From 6484f5d16f9d5368afac61091972242f3bd695a9 Mon Sep 17 00:00:00 2001 From: Johannes Thumshirn Date: Wed, 12 Jul 2017 15:38:56 +0200 Subject: nvme: also provide a UUID in the WWID sysfs attribute The WWID sysfs attribute can provide multiple means of a World Wide ID for a NVMe device. It can either be a NGUID, a EUI-64 or a concatenation of VID, Serial Number, Model and the Namespace ID in this order of preference. If the target also sends us a UUID use the UUID for identification and give it the highest priority. This eases generation of /dev/disk/by-* symlinks. Signed-off-by: Johannes Thumshirn Signed-off-by: Christoph Hellwig --- drivers/nvme/host/core.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'drivers') diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c index 3b77cfe5aa1e..4cacab331f2a 100644 --- a/drivers/nvme/host/core.c +++ b/drivers/nvme/host/core.c @@ -1995,6 +1995,9 @@ static ssize_t wwid_show(struct device *dev, struct device_attribute *attr, int serial_len = sizeof(ctrl->serial); int model_len = sizeof(ctrl->model); + if (!uuid_is_null(&ns->uuid)) + return sprintf(buf, "uuid.%pU\n", &ns->uuid); + if (memchr_inv(ns->nguid, 0, sizeof(ns->nguid))) return sprintf(buf, "eui.%16phN\n", ns->nguid); -- cgit v1.2.3 From 8b25f351929b5a5216ccb2c8882965134019679d Mon Sep 17 00:00:00 2001 From: James Smart Date: Tue, 18 Jul 2017 14:29:34 -0700 Subject: nvme-fc: address target disconnect race conditions in fcp io submit There are cases where threads are in the process of submitting new io when the LLDD calls in to remove the remote port. In some cases, the next io actually goes to the LLDD, who knows the remoteport isn't present and rejects it. To properly recovery/restart these i/o's we don't want to hard fail them, we want to treat them as temporary resource errors in which a delayed retry will work. Add a couple more checks on remoteport connectivity and commonize the busy response handling when it's seen. Signed-off-by: James Smart Signed-off-by: Christoph Hellwig --- drivers/nvme/host/fc.c | 19 +++++++++++-------- 1 file changed, 11 insertions(+), 8 deletions(-) (limited to 'drivers') diff --git a/drivers/nvme/host/fc.c b/drivers/nvme/host/fc.c index d666ada39a9b..5630ca46c3b5 100644 --- a/drivers/nvme/host/fc.c +++ b/drivers/nvme/host/fc.c @@ -1888,7 +1888,7 @@ nvme_fc_start_fcp_op(struct nvme_fc_ctrl *ctrl, struct nvme_fc_queue *queue, * the target device is present */ if (ctrl->rport->remoteport.port_state != FC_OBJSTATE_ONLINE) - return BLK_STS_IOERR; + goto busy; if (!nvme_fc_ctrl_get(ctrl)) return BLK_STS_IOERR; @@ -1958,22 +1958,25 @@ nvme_fc_start_fcp_op(struct nvme_fc_ctrl *ctrl, struct nvme_fc_queue *queue, queue->lldd_handle, &op->fcp_req); if (ret) { - if (op->rq) /* normal request */ + if (!(op->flags & FCOP_FLAGS_AEN)) nvme_fc_unmap_data(ctrl, op->rq, op); - /* else - aen. no cleanup needed */ nvme_fc_ctrl_put(ctrl); - if (ret != -EBUSY) + if (ctrl->rport->remoteport.port_state == FC_OBJSTATE_ONLINE && + ret != -EBUSY) return BLK_STS_IOERR; - if (op->rq) - blk_mq_delay_run_hw_queue(queue->hctx, NVMEFC_QUEUE_DELAY); - - return BLK_STS_RESOURCE; + goto busy; } return BLK_STS_OK; + +busy: + if (!