From: Sasha Levin <sashal@kernel.org>
To: stable@vger.kernel.org
Cc: Keith Busch <kbusch@kernel.org>,
Nilay Shroff <nilay@linux.ibm.com>,
Christoph Hellwig <hch@lst.de>, Sasha Levin <sashal@kernel.org>
Subject: [PATCH 6.1.y 2/3] nvme-pci: do not directly handle subsys reset fallout
Date: Tue, 20 Jan 2026 21:56:58 -0500 [thread overview]
Message-ID: <20260121025659.1157002-2-sashal@kernel.org> (raw)
In-Reply-To: <20260121025659.1157002-1-sashal@kernel.org>
From: Keith Busch <kbusch@kernel.org>
[ Upstream commit 210b1f6576e8b367907e7ff51ef425062e1468e4 ]
Scheduling reset_work after a nvme subsystem reset is expected to fail
on pcie, but this also prevents potential handling the platform's pcie
services may provide that might successfully recovering the link without
re-enumeration. Such examples include AER, DPC, and power's EEH.
Provide a pci specific operation that safely initiates a subsystem
reset, and instead of scheduling reset work, read back the status
register to trigger a pcie read error.
Since this only affects pci, the other fabrics drivers subscribe to a
generic nvmf subsystem reset that is exactly the same as before. The
loop fabric doesn't use it because nvmet doesn't support setting that
property anyway.
And since we're using the magic NSSR value in two places now, provide a
symbolic define for it.
Reported-by: Nilay Shroff <nilay@linux.ibm.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Keith Busch <kbusch@kernel.org>
Stable-dep-of: 0edb475ac0a7 ("nvme: fix PCIe subsystem reset controller state transition")
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
drivers/nvme/host/fabrics.c | 15 +++++++++++++++
drivers/nvme/host/fabrics.h | 1 +
drivers/nvme/host/fc.c | 1 +
drivers/nvme/host/nvme.h | 14 +++-----------
drivers/nvme/host/pci.c | 36 ++++++++++++++++++++++++++++++++++++
drivers/nvme/host/rdma.c | 1 +
drivers/nvme/host/tcp.c | 1 +
include/linux/nvme.h | 3 +++
8 files changed, 61 insertions(+), 11 deletions(-)
diff --git a/drivers/nvme/host/fabrics.c b/drivers/nvme/host/fabrics.c
index ce27276f552da..fe621028a082e 100644
--- a/drivers/nvme/host/fabrics.c
+++ b/drivers/nvme/host/fabrics.c
@@ -253,6 +253,21 @@ int nvmf_reg_write32(struct nvme_ctrl *ctrl, u32 off, u32 val)
}
EXPORT_SYMBOL_GPL(nvmf_reg_write32);
+int nvmf_subsystem_reset(struct nvme_ctrl *ctrl)
+{
+ int ret;
+
+ if (!nvme_wait_reset(ctrl))
+ return -EBUSY;
+
+ ret = ctrl->ops->reg_write32(ctrl, NVME_REG_NSSR, NVME_SUBSYS_RESET);
+ if (ret)
+ return ret;
+
+ return nvme_try_sched_reset(ctrl);
+}
+EXPORT_SYMBOL_GPL(nvmf_subsystem_reset);
+
/**
* nvmf_log_connect_error() - Error-parsing-diagnostic print out function for
* connect() errors.
