From 0b0747d507bffb827e40fc0f9fb5883fffc23477 Mon Sep 17 00:00:00 2001 From: Junxiao Bi Date: Mon, 28 Aug 2023 15:10:18 -0700 Subject: scsi: megaraid_sas: Fix deadlock on firmware crashdump The following processes run into a deadlock. CPU 41 was waiting for CPU 29 to handle a CSD request while holding spinlock "crashdump_lock", but CPU 29 was hung by that spinlock with IRQs disabled. PID: 17360 TASK: ffff95c1090c5c40 CPU: 41 COMMAND: "mrdiagd" !# 0 [ffffb80edbf37b58] __read_once_size at ffffffff9b871a40 include/linux/compiler.h:185:0 !# 1 [ffffb80edbf37b58] atomic_read at ffffffff9b871a40 arch/x86/include/asm/atomic.h:27:0 !# 2 [ffffb80edbf37b58] dump_stack at ffffffff9b871a40 lib/dump_stack.c:54:0 # 3 [ffffb80edbf37b78] csd_lock_wait_toolong at ffffffff9b131ad5 kernel/smp.c:364:0 # 4 [ffffb80edbf37b78] __csd_lock_wait at ffffffff9b131ad5 kernel/smp.c:384:0 # 5 [ffffb80edbf37bf8] csd_lock_wait at ffffffff9b13267a kernel/smp.c:394:0 # 6 [ffffb80edbf37bf8] smp_call_function_many at ffffffff9b13267a kernel/smp.c:843:0 # 7 [ffffb80edbf37c50] smp_call_function at ffffffff9b13279d kernel/smp.c:867:0 # 8 [ffffb80edbf37c50] on_each_cpu at ffffffff9b13279d kernel/smp.c:976:0 # 9 [ffffb80edbf37c78] flush_tlb_kernel_range at ffffffff9b085c4b arch/x86/mm/tlb.c:742:0 #10 [ffffb80edbf37cb8] __purge_vmap_area_lazy at ffffffff9b23a1e0 mm/vmalloc.c:701:0 #11 [ffffb80edbf37ce0] try_purge_vmap_area_lazy at ffffffff9b23a2cc mm/vmalloc.c:722:0 #12 [ffffb80edbf37ce0] free_vmap_area_noflush at ffffffff9b23a2cc mm/vmalloc.c:754:0 #13 [ffffb80edbf37cf8] free_unmap_vmap_area at ffffffff9b23bb3b mm/vmalloc.c:764:0 #14 [ffffb80edbf37cf8] remove_vm_area at ffffffff9b23bb3b mm/vmalloc.c:1509:0 #15 [ffffb80edbf37d18] __vunmap at ffffffff9b23bb8a mm/vmalloc.c:1537:0 #16 [ffffb80edbf37d40] vfree at ffffffff9b23bc85 mm/vmalloc.c:1612:0 #17 [ffffb80edbf37d58] megasas_free_host_crash_buffer [megaraid_sas] at ffffffffc020b7f2 drivers/scsi/megaraid/megaraid_sas_fusion.c:3932:0 #18 [ffffb80edbf37d80] fw_crash_state_store [megaraid_sas] at ffffffffc01f804d drivers/scsi/megaraid/megaraid_sas_base.c:3291:0 #19 [ffffb80edbf37dc0] dev_attr_store at ffffffff9b56dd7b drivers/base/core.c:758:0 #20 [ffffb80edbf37dd0] sysfs_kf_write at ffffffff9b326acf fs/sysfs/file.c:144:0 #21 [ffffb80edbf37de0] kernfs_fop_write at ffffffff9b325fd4 fs/kernfs/file.c:316:0 #22 [ffffb80edbf37e20] __vfs_write at ffffffff9b29418a fs/read_write.c:480:0 #23 [ffffb80edbf37ea8] vfs_write at ffffffff9b294462 fs/read_write.c:544:0 #24 [ffffb80edbf37ee8] SYSC_write at ffffffff9b2946ec fs/read_write.c:590:0 #25 [ffffb80edbf37ee8] SyS_write at ffffffff9b2946ec fs/read_write.c:582:0 #26 [ffffb80edbf37f30] do_syscall_64 at ffffffff9b003ca9 arch/x86/entry/common.c:298:0 #27 [ffffb80edbf37f58] entry_SYSCALL_64 at ffffffff9ba001b1 arch/x86/entry/entry_64.S:238:0 PID: 17355 TASK: ffff95c1090c3d80 CPU: 29 COMMAND: "mrdiagd" !