(op->flags & FCOP_FLAGS_AEN) && queue->hctx) + blk_mq_delay_run_hw_queue(queue->hctx, NVMEFC_QUEUE_DELAY); + + return BLK_STS_RESOURCE; } static blk_status_t -- cgit v1.2.3 From 9c5358e15ca12ed3dc3b1e51671dee5d155de8e0 Mon Sep 17 00:00:00 2001 From: James Smart Date: Mon, 17 Jul 2017 13:59:39 -0700 Subject: nvme-fc: revise TRADDR parsing The FC-NVME spec hasn't locked down on the format string for TRADDR. Currently the spec is lobbying for "nn-<16hexdigits>:pn-<16hexdigits>" where the wwn's are hex values but not prefixed by 0x. Most implementations so far expect a string format of "nn-0x<16hexdigits>:pn-0x<16hexdigits>" to be used. The transport uses the match_u64 parser which requires a leading 0x prefix to set the base properly. If it's not there, a match will either fail or return a base 10 value. The resolution in T11 is pushing out. Therefore, to fix things now and to cover any eventuality and any implementations already in the field, this patch adds support for both formats. The change consists of replacing the token matching routine with a routine that validates the fixed string format, and then builds a local copy of the hex name with a 0x prefix before calling the system parser. Note: the same parser routine exists in both the initiator and target transports. Given this is about the only "shared" item, we chose to replicate rather than create an interdendency on some shared code. Signed-off-by: James Smart Signed-off-by: Christoph Hellwig --- drivers/nvme/host/fc.c | 102 ++++++++++++++++++++++++----------------------- drivers/nvme/target/fc.c | 101 ++++++++++++++++++++++++---------------------- include/linux/nvme-fc.h | 19 +++++++++ 3 files changed, 125 insertions(+), 97 deletions(-) (limited to 'drivers') diff --git a/drivers/nvme/host/fc.c b/drivers/nvme/host/fc.c index 5630ca46c3b5..5c2a08ef08ba 100644 --- a/drivers/nvme/host/fc.c +++ b/drivers/nvme/host/fc.c @@ -2805,66 +2805,70 @@ out_fail: return ERR_PTR(ret); } -enum { - FCT_TRADDR_ERR = 0, - FCT_TRADDR_WWNN = 1 << 0, - FCT_TRADDR_WWPN = 1 << 1, -}; struct nvmet_fc_traddr { u64 nn; u64 pn; }; -static const match_table_t traddr_opt_tokens = { - { FCT_TRADDR_WWNN, "nn-%s" }, - { FCT_TRADDR_WWPN, "pn-%s" }, - { FCT_TRADDR_ERR, NULL } -}; - static int -nvme_fc_parse_address(struct nvmet_fc_traddr *traddr, char *buf) +__nvme_fc_parse_u64(substring_t *sstr, u64 *val) { - substring_t args[MAX_OPT_ARGS]; - char *options, *o, *p; - int token, ret = 0; u64 token64; - options = o = kstrdup(buf, GFP_KERNEL); - if (!options) - return -ENOMEM; + if (match_u64(sstr, &token64)) + return -EINVAL; + *val = token64; - while ((p = strsep(&o, ":\n")) != NULL) { - if (!