diff --git a/drivers/nvme/host/fabrics.h b/drivers/nvme/host/fabrics.h
index 60c238caf7a97..be2388ecc91be 100644
--- a/drivers/nvme/host/fabrics.h
+++ b/drivers/nvme/host/fabrics.h
@@ -199,6 +199,7 @@ static inline void nvmf_complete_timed_out_request(struct request *rq)
int nvmf_reg_read32(struct nvme_ctrl *ctrl, u32 off, u32 *val);
int nvmf_reg_read64(struct nvme_ctrl *ctrl, u32 off, u64 *val);
int nvmf_reg_write32(struct nvme_ctrl *ctrl, u32 off, u32 val);
+int nvmf_subsystem_reset(struct nvme_ctrl *ctrl);
int nvmf_connect_admin_queue(struct nvme_ctrl *ctrl);
int nvmf_connect_io_queue(struct nvme_ctrl *ctrl, u16 qid);
int nvmf_register_transport(struct nvmf_transport_ops *ops);
diff --git a/drivers/nvme/host/fc.c b/drivers/nvme/host/fc.c
index 71e22617c2874..dc84cade703db 100644
--- a/drivers/nvme/host/fc.c
+++ b/drivers/nvme/host/fc.c
@@ -3362,6 +3362,7 @@ static const struct nvme_ctrl_ops nvme_fc_ctrl_ops = {
.reg_read32 = nvmf_reg_read32,
.reg_read64 = nvmf_reg_read64,
.reg_write32 = nvmf_reg_write32,
+ .subsystem_reset = nvmf_subsystem_reset,
.free_ctrl = nvme_fc_free_ctrl,
.submit_async_event = nvme_fc_submit_async_event,
.delete_ctrl = nvme_fc_delete_ctrl,
diff --git a/drivers/nvme/host/nvme.h b/drivers/nvme/host/nvme.h
index 0f49b779dec65..8c97777ba629c 100644
--- a/drivers/nvme/host/nvme.h
+++ b/drivers/nvme/host/nvme.h
@@ -538,6 +538,7 @@ struct nvme_ctrl_ops {
int (*reg_read64)(struct nvme_ctrl *ctrl, u32 off, u64 *val);
void (*free_ctrl)(struct nvme_ctrl *ctrl);
void (*submit_async_event)(struct nvme_ctrl *ctrl);
+ int (*subsystem_reset)(struct nvme_ctrl *ctrl);
void (*delete_ctrl)(struct nvme_ctrl *ctrl);
void (*stop_ctrl)(struct nvme_ctrl *ctrl);
int (*get_address)(struct nvme_ctrl *ctrl, char *buf, int size);
@@ -636,18 +637,9 @@ int nvme_try_sched_reset(struct nvme_ctrl *ctrl);
static inline int nvme_reset_subsystem(struct nvme_ctrl *ctrl)
{
- int ret;
-
- if (!ctrl->subsystem)
+ if (!ctrl->subsystem || !ctrl->ops->subsystem_reset)
return -ENOTTY;
- if (!nvme_wait_reset(ctrl))
- return -EBUSY;
-
- ret = ctrl->ops->reg_write32(ctrl, NVME_REG_NSSR, 0x4E564D65);
- if (ret)
- return ret;
-
- return nvme_try_sched_reset(ctrl);
+ return ctrl->ops->subsystem_reset(ctrl);
}
/*
diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
index d162952194304..eeb170831dddb 100644
--- a/drivers/nvme/host/pci.c
+++ b/drivers/nvme/host/pci.c
@@ -1190,6 +1190,41 @@ static void nvme_pci_submit_async_event(struct nvme_ctrl *ctrl)
spin_unlock(&nvmeq->sq_lock);
}
+static int nvme_pci_subsystem_reset(struct nvme_ctrl *ctrl)
+{
+ struct nvme_dev *dev = to_nvme_dev(ctrl);
+ int ret = 0;
+
+ /*
+ * Taking the shutdown_lock ensures the BAR mapping is not being
+ * altered by reset_work. Holding this lock before the RESETTING state
+ * change, if successful, also ensures nvme_remove won't be able to
+ * proceed to iounmap until we're done.
+ */
+ mutex_lock(&dev->shutdown_lock);
+ if (!dev->bar_mapped_size) {
+ ret = -ENODEV;
+ goto unlock;
+ }
+
+ if (!nvme_change_ctrl_state(ctrl, NVME_CTRL_RESETTING)) {
+ ret = -EBUSY;
+ goto unlock;
+ }
+
+ writel(NVME_SUBSYS_RESET, dev->bar + NVME_REG_NSSR);
+ nvme_change_ctrl_state(ctrl, NVME_CTRL_LIVE);
+
+ /*
+ * Read controller status to flush the previous write and trigger a
+ * pcie read error.