# 0 [ffffb80f2d3c7d30] __read_once_size at ffffffff9b0f2ab0 include/linux/compiler.h:185:0 !# 1 [ffffb80f2d3c7d30] native_queued_spin_lock_slowpath at ffffffff9b0f2ab0 kernel/locking/qspinlock.c:368:0 # 2 [ffffb80f2d3c7d58] pv_queued_spin_lock_slowpath at ffffffff9b0f244b arch/x86/include/asm/paravirt.h:674:0 # 3 [ffffb80f2d3c7d58] queued_spin_lock_slowpath at ffffffff9b0f244b arch/x86/include/asm/qspinlock.h:53:0 # 4 [ffffb80f2d3c7d68] queued_spin_lock at ffffffff9b8961a6 include/asm-generic/qspinlock.h:90:0 # 5 [ffffb80f2d3c7d68] do_raw_spin_lock_flags at ffffffff9b8961a6 include/linux/spinlock.h:173:0 # 6 [ffffb80f2d3c7d68] __raw_spin_lock_irqsave at ffffffff9b8961a6 include/linux/spinlock_api_smp.h:122:0 # 7 [ffffb80f2d3c7d68] _raw_spin_lock_irqsave at ffffffff9b8961a6 kernel/locking/spinlock.c:160:0 # 8 [ffffb80f2d3c7d88] fw_crash_buffer_store [megaraid_sas] at ffffffffc01f8129 drivers/scsi/megaraid/megaraid_sas_base.c:3205:0 # 9 [ffffb80f2d3c7dc0] dev_attr_store at ffffffff9b56dd7b drivers/base/core.c:758:0 #10 [ffffb80f2d3c7dd0] sysfs_kf_write at ffffffff9b326acf fs/sysfs/file.c:144:0 #11 [ffffb80f2d3c7de0] kernfs_fop_write at ffffffff9b325fd4 fs/kernfs/file.c:316:0 #12 [ffffb80f2d3c7e20] __vfs_write at ffffffff9b29418a fs/read_write.c:480:0 #13 [ffffb80f2d3c7ea8] vfs_write at ffffffff9b294462 fs/read_write.c:544:0 #14 [ffffb80f2d3c7ee8] SYSC_write at ffffffff9b2946ec fs/read_write.c:590:0 #15 [ffffb80f2d3c7ee8] SyS_write at ffffffff9b2946ec fs/read_write.c:582:0 #16 [ffffb80f2d3c7f30] do_syscall_64 at ffffffff9b003ca9 arch/x86/entry/common.c:298:0 #17 [ffffb80f2d3c7f58] entry_SYSCALL_64 at ffffffff9ba001b1 arch/x86/entry/entry_64.S:238:0 The lock is used to synchronize different sysfs operations, it doesn't protect any resource that will be touched by an interrupt. Consequently it's not required to disable IRQs. Replace the spinlock with a mutex to fix the deadlock. Signed-off-by: Junxiao Bi Link: https://lore.kernel.org/r/20230828221018.19471-1-junxiao.bi@oracle.com Reviewed-by: Mike Christie Cc: stable@vger.kernel.org Signed-off-by: Martin K. Petersen --- drivers/scsi/megaraid/megaraid_sas.h | 2 +- drivers/scsi/megaraid/megaraid_sas_base.c | 21 +++++++++------------ 2 files changed, 10 insertions(+), 13 deletions(-) (limited to 'drivers/scsi') diff --git a/drivers/scsi/megaraid/megaraid_sas.h b/drivers/scsi/megaraid/megaraid_sas.h index 3554f6b07727..94abba57582d 100644 --- a/drivers/scsi/megaraid/megaraid_sas.h +++ b/drivers/scsi/megaraid/megaraid_sas.h @@ -2332,7 +2332,7 @@ struct megasas_instance { u32 support_morethan256jbod; /* FW support for more than 256 PD/JBOD */ bool use_seqnum_jbod_fp; /* Added for PD sequence */ bool smp_affinity_enable; - spinlock_t crashdump_lock; + struct mutex crashdump_lock; struct megasas_register_set __iomem *reg_set; u32 __iomem *reply_post_host_index_addr[MR_MAX_MSIX_REG_ARRAY]; diff --git a/drivers/scsi/megaraid/megaraid_sas_base.c b/drivers/scsi/megaraid/megaraid_sas_base.c index b9d46dcb5210..e1aa667dae66 100644 --- a/drivers/scsi/megaraid/megaraid_sas_base.c +++ b/drivers/scsi/megaraid/megaraid_sas_base.c @@ -3271,14 +3271,13 @@ fw_crash_buffer_store(struct device *cdev, struct megasas_instance *instance = (struct megasas_instance *) shost->hostdata; int val = 0; - unsigned long flags; if (kstrtoint(buf, 0, &val) != 0) return -EINVAL; - spin_lock_irqsave(&instance->crashdump_lock, flags); + mutex_lock(&instance->crashdump_lock); instance->fw_crash_buffer_offset = val; - spin_unlock_irqrestore(&instance->crashdump_lock, flags); + mutex_unlock(&instance->crashdump_lock); return strlen(buf); } @@ -3293,24 +3292,23 @@ fw_crash_buffer_show(struct device *cdev, unsigned long dmachunk = CRASH_DMA_BUF_SIZE; unsigned long chunk_left_bytes; unsigned long src_addr; - unsigned long flags; u32 buff_offset; - spin_lock_irqsave(&instance->crashdump_lock, flags); + mutex_lock(&instance->crashdump_lock); buff_offset = instance->fw_crash_buffer_offset; if (!instance->crash_dump_buf || !((instance->fw_crash_state == AVAILABLE) || (instance->fw_crash_state == COPYING))) { dev_err(&instance->pdev->dev, "Firmware crash dump is not available\n"); - spin_unlock_irqrestore(&instance->crashdump_lock, flags); + mutex_unlock(&instance->crashdump_lock); return -EINVAL; } if (buff_offset > (instance->fw_crash_buffer_size * dmachunk)) { dev_err(&instance->pdev->dev, "Firmware crash dump offset is out of range\n"); - spin_unlock_irqrestore(&instance->crashdump_lock, flags); + mutex_unlock(&instance->crashdump_lock); return 0; } @@ -3322,7 +3320,7 @@ fw_crash_buffer_show(struct device *cdev, src_addr = (unsigned long)instance->crash_buf[buff_offset / dmachunk] + (buff_offset % dmachunk); memcpy(buf, (void *)src_addr, size); - spin_unlock_irqrestore(&instance->crashdump_lock, flags); + mutex_unlock(&instance->crashdump_lock); return size; } @@ -3347,7 +3345,6 @@ fw_crash_state_store(struct device *cdev, struct megasas_instance *instance = (struct megasas_instance *) shost->hostdata; int val = 0; - unsigned long flags; if (kstrtoint(buf, 0, &val) != 0) return -EINVAL; @@ -3361,9 +3358,9 @@ fw_crash_state_store(struct device *cdev, instance->fw_crash_state = val; if ((val == COPIED) || (val == COPY_ERROR)) { - spin_lock_irqsave(&instance->crashdump_lock, flags); + mutex_lock(&instance->crashdump_lock); megasas_free_host_crash_buffer(instance); - spin_unlock_irqrestore(&instance->crashdump_lock, flags); + mutex_unlock(&instance->crashdump_lock); if (val == COPY_ERROR) dev_info(&instance->pdev->dev, "application failed to " "copy Firmware crash dump\n"); @@ -7422,7 +7419,7 @@ static inline void megasas_init_ctrl_params(struct megasas_instance *instance) init_waitqueue_head(&instance->int_cmd_wait_q); init_waitqueue_head(&instance->abort_cmd_wait_q); - spin_lock_init(&instance->crashdump_lock); + mutex_init(&instance->crashdump_lock); spin_lock_init(&instance->mfi_pool_lock); spin_lock_init(&instance->hba_lock); spin_lock_init(&instance->stream_lock); -- cgit v1.2.3 From 31a0865bf593e59c4433a3624b4c87c40049ed9a Mon Sep 17 00:00:00 2001 From: Alex Henrie Date: Wed, 30 Aug 2023 23:19:42 -0600 Subject: scsi: ppa: Fix accidentally reversed conditions for 16-bit and 32-bit EPP The conditions were correct in the ppa_in() function but not in the ppa_out() function. Fixes: 68a4f84a17c1 ("scsi: ppa: Add a module parameter for the transfer mode") Signed-off-by: Alex Henrie Link: https://lore.kernel.org/r/20230831051945.515476-1-alexhenrie24@gmail.com Signed-off-by: Martin K. Petersen --- drivers/scsi/ppa.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers/scsi') diff --git a/drivers/scsi/ppa.c b/drivers/scsi/ppa.c index 19f0b93fa3d8..d592ee9170c1 100644 --- a/drivers/scsi/ppa.c +++ b/drivers/scsi/ppa.c @@ -307,9 +307,9 @@ static int ppa_out(ppa_struct *dev, char *buffer, int len) case PPA_EPP_8: epp_reset(ppb); w_ctr(ppb, 0x4); - if (dev->mode == PPA_EPP_32 && !(((long) buffer | len) & 0x01)) + if (dev->mode == PPA_EPP_32 && !(((long) buffer | len) & 0x03)) outsl(ppb + 4, buffer, len >> 2); - else if (dev->mode == PPA_EPP_16 && !(((long) buffer | len) & 0x03)) + else if (dev->mode == PPA_EPP_16 && !(((long) buffer | len) & 0x01)) outsw(ppb + 4, buffer, len >> 1); else outsb(ppb + 4, buffer, len); -- cgit v1.2.3 From 0be7592885d7b4c20595c388adc13930b653b847 Mon Sep 17 00:00:00 2001 From: Nilesh Javali Date: Thu, 31 Aug 2023 16:51:45 +0530 Subject: scsi: qla2xxx: Correct endianness for rqstlen and rsplen rqstlen and rsplen were changed to __le32 to fix sparse warnings: drivers/scsi/qla2xxx/qla_nvme.c:402:30: warning: incorrect type in assignment (different base types) drivers/scsi/qla2xxx/qla_nvme.c:402:30: expected restricted __le32 [usertype] cmd_len drivers/scsi/qla2xxx/qla_nvme.c:402:30: got unsigned short [usertype] rsplen drivers/scsi/qla2xxx/qla_nvme.c:507:30: warning: incorrect type in assignment (different base types) drivers/scsi/qla2xxx/qla_nvme.c:507:30: expected restricted __le32 [usertype] cmd_len drivers/scsi/qla2xxx/qla_nvme.c:507:30: got unsigned int [usertype] rqstlen drivers/scsi/qla2xxx/qla_nvme.c:508:30: warning: incorrect type in assignment (different base types) drivers/scsi/qla2xxx/qla_nvme.c:508:30: expected restricted __le32 [usertype] rsp_len drivers/scsi/qla2xxx/qla_nvme.c:508:30: got unsigned int [usertype] rsplen Correct the endianness in qla2xxx driver thus avoiding changes in nvme-fc-driver.h. Fixes: 875386b98857 ("scsi: qla2xxx: Add Unsolicited LS Request and Response Support for NVMe") Signed-off-by: Nilesh Javali Link: https://lore.kernel.org/r/20230831112146.32595-1-njavali@marvell.com Signed-off-by: Martin K. Petersen --- drivers/scsi/qla2xxx/qla_nvme.