From 48379270fe6808cf4612ee094adc8da2b7a83baa Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Mon, 3 Nov 2014 19:36:40 +0100 Subject: scsi: only re-lock door after EH on devices that were reset Setups that use the blk-mq I/O path can lock up if a host with a single device that has its door locked enters EH. Make sure to only send the command to re-lock the door to devices that actually were reset and thus might have lost their state. Otherwise the EH code might be get blocked on blk_get_request as all requests for non-reset devices might be in use. Cc: stable@vger.kernel.org Signed-off-by: Christoph Hellwig Reported-by: Meelis Roos Tested-by: Meelis Roos Reviewed-by: Martin K. Petersen --- drivers/scsi/scsi_error.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c index 9a6f8468225f..c21e53071600 100644 --- a/drivers/scsi/scsi_error.c +++ b/drivers/scsi/scsi_error.c @@ -2001,8 +2001,10 @@ static void scsi_restart_operations(struct Scsi_Host *shost) * is no point trying to lock the door of an off-line device. */ shost_for_each_device(sdev, shost) { - if (scsi_device_online(sdev) && sdev->locked) + if (scsi_device_online(sdev) && sdev->was_reset && sdev->locked) { scsi_eh_lock_door(sdev); + sdev->was_reset = 0; + } } /* -- cgit v1.2.3 From e925cc431ac8285ad6e8b3fe09f6e3d8b3c30d56 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 6 Nov 2014 15:11:22 -0600 Subject: scsi: call device handler for failed TUR command Multipath devices using the TUR path checker need to see the sense code for a failed TUR command in their device handler. Since commit 14216561e164671ce147458653b1fea06a we always return success for mid layer issued TUR commands before calling the device handler, which stopped the TUR path checker from working. Move the call to the device handler check sense method before the early return for TUR commands to give the device handler a chance to intercept them. Signed-off-by: Christoph Hellwig Tested-by: Wen Xiong Reviewed-by: Hannes Reinecke --- drivers/scsi/scsi_error.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c index c21e53071600..bc5ff6ff9c79 100644 --- a/drivers/scsi/scsi_error.c +++ b/drivers/scsi/scsi_error.c @@ -459,14 +459,6 @@ static int scsi_check_sense(struct scsi_cmnd *scmd) if (! scsi_command_normalize_sense(scmd, &sshdr)) return FAILED; /* no valid sense data */ - if (scmd->cmnd[0] == TEST_UNIT_READY && scmd->scsi_done != scsi_eh_done) - /* - * nasty: for mid-layer issued TURs, we need to return the - * actual sense data without any recovery attempt. For eh - * issued ones, we need to try to recover and interpret - */ - return SUCCESS; - scsi_report_sense(sdev, &sshdr); if (scsi_sense_is_deferred(&sshdr)) @@ -482,6 +474,14 @@ static int scsi_check_sense(struct scsi_cmnd *scmd) /* handler does not care. Drop down to default handling */ } + if (scmd->cmnd[0] == TEST_UNIT_READY && scmd->scsi_done != scsi_eh_done) + /* + * nasty: for mid-layer issued TURs, we need to return the + * actual sense data without any recovery attempt. For eh + * issued ones, we need to try to recover and interpret + */ + return SUCCESS; + /* * Previous logic looked for FILEMARK, EOM or ILI which are * mainly associated with tapes and returned SUCCESS. -- cgit v1.2.3 From 333b2448cf5bd5a94c91a77136cac837d38fb984 Mon Sep 17 00:00:00 2001 From: "wenxiong@linux.vnet.ibm.com" Date: Thu, 6 Nov 2014 15:11:23 -0600 Subject: scsi: TUR path is down after adapter gets reset with multipath This patch fixes an issue with multipath ipr SAS devices which require a start unit command to be issued following an adapter reset. Without this patch, paths get marked failed following an adapter reset and since the error handler never gets invoked to issue the start unit, the paths are never recovered. Returning FAILED for this case ensures the error handler wakes up to issue the start unit. Signed-off-by: Brian King Tested-by: Wen Xiong Reviewed-by: Hannes Reinecke Signed-off-by: Christoph Hellwig --- drivers/scsi/device_handler/scsi_dh_alua.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/scsi/device_handler/scsi_dh_alua.c b/drivers/scsi/device_handler/scsi_dh_alua.c index e99507ed0e3c..fd78bdc53528 100644 --- a/drivers/scsi/device_handler/scsi_dh_alua.c +++ b/drivers/scsi/device_handler/scsi_dh_alua.c @@ -474,6 +474,13 @@ static int alua_check_sense(struct scsi_device *sdev, * LUN Not Ready -- Offline */ return SUCCESS; + if (sdev->allow_restart && + sense_hdr->asc == 0x04 && sense_hdr->ascq == 0x02) + /* + * if the device is not started, we need to wake + * the error handler to start the motor + */ + return FAILED; break; case UNIT_ATTENTION: if (sense_hdr->asc == 0x29 && sense_hdr->ascq == 0x00) -- cgit v1.2.3 From d8ae3c33599ac5c7bc758d88529debbbe8c09265 Mon Sep 17 00:00:00 2001 From: Anish Bhatt Date: Tue, 4 Nov 2014 15:19:22 -0800 Subject: cxgbi: add maintainer for cxgb3i/cxgb4i Signed-off-by: Anish Bhatt Signed-off-by: Christoph Hellwig --- MAINTAINERS | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index ea4d0058fd1b..84d9c5dea6bc 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2744,6 +2744,13 @@ W: http://www.chelsio.com S: Supported F: drivers/net/ethernet/chelsio/cxgb3/ +CXGB3 ISCSI DRIVER (CXGB3I) +M: Karen Xie +L: linux-scsi@vger.kernel.org +W: http://www.chelsio.com +S: Supported +F: drivers/scsi/cxgbi/cxgb3i + CXGB3 IWARP RNIC DRIVER (IW_CXGB3) M: Steve Wise L: linux-rdma@vger.kernel.org @@ -2758,6 +2765,13 @@ W: http://www.chelsio.com S: Supported F: drivers/net/ethernet/chelsio/cxgb4/ +CXGB4 ISCSI DRIVER (CXGB4I) +M: Karen Xie +L: linux-scsi@vger.kernel.org +W: http://www.chelsio.com +S: Supported +F: drivers/scsi/cxgbi/cxgb4i + CXGB4 IWARP RNIC DRIVER (IW_CXGB4) M: Steve Wise L: linux-rdma@vger.kernel.org -- cgit v1.2.3 From 7b07bf244ae33e4ce8d51c23f1359121942a6f3e Mon Sep 17 00:00:00 2001 From: Anish Bhatt Date: Thu, 6 Nov 2014 12:53:58 -0800 Subject: cxgb4i: send abort_rpl correctly Connection retries were not being cleaned up correctly if they failed as a result of link down. Applies on top of drivers-for-3.18. Signed-off-by: Anish Bhatt Signed-off-by: Karen Xie Signed-off-by: Christoph Hellwig --- drivers/scsi/cxgbi/cxgb4i/cxgb4i.c | 15 +++++++++------ drivers/scsi/cxgbi/libcxgbi.c | 18 ++++++++---------- 2 files changed, 17 insertions(+), 16 deletions(-) diff --git a/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c b/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c index 3e0a0d315f72..81bb3bd7909d 100644 --- a/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c +++ b/drivers/scsi/cxgbi/cxgb4i/cxgb4i.c @@ -936,20 +936,23 @@ static void do_abort_req_rss(struct