From 441b7195e2812f2d7a9dafe02f052f642957bd8f Mon Sep 17 00:00:00 2001 From: Erwan Velu Date: Fri, 1 Mar 2019 17:08:06 +0100 Subject: scsi: smartpqi: Reporting 'logical unit failure' When the HARDWARE_ERROR/0x3e/0x1 case is triggered, the logical volume is offlined. When reading the kernel log, the reason why the device got offlined isn't reported to the user. This situation makes it difficult for admins to root cause. Log a message when this condition occurs. [mkp: tweaked commit message] Signed-off-by: Erwan Velu Acked-by: Don Brace Signed-off-by: Martin K. Petersen --- drivers/scsi/smartpqi/smartpqi_init.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/scsi/smartpqi/smartpqi_init.c b/drivers/scsi/smartpqi/smartpqi_init.c index f6eaea2eadd1..c7b023683575 100644 --- a/drivers/scsi/smartpqi/smartpqi_init.c +++ b/drivers/scsi/smartpqi/smartpqi_init.c @@ -2764,6 +2764,12 @@ static void pqi_process_raid_io_error(struct pqi_io_request *io_request) sshdr.sense_key == HARDWARE_ERROR && sshdr.asc == 0x3e && sshdr.ascq == 0x1) { + struct pqi_ctrl_info *ctrl_info = shost_to_hba(scmd->device->host); + struct pqi_scsi_dev *device = scmd->device->hostdata; + + if (printk_ratelimit()) + scmd_printk(KERN_ERR, scmd, "received 'logical unit failure' from controller for scsi %d:%d:%d:%d\n", + ctrl_info->scsi_host->host_no, device->bus, device->target, device->lun); pqi_take_device_offline(scmd->device, "RAID"); host_byte = DID_NO_CONNECT; } -- cgit v1.2.3-59-g8ed1b From 3722e6a52174d7c3a00e6f5efd006ca093f346c1 Mon Sep 17 00:00:00 2001 From: Felipe Franciosi Date: Wed, 27 Feb 2019 16:10:34 +0000 Subject: scsi: virtio_scsi: don't send sc payload with tmfs The virtio scsi spec defines struct virtio_scsi_ctrl_tmf as a set of device-readable records and a single device-writable response entry: struct virtio_scsi_ctrl_tmf { // Device-readable part le32 type; le32 subtype; u8 lun[8]; le64 id; // Device-writable part u8 response; } The above should be organised as two descriptor entries (or potentially more if using VIRTIO_F_ANY_LAYOUT), but without any extra data after "le64 id" or after "u8 response". The Linux driver doesn't respect that, with virtscsi_abort() and virtscsi_device_reset() setting cmd->sc before calling virtscsi_tmf(). It results in the original scsi command payload (or writable buffers) added to the tmf. This fixes the problem by leaving cmd->sc zeroed out, which makes virtscsi_kick_cmd() add the tmf to the control vq without any payload. Cc: stable@vger.kernel.org Signed-off-by: Felipe Franciosi Reviewed-by: Paolo Bonzini Signed-off-by: Martin K. Petersen --- drivers/scsi/virtio_scsi.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/scsi/virtio_scsi.c b/drivers/scsi/virtio_scsi.c index 1a6f150cd2d8..8af01777d09c 100644 --- a/drivers/scsi/virtio_scsi.c +++ b/drivers/scsi/virtio_scsi.c @@ -586,7 +586,6 @@ static int virtscsi_device_reset(struct scsi_cmnd *sc) return FAILED; memset(cmd, 0, sizeof(*cmd)); - cmd->sc = sc; cmd->req.tmf = (struct virtio_scsi_ctrl_tmf_req){ .type = VIRTIO_SCSI_T_TMF, .subtype = cpu_to_virtio32(vscsi->vdev, @@ -645,7 +644,6 @@ static int virtscsi_abort(struct scsi_cmnd *sc) return FAILED; memset(cmd, 0, sizeof(*cmd)); - cmd->sc = sc; cmd->req.tmf = (struct virtio_scsi_ctrl_tmf_req){ .type = VIRTIO_SCSI_T_TMF, .subtype = VIRTIO_SCSI_T_TMF_ABORT_TASK, -- cgit v1.2.3-59-g8ed1b From 2ff717cdc693d532e4560f43858443aea7d88197 Mon Sep 17 00:00:00 2001 From: Cathy Avery Date: Thu, 28 Feb 2019 14:28:24 -0500 Subject: scsi: target: tcmu: wait for nl reply only if there are listeners or during an add genlmsg_multicast_allns now returns the correct statuses when a message is sent to a listener. However in the case of adding a device we want to wait for the listener otherwise we may miss the the device during startup. Signed-off-by: Cathy Avery Acked-by: Mike Christie Signed-off-by: Martin K. Petersen --- drivers/target/target_core_user.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/drivers/target/target_core_user.c b/drivers/target/target_core_user.c index 1e6d24943565..396c2fa1ec22 100644 --- a/drivers/target/target_core_user.c +++ b/drivers/target/target_core_user.c @@ -1764,11 +1764,12 @@ static int tcmu_netlink_event_send(struct tcmu_dev *udev, ret = genlmsg_multicast_allns(&tcmu_genl_family, skb, 0, TCMU_MCGRP_CONFIG, GFP_KERNEL); - /* We don't care if no one is listening */ - if (ret == -ESRCH) - ret = 0; - if (!ret) - ret = tcmu_wait_genl_cmd_reply(udev); + + /* Wait during an add as the listener may not be up yet */ + if (ret == 0 || + (ret == -ESRCH && cmd == TCMU_CMD_ADDED_DEVICE)) + return tcmu_wait_genl_cmd_reply(udev); + return ret; } -- cgit v1.2.3-59-g8ed1b From 31b6a05f86e690e1818116fd23c3be915cc9d9ed Mon Sep 17 00:00:00 2001 From: Steve Sistare Date: Fri, 1 Mar 2019 06:46:28 -0800 Subject: scsi: megaraid_sas: reduce module load time megaraid_sas takes 1+ seconds to load while waiting for firmware: [2.822603] megaraid_sas 0000:03:00.0: Waiting for FW to come to ready state [3.871003] megaraid_sas 0000:03:00.0: FW now in Ready state This is due to the following loop in megasas_transition_to_ready(), which waits a minimum of 1 second, even though the FW becomes ready in tens of millisecs: /* * The cur_state should not last for more than max_wait secs */ for (i = 0; i < max_wait; i++) { ... msleep(1000); ... dev_info(&instance->pdev->dev, "FW now in Ready state\n"); This is a regression, caused by a change of the msleep granularity from 1 to 1000 due to concern about waiting too long on systems with coarse jiffies. To fix, increase iterations and use msleep(20), which results in: [2.670627] megaraid_sas 0000:03:00.0: Waiting for FW to come to ready state [2.739386] megaraid_sas 0000:03:00.0: FW now in Ready state Fixes: fb2f3e96d80f ("scsi: megaraid_sas: Fix msleep granularity") Signed-off-by: Steve Sistare Acked-by: Sumit Saxena Signed-off-by: Martin K. Petersen --- drivers/scsi/megaraid/megaraid_sas_base.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/scsi/megaraid/megaraid_sas_base.c b/drivers/scsi/megaraid/megaraid_sas_base.c index 8a2ee4cf26b0..387fb930f799 100644 --- a/drivers/scsi/megaraid/megaraid_sas_base.c +++ b/drivers/scsi/megaraid/megaraid_sas_base.c @@ -3924,12 +3924,12 @@ megasas_transition_to_ready(struct megasas_instance *instance, int ocr) /* * The cur_state should not last for more than max_wait secs */ - for (i = 0; i < max_wait; i++) { + for (i = 0; i < max_wait * 50; i++) { curr_abs_state = instance->instancet-> read_fw_status_reg(instance); if (abs_state == curr_abs_state) { - msleep(1000); + msleep(20); } else break; } -- cgit v1.2.3-59-g8ed1b From db0f166e9a37215b15d5d732c98fa15219adccf0 Mon Sep 17 00:00:00 2001 From: Bill Kuzeja Date: Mon, 4 Mar 2019 08:25:46 -0500 Subject: scsi: qla2xxx: Fix panic in qla_dfs_tgt_counters_show When trying to display tgt_counters in the debugfs, a panic can result. There is no null check for qpair after it is assigned in the for-loop. Unless vha->hw->queue_pair_map array is completely filled with entries, the system will panic dereferencing a null pointer. Signed-off-by: Bill Kuzeja Acked-by: Himanshu Madhani Signed-off-by: Martin K. Petersen --- drivers/scsi/qla2xxx/qla_dfs.