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'drivers/scsi') diff --git a/drivers/scsi/qla2xxx/qla_nvme.c b/drivers/scsi/qla2xxx/qla_nvme.c index db753d712991..a8ddf356e662 100644 --- a/drivers/scsi/qla2xxx/qla_nvme.c +++ b/drivers/scsi/qla2xxx/qla_nvme.c @@ -399,14 +399,14 @@ static int qla_nvme_xmt_ls_rsp(struct nvme_fc_local_port *lport, nvme->u.nvme.dl = 0; nvme->u.nvme.timeout_sec = 0; nvme->u.nvme.cmd_dma = fd_resp->rspdma; - nvme->u.nvme.cmd_len = fd_resp->rsplen; + nvme->u.nvme.cmd_len = cpu_to_le32(fd_resp->rsplen); nvme->u.nvme.rsp_len = 0; nvme->u.nvme.rsp_dma = 0; nvme->u.nvme.exchange_address = uctx->exchange_address; nvme->u.nvme.nport_handle = uctx->nport_handle; nvme->u.nvme.ox_id = uctx->ox_id; dma_sync_single_for_device(&ha->pdev->dev, nvme->u.nvme.cmd_dma, - le32_to_cpu(fd_resp->rsplen), DMA_TO_DEVICE); + fd_resp->rsplen, DMA_TO_DEVICE); ql_dbg(ql_dbg_unsol, vha, 0x2122, "Unsol lsreq portid=%06x %8phC exchange_address 0x%x ox_id 0x%x hdl 0x%x\n", @@ -504,13 +504,13 @@ static int qla_nvme_ls_req(struct nvme_fc_local_port *lport, nvme->u.nvme.desc = fd; nvme->u.nvme.dir = 0; nvme->u.nvme.dl = 0; - nvme->u.nvme.cmd_len = fd->rqstlen; - nvme->u.nvme.rsp_len = fd->rsplen; + nvme->u.nvme.cmd_len = cpu_to_le32(fd->rqstlen); + nvme->u.nvme.rsp_len = cpu_to_le32(fd->rsplen); nvme->u.nvme.rsp_dma = fd->rspdma; nvme->u.nvme.timeout_sec = fd->timeout; nvme->u.nvme.cmd_dma = fd->rqstdma; dma_sync_single_for_device(&ha->pdev->dev, nvme->u.nvme.cmd_dma, - le32_to_cpu(fd->rqstlen), DMA_TO_DEVICE); + fd->rqstlen, DMA_TO_DEVICE); rval = qla2x00_start_sp(sp); if (rval != QLA_SUCCESS) { -- cgit v1.2.3 From 59f10a05b5c7b675256a66e3161741239889ff80 Mon Sep 17 00:00:00 2001 From: Nilesh Javali Date: Thu, 31 Aug 2023 16:51:46 +0530 Subject: scsi: qla2xxx: Use raw_smp_processor_id() instead of smp_processor_id() The following call trace was observed: localhost kernel: nvme nvme0: NVME-FC{0}: controller connect complete localhost kernel: BUG: using smp_processor_id() in preemptible [00000000] code: kworker/u129:4/75092 localhost kernel: nvme nvme0: NVME-FC{0}: new ctrl: NQN "nqn.1992-08.com.netapp:sn.b42d198afb4d11ecad6d00a098d6abfa:subsystem.PR_Channel2022_RH84_subsystem_291" localhost kernel: caller is qla_nvme_post_cmd+0x216/0x1380 [qla2xxx] localhost kernel: CPU: 6 PID: 75092 Comm: kworker/u129:4 Kdump: loaded Tainted: G B W OE --------- --- 5.14.0-70.22.1.el9_0.x86_64+debug #1 localhost kernel: Hardware name: HPE ProLiant XL420 Gen10/ProLiant XL420 Gen10, BIOS U39 01/13/2022 localhost kernel: Workqueue: nvme-wq nvme_async_event_work [nvme_core] localhost kernel: Call Trace: localhost kernel: dump_stack_lvl+0x57/0x7d localhost kernel: check_preemption_disabled+0xc8/0xd0 localhost