From: Jesse Taube <jtaubepe@redhat.com>
To: linux-scsi@vger.kernel.org
Cc: Jonathan Corbet <corbet@lwn.net>,
Shuah Khan <skhan@linuxfoundation.org>,
Keith Busch <kbusch@kernel.org>, Jens Axboe <axboe@kernel.dk>,
Christoph Hellwig <hch@lst.de>, Sagi Grimberg <sagi@grimberg.me>,
Justin Tee <justin.tee@broadcom.com>,
Naresh Gottumukkala <nareshgottumukkala83@gmail.com>,
Paul Ely <paul.ely@broadcom.com>,
Chaitanya Kulkarni <kch@nvidia.com>,
"James E.J. Bottomley" <James.Bottomley@HansenPartnership.com>,
"Martin K. Petersen" <martin.petersen@oracle.com>,
Nilesh Javali <njavali@marvell.com>,
GR-QLogic-Storage-Upstream@marvell.com,
Hannes Reinecke <hare@suse.de>, Jesse Taube <jtaubepe@redhat.com>,
"Gustavo A. R. Silva" <gustavoars@kernel.org>,
John Meneghini <jmeneghi@redhat.com>,
Bryan Gurney <bgurney@redhat.com>,
Chris Leech <cleech@redhat.com>,
"Ewan D . Milne" <emilne@redhat.com>,
shinichiro.kawasaki@wdc.com, linux-doc@vger.kernel.org,
linux-kernel@vger.kernel.org, linux-nvme@lists.infradead.org,
linux-block@vger.kernel.org
Subject: [PATCH 03/13] nvme-multipath: numa support for marginal paths
Date: Wed, 12 Aug 2026 14:12:50 -0400 [thread overview]
Message-ID: <20260812181300.3712426-4-jtaubepe@redhat.com> (raw)
In-Reply-To: <20260812181300.3712426-1-jtaubepe@redhat.com>
FPIN LI (link integrity) messages are received when the attached
fabric detects hardware errors. In response to these messages I/O
should be directed away from the affected ports, and only used
if no other non-marginal paths are available.
To handle this a new controller flag 'NVME_CTRL_MARGINAL' is added
which will cause the multipath scheduler to skip these paths when
checking for 'optimized' paths.
Signed-off-by: Jesse Taube <jtaubepe@redhat.com>
---
This is a distinct change from the previous commit
which treated marginal paths as non-optimized but still usable.
This changes the priority of marginal paths to be lower than
non-optimized paths.
V10 -> V11:
- New commit
---
drivers/nvme/host/multipath.c | 53 +++++++++++++++++++++++++++++++++--
1 file changed, 50 insertions(+), 3 deletions(-)
diff --git a/drivers/nvme/host/multipath.c b/drivers/nvme/host/multipath.c
index 9b9a657fa330..02936c580a37 100644
--- a/drivers/nvme/host/multipath.c
+++ b/drivers/nvme/host/multipath.c
@@ -305,10 +305,44 @@ static bool nvme_path_is_disabled(struct nvme_ns *ns)
return false;
}
+/*
+ * Returns true if the new distance is better than the old one.
+ */
+static bool is_best_distance(bool found_is_marginal, bool marginal,
+ int old_distance, int distance)
+{
+ if (found_is_marginal) {
+ if (marginal) {
+ /*
+ * A marginal path has already been found,
+ * or this is the first path found.
+ * This one is also marginal, but closer
+ * to the NUMA node, so prefer it.
+ */
+ if (distance < old_distance)
+ return true;
+ } else {
+ /* Found a non-marginal path, use it over a marginal one. */
+ return true;
+ }
+ } else {
+ /* A non-marginal path has already found. This one is marginal, so skip it. */
+ if (marginal)
+ return false;
+
+ /* Found a closer non-marginal path, use it. */
+ if (distance < old_distance)
+ return true;
+ }
+
+ return false;
+}
+
static struct nvme_ns *__nvme_find_path(struct nvme_ns_head *head, int node)
{
int found_distance = INT_MAX, fallback_distance = INT_MAX, distance;
struct nvme_ns *found = NULL, *fallback = NULL, *ns;
+ bool found_is_marginal = true, fallback_is_marginal = true;
list_for_each_entry_srcu(ns, &head->list, siblings,
srcu_read_lock_held(&head->srcu)) {
@@ -323,15 +357,19 @@ static struct nvme_ns *__nvme_find_path(struct nvme_ns_head *head, int node)
switch (ns->ana_state) {
case NVME_ANA_OPTIMIZED:
- if (distance < found_distance) {
+ if (is_best_distance(found_is_marginal, nvme_ctrl_is_marginal(ns->ctrl),
+ found_distance, distance)) {
found_distance = distance;
found = ns;
+ found_is_marginal = nvme_ctrl_is_marginal(ns->ctrl);
}
break;
case NVME_ANA_NONOPTIMIZED:
- if (distance < fallback_distance) {
+ if (is_best_distance(fallback_is_marginal, nvme_ctrl_is_marginal(ns->ctrl),
+ fallback_distance, distance)) {
fallback_distance = distance;
fallback = ns;
+ fallback_is_marginal = nvme_ctrl_is_marginal(ns->ctrl);
}
break;
default:
@@ -339,6 +377,14 @@ static struct nvme_ns *__nvme_find_path(struct nvme_ns_head *head, int node)
}
}
+ /*
+ * Use non-optimized path only if it is not marginal
+ * and no optimized path is marginal.