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/scsi/qla2xxx/qla_dfs.c b/drivers/scsi/qla2xxx/qla_dfs.c index ead17288e2a7..5819a45ac5ef 100644 --- a/drivers/scsi/qla2xxx/qla_dfs.c +++ b/drivers/scsi/qla2xxx/qla_dfs.c @@ -193,6 +193,8 @@ qla_dfs_tgt_counters_show(struct seq_file *s, void *unused) for (i = 0; i < vha->hw->max_qpairs; i++) { qpair = vha->hw->queue_pair_map[i]; + if (!qpair) + continue; qla_core_sbt_cmd += qpair->tgt_counters.qla_core_sbt_cmd; core_qla_que_buf += qpair->tgt_counters.core_qla_que_buf; qla_core_ret_ctio += qpair->tgt_counters.qla_core_ret_ctio; -- cgit v1.2.3-59-g8ed1b From 176eb927744201ba95ca9a85c2f9af457559374f Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Mon, 4 Mar 2019 20:39:11 +0100 Subject: scsi: ufs: hisi: fix ufs_hba_variant_ops passing Without CONFIG_OF, the of_match_node() helper does not evaluate its argument, and the compiler warns about the unused variable: drivers/scsi/ufs/ufs-hisi.c: In function 'ufs_hisi_probe': drivers/scsi/ufs/ufs-hisi.c:673:17: error: unused variable 'dev' [-Werror=unused-variable] Rework this code to pass the data directly, and while we're at it, correctly handle the const pointers. Fixes: 653fcb07d95e ("scsi: ufs: Add HI3670 SoC UFS driver support") Signed-off-by: Arnd Bergmann Reviewed-by: Avri Altman Signed-off-by: Martin K. Petersen --- drivers/scsi/ufs/ufs-hisi.c | 11 ++++------- drivers/scsi/ufs/ufshcd-pltfrm.c | 2 +- drivers/scsi/ufs/ufshcd-pltfrm.h | 2 +- drivers/scsi/ufs/ufshcd.h | 2 +- 4 files changed, 7 insertions(+), 10 deletions(-) diff --git a/drivers/scsi/ufs/ufs-hisi.c b/drivers/scsi/ufs/ufs-hisi.c index f2d3df357a97..0e855b5afe82 100644 --- a/drivers/scsi/ufs/ufs-hisi.c +++ b/drivers/scsi/ufs/ufs-hisi.c @@ -640,7 +640,7 @@ static int ufs_hi3670_init(struct ufs_hba *hba) return 0; } -static struct ufs_hba_variant_ops ufs_hba_hi3660_vops = { +static const struct ufs_hba_variant_ops ufs_hba_hi3660_vops = { .name = "hi3660", .init = ufs_hi3660_init, .link_startup_notify = ufs_hisi_link_startup_notify, @@ -649,7 +649,7 @@ static struct ufs_hba_variant_ops ufs_hba_hi3660_vops = { .resume = ufs_hisi_resume, }; -static struct ufs_hba_variant_ops ufs_hba_hi3670_vops = { +static const struct ufs_hba_variant_ops ufs_hba_hi3670_vops = { .name = "hi3670", .init = ufs_hi3670_init, .link_startup_notify = ufs_hisi_link_startup_notify, @@ -669,13 +669,10 @@ MODULE_DEVICE_TABLE(of, ufs_hisi_of_match); static int ufs_hisi_probe(struct platform_device *pdev) { const struct of_device_id *of_id; - struct ufs_hba_variant_ops *vops; - struct device *dev = &pdev->dev; - of_id = of_match_node(ufs_hisi_of_match, dev->of_node); - vops = (struct ufs_hba_variant_ops *)of_id->data; + of_id = of_match_node(ufs_hisi_of_match, pdev->dev.of_node); - return ufshcd_pltfrm_init(pdev, vops); + return ufshcd_pltfrm_init(pdev, of_id->data); } static int ufs_hisi_remove(struct platform_device *pdev) diff --git a/drivers/scsi/ufs/ufshcd-pltfrm.c b/drivers/scsi/ufs/ufshcd-pltfrm.c index 895a9b5ac989..27213676329c 100644 --- a/drivers/scsi/ufs/ufshcd-pltfrm.c +++ b/drivers/scsi/ufs/ufshcd-pltfrm.c @@ -297,7 +297,7 @@ static void ufshcd_init_lanes_per_dir(struct ufs_hba *hba) * Returns 0 on success, non-zero value on failure */ int ufshcd_pltfrm_init(struct platform_device *pdev, - struct ufs_hba_variant_ops *vops) + const struct ufs_hba_variant_ops *vops) { struct ufs_hba *hba; void __iomem *mmio_base; diff --git a/drivers/scsi/ufs/ufshcd-pltfrm.h b/drivers/scsi/ufs/ufshcd-pltfrm.h index df64c4180340..1f29e1fd6d52 100644 --- a/drivers/scsi/ufs/ufshcd-pltfrm.h +++ b/drivers/scsi/ufs/ufshcd-pltfrm.h @@ -17,7 +17,7 @@ #include "ufshcd.h" int ufshcd_pltfrm_init(struct platform_device *pdev, - struct ufs_hba_variant_ops *vops); + const struct ufs_hba_variant_ops *vops); void ufshcd_pltfrm_shutdown(struct platform_device *pdev); #ifdef CONFIG_PM diff --git a/drivers/scsi/ufs/ufshcd.h b/drivers/scsi/ufs/ufshcd.h index 69ba7445d2b3..ecfa898b9ccc 100644 --- a/drivers/scsi/ufs/ufshcd.h +++ b/drivers/scsi/ufs/ufshcd.h @@ -546,7 +546,7 @@ struct ufs_hba { int nutrs; int nutmrs; u32 ufs_version; - struct ufs_hba_variant_ops *vops; + const struct ufs_hba_variant_ops *vops; void *priv; unsigned int irq; bool is_irq_enabled; -- cgit v1.2.3-59-g8ed1b From cda7fa18653c47865d2a048d5bfa211d464b23e1 Mon Sep 17 00:00:00 2001 From: James Smart Date: Mon, 4 Mar 2019 15:15:43 -0800 Subject: scsi: lpfc: Correct __lpfc_sli_issue_iocb_s4 lockdep check The outer routine lpfc_sli_issue_iocb(), which decomposes into the SLI3 (s3) or SLI4 (s4) subroutines takes out the locks. For s3, it takes out the hbalock. For s4, it takes out the ring_lock. The lockdep check in the s3 and s4 subroutines both check hbalock, which is incorrect for s4. Revise the s4 subroutine to lockdep check the ring_lock. Reported-by: Bart Van Assche Signed-off-by: Dick Kennedy Signed-off-by: James Smart Signed-off-by: Martin K. Petersen --- drivers/scsi/lpfc/lpfc_sli.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c index 62957d682d3f..f29f7cbb959d 100644 --- a/drivers/scsi/lpfc/lpfc_sli.c +++ b/drivers/scsi/lpfc/lpfc_sli.c @@ -9881,7 +9881,7 @@ __lpfc_sli_issue_iocb_s4(struct lpfc_hba *phba, uint32_t ring_number, * The WQE can be either 64 or 128 bytes, */ - lockdep_assert_held(&phba->hbalock); + lockdep_assert_held(&pring->ring_lock); if (piocb->sli4_xritag == NO_XRI) { if (piocb->iocb.ulpCommand == CMD_ABORT_XRI_CN || -- cgit v1.2.3-59-g8ed1b From 1ffdd2c0440dfda533ca9477a89550c9f48f4b35 Mon Sep 17 00:00:00 2001 From: James Smart Date: Mon, 4 Mar 2019 15:27:51 -0800 Subject: scsi: lpfc: resolve static checker warning in lpfc_sli4_hba_unset The patch that replaced io channels for hdw_queues now reports the following static checker warning: drivers/scsi/lpfc/lpfc_init.c:11136 lpfc_sli4_hba_unset() error: we previously assumed 'phba->pport' could be null (see line 11074) Resolve by adding a pport NULL check. [mkp: tag tweak] Fixes: cdb42becdd40 ("scsi: lpfc: Replace io_channels for nvme and fcp with general hdw_queues per cpu"_ Reported-by: Dan Carpenter Signed-off-by: Dick Kennedy Signed-off-by: James Smart Signed-off-by: Martin K. Petersen --- drivers/scsi/lpfc/lpfc_init.