From 59e330f8ff7ada7aa64fa422f6adf22a45152a7e Mon Sep 17 00:00:00 2001 From: Keith Busch Date: Thu, 17 Sep 2020 08:50:25 -0700 Subject: [PATCH 1/5] nvme: return errors for hwmon init Initializing the nvme hwmon retrieves a log from the controller. If the controller is broken, we need to return the appropriate error so that subsequent initialization doesn't attempt to continue. Reported-by: Tong Zhang Signed-off-by: Keith Busch Signed-off-by: Christoph Hellwig --- drivers/nvme/host/core.c | 7 +++++-- drivers/nvme/host/hwmon.c | 14 ++++++-------- drivers/nvme/host/nvme.h | 7 +++++-- 3 files changed, 16 insertions(+), 12 deletions(-) diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c index c013eb52fdc8..4cea14c18a6d 100644 --- a/drivers/nvme/host/core.c +++ b/drivers/nvme/host/core.c @@ -3236,8 +3236,11 @@ int nvme_init_identify(struct nvme_ctrl *ctrl) if (ret < 0) return ret; - if (!ctrl->identified) - nvme_hwmon_init(ctrl); + if (!ctrl->identified) { + ret = nvme_hwmon_init(ctrl); + if (ret < 0) + return ret; + } ctrl->identified = true; diff --git a/drivers/nvme/host/hwmon.c b/drivers/nvme/host/hwmon.c index 412a6c97c0d8..552dbc04567b 100644 --- a/drivers/nvme/host/hwmon.c +++ b/drivers/nvme/host/hwmon.c @@ -59,12 +59,8 @@ static int nvme_set_temp_thresh(struct nvme_ctrl *ctrl, int sensor, bool under, static int nvme_hwmon_get_smart_log(struct nvme_hwmon_data *data) { - int ret; - - ret = nvme_get_log(data->ctrl, NVME_NSID_ALL, NVME_LOG_SMART, 0, + return nvme_get_log(data->ctrl, NVME_NSID_ALL, NVME_LOG_SMART, 0, NVME_CSI_NVM, &data->log, sizeof(data->log), 0); - - return ret <= 0 ? ret : -EIO; } static int nvme_hwmon_read(struct device *dev, enum hwmon_sensor_types type, @@ -225,7 +221,7 @@ static const struct hwmon_chip_info nvme_hwmon_chip_info = { .info = nvme_hwmon_info, }; -void nvme_hwmon_init(struct nvme_ctrl *ctrl) +int nvme_hwmon_init(struct nvme_ctrl *ctrl) { struct device *dev = ctrl->dev; struct nvme_hwmon_data *data; @@ -234,7 +230,7 @@ void nvme_hwmon_init(struct nvme_ctrl *ctrl) data = devm_kzalloc(dev, sizeof(*data), GFP_KERNEL); if (!data) - return; + return 0; data->ctrl = ctrl; mutex_init(&data->read_lock); @@ -244,7 +240,7 @@ void nvme_hwmon_init(struct nvme_ctrl *ctrl) dev_warn(ctrl->device, "Failed to read smart log (error %d)\n", err); devm_kfree(dev, data); - return; + return err; } hwmon = devm_hwmon_device_register_with_info(dev, "nvme", data, @@ -254,4 +250,6 @@ void nvme_hwmon_init(struct nvme_ctrl *ctrl) dev_warn(dev, "Failed to instantiate hwmon device\n"); devm_kfree(dev, data); } + + return 0; } diff --git a/drivers/nvme/host/nvme.h b/drivers/nvme/host/nvme.h index 9fd45ff656da..2aaedfa43ed8 100644 --- a/drivers/nvme/host/nvme.h +++ b/drivers/nvme/host/nvme.h @@ -827,9 +827,12 @@ static inline struct nvme_ns *nvme_get_ns_from_dev(struct device *dev) } #ifdef CONFIG_NVME_HWMON -void nvme_hwmon_init(struct nvme_ctrl *ctrl); +int nvme_hwmon_init(struct nvme_ctrl *ctrl); #else -static inline void nvme_hwmon_init(struct nvme_ctrl *ctrl) { } +static inline int nvme_hwmon_init(struct