*p) - continue; + return 0; +} - token = match_token(p, traddr_opt_tokens, args); - switch (token) { - case FCT_TRADDR_WWNN: - if (match_u64(args, &token64)) { - ret = -EINVAL; - goto out; - } - traddr->nn = token64; - break; - case FCT_TRADDR_WWPN: - if (match_u64(args, &token64)) { - ret = -EINVAL; - goto out; - } - traddr->pn = token64; - break; - default: - pr_warn("unknown traddr token or missing value '%s'\n", - p); - ret = -EINVAL; - goto out; - } - } +/* + * This routine validates and extracts the WWN's from the TRADDR string. + * As kernel parsers need the 0x to determine number base, universally + * build string to parse with 0x prefix before parsing name strings. + */ +static int +nvme_fc_parse_traddr(struct nvmet_fc_traddr *traddr, char *buf, size_t blen) +{ + char name[2 + NVME_FC_TRADDR_HEXNAMELEN + 1]; + substring_t wwn = { name, &name[sizeof(name)-1] }; + int nnoffset, pnoffset; + + /* validate it string one of the 2 allowed formats */ + if (strnlen(buf, blen) == NVME_FC_TRADDR_MAXLENGTH && + !strncmp(buf, "nn-0x", NVME_FC_TRADDR_OXNNLEN) && + !strncmp(&buf[NVME_FC_TRADDR_MAX_PN_OFFSET], + "pn-0x", NVME_FC_TRADDR_OXNNLEN)) { + nnoffset = NVME_FC_TRADDR_OXNNLEN; + pnoffset = NVME_FC_TRADDR_MAX_PN_OFFSET + + NVME_FC_TRADDR_OXNNLEN; + } else if ((strnlen(buf, blen) == NVME_FC_TRADDR_MINLENGTH && + !strncmp(buf, "nn-", NVME_FC_TRADDR_NNLEN) && + !strncmp(&buf[NVME_FC_TRADDR_MIN_PN_OFFSET], + "pn-", NVME_FC_TRADDR_NNLEN))) { + nnoffset = NVME_FC_TRADDR_NNLEN; + pnoffset = NVME_FC_TRADDR_MIN_PN_OFFSET + NVME_FC_TRADDR_NNLEN; + } else + goto out_einval; -out: - kfree(options); - return ret; + name[0] = '0'; + name[1] = 'x'; + name[2 + NVME_FC_TRADDR_HEXNAMELEN] = 0; + + memcpy(&name[2], &buf[nnoffset], NVME_FC_TRADDR_HEXNAMELEN); + if (__nvme_fc_parse_u64(&wwn, &traddr->nn)) + goto out_einval; + + memcpy(&name[2], &buf[pnoffset], NVME_FC_TRADDR_HEXNAMELEN); + if (__nvme_fc_parse_u64(&wwn, &traddr->pn)) + goto out_einval; + + return 0; + +out_einval: + pr_warn("%s: bad traddr string\n", __func__); + return -EINVAL; } static struct nvme_ctrl * @@ -2878,11 +2882,11 @@ nvme_fc_create_ctrl(struct device *dev, struct nvmf_ctrl_options *opts) unsigned long flags; int ret; - ret = nvme_fc_parse_address(&raddr, opts->traddr); + ret = nvme_fc_parse_traddr(&raddr, opts->traddr, NVMF_TRADDR_SIZE); if (ret || !raddr.nn || !raddr.pn) return ERR_PTR(-EINVAL); - ret = nvme_fc_parse_address(&laddr, opts->host_traddr); + ret = nvme_fc_parse_traddr(&laddr, opts->host_traddr, NVMF_TRADDR_SIZE); if (ret || !laddr.nn || !laddr.pn) return ERR_PTR(-EINVAL); diff --git a/drivers/nvme/target/fc.c b/drivers/nvme/target/fc.c index d5801c150b1c..31ca55dfcb1d 100644 --- a/drivers/nvme/target/fc.c +++ b/drivers/nvme/target/fc.c @@ -2293,66 +2293,70 @@ nvmet_fc_rcv_fcp_abort(struct nvmet_fc_target_port *target_port, } EXPORT_SYMBOL_GPL(nvmet_fc_rcv_fcp_abort); -enum { - FCT_TRADDR_ERR = 0, - FCT_TRADDR_WWNN = 1 << 0, - FCT_TRADDR_WWPN = 1 << 1, -}; struct nvmet_fc_traddr { u64 nn; u64 pn; }; -static const match_table_t traddr_opt_tokens = { - { FCT_TRADDR_WWNN, "nn-%s" }, - { FCT_TRADDR_WWPN, "pn-%s" }, - { FCT_TRADDR_ERR, NULL } -}; - static int -nvmet_fc_parse_traddr(struct nvmet_fc_traddr *traddr, char *buf) +__nvme_fc_parse_u64(substring_t *sstr, u64 *val) { - substring_t args[MAX_OPT_ARGS]; - char *options, *o, *p; - int token, ret = 0; u64 token64; - options = o = kstrdup(buf, GFP_KERNEL); - if (!options) - return -ENOMEM; + if (match_u64(sstr, &token64)) + return -EINVAL; + *val = token64; - while ((p = strsep(&o, ":\n")) != NULL) { - if (!