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/scsi/lpfc/lpfc_init.c b/drivers/scsi/lpfc/lpfc_init.c index 3eb04c3be300..cd09b2120abd 100644 --- a/drivers/scsi/lpfc/lpfc_init.c +++ b/drivers/scsi/lpfc/lpfc_init.c @@ -11132,7 +11132,8 @@ lpfc_sli4_hba_unset(struct lpfc_hba *phba) lpfc_sli4_ras_dma_free(phba); /* Stop the SLI4 device port */ - phba->pport->work_port_events = 0; + if (phba->pport) + phba->pport->work_port_events = 0; } /** -- cgit v1.2.3-59-g8ed1b From 038d710fca5bb149d3af2e0b71f1284f8430a979 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Mon, 4 Mar 2019 20:39:10 +0100 Subject: scsi: qla2xxx: avoid printf format warning Depending on the target architecture and configuration, both phys_addr_t and dma_addr_t may be smaller than 'long long', so we get a warning when printing either of them using the %llx format string: drivers/scsi/qla2xxx/qla_iocb.c: In function 'qla24xx_walk_and_build_prot_sglist': drivers/scsi/qla2xxx/qla_iocb.c:1140:46: error: format '%llx' expects argument of type 'long long unsigned int', but argument 6 has type 'dma_addr_t' {aka 'unsigned int'} [-Werror=format=] "%s: page boundary crossing (phys=%llx len=%x)\n", ~~~^ %x __func__, sle_phys, sg->length); ~~~~~~~~ drivers/scsi/qla2xxx/qla_iocb.c:1180:29: error: format '%llx' expects argument of type 'long long unsigned int', but argument 7 has type 'dma_addr_t' {aka 'unsigned int'} [-Werror=format=] "%s: sg[%x] (phys=%llx sglen=%x) ldma_sg_len: %x dif_bundl_len: %x ldma_needed: %x\n", ~~~^ There are special %pad and %pap format strings in Linux that we could use here, but since the driver already does 64-bit arithmetic on the values, using a plain 'u64' seems more consistent here. Note: A possible related issue may be that the driver possibly checks the wrong kind of overflow: when an IOMMU is in use, buffers that cross a 32-bit boundary in physical addresses would still be mapped into dma addresses within the low 4GB space, so I suspect that we actually want to check sg_dma_address() instead of sg_phys() here. Fixes: 50b812755e97 ("scsi: qla2xxx: Fix DMA error when the DIF sg buffer crosses 4GB boundary") Signed-off-by: Arnd Bergmann Acked-by: Himanshu Madhani Signed-off-by: Martin K. Petersen --- drivers/scsi/qla2xxx/qla_iocb.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/scsi/qla2xxx/qla_iocb.c b/drivers/scsi/qla2xxx/qla_iocb.c index 63f8e3c19841..456a41d2e2c6 100644 --- a/drivers/scsi/qla2xxx/qla_iocb.c +++ b/drivers/scsi/qla2xxx/qla_iocb.c @@ -1132,7 +1132,7 @@ qla24xx_walk_and_build_prot_sglist(struct qla_hw_data *ha, srb_t *sp, /* if initiator doing write or target doing read */ if (direction_to_device) { for_each_sg(sgl, sg, tot_dsds, i) { - dma_addr_t sle_phys = sg_phys(sg); + u64 sle_phys = sg_phys(sg); /* If SGE addr + len flips bits in upper 32-bits */ if (MSD(sle_phys + sg->length) ^ MSD(sle_phys)) { @@ -1178,7 +1178,7 @@ qla24xx_walk_and_build_prot_sglist(struct qla_hw_data *ha, srb_t *sp, ql_dbg(ql_dbg_tgt + ql_dbg_verbose, vha, 0xe023, "%s: sg[%x] (phys=%llx sglen=%x) ldma_sg_len: %x dif_bundl_len: %x ldma_needed: %x\n", - __func__, i, sg_phys(sg), sglen, ldma_sg_len, + __func__, i, (u64)sg_phys(sg), sglen, ldma_sg_len, difctx->dif_bundl_len, ldma_needed); while (sglen) { -- cgit v1.2.3-59-g8ed1b From 08b11eaccfcf86a3bac6755625d933ac15ccc27a Mon Sep 17 00:00:00 2001 From: Vasily Averin Date: Thu, 21 Feb 2019 18:23:17 +0300 Subject: scsi: libiscsi: fall back to sendmsg for slab pages In "XFS over network block device" scenario XFS can create IO requests with slab-based XFS metadata. During processing such requests tcp_sendpage() can merge skb fragments with neighbour slab objects. If receiving side is located on the same host tcp_recvmsg() can trigger BUG_ON in hardening check and crash the host with following message: usercopy: kernel memory exposure attempt detected from XXXXXXXX (kmalloc-512) (1024 bytes) This patch redirect such requests from sednpage to sendmsg path. The problem is similar to one described in recent commit 7e241f647dc7 ("libceph: fall back to sendmsg for slab pages") Signed-off-by: Vasily Averin Acked-by: Chris Leech Signed-off-by: Martin K. Petersen --- drivers/scsi/libiscsi_tcp.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/scsi/libiscsi_tcp.c b/drivers/scsi/libiscsi_tcp.c index 9923e9e3b884..c3fe3f3a78f5 100644 --- a/drivers/scsi/libiscsi_tcp.c +++ b/drivers/scsi/libiscsi_tcp.c @@ -129,12 +129,17 @@ static void iscsi_tcp_segment_map(struct iscsi_segment *segment, int recv) BUG_ON(sg->length == 0); /* + * We always map for the recv path. + * * If the page count is greater than one it is ok to send * to the network layer's zero copy send path. If not we - * have to go the slow sendmsg path. We always map for the - * recv path. + * have to go the slow sendmsg path. + * + * Same goes for slab pages: skb_can_coalesce() allows + * coalescing neighboring slab objects into a single frag which + * triggers one of hardened usercopy checks. */ - if (page_count(sg_page(sg)) >= 1 && !recv) + if (!recv && page_count(sg_page(sg)) >= 1 && !PageSlab(sg_page(sg))) return; if (recv) { -- cgit v1.2.3-59-g8ed1b From 98effe4746c6e64b76703bbf5fd64e304e3e5396 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Mon, 4 Mar 2019 11:44:34 +0200 Subject: scsi: target: tcmu: Switch to bitmap_zalloc() Switch to bitmap_zalloc() to show clearly what we are allocating. Besides that it returns pointer of bitmap type instead of opaque void *. Signed-off-by: Andy Shevchenko Reviewed-by: Bart Van Assche Acked-by: Mike Christie Signed-off-by: Martin K. Petersen --- drivers/target/target_core_user.