cxgbi_device *cdev, struct sk_buff *skb) cxgbi_sock_get(csk); spin_lock_bh(&csk->lock); - if (!cxgbi_sock_flag(csk, CTPF_ABORT_REQ_RCVD)) { - cxgbi_sock_set_flag(csk, CTPF_ABORT_REQ_RCVD); - cxgbi_sock_set_state(csk, CTP_ABORTING); - goto done; + cxgbi_sock_clear_flag(csk, CTPF_ABORT_REQ_RCVD); + + if (!cxgbi_sock_flag(csk, CTPF_TX_DATA_SENT)) { + send_tx_flowc_wr(csk); + cxgbi_sock_set_flag(csk, CTPF_TX_DATA_SENT); } - cxgbi_sock_clear_flag(csk, CTPF_ABORT_REQ_RCVD); + cxgbi_sock_set_flag(csk, CTPF_ABORT_REQ_RCVD); + cxgbi_sock_set_state(csk, CTP_ABORTING); + send_abort_rpl(csk, rst_status); if (!cxgbi_sock_flag(csk, CTPF_ABORT_RPL_PENDING)) { csk->err = abort_status_to_errno(csk, req->status, &rst_status); cxgbi_sock_closed(csk); } -done: + spin_unlock_bh(&csk->lock); cxgbi_sock_put(csk); rel_skb: diff --git a/drivers/scsi/cxgbi/libcxgbi.c b/drivers/scsi/cxgbi/libcxgbi.c index 674d498b46ab..13d869a92248 100644 --- a/drivers/scsi/cxgbi/libcxgbi.c +++ b/drivers/scsi/cxgbi/libcxgbi.c @@ -905,18 +905,16 @@ void cxgbi_sock_rcv_abort_rpl(struct cxgbi_sock *csk) { cxgbi_sock_get(csk); spin_lock_bh(&csk->lock); + + cxgbi_sock_set_flag(csk, CTPF_ABORT_RPL_RCVD); if (cxgbi_sock_flag(csk, CTPF_ABORT_RPL_PENDING)) { - if (!cxgbi_sock_flag(csk, CTPF_ABORT_RPL_RCVD)) - cxgbi_sock_set_flag(csk, CTPF_ABORT_RPL_RCVD); - else { - cxgbi_sock_clear_flag(csk, CTPF_ABORT_RPL_RCVD); - cxgbi_sock_clear_flag(csk, CTPF_ABORT_RPL_PENDING); - if (cxgbi_sock_flag(csk, CTPF_ABORT_REQ_RCVD)) - pr_err("csk 0x%p,%u,0x%lx,%u,ABT_RPL_RSS.\n", - csk, csk->state, csk->flags, csk->tid); - cxgbi_sock_closed(csk); - } + cxgbi_sock_clear_flag(csk, CTPF_ABORT_RPL_PENDING); + if (cxgbi_sock_flag(csk, CTPF_ABORT_REQ_RCVD)) + pr_err("csk 0x%p,%u,0x%lx,%u,ABT_RPL_RSS.\n", + csk, csk->state, csk->flags, csk->tid); + cxgbi_sock_closed(csk); } + spin_unlock_bh(&csk->lock); cxgbi_sock_put(csk); } -- cgit v1.2.3 From c12de882e7765585ae3a2ae22aa569285951613a Mon Sep 17 00:00:00 2001 From: Jiang Liu Date: Mon, 3 Nov 2014 20:44:20 +0800 Subject: megaraid_sas: fix bug in handling return value of pci_enable_msix_range() Function pci_enable_msix_range() may return negative values for error conditions. So it's a bug by checking (pci_enable_msix_range() != 0) for success and causes failure to megaraid driver when MSI is disabled. [ 16.487267] megaraid_sas 0000:02:00.0: Controller type: iMR [ 16.487275] genirq: Flags mismatch irq 0. 00000000 (megasas) vs. 00015a00 (tii mer) [ 16.487347] megasas: Failed to register IRQ for vector 0. Fixes: 8ae80ed1734b "megaraid: Use pci_enable_msix_range() instead of pci_enable_msix()" Signed-off-by: Jiang Liu Acked-by: Sumit Saxena Signed-off-by: Christoph Hellwig Cc: # 3.17 --- drivers/scsi/megaraid/megaraid_sas_base.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/scsi/megaraid/megaraid_sas_base.c b/drivers/scsi/megaraid/megaraid_sas_base.c index f6a69a3b1b3f..5640ad1c8214 100644 --- a/drivers/scsi/megaraid/megaraid_sas_base.c +++ b/drivers/scsi/megaraid/megaraid_sas_base.c @@ -4453,7 +4453,7 @@ static int megasas_init_fw(struct megasas_instance *instance) instance->msixentry[i].entry = i; i = pci_enable_msix_range(instance->pdev, instance->msixentry, 1, instance->msix_vectors); - if (i) + if (i > 0) instance->msix_vectors = i; else instance->msix_vectors = 0; -- cgit