kernel: qla_nvme_post_cmd+0x216/0x1380 [qla2xxx] Use raw_smp_processor_id() instead of smp_processor_id(). Also use queue_work() across the driver instead of queue_work_on() thus avoiding usage of smp_processor_id() when CONFIG_DEBUG_PREEMPT is enabled. Cc: stable@vger.kernel.org Suggested-by: John Garry Signed-off-by: Nilesh Javali Link: https://lore.kernel.org/r/20230831112146.32595-2-njavali@marvell.com Signed-off-by: Martin K. Petersen --- drivers/scsi/qla2xxx/qla_inline.h | 2 +- drivers/scsi/qla2xxx/qla_isr.c | 6 +++--- drivers/scsi/qla2xxx/qla_target.c | 3 +-- drivers/scsi/qla2xxx/tcm_qla2xxx.c | 4 ++-- 4 files changed, 7 insertions(+), 8 deletions(-) (limited to 'drivers/scsi') diff --git a/drivers/scsi/qla2xxx/qla_inline.h b/drivers/scsi/qla2xxx/qla_inline.h index 0556969f6dc1..a4a56ab0ba74 100644 --- a/drivers/scsi/qla2xxx/qla_inline.h +++ b/drivers/scsi/qla2xxx/qla_inline.h @@ -577,7 +577,7 @@ fcport_is_bigger(fc_port_t *fcport) static inline struct qla_qpair * qla_mapq_nvme_select_qpair(struct qla_hw_data *ha, struct qla_qpair *qpair) { - int cpuid = smp_processor_id(); + int cpuid = raw_smp_processor_id(); if (qpair->cpuid != cpuid && ha->qp_cpu_map[cpuid]) { diff --git a/drivers/scsi/qla2xxx/qla_isr.c b/drivers/scsi/qla2xxx/qla_isr.c index e98788191897..d48007e18288 100644 --- a/drivers/scsi/qla2xxx/qla_isr.c +++ b/drivers/scsi/qla2xxx/qla_isr.c @@ -3965,7 +3965,7 @@ void qla24xx_process_response_queue(struct scsi_qla_host *vha, if (!ha->flags.fw_started) return; - if (rsp->qpair->cpuid != smp_processor_id() || !rsp->qpair->rcv_intr) { + if (rsp->qpair->cpuid != raw_smp_processor_id() || !rsp->qpair->rcv_intr) { rsp->qpair->rcv_intr = 1; if (!rsp->qpair->cpu_mapped) @@ -4468,7 +4468,7 @@ qla2xxx_msix_rsp_q(int irq, void *dev_id) } ha = qpair->hw; - queue_work_on(smp_processor_id(), ha->wq, &qpair->q_work); + queue_work(ha->wq, &qpair->q_work); return IRQ_HANDLED; } @@ -4494,7 +4494,7 @@ qla2xxx_msix_rsp_q_hs(int irq, void *dev_id) wrt_reg_dword(®->hccr, HCCRX_CLR_RISC_INT); spin_unlock_irqrestore(&ha->hardware_lock, flags); - queue_work_on(smp_processor_id(), ha->wq, &qpair->q_work); + queue_work(ha->wq, &qpair->q_work); return IRQ_HANDLED; } diff --git a/drivers/scsi/qla2xxx/qla_target.c b/drivers/scsi/qla2xxx/qla_target.c index 2b815a9928ea..2ef2dbac0db2 100644 --- a/drivers/scsi/qla2xxx/qla_target.c +++ b/drivers/scsi/qla2xxx/qla_target.c @@ -4425,8 +4425,7 @@ static int qlt_handle_cmd_for_atio(struct scsi_qla_host *vha, queue_work_on(cmd->se_cmd.cpuid, qla_tgt_wq, &cmd->work); } else if (ha->msix_count) { if (cmd->atio.u.isp24.fcp_cmnd.rddata) - queue_work_on(smp_processor_id(), qla_tgt_wq, - &cmd->work); + queue_work(qla_tgt_wq, &cmd->work); else queue_work_on(cmd->se_cmd.cpuid, qla_tgt_wq, &cmd->work); diff --git a/drivers/scsi/qla2xxx/tcm_qla2xxx.c b/drivers/scsi/qla2xxx/tcm_qla2xxx.c index 