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/drivers/target/target_core_user.c b/drivers/target/target_core_user.c index 396c2fa1ec22..4eeeaaf3796b 100644 --- a/drivers/target/target_core_user.c +++ b/drivers/target/target_core_user.c @@ -1633,7 +1633,7 @@ static void tcmu_dev_kref_release(struct kref *kref) WARN_ON(!all_expired); tcmu_blocks_release(&udev->data_blocks, 0, udev->dbi_max + 1); - kfree(udev->data_bitmap); + bitmap_free(udev->data_bitmap); mutex_unlock(&udev->cmdr_lock); call_rcu(&dev->rcu_head, tcmu_dev_call_rcu); @@ -1841,9 +1841,7 @@ static int tcmu_configure_device(struct se_device *dev) info = &udev->uio_info; mutex_lock(&udev->cmdr_lock); - udev->data_bitmap = kcalloc(BITS_TO_LONGS(udev->max_blocks), - sizeof(unsigned long), - GFP_KERNEL); + udev->data_bitmap = bitmap_zalloc(udev->max_blocks, GFP_KERNEL); mutex_unlock(&udev->cmdr_lock); if (!udev->data_bitmap) { ret = -ENOMEM; @@ -1930,7 +1928,7 @@ err_register: vfree(udev->mb_addr); udev->mb_addr = NULL; err_vzalloc: - kfree(udev->data_bitmap); + bitmap_free(udev->data_bitmap); udev->data_bitmap = NULL; err_bitmap_alloc: kfree(info->name); -- cgit v1.2.3-59-g8ed1b From 352b205a3bac6bb68f27228768a5c12294e650bf Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Mon, 4 Mar 2019 20:39:08 +0100 Subject: scsi: lpfc: fix unused variable warning The newly introduced 'cpu' variable is only used inside of an optional block, so we get a warning without CONFIG_SCSI_LPFC_DEBUG_FS: drivers/scsi/lpfc/lpfc_nvme.c: In function 'lpfc_nvme_io_cmd_wqe_cmpl': drivers/scsi/lpfc/lpfc_nvme.c:968:30: error: unused variable 'cpu' [-Werror=unused-variable] uint32_t code, status, idx, cpu; Move the declaration into the same block to avoid the warning. Fixes: 63df6d637e33 ("scsi: lpfc: Adapt cpucheck debugfs logic to Hardware Queues") Signed-off-by: Arnd Bergmann Signed-off-by: Martin K. Petersen --- drivers/scsi/lpfc/lpfc_nvme.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/scsi/lpfc/lpfc_nvme.c b/drivers/scsi/lpfc/lpfc_nvme.c index e73895674f39..4aa5c004785e 100644 --- a/drivers/scsi/lpfc/lpfc_nvme.c +++ b/drivers/scsi/lpfc/lpfc_nvme.c @@ -964,7 +964,7 @@ lpfc_nvme_io_cmd_wqe_cmpl(struct lpfc_hba *phba, struct lpfc_iocbq *pwqeIn, struct lpfc_nodelist *ndlp; struct lpfc_nvme_fcpreq_priv *freqpriv; struct lpfc_nvme_lport *lport; - uint32_t code, status, idx, cpu; + uint32_t code, status, idx; uint16_t cid, sqhd, data; uint32_t *ptr; @@ -1137,6 +1137,7 @@ out_err: lpfc_nvme_ktime(phba, lpfc_ncmd); } if (phba->cpucheck_on & LPFC_CHECK_NVME_IO) { + uint32_t cpu; idx = lpfc_ncmd->cur_iocbq.hba_wqidx; cpu = smp_processor_id(); if (cpu < LPFC_CHECK_CPU_CNT) { -- cgit v1.2.3-59-g8ed1b From f996861be108f6152d60d7357301697c3309c77f Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Mon, 4 Mar 2019 20:39:09 +0100 Subject: scsi: lpfc: fix 32-bit format string warning On 32-bit architectures, we see a warning when %ld is used to print a size_t: In file included from drivers/scsi/lpfc/lpfc_init.c:62: drivers/scsi/lpfc/lpfc_init.c: In function 'lpfc_new_io_buf': drivers/scsi/lpfc/lpfc_logmsg.h:62:45: error: format '%ld' expects argument of type 'long int', but argument 5 has type 'unsigned int' [-Werror=format=] This is harmless, but portable code should just use %zd to avoid the warning. Fixes: 0794d601d174 ("scsi: lpfc: Implement common IO buffers between NVME and SCSI") Signed-off-by: Arnd Bergmann Signed-off-by: Martin K. Petersen --- drivers/scsi/lpfc/lpfc_init.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/scsi/lpfc/lpfc_init.c b/drivers/scsi/lpfc/lpfc_init.c index cd09b2120abd..707dbe7a8d15 100644 --- a/drivers/scsi/lpfc/lpfc_init.c +++ b/drivers/scsi/lpfc/lpfc_init.c @@ -4090,7 +4090,7 @@ lpfc_new_io_buf(struct lpfc_hba *phba, int num_to_alloc) /* Sanity check to ensure our sizing is right for both SCSI and NVME */ if (sizeof(struct lpfc_io_buf) > LPFC_COMMON_IO_BUF_SZ) { lpfc_printf_log(phba, KERN_ERR, LOG_FCP, - "6426 Common buffer size %ld exceeds %d\n", + "6426 Common buffer size %zd exceeds %d\n", sizeof(struct lpfc_io_buf), LPFC_COMMON_IO_BUF_SZ); return 0; -- cgit v1.2.3-59-g8ed1b From b8870ec63676aba1d823f0b36c5f7e9929e57d23 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 20 Feb 2019 08:39:13 +0300 Subject: scsi: qla2xxx: check for kstrtol() failure The error handling was unintentionally left out so it introduces a Smatch static checker warning: drivers/scsi/qla2xxx/qla_attr.c:1655 qla2x00_port_speed_store() error: uninitialized symbol 'type'. Fixes: a7b9ca7fc87a ("scsi: qla2xxx: Add support for setting port speed") Signed-off-by: Dan Carpenter Signed-off-by: Martin K. Petersen --- drivers/scsi/qla2xxx/qla_attr.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/scsi/qla2xxx/qla_attr.c b/drivers/scsi/qla2xxx/qla_attr.c index f8fd482a06e0..41734efe4bf9 100644 --- a/drivers/scsi/qla2xxx/qla_attr.c +++ b/drivers/scsi/qla2xxx/qla_attr.c @@ -1652,6 +1652,8 @@ qla2x00_port_speed_store(struct device *dev, struct device_attribute *attr, } rval = kstrtol(buf, 10, &type); + if (rval) + return rval; speed = type; if (type == 40 || type == 80 || type == 160 || type == 320) { -- cgit v1.2.3-59-g8ed1b From fba770c6682447f1998267f4143595e77be4f112 Mon Sep 17 00:00:00 2001 From: Xiang Chen Date: Thu, 28 Feb 2019 22:50:57 +0800 Subject: scsi: hisi_sas: Change return variable type in phy_up_v3_hw() According to the tool fortify, phy_up_v3_hw() returns signed value, while it should return an unsigned value. So change variable "res" from int to irq_return_t. Signed-off-by: Xiang Chen Signed-off-by: John Garry Signed-off-by: Martin K. Petersen --- drivers/scsi/hisi_sas/hisi_sas_v3_hw.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c index 00738d0673fe..151a102c4cc7 100644 --- a/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c +++ b/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c @@ -1344,7 +1344,8 @@ static void prep_abort_v3_hw(struct hisi_hba *hisi_hba, static irqreturn_t phy_up_v3_hw(int phy_no, struct hisi_hba *hisi_hba) { - int i, res; + int i; + irqreturn_t res; u32 context, port_id, link_rate; struct hisi_sas_phy *phy = &hisi_hba->phy[phy_no]; struct asd_sas_phy *sas_phy = &phy->sas_phy; -- cgit v1.2.3-59-g8ed1b From 4790595723d4b833b18c994973d39f9efb842887 Mon Sep 17 00:00:00 2001 From: Xiang Chen Date: Thu, 28 Feb 2019 22:50:58 +0800 Subject: scsi: hisi_sas: Fix a timeout race of driver internal and SMP IO For internal IO and SMP IO, there is a time-out timer for them. In the timer handler, it checks whether IO is done according to the flag task->task_state_lock. There is an issue which may cause system suspended: internal IO or SMP IO is sent, but at that time because of hardware exception (such as inject 2Bit ECC error), so IO is not completed and also not timeout. But, at that time, the SAS controller reset occurs to recover system. It will release the resource and set the status of IO to be SAS_TASK_STATE_DONE, so when IO timeout, it will never complete the completion of IO and wait for ever. [ 729.123632] Call trace: [ 729.126791] [] __switch_to+0x94/0xa8 [ 729.133106] [] __schedule+0x1e8/0x7fc [ 729.138975] [] schedule+0x34/0x8c [ 729.144401] [] schedule_timeout+0x1d8/0x3cc [ 729.150690] [] wait_for_common+0xdc/0x1a0 [ 729.157101] [] wait_for_completion+0x28/0x34 [ 729.165973] [] hisi_sas_internal_task_abort+0x2a0/0x424 [hisi_sas_test_main] [ 729.176447] [] hisi_sas_abort_task+0x244/0x2d8 [hisi_sas_test_main] [ 729.185258] [] sas_eh_handle_sas_errors+0x1c8/0x7b8 [ 729.192391] [] sas_scsi_recover_host+0x130/0x398 [ 729.199237] [] scsi_error_handler+0x148/0x5c0 [ 729.206009] [] kthread+0x10c/0x138 [ 729.211563] [] ret_from_fork+0x10/0x18 To solve the issue, callback function task_done of those IOs need to be called when on SAS controller reset. Signed-off-by: Xiang Chen Signed-off-by: John Garry Signed-off-by: Martin K. Petersen --- drivers/scsi/hisi_sas/hisi_sas_main.