, dev, "nvme%d",
- dev->instance))) {
- dev_err(dev->dev,
- "Failed to start controller remove task\n");
- kref_put(&dev->kref, nvme_free_dev);
- }
+ nvme_remove_dead_ctrl(dev);
}
static void nvme_reset_work(struct work_struct *ws)
@@ -3073,7 +3058,7 @@ static void nvme_reset_work(struct work_struct *ws)
/* Fail this device if reset occured during probe to avoid
* infinite initialization loops. */
if (in_probe) {
- nvme_dead_ctrl(dev);
+ nvme_remove_dead_ctrl(dev);
return;
}
/* Schedule device resume asynchronously so the reset work is available
--
1.9.1
next prev parent reply other threads:[~2015-10-02 17:58 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-10-02 17:58 nvme reset and probe updates Christoph Hellwig
2015-10-02 17:58 ` [PATCH 1/7] nvme: delete dev from dev_list in nvme_reset Christoph Hellwig
2015-10-02 21:13 ` Keith Busch
2015-10-03 10:01 ` Christoph Hellwig
2015-10-02 17:58 ` [PATCH 2/7] nvme: merge nvme_dev_reset into nvme_reset_failed_dev Christoph Hellwig
2015-10-02 21:15 ` Keith Busch
2015-10-02 17:58 ` [PATCH 3/7] nvme: factor reset code into a common helper Christoph Hellwig
2015-10-02 21:19 ` Keith Busch
2015-10-02 17:58 ` [PATCH 4/7] nvme: merge nvme_dev_start, nvme_dev_resume and nvme_async_probe Christoph Hellwig
2015-10-02 21:27 ` Keith Busch
2015-10-03 10:02 ` Christoph Hellwig
2015-10-02 17:58 ` [PATCH 5/7] nvme: remove duplicate call to nvme_set_irq_hints Christoph Hellwig
2015-10-02 17:58 ` [PATCH 6/7] nvme: properly handle partially initialized queues in nvme_create_io_queues Christoph Hellwig
2015-10-02 21:36 ` Keith Busch
2015-10-03 10:03 ` Christoph Hellwig
2015-10-02 17:58 ` Christoph Hellwig [this message]
2015-10-02 21:54 ` [PATCH 7/7] nvme: remove failed controllers directly from probe/reset context Keith Busch
2015-10-03 10:03 ` 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=1443808699-4738-8-git-send-email-hch@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).