v1.2.3 From dc6311dd2a8fb9a65c1b5b94abec97bdd94cfae6 Mon Sep 17 00:00:00 2001 From: Maurizio Lombardi Date: Fri, 7 Nov 2014 12:55:39 +0100 Subject: bnx2fc: fix tgt spinlock locking bnx2fc_queuecommand(): when allocating a new io_req, the tgt_lock spinlock must be locked before calling bnx2fc_cmd_alloc(). The spinlock should also be kept locked until bnx2fc_post_io_req() has been completed. If not, a kernel thread may call bnx2fc_process_cq_compl() that extracts the newly allocated io_req from hba->cmd_mgr->cmds and destroys it while it is still being used by bnx2fc_post_io_req(). BUG: unable to handle kernel NULL pointer dereference at 000000000000004c IP: [] bnx2fc_init_task+0x6a/0x230 [bnx2fc] PGD 0 Oops: 0000 [#1] SMP last sysfs file: /sys/devices/pci0000:00/0000:00:02.0/0000:04:00.3/net/eth3/type CPU 33 Modules linked in: autofs4 target_core_iblock target_core_file target_core_pscsi target_core_mod configfs bnx2fc cnic uio fcoe libfcoe libfc scsi_transport_fc 8021q garp scsi_tgt stp llc cpufreq_ondemand freq_table pcc_cpufreq ipt_REJECT nf_conntrack_ipv4 nf_defrag_ipv4 iptable_filter ip_tables ip6t_REJECT nf_conntrack_ipv6 nf_defrag_ipv6 xt_state nf_conntrack ip6table_filter ip6_tables ipv6 power_meter microcode iTCO_wdt iTCO_vendor_support hpilo hpwdt sg bnx2x libcrc32c mdio serio_raw lpc_ich mfd_core shpchp ext4 jbd2 mbcache sd_mod crc_t10dif hpsa video output dm_mirror dm_region_hash dm_log dm_mod [last unloaded: scsi_wait_scan] Pid: 7355, comm: bnx2fc_thread/3 Not tainted 2.6.32-431.el6.x86_64 #1 HP ProLiant BL460c Gen8 RIP: 0010:[] [] bnx2fc_init_task+0x6a/0x230 [bnx2fc] RSP: 0018:ffff8820b0da3b68 EFLAGS: 00010246 RAX: 0000000000000000 RBX: ffff882003801080 RCX: 0000000000000000 RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff882003801100 RBP: ffff8820b0da3bc8 R08: ffffffff8160d4e8 R09: 0000000000000040 R10: 0000000000000000 R11: 0000000000000000 R12: ffff88400e600e00 R13: ffff8840108fbe40 R14: ffff88200ffe5400 R15: 0000000000000000 FS: 0000000000000000(0000) GS:ffff8820b0da0000(0000) knlGS:0000000000000000 CS: 0010 DS: 0018 ES: 0018 CR0: 000000008005003b CR2: 000000000000004c CR3: 0000002010b67000 CR4: 00000000001407e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 Process bnx2fc_thread/3 (pid: 7355, threadinfo ffff88401f940000, task ffff884012f5f540) Stack: ffff8820b0da3bc8 ffffffff81527303 ffff884000000020 ffff8820b0da3bd8 ffff8820b0da3b98 000000028138931a ffff88400f506078 ffff88400e600e00 ffff88200ffe5400 ffff88200ffe5590 0000000000000021 0000000000000002 Call Trace: [] ? printk+0x41/0x46 [] bnx2fc_post_io_req+0x11c/0x440 [bnx2fc] [] ? cpumask_next_and+0x29/0x50 [] ? scsi_done+0x0/0x60 [] bnx2fc_queuecommand+0x117/0x140 [bnx2fc] [] scsi_dispatch_cmd+0xe5/0x310 [] scsi_request_fn+0x5ee/0x7a0 [] __blk_run_queue+0x31/0x40 [] blk_run_queue+0x30/0x50 [] scsi_run_queue+0xc6/0x270 [] ? elv_requeue_request+0x52/0xa0 [] scsi_requeue_command+0x90/0xb0 [] scsi_io_completion+0x154/0x6c0 [] scsi_finish_command+0xc2/0x130 [] scsi_softirq_done+0x145/0x170 [] blk_done_softirq+0x85/0xa0 [] __do_softirq+0xc1/0x1e0 [] ? call_softirq+0x1c/0x30 [] call_softirq+0x1c/0x30 [] ? do_softirq+0x65/0xa0 [] local_bh_enable_ip+0x9a/0xb0 [] _spin_unlock_bh+0x1b/0x20 [] bnx2fc_process_cq_compl+0x257/0x2b0 [bnx2fc] [] bnx2fc_percpu_io_thread+0xea/0x160 [bnx2fc] [] ? bnx2fc_percpu_io_thread+0x0/0x160 [bnx2fc] [] kthread+0x96/0xa0 [] child_rip+0xa/0x20 [] ? kthread+0x0/0xa0 [] ? child_rip+0x0/0x20 Code: 89 df 45 8b 7e 30 0f 85 75 01 00 00 89 d1 31 c0 c1 e9 03 83 e2 04 89 c9 f3 48 ab 74 06 c7 07 00 00 00 00 49 89 9c 24 88 01 00 00 <83> 7e 4c 01 b8 01 00 00 00 0f 84 e7 00 00 00 89 c2 0a 53 38 41 RIP [] bnx2fc_init_task+0x6a/0x230 [bnx2fc] RSP CR2: 000000000000004c Signed-off-by: Maurizio Lombardi Acked-by: Chad Dupuis Signed-off-by: Christoph Hellwig --- drivers/scsi/bnx2fc/bnx2fc_els.c | 2 -- drivers/scsi/bnx2fc/bnx2fc_io.c | 19 ++++++++++--------- 2 files changed, 10 insertions(+), 11 deletions(-) diff --git a/drivers/scsi/bnx2fc/bnx2fc_els.c b/drivers/scsi/bnx2fc/bnx2fc_els.c index ca75c7ca2559..ef355c13ccc4 100644 --- a/drivers/scsi/bnx2fc/bnx2fc_els.c +++ b/drivers/scsi/bnx2fc/bnx2fc_els.c @@ -480,9 +480,7 @@ void bnx2fc_rec_compl(struct bnx2fc_els_cb_arg *cb_arg) bnx2fc_initiate_cleanup(orig_io_req); /* Post a new IO req with the same sc_cmd */ BNX2FC_IO_DBG(rec_req, "Post IO request again\n"); - spin_unlock_bh(&tgt->tgt_lock); rc = bnx2fc_post_io_req(tgt, new_io_req); - spin_lock_bh(&tgt->tgt_lock); if (!rc) goto free_frame; BNX2FC_IO_DBG(rec_req, "REC: io post err\n"); diff --git a/drivers/scsi/bnx2fc/bnx2fc_io.c b/drivers/scsi/bnx2fc/bnx2fc_io.c index 0679782d9d15..5b99844ef6bf 100644 --- a/drivers/scsi/bnx2fc/bnx2fc_io.c +++ b/drivers/scsi/bnx2fc/bnx2fc_io.c @@ -1894,18 +1894,24 @@ int bnx2fc_queuecommand(struct Scsi_Host *host, goto exit_qcmd; } } + + spin_lock_bh(&tgt->tgt_lock); + io_req = bnx2fc_cmd_alloc(tgt); if (!io_req) { rc = SCSI_MLQUEUE_HOST_BUSY; - goto exit_qcmd; + goto exit_qcmd_tgtlock; } io_req->sc_cmd = sc_cmd; if (bnx2fc_post_io_req(tgt, io_req)) { printk(KERN_ERR PFX "Unable to post io_req\n"); rc = SCSI_MLQUEUE_HOST_BUSY; - goto exit_qcmd; + goto exit_qcmd_tgtlock; } + +exit_qcmd_tgtlock: + spin_unlock_bh(&tgt->tgt_lock); exit_qcmd: return rc; } @@ -2020,6 +2026,8 @@ int bnx2fc_post_io_req(struct bnx2fc_rport *tgt, int task_idx, index; u16 xid; + /* bnx2fc_post_io_req() is called with the tgt_lock held */ + /* Initialize rest of io_req fields */ io_req->cmd_type = BNX2FC_SCSI_CMD; io_req->port = port; @@ -2047,9 +2055,7 @@ int bnx2fc_post_io_req(struct bnx2fc_rport *tgt, /* Build buffer descriptor list for firmware from sg list */ if (bnx2fc_build_bd_list_from_sg(io_req)) { printk(KERN_ERR PFX "BD list creation failed\n"); - spin_lock_bh(&tgt->tgt_lock); kref_put(&io_req->refcount, bnx2fc_cmd_release); - spin_unlock_bh(&tgt->tgt_lock); return -EAGAIN; } @@ -2061,19 +2067,15 @@ int bnx2fc_post_io_req(struct bnx2fc_rport *tgt, task = &(task_page[index]); bnx2fc_init_task(io_req, task); - spin_lock_bh(&tgt->tgt_lock); - if (tgt->flush_in_prog) { printk(KERN_ERR PFX "Flush in progress..Host Busy\n"); kref_put(&io_req->refcount, bnx2fc_cmd_release); - spin_unlock_bh(&tgt->tgt_lock); return -EAGAIN; } if (!test_bit(BNX2FC_FLAG_SESSION_READY, &tgt->flags)) { printk(KERN_ERR PFX "Session not ready...post_io\n"); kref_put(&io_req->refcount, bnx2fc_cmd_release); - spin_unlock_bh(&tgt->tgt_lock); return -EAGAIN; } @@ -2091,6 +2093,5 @@ int bnx2fc_post_io_req(struct bnx2fc_rport *tgt, /* Ring doorbell */ bnx2fc_ring_doorbell(tgt); - spin_unlock_bh(&tgt->tgt_lock); return 0; } -- cgit v1.2.3