All of lore.kernel.org
 help / color / mirror / Atom feed
From: hch@lst.de (Christoph Hellwig)
Subject: [PATCH 02/26] nvme: only ignore hardware errors in nvme_create_io_queues
Date: Tue,  1 Dec 2015 19:52:37 +0100	[thread overview]
Message-ID: <1448995981-11982-3-git-send-email-hch@lst.de> (raw)
In-Reply-To: <1448995981-11982-1-git-send-email-hch@lst.de>

Half initialized queues due to kernel error returns or timeout are still a
good reason to give up on initializing a controller.

Signed-off-by: Christoph Hellwig <hch at lst.de>
---
 drivers/nvme/host/pci.c | 35 ++++++++++++++++++++---------------
 1 file changed, 20 insertions(+), 15 deletions(-)

diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
index a64d0ba..1f92b32 100644
--- a/drivers/nvme/host/pci.c
+++ b/drivers/nvme/host/pci.c
@@ -1537,26 +1537,33 @@ static int nvme_kthread(void *data)
 	return 0;
 }
 
-/*
- * Create I/O queues.  Failing to create an I/O queue is not an issue,
- * we can continue with less than the desired amount of queues, and
- * even a controller without I/O queues an still be used to issue
- * admin commands.  This might be useful to upgrade a buggy firmware
- * for example.
- */
-static void nvme_create_io_queues(struct nvme_dev *dev)
+static int nvme_create_io_queues(struct nvme_dev *dev)
 {
 	unsigned i;
+	int ret = 0;
 
-	for (i = dev->queue_count; i <= dev->max_qid; i++)
-		if (!nvme_alloc_queue(dev, i, dev->q_depth))
+	for (i = dev->queue_count; i <= dev->max_qid; i++) {
+		if (!nvme_alloc_queue(dev, i, dev->q_depth)) {
+			ret = -ENOMEM;
 			break;
+		}
+	}
 
-	for (i = dev->online_queues; i <= dev->queue_count - 1; i++)
-		if (nvme_create_queue(dev->queues[i], i)) {
+	for (i = dev->online_queues; i <= dev->queue_count - 1; i++) {
+		ret = nvme_create_queue(dev->queues[i], i);
+		if (ret) {
 			nvme_free_queues(dev, i);
 			break;
 		}
+	}
+
+	/*
+	 * Ignore failing Create SQ/CQ commands, we can continue with less
+	 * than the desired aount of queues, and even a controller without
+	 * I/O queues an still be used to issue admin commands.  This might
+	 * be useful to upgrade a buggy firmware for example.
+	 */
+	return ret >= 0 ? 0 : ret;
 }
 
 static void __iomem *nvme_map_cmb(struct nvme_dev *dev)
@@ -1702,9 +1709,7 @@ static int nvme_setup_io_queues(struct nvme_dev *dev)
 
 	/* Free previously allocated queues that are no longer usable */
 	nvme_free_queues(dev, nr_io_queues + 1);
-	nvme_create_io_queues(dev);
-
-	return 0;
+	return nvme_create_io_queues(dev);
 
  free_queues:
 	nvme_free_queues(dev, 1);
-- 
1.9.1

  parent reply	other threads:[~2015-12-01 18:52 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-12-01 18:52 NVMe reset fixes and completion path optimizations Christoph Hellwig
2015-12-01 18:52 ` [PATCH 01/26] block: defer timeouts to a workqueue Christoph Hellwig
2015-12-01 18:52 ` Christoph Hellwig [this message]
2015-12-01 18:52 ` [PATCH 03/26] nvme: only add a controller to dev_list after it's been fully initialized Christoph Hellwig
2015-12-01 18:52 ` [PATCH 04/26] nvme: protect against simultaneous shutdown invocations Christoph Hellwig
2015-12-01 18:52 ` [PATCH 05/26] nvme: don't take the I/O queue q_lock in nvme_timeout Christoph Hellwig
2015-12-01 18:52 ` [PATCH 06/26] nvme: merge nvme_abort_req and nvme_timeout Christoph Hellwig
2015-12-01 18:52 ` [PATCH 07/26] nvme: add NVME_SC_CANCELLED Christoph Hellwig
2015-12-01 18:52 ` [PATCH 08/26] nvme: simplify resets Christoph Hellwig
2015-12-01 18:52 ` [PATCH 09/26] nvme: do not restart the request timeout if we're resetting the controller Christoph Hellwig
2015-12-01 18:52 ` [PATCH 10/26] nvme: merge probe_work and reset_work Christoph Hellwig
2015-12-01 18:52 ` [PATCH 11/26] nvme: remove dead controllers from a work item Christoph Hellwig
2015-12-01 18:52 ` [PATCH 12/26] nvme: switch abort_limit to an atomic_t Christoph Hellwig
2015-12-01 18:52 ` [PATCH 13/26] NVMe: Implement namespace list scanning Christoph Hellwig
2015-12-01 18:52 ` [PATCH 14/26] NVMe: Use unbounded work queue for all work Christoph Hellwig
2015-12-01 18:52 ` [PATCH 15/26] NVMe: Remove device management handles on remove Christoph Hellwig
2015-12-01 18:52 ` [PATCH 16/26] NVMe: Simplify metadata setup Christoph Hellwig
2015-12-01 18:52 ` [PATCH 17/26] nvme: fix admin queue depth Christoph Hellwig
2015-12-01 18:52 ` [PATCH 18/26] nvme: factor out a few helpers from req_completion 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=1448995981-11982-3-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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.