3b5ba4b47b3b..68a0e6a2fb6e 100644 --- a/drivers/scsi/qla2xxx/tcm_qla2xxx.c +++ b/drivers/scsi/qla2xxx/tcm_qla2xxx.c @@ -310,7 +310,7 @@ static void tcm_qla2xxx_free_cmd(struct qla_tgt_cmd *cmd) cmd->trc_flags |= TRC_CMD_DONE; INIT_WORK(&cmd->work, tcm_qla2xxx_complete_free); - queue_work_on(smp_processor_id(), tcm_qla2xxx_free_wq, &cmd->work); + queue_work(tcm_qla2xxx_free_wq, &cmd->work); } /* @@ -547,7 +547,7 @@ static void tcm_qla2xxx_handle_data(struct qla_tgt_cmd *cmd) cmd->trc_flags |= TRC_DATA_IN; cmd->cmd_in_wq = 1; INIT_WORK(&cmd->work, tcm_qla2xxx_handle_data_work); - queue_work_on(smp_processor_id(), tcm_qla2xxx_free_wq, &cmd->work); + queue_work(tcm_qla2xxx_free_wq, &cmd->work); } static int tcm_qla2xxx_chk_dif_tags(uint32_t tag) -- cgit v1.2.3 From d0b0822e32dbae80bbcb3cc86f34d28539d913df Mon Sep 17 00:00:00 2001 From: Jinjie Ruan Date: Thu, 31 Aug 2023 22:09:29 +0800 Subject: scsi: qla2xxx: Fix NULL vs IS_ERR() bug for debugfs_create_dir() Since both debugfs_create_dir() and debugfs_create_file() return ERR_PTR and never NULL, use IS_ERR() instead of checking for NULL. Fixes: 1e98fb0f9208 ("scsi: qla2xxx: Setup debugfs entries for remote ports") Signed-off-by: Jinjie Ruan Link: https://lore.kernel.org/r/20230831140930.3166359-1-ruanjinjie@huawei.com Signed-off-by: Martin K. Petersen --- drivers/scsi/qla2xxx/qla_dfs.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'drivers/scsi') diff --git a/drivers/scsi/qla2xxx/qla_dfs.c b/drivers/scsi/qla2xxx/qla_dfs.c index f060e593685d..a7a364760b80 100644 --- a/drivers/scsi/qla2xxx/qla_dfs.c +++ b/drivers/scsi/qla2xxx/qla_dfs.c @@ -116,7 +116,7 @@ qla2x00_dfs_create_rport(scsi_qla_host_t *vha, struct fc_port *fp) sprintf(wwn, "pn-%016llx", wwn_to_u64(fp->port_name)); fp->dfs_rport_dir = debugfs_create_dir(wwn, vha->dfs_rport_root); - if (!fp->dfs_rport_dir) + if (IS_ERR(fp->dfs_rport_dir)) return; if (NVME_TARGET(vha->hw, fp)) debugfs_create_file("dev_loss_tmo", 0600, fp->dfs_rport_dir, @@ -708,14 +708,14 @@ create_nodes: if (IS_QLA27XX(ha) || IS_QLA83XX(ha) || IS_QLA28XX(ha)) { ha->tgt.dfs_naqp = debugfs_create_file("naqp", 0400, ha->dfs_dir, vha, &dfs_naqp_ops); - if (!ha->tgt.dfs_naqp) { + if (IS_ERR(ha->tgt.dfs_naqp)) { ql_log(ql_log_warn, vha, 0xd011, "Unable to create debugFS naqp node.\n"); goto out; } } vha->dfs_rport_root = debugfs_create_dir("rports", ha->dfs_dir); - if (!vha->dfs_rport_root) { + if (IS_ERR(vha->dfs_rport_root)) { ql_log(ql_log_warn, vha, 0xd012, "Unable to create debugFS rports node.