*p) - continue; + return 0; +} - token = match_token(p, traddr_opt_tokens, args); - switch (token) { - case FCT_TRADDR_WWNN: - if (match_u64(args, &token64)) { - ret = -EINVAL; - goto out; - } - traddr->nn = token64; - break; - case FCT_TRADDR_WWPN: - if (match_u64(args, &token64)) { - ret = -EINVAL; - goto out; - } - traddr->pn = token64; - break; - default: - pr_warn("unknown traddr token or missing value '%s'\n", - p); - ret = -EINVAL; - goto out; - } - } +/* + * This routine validates and extracts the WWN's from the TRADDR string. + * As kernel parsers need the 0x to determine number base, universally + * build string to parse with 0x prefix before parsing name strings. + */ +static int +nvme_fc_parse_traddr(struct nvmet_fc_traddr *traddr, char *buf, size_t blen) +{ + char name[2 + NVME_FC_TRADDR_HEXNAMELEN + 1]; + substring_t wwn = { name, &name[sizeof(name)-1] }; + int nnoffset, pnoffset; + + /* validate it string one of the 2 allowed formats */ + if (strnlen(buf, blen) == NVME_FC_TRADDR_MAXLENGTH && + !strncmp(buf, "nn-0x", NVME_FC_TRADDR_OXNNLEN) && + !strncmp(&buf[NVME_FC_TRADDR_MAX_PN_OFFSET], + "pn-0x", NVME_FC_TRADDR_OXNNLEN)) { + nnoffset = NVME_FC_TRADDR_OXNNLEN; + pnoffset = NVME_FC_TRADDR_MAX_PN_OFFSET + + NVME_FC_TRADDR_OXNNLEN; + } else if ((strnlen(buf, blen) == NVME_FC_TRADDR_MINLENGTH && + !strncmp(buf, "nn-", NVME_FC_TRADDR_NNLEN) && + !strncmp(&buf[NVME_FC_TRADDR_MIN_PN_OFFSET], + "pn-", NVME_FC_TRADDR_NNLEN))) { + nnoffset = NVME_FC_TRADDR_NNLEN; + pnoffset = NVME_FC_TRADDR_MIN_PN_OFFSET + NVME_FC_TRADDR_NNLEN; + } else + goto out_einval; + + name[0] = '0'; + name[1] = 'x'; + name[2 + NVME_FC_TRADDR_HEXNAMELEN] = 0; + + memcpy(&name[2], &buf[nnoffset], NVME_FC_TRADDR_HEXNAMELEN); + if (__nvme_fc_parse_u64(&wwn, &traddr->nn)) + goto out_einval; + + memcpy(&name[2], &buf[pnoffset], NVME_FC_TRADDR_HEXNAMELEN); + if (__nvme_fc_parse_u64(&wwn, &traddr->pn)) + goto out_einval; -out: - kfree(options); - return ret; + return 0; + +out_einval: + pr_warn("%s: bad traddr string\n", __func__); + return -EINVAL; } static int @@ -2370,7 +2374,8 @@ nvmet_fc_add_port(struct nvmet_port *port) /* map the traddr address info to a target port */ - ret = nvmet_fc_parse_traddr(&traddr, port->disc_addr.traddr); + ret = nvme_fc_parse_traddr(&traddr, port->disc_addr.traddr, + sizeof(port->disc_addr.traddr)); if (ret) return ret; diff --git a/include/linux/nvme-fc.h b/include/linux/nvme-fc.h index 21c37e39e41a..36cca93a5ff2 100644 --- a/include/linux/nvme-fc.h +++ b/include/linux/nvme-fc.h @@ -334,5 +334,24 @@ struct fcnvme_ls_disconnect_acc { #define NVME_FC_LS_TIMEOUT_SEC 2 /* 2 seconds */ #define NVME_FC_TGTOP_TIMEOUT_SEC 2 /* 2 seconds */ +/* + * TRADDR string must be of form "nn-<16hexdigits>:pn-<16hexdigits>" + * the string is allowed to be specified with or without a "0x" prefix + * infront of the <16hexdigits>. Without is considered the "min" string + * and with is considered the "max" string. The hexdigits may be upper + * or lower case. + */ +#define NVME_FC_TRADDR_NNLEN 3 /* "?n-" */ +#define NVME_FC_TRADDR_OXNNLEN 5 /* "?n-0x" */ +#define NVME_FC_TRADDR_HEXNAMELEN 16 +#define NVME_FC_TRADDR_MINLENGTH \ + (2 * (NVME_FC_TRADDR_NNLEN + NVME_FC_TRADDR_HEXNAMELEN) + 1) +#define NVME_FC_TRADDR_MAXLENGTH \ + (2 * (NVME_FC_TRADDR_OXNNLEN + NVME_FC_TRADDR_HEXNAMELEN) + 1) +#define NVME_FC_TRADDR_MIN_PN_OFFSET \ + (NVME_FC_TRADDR_NNLEN + NVME_FC_TRADDR_HEXNAMELEN + 1) +#define NVME_FC_TRADDR_MAX_PN_OFFSET \ + (NVME_FC_TRADDR_OXNNLEN + NVME_FC_TRADDR_HEXNAMELEN + 1) + #endif /* _NVME_FC_H */ -- cgit v1.2.3 From 50cdb7c61b019a732fe34635a7cbf2a7487f5e90 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 25 Jul 2017 17:39:07 +0200 Subject: nvme-pci: fix HMB size calculation It's possible the preferred HMB size may not be a multiple of the chunk_size. This patch moves len to function scope and uses that in the for loop increment so the last iteration doesn't cause the total size to exceed the allocated HMB size. Based on an earlier patch from Keith Busch. Signed-off-by: Christoph Hellwig Reported-by: Dan Carpenter Reviewed-by: Keith Busch Fixes: 87ad72a59a38 ("nvme-pci: implement host memory buffer support") --- drivers/nvme/host/pci.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'drivers') diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c index 8569ee771269..cd888a47d0fc 100644 --- a/drivers/nvme/host/pci.c +++ b/drivers/nvme/host/pci.c @@ -1619,7 +1619,7 @@ static void nvme_free_host_mem(struct nvme_dev *dev) static int nvme_alloc_host_mem(struct nvme_dev *dev, u64 min, u64 preferred) { struct nvme_host_mem_buf_desc *descs; - u32 chunk_size, max_entries; + u32 chunk_size, max_entries, len; int i = 0; void **bufs; u64 size = 0, tmp; @@ -1638,10 +1638,10 @@ retry: if (!bufs) goto out_free_descs; - for (size = 0; size < preferred; size += chunk_size) { - u32 len = min_t(u64, chunk_size, preferred - size); + for (size = 0; size < preferred; size += len) { dma_addr_t dma_addr; + len = min_t(u64, chunk_size, preferred - size); bufs[i] = dma_alloc_attrs(dev->dev, len, &dma_addr, GFP_KERNEL, DMA_ATTR_NO_KERNEL_MAPPING | DMA_ATTR_NO_WARN); if (!bufs[i]) -- cgit v1.2.3 From 7dd1ab163c17e11473a65b11f7e748db30618ebb Mon Sep 17 00:00:00 2001 From: Scott Bauer Date: Tue, 25 Jul 2017 10:27:06 -0600 Subject: nvme: validate admin queue before unquiesce With a misbehaving controller it's possible we'll never enter the live state and create an admin queue. When we fail out of reset work it's possible we failed out early enough without setting up the admin queue. We tear down queues after a failed reset, but needed to do some more