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/scsi/hisi_sas/hisi_sas_main.c b/drivers/scsi/hisi_sas/hisi_sas_main.c index 923296653ed7..dd03dcbd3786 100644 --- a/drivers/scsi/hisi_sas/hisi_sas_main.c +++ b/drivers/scsi/hisi_sas/hisi_sas_main.c @@ -980,7 +980,8 @@ static void hisi_sas_do_release_task(struct hisi_hba *hisi_hba, struct sas_task spin_lock_irqsave(&task->task_state_lock, flags); task->task_state_flags &= ~(SAS_TASK_STATE_PENDING | SAS_TASK_AT_INITIATOR); - task->task_state_flags |= SAS_TASK_STATE_DONE; + if (!slot->is_internal && task->task_proto != SAS_PROTOCOL_SMP) + task->task_state_flags |= SAS_TASK_STATE_DONE; spin_unlock_irqrestore(&task->task_state_lock, flags); } -- cgit v1.2.3-59-g8ed1b From aaeb82323d12f3b61015878e035d74600e1f2c85 Mon Sep 17 00:00:00 2001 From: Xiaofei Tan Date: Thu, 28 Feb 2019 22:50:59 +0800 Subject: scsi: hisi_sas: print PHY RX errors count for later revision of v3 hw The later revision of v3 hw has added an function of interrupt coalesce according to time for PHY RX errors. We set the coalesce time to 1s. Then we print PHY RX errors count when PHY RX errors happen, and don't need to worry that there may be too much log prints. Besides, we use hisi_sas_phy.lock to protect error count value. Because we update them by calling phy_get_events_v3_hw(), which is also used by core driver (for get PHY events function). We relocate phy_get_events_v3_hw() to avoid a further declaration. Signed-off-by: Xiaofei Tan Signed-off-by: John Garry Signed-off-by: Martin K. Petersen --- drivers/scsi/hisi_sas/hisi_sas.h | 1 + drivers/scsi/hisi_sas/hisi_sas_v3_hw.c | 84 ++++++++++++++++++++++++---------- 2 files changed, 60 insertions(+), 25 deletions(-) diff --git a/drivers/scsi/hisi_sas/hisi_sas.h b/drivers/scsi/hisi_sas/hisi_sas.h index 6c87bd34509a..515aee9318a4 100644 --- a/drivers/scsi/hisi_sas/hisi_sas.h +++ b/drivers/scsi/hisi_sas/hisi_sas.h @@ -161,6 +161,7 @@ struct hisi_sas_phy { u8 in_reset; u8 reserved[2]; u32 phy_type; + u32 code_violation_err_count; enum sas_linkrate minimum_linkrate; enum sas_linkrate maximum_linkrate; }; diff --git a/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c index 151a102c4cc7..720721196b12 100644 --- a/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c +++ b/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c @@ -181,6 +181,8 @@ #define CHL_INT1_DMAC_RX_AXI_RD_ERR_OFF 22 #define CHL_INT2 (PORT_BASE + 0x1bc) #define CHL_INT2_SL_IDAF_TOUT_CONF_OFF 0 +#define CHL_INT2_RX_DISP_ERR_OFF 28 +#define CHL_INT2_RX_CODE_ERR_OFF 29 #define CHL_INT2_RX_INVLD_DW_OFF 30 #define CHL_INT2_STP_LINK_TIMEOUT_OFF 31 #define CHL_INT0_MSK (PORT_BASE + 0x1c0) @@ -544,6 +546,8 @@ static void init_reg_v3_hw(struct hisi_hba *hisi_hba) hisi_sas_phy_write32(hisi_hba, i, STP_LINK_TIMER, 0x7f7a120); hisi_sas_phy_write32(hisi_hba, i, CON_CFG_DRIVER, 0x2a0a01); hisi_sas_phy_write32(hisi_hba, i, SAS_SSP_CON_TIMER_CFG, 0x32); + hisi_sas_phy_write32(hisi_hba, i, SAS_EC_INT_COAL_TIME, + 0x30f4240); /* used for 12G negotiate */ hisi_sas_phy_write32(hisi_hba, i, COARSETUNE_TIME, 0x1e); hisi_sas_phy_write32(hisi_hba, i, AIP_LIMIT, 0x2ffff); @@ -1576,6 +1580,39 @@ static void handle_chl_int1_v3_hw(struct hisi_hba *hisi_hba, int phy_no) hisi_sas_phy_write32(hisi_hba, phy_no, CHL_INT1, irq_value); } +static void phy_get_events_v3_hw(struct hisi_hba *hisi_hba, int phy_no) +{ + struct hisi_sas_phy *phy = &hisi_hba->phy[phy_no]; + struct asd_sas_phy *sas_phy = &phy->sas_phy; + struct sas_phy *sphy = sas_phy->phy; + unsigned long flags; + u32 reg_value; + + spin_lock_irqsave(&phy->lock, flags); + + /* loss dword sync */ + reg_value = hisi_sas_phy_read32(hisi_hba, phy_no, ERR_CNT_DWS_LOST); + sphy->loss_of_dword_sync_count += reg_value; + + /* phy reset problem */ + reg_value = hisi_sas_phy_read32(hisi_hba, phy_no, ERR_CNT_RESET_PROB); + sphy->phy_reset_problem_count += reg_value; + + /* invalid dword */ + reg_value = hisi_sas_phy_read32(hisi_hba, phy_no, ERR_CNT_INVLD_DW); + sphy->invalid_dword_count += reg_value; + + /* disparity err */ + reg_value = hisi_sas_phy_read32(hisi_hba, phy_no, ERR_CNT_DISP_ERR); + sphy->running_disparity_error_count += reg_value; + + /* code violation error */ + reg_value = hisi_sas_phy_read32(hisi_hba, phy_no, ERR_CNT_CODE_ERR); + phy->code_violation_err_count += reg_value; + + spin_unlock_irqrestore(&phy->lock, flags); +} + static void handle_chl_int2_v3_hw(struct hisi_hba *hisi_hba, int phy_no) { u32 irq_msk = hisi_sas_phy_read32(hisi_hba, phy_no, CHL_INT2_MSK); @@ -1583,6 +1620,9 @@ static void handle_chl_int2_v3_hw(struct hisi_hba *hisi_hba, int phy_no) struct hisi_sas_phy *phy = &hisi_hba->phy[phy_no]; struct pci_dev *pci_dev = hisi_hba->pci_dev; struct device *dev = hisi_hba->dev; + static const u32 msk = BIT(CHL_INT2_RX_DISP_ERR_OFF) | + BIT(CHL_INT2_RX_CODE_ERR_OFF) | + BIT(CHL_INT2_RX_INVLD_DW_OFF); irq_value &= ~irq_msk; if (!irq_value) @@ -1603,6 +1643,25 @@ static void handle_chl_int2_v3_hw(struct hisi_hba *hisi_hba, int phy_no) hisi_sas_notify_phy_event(phy, HISI_PHYE_LINK_RESET); } + if (pci_dev->revision > 0x20 && (irq_value & msk)) { + struct asd_sas_phy *sas_phy = &phy->sas_phy; + struct sas_phy *sphy = sas_phy->phy; + + phy_get_events_v3_hw(hisi_hba, phy_no); + + if (irq_value & BIT(CHL_INT2_RX_INVLD_DW_OFF)) + dev_info(dev, "phy%d invalid dword cnt: %u\n", phy_no, + sphy->invalid_dword_count); + + if (irq_value & BIT(CHL_INT2_RX_CODE_ERR_OFF)) + dev_info(dev, "phy%d code violation cnt: %u\n", phy_no, + phy->code_violation_err_count); + + if (irq_value & BIT(CHL_INT2_RX_DISP_ERR_OFF)) + dev_info(dev, "phy%d disparity error cnt: %u\n", phy_no, + sphy->running_disparity_error_count); + } + if ((irq_value & BIT(CHL_INT2_RX_INVLD_DW_OFF)) && (pci_dev->revision == 0x20)) { u32 reg_value; @@ -2231,31 +2290,6 @@ static u32 get_phys_state_v3_hw(struct hisi_hba *hisi_hba) return hisi_sas_read32(hisi_hba, PHY_STATE); } -static void phy_get_events_v3_hw(struct hisi_hba *hisi_hba, int phy_no) -{ - struct hisi_sas_phy *phy = &hisi_hba->phy[phy_no]; - struct asd_sas_phy *sas_phy = &phy->sas_phy; - struct sas_phy *sphy = sas_phy->phy; - u32 reg_value; - - /* loss dword sync */ - reg_value = hisi_sas_phy_read32(hisi_hba, phy_no, ERR_CNT_DWS_LOST); - sphy->loss_of_dword_sync_count += reg_value; - - /* phy