\n"); goto out; -- cgit v1.2.3 From 7df0b2605489bef3f4223ad66f1f9bb8d50d4cd2 Mon Sep 17 00:00:00 2001 From: Javed Hasan Date: Fri, 1 Sep 2023 11:36:46 +0530 Subject: scsi: qedf: Add synchronization between I/O completions and abort Avoid race condition between I/O completion and abort processing by protecting the cmd_type with the rport lock. Signed-off-by: Javed Hasan Signed-off-by: Saurav Kashyap Link: https://lore.kernel.org/r/20230901060646.27885-1-skashyap@marvell.com Signed-off-by: Martin K. Petersen --- drivers/scsi/qedf/qedf_io.c | 10 ++++++++-- drivers/scsi/qedf/qedf_main.c | 7 ++++++- 2 files changed, 14 insertions(+), 3 deletions(-) (limited to 'drivers/scsi') diff --git a/drivers/scsi/qedf/qedf_io.c b/drivers/scsi/qedf/qedf_io.c index 4750ec5789a8..10fe3383855c 100644 --- a/drivers/scsi/qedf/qedf_io.c +++ b/drivers/scsi/qedf/qedf_io.c @@ -1904,6 +1904,7 @@ int qedf_initiate_abts(struct qedf_ioreq *io_req, bool return_scsi_cmd_on_abts) goto drop_rdata_kref; } + spin_lock_irqsave(&fcport->rport_lock, flags); if (!test_bit(QEDF_CMD_OUTSTANDING, &io_req->flags) || test_bit(QEDF_CMD_IN_CLEANUP, &io_req->flags) || test_bit(QEDF_CMD_IN_ABORT, &io_req->flags)) { @@ -1911,17 +1912,20 @@ int qedf_initiate_abts(struct qedf_ioreq *io_req, bool return_scsi_cmd_on_abts) "io_req xid=0x%x sc_cmd=%p already in cleanup or abort processing or already completed.\n", io_req->xid, io_req->sc_cmd); rc = 1; + spin_unlock_irqrestore(&fcport->rport_lock, flags); goto drop_rdata_kref; } + /* Set the command type to abort */ + io_req->cmd_type = QEDF_ABTS; + spin_unlock_irqrestore(&fcport->rport_lock, flags); + kref_get(&io_req->refcount); xid = io_req->xid; qedf->control_requests++; qedf->packet_aborts++; - /* Set the command type to abort */ - io_req->cmd_type = QEDF_ABTS; io_req->return_scsi_cmd_on_abts = return_scsi_cmd_on_abts; set_bit(QEDF_CMD_IN_ABORT, &io_req->flags); @@ -2210,7 +2214,9 @@ process_els: refcount, fcport, fcport->rdata->ids.port_id); /* Cleanup cmds re-use the same TID as the original I/O */ + spin_lock_irqsave(&fcport->rport_lock, flags); io_req->cmd_type = QEDF_CLEANUP; + spin_unlock_irqrestore(&fcport->rport_lock, flags); io_req->return_scsi_cmd_on_abts = return_scsi_cmd_on_abts; init_completion(&io_req->cleanup_done); diff --git a/drivers/scsi/qedf/qedf_main.c b/drivers/scsi/qedf/qedf_main.c index 7825765c936c..91f3f1d7098e 100644 --- a/drivers/scsi/qedf/qedf_main.c +++ b/drivers/scsi/qedf/qedf_main.c @@ -2805,6 +2805,8 @@ void qedf_process_cqe(struct qedf_ctx *qedf, struct fcoe_cqe *cqe) struct qedf_ioreq *io_req; struct qedf_rport *fcport; u32 comp_type; + u8 io_comp_type; + unsigned long flags; comp_type = (cqe->cqe_data >> FCOE_CQE_CQE_TYPE_SHIFT) & FCOE_CQE_CQE_TYPE_MASK; @@ -2838,11 +2840,14 @@ void qedf_process_cqe(struct qedf_ctx *qedf, struct fcoe_cqe *cqe) return; } + spin_lock_irqsave(&fcport->rport_lock, flags); + io_comp_type = io_req->cmd_type; + spin_unlock_irqrestore(&fcport->rport_lock, flags); switch (comp_type) { case FCOE_GOOD_COMPLETION_CQE_TYPE: atomic_inc(&fcport->free_sqes); - switch (io_req->cmd_type) { + switch (io_comp_type) { case QEDF_SCSI_CMD: qedf_scsi_completion(qedf, cqe, io_req); break; -- cgit v1.2.3