From 886d98fa48580cd5d3406f4e6f258bb990bb1531 Mon Sep 17 00:00:00 2001 From: Keith Busch Date: Tue, 20 Jan 2026 22:02:37 -0500 Subject: [PATCH] nvme-pci: do not directly handle subsys reset fallout [ 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 Reviewed-by: Christoph Hellwig Signed-off-by: Keith Busch Stable-dep-of: 0edb475ac0a7 ("nvme: fix PCIe subsystem reset controller state transition") Signed-off-by: Sasha Levin Signed-off-by: Greg Kroah-Hartman --- 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 668c6bb7a567..d4028c669a2b 100644 --- a/drivers/nvme/host/fabrics.c +++ b/drivers/nvme/host/fabrics.c @@ -254,6 +254,21 @@ 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 561c2abd3892..f727f846ec34 100644 --- a/drivers/nvme/host/fabrics.h +++ b/drivers/nvme/host/fabrics.h @@ -182,6 +182,7 @@ nvmf_ctlr_matches_baseopts(struct nvme_ctrl *ctrl, 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 ebca6244a96c..ac1e4482011e 100644 --- a/drivers/nvme/host/fc.c +++ b/drivers/nvme/host/fc.c @@ -3361,6 +3361,7 @@ 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 c8ec0e146c8c..c6cd3d65065c 100644 --- a/drivers/nvme/host/nvme.h +++ b/drivers/nvme/host/nvme.h @@ -514,6 +514,7 @@ 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); @@ -583,18 +584,9 @@ 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 cd383ddf53f8..62d76245f3a9 100644 --- a/drivers/nvme/host/pci.c +++ b/drivers/nvme/host/pci.c @@ -1121,6 +1121,41 @@ 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 = { }; @@ -2905,6 +2940,7 @@ 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, }; diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c index 6e92bdf459fe..00c6924bf5f9 100644 --- a/drivers/nvme/host/rdma.c +++ b/drivers/nvme/host/rdma.c @@ -2287,6 +2287,7 @@ 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 99bf17f2dcfc..1be8f7867a06 100644 --- a/drivers/nvme/host/tcp.c +++ b/drivers/nvme/host/tcp.c @@ -2559,6 +2559,7 @@ 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 537cc5b7e050..94fa0184c153 100644 --- a/include/linux/nvme.h +++ b/include/linux/nvme.h @@ -27,6 +27,9 @@ #define NVME_NSID_ALL 0xffffffff +/* Special NSSR value, 'NVMe' */ +#define NVME_SUBSYS_RESET 0x4E564D65 + enum nvme_subsys_type { NVME_NQN_DISC = 1, /* Discovery type target subsystem */ NVME_NQN_NVME = 2, /* NVME type target subsystem */