reset problem */ - reg_value = hisi_sas_phy_read32(hisi_hba, phy_no, ERR_CNT_RESET_PROB); - sphy->phy_reset_problem_count += reg_value; - - /* invalid dword */ - reg_value = hisi_sas_phy_read32(hisi_hba, phy_no, ERR_CNT_INVLD_DW); - sphy->invalid_dword_count += reg_value; - - /* disparity err */ - reg_value = hisi_sas_phy_read32(hisi_hba, phy_no, ERR_CNT_DISP_ERR); - sphy->running_disparity_error_count += reg_value; - -} - static int disable_host_v3_hw(struct hisi_hba *hisi_hba) { struct device *dev = hisi_hba->dev; -- cgit v1.2.3-59-g8ed1b From efdcad62e7b8a02fcccc5ccca57806dce1482ac8 Mon Sep 17 00:00:00 2001 From: John Garry Date: Thu, 28 Feb 2019 22:51:00 +0800 Subject: scsi: hisi_sas: Set PHY linkrate when disconnected When the PHY comes down, we currently do not set the negotiated linkrate: root@(none)$ pwd /sys/class/sas_phy/phy-0:0 root@(none)$ more enable 1 root@(none)$ more negotiated_linkrate 12.0 Gbit root@(none)$ echo 0 > enable root@(none)$ more negotiated_linkrate 12.0 Gbit root@(none)$ This patch fixes the driver code to set it properly when the PHY comes down. If the PHY had been enabled, then set unknown; otherwise, flag as disabled. The logical place to set the negotiated linkrate for this scenario is PHY down routine, which is called from the PHY down ISR. However, it is not possible to know if the PHY comes down due to PHY disable or loss of link, as sas_phy.enabled member is not set until after the transport disable routine is complete, which races with the PHY down ISR. As an imperfect solution, use sas_phy_data.enable as the flag to know if the PHY is down due to disable. It's imperfect, as sas_phy_data is internal to libsas. I can't see another way without adding a new field to hisi_sas_phy and managing it, or changing SCSI SAS transport. Signed-off-by: John Garry Signed-off-by: Martin K. Petersen --- drivers/scsi/hisi_sas/hisi_sas_main.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/drivers/scsi/hisi_sas/hisi_sas_main.c b/drivers/scsi/hisi_sas/hisi_sas_main.c index dd03dcbd3786..d8204bc3931b 100644 --- a/drivers/scsi/hisi_sas/hisi_sas_main.c +++ b/drivers/scsi/hisi_sas/hisi_sas_main.c @@ -10,6 +10,7 @@ */ #include "hisi_sas.h" +#include "../libsas/sas_internal.h" #define DRV_NAME "hisi_sas" #define DEV_IS_GONE(dev) \ @@ -2126,9 +2127,18 @@ static int hisi_sas_write_gpio(struct sas_ha_struct *sha, u8 reg_type, static void hisi_sas_phy_disconnected(struct hisi_sas_phy *phy) { + struct asd_sas_phy *sas_phy = &phy->sas_phy; + struct sas_phy *sphy = sas_phy->phy; + struct sas_phy_data *d = sphy->hostdata; + phy->phy_attached = 0; phy->phy_type = 0; phy->port = NULL; + + if (d->enable) + sphy->negotiated_linkrate = SAS_LINK_RATE_UNKNOWN; + else + sphy->negotiated_linkrate = SAS_PHY_DISABLED; } void hisi_sas_phy_down(struct hisi_hba *hisi_hba, int phy_no, int rdy) -- cgit v1.2.3-59-g8ed1b From 57dbb2b218eb6b4faa39025e3e5974742a4e8986 Mon Sep 17 00:00:00 2001 From: Xiang Chen Date: Thu, 28 Feb 2019 22:51:01 +0800 Subject: scsi: hisi_sas: Send HARD RESET to clear the previous affiliation of STP target port If we exchange SAS expander from one SAS controller to other SAS controller without powering it down, the STP target port will maintain previous affiliation and reject all subsequent connection requests from other STP initiator ports with OPEN_REJECT (STP RESOURCES BUSY). To solve this issue, send HARD RESET to clear the previous affiliation of STP target port according to SPL (chapter 6.19.4). We (re-)introduce dev status flag to know if to sleep in NEXUS reset code or not for remote PHYs. The idea is that if the device is being initialised, we don't require the delay, and caller would wait for link to be established, cf. sas_ata_hard_reset(). Co-developed-by: Luo Jiaxing Signed-off-by: Xiang Chen Signed-off-by: John Garry Signed-off-by: Martin K. Petersen --- drivers/scsi/hisi_sas/hisi_sas.h | 7 ++++++ drivers/scsi/hisi_sas/hisi_sas_main.c | 44 +++++++++++++++++++++++++++++++--- drivers/scsi/hisi_sas/hisi_sas_v2_hw.c | 1 + 3 files changed, 49 insertions(+), 3 deletions(-) diff --git a/drivers/scsi/hisi_sas/hisi_sas.h b/drivers/scsi/hisi_sas/hisi_sas.h index 515aee9318a4..9bfa9f12d81e 100644 --- a/drivers/scsi/hisi_sas/hisi_sas.h +++ b/drivers/scsi/hisi_sas/hisi_sas.h @@ -18,6 +18,7 @@ #include #include #include +#include #include #include #include @@ -94,6 +95,11 @@ enum { PORT_TYPE_SATA = (1U << 0), }; +enum dev_status { + HISI_SAS_DEV_INIT, + HISI_SAS_DEV_NORMAL, +}; + enum { HISI_SAS_INT_ABT_CMD = 0, HISI_SAS_INT_ABT_DEV = 1, @@ -195,6 +201,7 @@ struct hisi_sas_device { struct hisi_sas_dq *dq; struct list_head list; enum sas_device_type dev_type; + enum dev_status dev_status; int device_id; int sata_idx; spinlock_t lock; /* For protecting slots */ diff --git a/drivers/scsi/hisi_sas/hisi_sas_main.c b/drivers/scsi/hisi_sas/hisi_sas_main.c index d8204bc3931b..d12924256964 100644 --- a/drivers/scsi/hisi_sas/hisi_sas_main.c +++ b/drivers/scsi/hisi_sas/hisi_sas_main.c @@ -708,6 +708,7 @@ static struct hisi_sas_device *hisi_sas_alloc_dev(struct domain_device *device) hisi_hba->devices[i].device_id = i; sas_dev = &hisi_hba->devices[i]; + sas_dev->dev_status = HISI_SAS_DEV_INIT; sas_dev->dev_type = device->dev_type; sas_dev->hisi_hba = hisi_hba; sas_dev->sas_device = device; @@ -732,6 +733,8 @@ static int hisi_sas_init_device(struct domain_device *device) struct hisi_sas_tmf_task tmf_task; int retry = HISI_SAS_SRST_ATA_DISK_CNT; struct hisi_hba *hisi_hba = dev_to_hisi_hba(device); + struct device *dev = hisi_hba->dev; + struct sas_phy *local_phy; switch (device->dev_type) { case SAS_END_DEVICE: @@ -747,6 +750,31 @@ static int hisi_sas_init_device(struct domain_device *device) case SAS_SATA_PM: case SAS_SATA_PM_PORT: case SAS_SATA_PENDING: + /* + * send HARD RESET to clear previous affiliation of + * STP target port + */ + local_phy = sas_get_local_phy(device); + if (!scsi_is_sas_phy_local(local_phy)) { + unsigned long deadline = ata_deadline(jiffies, 20000); + struct sata_device *sata_dev = &device->sata_dev; + struct ata_host *ata_host = sata_dev->ata_host; + struct ata_port_operations *ops = ata_host->ops; + struct ata_port *ap = sata_dev->ap; + struct ata_link *link; + unsigned int classes; + + ata_for_each_link(link, ap, EDGE) + rc = ops->hardreset(link, &classes, + deadline); + } + sas_put_local_phy(local_phy); + if (rc) { + dev_warn(dev, "SATA disk hardreset fail: 0x%x\n", + rc); + return rc; + } + while (retry-- > 0) { rc = hisi_sas_softreset_ata_disk(device); if (!rc) @@ -809,6 +837,7 @@ static int hisi_sas_dev_found(struct domain_device *device) rc = hisi_sas_init_device(device); if (rc) goto err_out; + sas_dev->dev_status = HISI_SAS_DEV_NORMAL; return 0; err_out: @@ -1715,20 +1744,23 @@ static int hisi_sas_clear_aca(struct domain_device *device, u8 *lun) static int hisi_sas_debug_I_T_nexus_reset(struct domain_device *device) { struct sas_phy *local_phy = sas_get_local_phy(device); - int rc, reset_type = (device->dev_type == SAS_SATA_DEV || - (device->tproto & SAS_PROTOCOL_STP)) ? 