+ */
+ readl(dev->bar + NVME_REG_CSTS);
+unlock:
+ mutex_unlock(&dev->shutdown_lock);
+ return ret;
+}
+
static int adapter_delete_queue(struct nvme_dev *dev, u8 opcode, u16 id)
{
struct nvme_command c = { };
@@ -3033,6 +3068,7 @@ static const struct nvme_ctrl_ops nvme_pci_ctrl_ops = {
.reg_read64 = nvme_pci_reg_read64,
.free_ctrl = nvme_pci_free_ctrl,
.submit_async_event = nvme_pci_submit_async_event,
+ .subsystem_reset = nvme_pci_subsystem_reset,
.get_address = nvme_pci_get_address,
.print_device_info = nvme_pci_print_device_info,
.supports_pci_p2pdma = nvme_pci_supports_pci_p2pdma,
diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c
index aa1734e2fd44e..47fbf561c01e6 100644
--- a/drivers/nvme/host/rdma.c
+++ b/drivers/nvme/host/rdma.c
@@ -2256,6 +2256,7 @@ static const struct nvme_ctrl_ops nvme_rdma_ctrl_ops = {
.reg_read32 = nvmf_reg_read32,
.reg_read64 = nvmf_reg_read64,
.reg_write32 = nvmf_reg_write32,
+ .subsystem_reset = nvmf_subsystem_reset,
.free_ctrl = nvme_rdma_free_ctrl,
.submit_async_event = nvme_rdma_submit_async_event,
.delete_ctrl = nvme_rdma_delete_ctrl,
diff --git a/drivers/nvme/host/tcp.c b/drivers/nvme/host/tcp.c
index 4e1b91c0416b9..441cacad4498b 100644
--- a/drivers/nvme/host/tcp.c
+++ b/drivers/nvme/host/tcp.c
@@ -2612,6 +2612,7 @@ static const struct nvme_ctrl_ops nvme_tcp_ctrl_ops = {
.reg_read32 = nvmf_reg_read32,
.reg_read64 = nvmf_reg_read64,
.reg_write32 = nvmf_reg_write32,
+ .subsystem_reset = nvmf_subsystem_reset,
.free_ctrl = nvme_tcp_free_ctrl,
.submit_async_event = nvme_tcp_submit_async_event,
.delete_ctrl = nvme_tcp_delete_ctrl,
diff --git a/include/linux/nvme.h b/include/linux/nvme.h
index 15086715632e0..0d6153fc9b01a 100644
--- a/include/linux/nvme.h
+++ b/include/linux/nvme.h
@@ -28,6 +28,9 @@
#define NVME_NSID_ALL 0xffffffff
+/* Special NSSR value, 'NVMe' */
+#define NVME_SUBSYS_RESET 0x4E564D65
+
enum nvme_subsys_type {
/* Referral to another discovery type target subsystem */
NVME_NQN_DISC = 1,
--
2.51.0
next prev parent reply other threads:[~2026-01-21 2:57 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-01-20 13:07 FAILED: patch "[PATCH] nvme: fix PCIe subsystem reset controller state transition" failed to apply to 6.1-stable tree gregkh
2026-01-21 2:56 ` [PATCH 6.1.y 1/3] nvme-fc: rename free_ctrl callback to match name pattern Sasha Levin
2026-01-21 2:56 ` Sasha Levin [this message]
2026-01-21 2:56 ` [PATCH 6.1.y 3/3] nvme: fix PCIe subsystem reset controller state transition Sasha Levin
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=20260121025659.1157002-2-sashal@kernel.org \
--to=sashal@kernel.org \
--cc=hch@lst.de \
--cc=kbusch@kernel.org \
--cc=nilay@linux.ibm.com \
--cc=stable@vger.kernel.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