sanitization. Fixes 443bd90f2cca: "nvme: host: unquiesce queue in nvme_kill_queues()" [ 189.650995] nvme nvme1: pci function 0000:0b:00.0 [ 317.680055] nvme nvme0: Device not ready; aborting reset [ 317.680183] nvme nvme0: Removing after probe failure status: -19 [ 317.681258] kasan: GPF could be caused by NULL-ptr deref or user memory access [ 317.681397] general protection fault: 0000 [#1] SMP KASAN [ 317.682984] CPU: 3 PID: 477 Comm: kworker/3:2 Not tainted 4.13.0-rc1+ #5 [ 317.683112] Hardware name: Gigabyte Technology Co., Ltd. Z170X-UD5/Z170X-UD5-CF, BIOS F5 03/07/2016 [ 317.683284] Workqueue: events nvme_remove_dead_ctrl_work [nvme] [ 317.683398] task: ffff8803b0990000 task.stack: ffff8803c2ef0000 [ 317.683516] RIP: 0010:blk_mq_unquiesce_queue+0x2b/0xa0 [ 317.683614] RSP: 0018:ffff8803c2ef7d40 EFLAGS: 00010282 [ 317.683716] RAX: dffffc0000000000 RBX: 0000000000000000 RCX: 1ffff1006fbdcde3 [ 317.683847] RDX: 0000000000000038 RSI: 1ffff1006f5a9245 RDI: 0000000000000000 [ 317.683978] RBP: ffff8803c2ef7d58 R08: 1ffff1007bcdc974 R09: 0000000000000000 [ 317.684108] R10: 1ffff1007bcdc975 R11: 0000000000000000 R12: 00000000000001c0 [ 317.684239] R13: ffff88037ad49228 R14: ffff88037ad492d0 R15: ffff88037ad492e0 [ 317.684371] FS: 0000000000000000(0000) GS:ffff8803de6c0000(0000) knlGS:0000000000000000 [ 317.684519] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 317.684627] CR2: 0000002d1860c000 CR3: 000000045b40d000 CR4: 00000000003406e0 [ 317.684758] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 317.684888] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 317.685018] Call Trace: [ 317.685084] nvme_kill_queues+0x4d/0x170 [nvme_core] [ 317.685185] nvme_remove_dead_ctrl_work+0x3a/0x90 [nvme] [ 317.685289] process_one_work+0x771/0x1170 [ 317.685372] worker_thread+0xde/0x11e0 [ 317.685452] ? pci_mmcfg_check_reserved+0x110/0x110 [ 317.685550] kthread+0x2d3/0x3d0 [ 317.685617] ? process_one_work+0x1170/0x1170 [ 317.685704] ? kthread_create_on_node+0xc0/0xc0 [ 317.685785] ret_from_fork+0x25/0x30 [ 317.685798] Code: 0f 1f 44 00 00 55 48 b8 00 00 00 00 00 fc ff df 48 89 e5 41 54 4c 8d a7 c0 01 00 00 53 48 89 fb 4c 89 e2 48 c1 ea 03 48 83 ec 08 <80> 3c 02 00 75 50 48 8b bb c0 01 00 00 e8 33 8a f9 00 0f ba b3 [ 317.685872] RIP: blk_mq_unquiesce_queue+0x2b/0xa0 RSP: ffff8803c2ef7d40 [ 317.685908] ---[ end trace a3f8704150b1e8b4 ]--- Signed-off-by: Scott Bauer Signed-off-by: Christoph Hellwig --- drivers/nvme/host/core.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'drivers') diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c index 4cacab331f2a..c49f1f8b2e57 100644 --- a/drivers/nvme/host/core.c +++ b/drivers/nvme/host/core.c @@ -2712,7 +2712,8 @@ void nvme_kill_queues(struct nvme_ctrl *ctrl) mutex_lock(&ctrl->namespaces_mutex); /* Forcibly unquiesce queues to avoid blocking dispatch */ - blk_mq_unquiesce_queue(ctrl->admin_q); + if (ctrl->admin_q) + blk_mq_unquiesce_queue(ctrl->admin_q); list_for_each_entry(ns, &ctrl->namespaces, list) { /* -- cgit v1.2.3