0 : 1; + struct hisi_sas_device *sas_dev = device->lldd_dev; struct hisi_hba *hisi_hba = dev_to_hisi_hba(device); struct sas_ha_struct *sas_ha = &hisi_hba->sha; struct asd_sas_phy *sas_phy = sas_ha->sas_phy[local_phy->number]; struct hisi_sas_phy *phy = container_of(sas_phy, struct hisi_sas_phy, sas_phy); DECLARE_COMPLETION_ONSTACK(phyreset); + int rc, reset_type; if (scsi_is_sas_phy_local(local_phy)) { phy->in_reset = 1; phy->reset_completion = &phyreset; } + reset_type = (sas_dev->dev_status == HISI_SAS_DEV_INIT || + !dev_is_sata(device)) ? 1 : 0; + rc = sas_phy_reset(local_phy, reset_type); sas_put_local_phy(local_phy); @@ -1744,8 +1776,13 @@ static int hisi_sas_debug_I_T_nexus_reset(struct domain_device *device) /* report PHY down if timed out */ if (!ret) hisi_sas_phy_down(hisi_hba, sas_phy->id, 0); - } else + } else if (sas_dev->dev_status != HISI_SAS_DEV_INIT) { + /* + * If in init state, we rely on caller to wait for link to be + * ready; otherwise, delay. + */ msleep(2000); + } return rc; } @@ -2264,6 +2301,7 @@ int hisi_sas_alloc(struct hisi_hba *hisi_hba) for (i = 0; i < HISI_SAS_MAX_DEVICES; i++) { hisi_hba->devices[i].dev_type = SAS_PHY_UNUSED; hisi_hba->devices[i].device_id = i; + hisi_hba->devices[i].dev_status = HISI_SAS_DEV_INIT; } for (i = 0; i < hisi_hba->queue_count; i++) { diff --git a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c index e40cc6b3b67b..89160ab3efb0 100644 --- a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c +++ b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c @@ -868,6 +868,7 @@ hisi_sas_device *alloc_dev_quirk_v2_hw(struct domain_device *device) hisi_hba->devices[i].device_id = i; sas_dev = &hisi_hba->devices[i]; + sas_dev->dev_status = HISI_SAS_DEV_INIT; sas_dev->dev_type = device->dev_type; sas_dev->hisi_hba = hisi_hba; sas_dev->sas_device = device; -- cgit v1.2.3-59-g8ed1b From cf9efd5d92365696580a45e0351208eef0ea1562 Mon Sep 17 00:00:00 2001 From: Xiang Chen Date: Thu, 28 Feb 2019 22:51:02 +0800 Subject: scsi: hisi_sas: Change SERDES_CFG init value to increase reliability of HiLink With default value of register SERDES_CFG, the link is not stable for some special disks when running IO. According to HW guys' suggestion, need to make the bit10~19 value of register SERDES_CFG the max value to increase the reliability of the HiLink. Signed-off-by: Xiang Chen Reviewed-by: Yupeng Zhou Signed-off-by: John Garry Signed-off-by: Martin K. Petersen --- drivers/scsi/hisi_sas/hisi_sas_v3_hw.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c index 720721196b12..e2f2c04355b9 100644 --- a/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c +++ b/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c @@ -129,6 +129,7 @@ #define PHY_CTRL_RESET_MSK (0x1 << PHY_CTRL_RESET_OFF) #define CMD_HDR_PIR_OFF 8 #define CMD_HDR_PIR_MSK (0x1 << CMD_HDR_PIR_OFF) +#define SERDES_CFG (PORT_BASE + 0x1c) #define SL_CFG (PORT_BASE + 0x84) #define AIP_LIMIT (PORT_BASE + 0x90) #define SL_CONTROL (PORT_BASE + 0x94) @@ -525,6 +526,7 @@ static void init_reg_v3_hw(struct hisi_hba *hisi_hba) } hisi_sas_phy_write32(hisi_hba, i, PROG_PHY_LINK_RATE, prog_phy_link_rate); + hisi_sas_phy_write32(hisi_hba, i, SERDES_CFG, 0xffc00); hisi_sas_phy_write32(hisi_hba, i, SAS_RX_TRAIN_TIMER, 0x13e80); hisi_sas_phy_write32(hisi_hba, i, CHL_INT0, 0xffffffff); hisi_sas_phy_write32(hisi_hba, i, CHL_INT1, 0xffffffff); -- cgit v1.2.3-59-g8ed1b From a656183e6c58a3a4d818278afab522b30a17014c Mon Sep 17 00:00:00 2001 From: Lee Duncan Date: Mon, 25 Feb 2019 09:41:30 -0800 Subject: scsi: libiscsi: Hold back_lock when calling iscsi_complete_task If there is an error queueing an iscsi command in iscsi_queuecommand(), for example if the transport fails to take the command in sessuin->tt->xmit_task(), then the error path can call iscsi_complete_task() without first aquiring the back_lock as required. This can lead to things like ITT pool can get corrupt, resulting in duplicate ITTs being sent out. The solution is to hold the back_lock around iscsi_complete_task() calls, and to add a little commenting to help others understand when back_lock must be held. Signed-off-by: Lee Duncan Acked-by: Chris Leech Signed-off-by: Martin K. Petersen --- drivers/scsi/libiscsi.c | 22 ++++++++++++++++++++-- 1 file changed, 20 insertions(+), 2 deletions(-) diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c index dd314d2b1111..418dbed7618f 100644 --- a/drivers/scsi/libiscsi.c +++ b/drivers/scsi/libiscsi.c @@ -798,7 +798,7 @@ EXPORT_SYMBOL_GPL(iscsi_conn_send_pdu); * @datalen: len of buffer * * iscsi_cmd_rsp sets up the scsi_cmnd fields based on the PDU and - * then completes the command and task. + * then completes the command and task. called under back_lock **/ static void iscsi_scsi_cmd_rsp(struct iscsi_conn *conn, struct iscsi_hdr *hdr, struct iscsi_task *task, char *data, @@ -894,6 +894,9 @@ out: * @conn: iscsi connection * @hdr: iscsi pdu * @task: scsi command task + * + * iscsi_data_in_rsp sets up the scsi_cmnd fields based on the data received + * then completes the command and task. called under back_lock **/ static void iscsi_data_in_rsp(struct iscsi_conn *conn, struct iscsi_hdr *hdr, @@ -978,6 +981,16 @@ static int iscsi_send_nopout(struct iscsi_conn *conn, struct iscsi_nopin *rhdr) return 0; } +/** + * iscsi_nop_out_rsp - SCSI NOP Response processing + * @task: scsi command task + * @nop: the nop structure + * @data: where to put the data + * @datalen: length of data + * + * iscsi_nop_out_rsp handles nop response from use or + * from user space. called under back_lock + **/ static int iscsi_nop_out_rsp(struct iscsi_task *task, struct iscsi_nopin *nop, char *data, int datalen) { @@ -1744,7 +1757,9 @@ int iscsi_queuecommand(struct Scsi_Host *host, struct scsi_cmnd *sc) return 0; prepd_reject: + spin_lock_bh(&session->back_lock); iscsi_complete_task(task, ISCSI_TASK_REQUEUE_SCSIQ); + spin_unlock_bh(&session->back_lock); reject: spin_unlock_bh(&session->frwd_lock); ISCSI_DBG_SESSION(session, "cmd 0x%x rejected (%d)\n", @@ -1752,7 +1767,9 @@ reject: return SCSI_MLQUEUE_TARGET_BUSY; prepd_fault: + spin_lock_bh(&session->back_lock); iscsi_complete_task(task, ISCSI_TASK_REQUEUE_SCSIQ); + spin_unlock_bh(&session->back_lock); fault: spin_unlock_bh(&session->frwd_lock); ISCSI_DBG_SESSION(session, "iscsi: cmd 0x%x is not queued (%d)\n", @@ -3069,8 +3086,9 @@ fail_mgmt_tasks(struct iscsi_session *session, struct iscsi_conn *conn) state = ISCSI_TASK_ABRT_SESS_RECOV; if (task->state == ISCSI_TASK_PENDING) state = ISCSI_TASK_COMPLETED; + spin_lock_bh(&session->back_lock); iscsi_complete_task(task, state); - + spin_unlock_bh(&session->back_lock); } } -- cgit v1.2.3-59-g8ed1b From 3a487ff78c4b2bf5686dbb29178cfd78b3eed053 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Thu, 7 Mar 2019 08:33:44 +0300 Subject: scsi: lpfc: Fix error codes in lpfc_sli4_pci_mem_setup() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit It used to be that "error" was set to -ENODEV at the start of the function but we shifted some code around an now "error" is set to zero for most error paths. There is a mix of direct returns and "goto out" but I changed everything to direct returns for consistency. Fixes: 56de8357049c ("scsi: lpfc: fix calls to dma_set_mask_and_coherent()") Signed-off-by: Dan Carpenter Acked-by: James Smart  Signed-off-by: Martin K. Petersen --- drivers/scsi/lpfc/lpfc_init.c | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) diff --git a/drivers/scsi/lpfc/lpfc_init.c b/drivers/scsi/lpfc/lpfc_init.c index 707dbe7a8d15..bbc2815fc012 100644 --- a/drivers/scsi/lpfc/lpfc_init.c +++ b/drivers/scsi/lpfc/lpfc_init.c @@ -10049,7 +10049,7 @@ lpfc_sli4_pci_mem_setup(struct lpfc_hba *phba) { struct pci_dev *pdev = phba->pcidev; unsigned long bar0map_len, bar1map_len, bar2map_len; - int error = -ENODEV; + int error; uint32_t if_type; if (!pdev) @@ -10066,7 +10066,7 @@ lpfc_sli4_pci_mem_setup(struct lpfc_hba *phba) */ if (pci_read_config_dword(pdev, LPFC_SLI_INTF, &phba->sli4_hba.sli_intf.word0)) { - return error; + return -ENODEV; } /* There is no SLI3 failback for SLI4 devices. */ @@ -10076,7 +10076,7 @@ lpfc_sli4_pci_mem_setup(struct lpfc_hba *phba) "2894 SLI_INTF reg contents invalid " "sli_intf reg 0x%x\n", phba->sli4_hba.sli_intf.word0); - return error; + return -ENODEV; } if_type = bf_get(lpfc_sli_intf_if_type, &phba->sli4_hba.sli_intf); @@ -10100,7 +10100,7 @@ lpfc_sli4_pci_mem_setup(struct lpfc_hba *phba) dev_printk(KERN_ERR, &pdev->dev, "ioremap failed for SLI4 PCI config " "registers.\n"); - goto out; + return -ENODEV; } phba->pci_bar0_memmap_p = phba->sli4_hba.conf_regs_memmap_p; /* Set up BAR0 PCI config space register memory map */ @@ -10111,7 +10111,7 @@ lpfc_sli4_pci_mem_setup(struct lpfc_hba *phba) if (if_type >= LPFC_SLI_INTF_IF_TYPE_2) { dev_printk(KERN_ERR, &pdev->dev, "FATAL - No BAR0 mapping for SLI4, if_type 2\n"); - goto out; + return -ENODEV; } phba->sli4_hba.conf_regs_memmap_p = ioremap(phba->pci_bar0_map, bar0map_len); @@ -10119,7 +10119,7 @@ lpfc_sli4_pci_mem_setup(struct lpfc_hba *phba) dev_printk(KERN_ERR, &pdev->dev, "ioremap failed for SLI4 PCI config " "registers.\n"); - goto out; + return -ENODEV; } lpfc_sli4_bar0_register_memmap(phba, if_type); } @@ -10165,6 +10165,7 @@ lpfc_sli4_pci_mem_setup(struct lpfc_hba *phba) if (!phba->sli4_hba.drbl_regs_memmap_p) { dev_err(&pdev->dev, "ioremap failed for SLI4 HBA doorbell registers.\n"); + error = -ENOMEM; goto out_iounmap_conf; } phba->pci_bar2_memmap_p = phba->sli4_hba.drbl_regs_memmap_p; @@ -10214,6 +10215,7 @@ lpfc_sli4_pci_mem_setup(struct lpfc_hba *phba) if (!phba->sli4_hba.dpp_regs_memmap_p) { dev_err(&pdev->dev, "ioremap failed for SLI4 HBA dpp registers.\n"); + error = -ENOMEM; goto out_iounmap_ctrl; } phba->pci_bar4_memmap_p = phba->sli4_hba.dpp_regs_memmap_p; @@ -10244,7 +10246,7 @@ out_iounmap_ctrl: iounmap(phba->sli4_hba.ctrl_regs_memmap_p); out_iounmap_conf: iounmap(phba->sli4_hba.conf_regs_memmap_p); -out: + return error; } -- cgit v1.2.3-59-g8ed1b From 0015437cc046e5ec2b57b00ff8312b8d432eac7c Mon Sep 17 00:00:00 2001 From: Sagar Biradar Date: Thu, 7 Mar 2019 23:26:41 -0800 Subject: scsi: aacraid: Fix performance issue on logical drives Fix performance issue where the queue depth for SmartIOC logical volumes is set to 1, and allow the usual logical volume code to be executed Fixes: a052865fe287 (aacraid: Set correct Queue Depth for HBA1000 RAW disks) Cc: stable@vger.kernel.org Signed-off-by: Sagar Biradar Reviewed-by: Dave Carroll Signed-off-by: Martin K. Petersen --- drivers/scsi/aacraid/linit.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/drivers/scsi/aacraid/linit.c b/drivers/scsi/aacraid/linit.c index 22ecacffeca6..598d6efae53e 100644 --- a/drivers/scsi/aacraid/linit.c +++ b/drivers/scsi/aacraid/linit.c @@ -413,13 +413,16 @@ static int aac_slave_configure(struct scsi_device *sdev) if (chn < AAC_MAX_BUSES && tid < AAC_MAX_TARGETS && aac->sa_firmware) { devtype = aac->hba_map[chn][tid].devtype; - if (devtype == AAC_DEVTYPE_NATIVE_RAW) + if (devtype == AAC_DEVTYPE_NATIVE_RAW) { depth = aac->hba_map[chn][tid].qd_limit; - else if (devtype == AAC_DEVTYPE_ARC_RAW) + set_timeout = 1; + goto common_config; + } + if (devtype == AAC_DEVTYPE_ARC_RAW) { set_qd_dev_type = true; - - set_timeout = 1; - goto common_config; + set_timeout = 1; + goto common_config; + } } if (aac->jbod && (sdev->type == TYPE_DISK)) -- cgit v1.2.3-59-g8ed1b From 52eaa798f4f4e983c711eaa1c13d8859a52946e8 Mon Sep 17 00:00:00 2001 From: Hannes Reinecke Date: Tue, 12 Mar 2019 09:08:12 +0100 Subject: scsi: ia64: simscsi: use request tag instead of serial_number Use the request tag for logging instead of the scsi command serial number. Signed-off-by: Hannes Reinecke Reviewed-by: Christoph Hellwig Signed-off-by: Martin K. Petersen [jejb: fix commit oneliner] Signed-off-by: James Bottomley --- arch/ia64/hp/sim/simscsi.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/arch/ia64/hp/sim/simscsi.c b/arch/ia64/hp/sim/simscsi.c index f86844fc0725..0a8a74271173 100644 --- a/arch/ia64/hp/sim/simscsi.c +++ b/arch/ia64/hp/sim/simscsi.c @@ -105,7 +105,8 @@ simscsi_interrupt (unsigned long val) atomic_dec(&num_reqs); queue[rd].sc = NULL; if (DBG) - printk("simscsi_interrupt: done with %ld\n", sc->serial_number); + printk("simscsi_interrupt: done with %u\n", + sc->request->tag); (*sc->scsi_done)(sc); rd = (rd + 1) % SIMSCSI_REQ_QUEUE_LEN; } @@ -214,8 +215,8 @@ simscsi_queuecommand_lck (struct scsi_cmnd *sc, void (*done)(struct scsi_cmnd *) register long sp asm ("sp"); if (DBG) - printk("simscsi_queuecommand: target=%d,cmnd=%u,sc=%lu,sp=%lx,done=%p\n", - target_id, sc->cmnd[0], sc->serial_number, sp, done); + printk("simscsi_queuecommand: target=%d,cmnd=%u,sc=%u,sp=%lx,done=%p\n", + target_id, sc->cmnd[0], sc->request->tag, sp, done); #endif sc->result = DID_BAD_TARGET << 16; -- cgit v1.2.3-59-g8ed1b