nvme_ctrl *ctrl) +{ + return 0; +} #endif u32 nvme_command_effects(struct nvme_ctrl *ctrl, struct nvme_ns *ns, From 50b7c24390a53c78de546215282fb52980f1d7b7 Mon Sep 17 00:00:00 2001 From: Xianting Tian Date: Tue, 22 Sep 2020 14:25:17 +0800 Subject: [PATCH 2/5] nvme-pci: fix NULL req in completion handler Currently, we use nvmeq->q_depth as the upper limit for a valid tag in nvme_handle_cqe(), it is not correct. Because the available tag number is recorded in tagset, which is not equal to nvmeq->q_depth. The nvme driver registers interrupts for queues before initializing the tagset, because it uses the number of successful request_irq() calls to configure the tagset parameters. This allows a race condition with the current tag validity check if the controller happens to produce an interrupt with a corrupted CQE before the tagset is initialized. Replace the driver's indirect tag check with the one already provided by the block layer. Signed-off-by: Xianting Tian Reviewed-by: Keith Busch Signed-off-by: Christoph Hellwig --- drivers/nvme/host/pci.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c index d31e298669a9..d286c267869d 100644 --- a/drivers/nvme/host/pci.c +++ b/drivers/nvme/host/pci.c @@ -940,13 +940,6 @@ static inline void nvme_handle_cqe(struct nvme_queue *nvmeq, u16 idx) struct nvme_completion *cqe = &nvmeq->cqes[idx]; struct request *req; - if (unlikely(cqe->command_id >= nvmeq->q_depth)) { - dev_warn(nvmeq->dev->ctrl.device, - "invalid id %d completed on queue %d\n", - cqe->command_id, le16_to_cpu(cqe->sq_id)); - return; - } - /* * AEN requests are special as they don't time out and can * survive any kind of queue freeze and often don't respond to @@ -960,6 +953,13 @@ static inline void nvme_handle_cqe(struct nvme_queue *nvmeq, u16 idx) } req = blk_mq_tag_to_rq(nvme_queue_tagset(nvmeq), cqe->command_id); + if (unlikely(!req)) { + dev_warn(nvmeq->dev->ctrl.device, + "invalid id %d completed on queue %d\n", + cqe->command_id, le16_to_cpu(cqe->sq_id)); + return; + } + trace_nvme_sq(req, cqe->sq_head, nvmeq->sq_tail); if (!nvme_try_complete_req(req, cqe->status, cqe->result)) nvme_pci_complete_rq(req); From 9e0e8dac985d4bd07d9e62922b9d189d3ca2fccf Mon Sep 17 00:00:00 2001 From: James Smart Date: Thu, 17 Sep 2020 13:33:22 -0700 Subject: [PATCH 3/5] nvme-fc: fail new connections to a deleted host or remote port The lldd may have made calls to delete a remote port or local port and the delete is in progress when the cli then attempts to create a new controller. Currently, this proceeds without error although it can't be very successful. Fix this by validating that both the host port and remote port are present when a new controller is to be created. Signed-off-by: James Smart Reviewed-by: Himanshu Madhani Signed-off-by: Christoph Hellwig --- drivers/nvme/host/fc.