+ */
+ if (found_is_marginal && !fallback_is_marginal)
+ found = fallback;
+
+ /* No optimized path found, use the fallback */
if (!found)
found = fallback;
if (found)
@@ -444,7 +490,8 @@ static struct nvme_ns *nvme_queue_depth_path(struct nvme_ns_head *head)
static inline bool nvme_path_is_optimized(struct nvme_ns *ns)
{
return nvme_ctrl_state(ns->ctrl) == NVME_CTRL_LIVE &&
- ns->ana_state == NVME_ANA_OPTIMIZED;
+ ns->ana_state == NVME_ANA_OPTIMIZED &&
+ !nvme_ctrl_is_marginal(ns->ctrl);
}
static struct nvme_ns *nvme_numa_path(struct nvme_ns_head *head)
--
2.54.0
next prev parent reply other threads:[~2026-08-12 18:13 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-08-12 18:12 [PATCH 00/13] nvme-fc: FPIN link integrity handling Jesse Taube
2026-08-12 18:12 ` [PATCH 01/13] fc_els: use 'union fc_tlv_desc' Jesse Taube
2026-08-12 18:12 ` [PATCH 02/13] nvme: add NVME_CTRL_MARGINAL flag Jesse Taube
2026-08-12 18:12 ` Jesse Taube [this message]
2026-08-12 18:12 ` [PATCH 04/13] nvme-multipath: queue-depth support for marginal paths Jesse Taube
2026-08-12 18:12 ` [PATCH 05/13] nvme-multipath: round-robin " Jesse Taube
2026-08-12 18:12 ` [PATCH 06/13] nvme: sysfs: emit the marginal path state in show_state() Jesse Taube
2026-08-12 18:12 ` [PATCH 07/13] scsi: scsi_transport_fc: Add set_rport_marginal to fc_function_template Jesse Taube
2026-08-12 18:12 ` [PATCH 08/13] scsi: scsi_transport_fc: user support for clearing NVME_CTRL_MARGINAL Jesse Taube
2026-08-12 18:12 ` [PATCH 09/13] nvme-fc: add nvme_fc_set_remoteport_fpin() Jesse Taube
2026-08-12 18:12 ` [PATCH 10/13] scsi: qla2xxx: enable FPIN notification for NVMe Jesse Taube
2026-08-12 18:12 ` [PATCH 11/13] scsi: lpfc: " Jesse Taube
2026-08-12 18:12 ` [PATCH 12/13] nvme: fcloop: Add set_rport_marginal to sysfs Jesse Taube
2026-08-12 18:34 ` Jesse Taube
2026-08-12 18:13 ` [PATCH 13/13] docs: nvme-multipath: Add FC-NVMe marginal state Jesse Taube
2026-08-12 18:46 ` Randy Dunlap
2026-08-12 18:50 ` Randy Dunlap
2026-08-13 4:53 ` [EXTERNAL] [PATCH 00/13] nvme-fc: FPIN link integrity handling Nilesh Javali
2026-08-13 16:24 ` Jesse Taube
2026-08-14 6:36 ` 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=20260812181300.3712426-4-jtaubepe@redhat.com \
--to=jtaubepe@redhat.com \
--cc=GR-QLogic-Storage-Upstream@marvell.com \
--cc=James.Bottomley@HansenPartnership.com \
--cc=axboe@kernel.dk \
--cc=bgurney@redhat.com \
--cc=cleech@redhat.com \
--cc=corbet@lwn.net \
--cc=emilne@redhat.com \
--cc=gustavoars@kernel.org \
--cc=hare@suse.de \
--cc=hch@lst.de \
--cc=jmeneghi@redhat.com \
--cc=justin.tee@broadcom.com \
--cc=kbusch@kernel.org \
--cc=kch@nvidia.com \
--cc=linux-block@vger.kernel.org \
--cc=linux-doc@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-nvme@lists.infradead.org \
--cc=linux-scsi@vger.kernel.org \
--cc=martin.petersen@oracle.com \
--cc=nareshgottumukkala83@gmail.com \
--cc=njavali@marvell.com \
--cc=paul.ely@broadcom.com \
--cc=sagi@grimberg.me \
--cc=shinichiro.kawasaki@wdc.com \
--cc=skhan@linuxfoundation.org \
/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