Menu
▾
▴
[Scst-svn] [master] qla2x00t-32gbit: Use raw_smp_processor_id() instead of smp_processor_id()
|
From: Gleb C. <lna...@ya...> - 2023-10-10 12:10:45
|
Commit: bea4580 GitHub URL: https://github.com/SCST-project/scst/commit/bea458054e05daeabe1b4f6749818191a8055110 Author: Gleb Chesnokov Date: 2023-10-10T15:10:02+03:00 Log Message: ----------- qla2x00t-32gbit: 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: st...@vg... Suggested-by: John Garry <joh...@or...> Signed-off-by: Nilesh Javali <nj...@ma...> Link: https://lore.kernel.org/r/202...@ma... Signed-off-by: Martin K. Petersen <mar...@or...> [ commit 59f10a05b5c7 upstream ] Modified Paths: -------------- qla2x00t-32gbit/qla_inline.h | 2 +- qla2x00t-32gbit/qla_isr.c | 4 ++-- qla2x00t-32gbit/qla_target.c | 3 +-- 3 files changed, 4 insertions(+), 5 deletions(-) =================================================================== diff --git a/qla2x00t-32gbit/qla_inline.h b/qla2x00t-32gbit/qla_inline.h index 6ce9094..8746184 100644 --- a/qla2x00t-32gbit/qla_inline.h +++ b/qla2x00t-32gbit/qla_inline.h @@ -585,7 +585,7 @@ 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/qla2x00t-32gbit/qla_isr.c b/qla2x00t-32gbit/qla_isr.c index 944e8aa..9b89e7e 100644 --- a/qla2x00t-32gbit/qla_isr.c +++ b/qla2x00t-32gbit/qla_isr.c @@ -4349,7 +4349,7 @@ 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; } @@ -4375,7 +4375,7 @@ 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/qla2x00t-32gbit/qla_target.c b/qla2x00t-32gbit/qla_target.c index 564819a..6ef76bd 100644 --- a/qla2x00t-32gbit/qla_target.c +++ b/qla2x00t-32gbit/qla_target.c @@ -4481,8 +4481,7 @@ static int qlt_handle_cmd_for_atio(struct scsi_qla_host *vha, #endif } else if (ha->msix_count) { if (cmd->atio.u.isp24.fcp_cmnd.rddata) - queue_work_on(raw_smp_processor_id(), qla_tgt_wq, - &cmd->work); + queue_work(qla_tgt_wq, &cmd->work); else #if HAVE_SE_CMD_CPUID queue_work_on(cmd->se_cmd.cpuid, qla_tgt_wq, |