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/nvme/host/fc.c b/drivers/nvme/host/fc.c index e8ef42b9d50c..e2e09e25c056 100644 --- a/drivers/nvme/host/fc.c +++ b/drivers/nvme/host/fc.c @@ -3671,12 +3671,14 @@ nvme_fc_create_ctrl(struct device *dev, struct nvmf_ctrl_options *opts) spin_lock_irqsave(&nvme_fc_lock, flags); list_for_each_entry(lport, &nvme_fc_lport_list, port_list) { if (lport->localport.node_name != laddr.nn || - lport->localport.port_name != laddr.pn) + lport->localport.port_name != laddr.pn || + lport->localport.port_state != FC_OBJSTATE_ONLINE) continue; list_for_each_entry(rport, &lport->endp_list, endp_list) { if (rport->remoteport.node_name != raddr.nn || - rport->remoteport.port_name != raddr.pn) + rport->remoteport.port_name != raddr.pn || + rport->remoteport.port_state != FC_OBJSTATE_ONLINE) continue; /* if fail to get reference fall through. Will error */ From 46d2613eae51d527ecaf0e8248a9bfcc0b92aa7e Mon Sep 17 00:00:00 2001 From: Chaitanya Kulkarni Date: Tue, 22 Sep 2020 12:49:38 -0700 Subject: [PATCH 4/5] nvme-core: don't use NVME_NSID_ALL for command effects and supported log In the function nvme_get_effects_log() it uses NVME_NSID_ALL which has namespace scope. The command effect log page is controller specific. Replace NVME_NSID_ALL with 0x00 which specifies the controller scope instead of namespace scope. Fixes: 84fef62d135b ("nvme: check admin passthru command effects") Link: https://bugzilla.kernel.org/show_bug.cgi?id=209287 Reported-by: Huai-Cheng Kuo Signed-off-by: Chaitanya Kulkarni Signed-off-by: Christoph Hellwig --- drivers/nvme/host/core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c index 4cea14c18a6d..53c93836c7c6 100644 --- a/drivers/nvme/host/core.c +++ b/drivers/nvme/host/core.c @@ -3041,7 +3041,7 @@ static int nvme_get_effects_log(struct nvme_ctrl *ctrl, u8 csi, if (!cel) return -ENOMEM; - ret = nvme_get_log(ctrl, NVME_NSID_ALL, NVME_LOG_CMD_EFFECTS, 0, csi, + ret = nvme_get_log(ctrl, 0x00, NVME_LOG_CMD_EFFECTS, 0, csi, &cel->log, sizeof(cel->log), 0); if (ret) { kfree(cel); From 3aab91774bbd8e571cfaddaf839aafd07718333c Mon Sep 17 00:00:00 2001 From: Jeffle Xu Date: Fri, 25 Sep 2020 14:00:31 +0800 Subject: [PATCH 5/5] block: remove unused BLK_QC_T_EAGAIN flag commit 7b6620d7db56 ("block: remove REQ_NOWAIT_INLINE") removed the REQ_NOWAIT_INLINE related code, but the diff wasn't applied to blk_types.h somehow. Then commit 2771cefeac49 ("block: remove the REQ_NOWAIT_INLINE flag") removed the REQ_NOWAIT_INLINE flag while the BLK_QC_T_EAGAIN flag still remains. Fixes: 7b6620d7db56 ("block: remove REQ_NOWAIT_INLINE") Signed-off-by: Jeffle Xu Signed-off-by: Jens Axboe --- include/linux/blk_types.h | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h index 4ecf4fed171f..b3fc5d3dd8ea 100644 --- a/include/linux/blk_types.h +++ b/include/linux/blk_types.h @@ -497,13 +497,12 @@ static inline int op_stat_group(unsigned int op) typedef unsigned int blk_qc_t; #define BLK_QC_T_NONE -1U -#define BLK_QC_T_EAGAIN -2U #define BLK_QC_T_SHIFT 16 #define BLK_QC_T_INTERNAL (1U << 31) static inline bool blk_qc_t_valid(blk_qc_t cookie) { - return cookie != BLK_QC_T_NONE && cookie != BLK_QC_T_EAGAIN; + return cookie != BLK_QC_T_NONE; } static inline unsigned int blk_qc_t_to_queue_num(blk_qc_t cookie)