diff options
| author | Dick Kennedy <dick.kennedy@broadcom.com> | 2017-08-23 16:55:42 -0700 |
|---|---|---|
| committer | Martin K. Petersen <martin.petersen@oracle.com> | 2017-08-24 22:29:41 -0400 |
| commit | 66d7ce93a0f5b991d6bf068f797dec49eb8e5c57 (patch) | |
| tree | e232b65e13c122844d6842477eb6e2f8242953f8 /drivers/scsi/lpfc/lpfc_init.c | |
| parent | e3e2863def0262782aec6745bb4c0a86b3dfdd3b (diff) | |
| download | linux-66d7ce93a0f5b991d6bf068f797dec49eb8e5c57.tar.bz2 | |
scsi: lpfc: Fix MRQ > 1 context list handling
Various oops including cpu LOCKUPs were seen.
For asynchronously received ius where the driver must assign exchange
resources, the resources were on a single get (free) list and put list
(finished, waiting to be put on get list). As all cpus are sharing the
lists, an interrupt for a receive frame may have to wait for all the
other cpus to place their done work onto the put list before it can
acquire the lock to pull from the list.
Fix by breaking the resource lists into per-cpu lists or at least more
than 1 list with cpu's sharing the lists). A cpu would allocate from the
free list for its own cpu, and put its done work on the its own put list
- avoiding the contention. As cpu load may vary, when empty, a cpu may
grab from another cpu, thereby changing resource distribution. But
searching for a resource only occurs on 1 or a few cpus until a single
resource can be allocated. if the condition reoccurs, it starts looking
at a different cpu.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Diffstat (limited to 'drivers/scsi/lpfc/lpfc_init.c')
| -rw-r--r-- | drivers/scsi/lpfc/lpfc_init.c | 17 |
1 files changed, 6 insertions, 11 deletions
diff --git a/drivers/scsi/lpfc/lpfc_init.c b/drivers/scsi/lpfc/lpfc_init.c index f82618a8918f..c22b88a08c1b 100644 --- a/drivers/scsi/lpfc/lpfc_init.c +++ b/drivers/scsi/lpfc/lpfc_init.c @@ -1253,6 +1253,7 @@ lpfc_hb_timeout_handler(struct lpfc_hba *phba) unsigned long time_elapsed; uint32_t tick_cqe, max_cqe, val; uint64_t tot, data1, data2, data3; + struct lpfc_nvmet_tgtport *tgtp; struct lpfc_register reg_data; void __iomem *eqdreg = phba->sli4_hba.u.if_type2.EQDregaddr; @@ -1281,13 +1282,11 @@ lpfc_hb_timeout_handler(struct lpfc_hba *phba) /* Check outstanding IO count */ if (phba->cfg_enable_fc4_type & LPFC_ENABLE_NVME) { if (phba->nvmet_support) { - spin_lock(&phba->sli4_hba.nvmet_ctx_get_lock); - spin_lock(&phba->sli4_hba.nvmet_ctx_put_lock); - tot = phba->sli4_hba.nvmet_xri_cnt - - (phba->sli4_hba.nvmet_ctx_get_cnt + - phba->sli4_hba.nvmet_ctx_put_cnt); - spin_unlock(&phba->sli4_hba.nvmet_ctx_put_lock); - spin_unlock(&phba->sli4_hba.nvmet_ctx_get_lock); + tgtp = phba->targetport->private; + /* Calculate outstanding IOs */ + tot = atomic_read(&tgtp->rcv_fcp_cmd_drop); + tot += atomic_read(&tgtp->xmt_fcp_release); + tot = atomic_read(&tgtp->rcv_fcp_cmd_in) - tot; } else { tot = atomic_read(&phba->fc4NvmeIoCmpls); data1 = atomic_read( @@ -5937,8 +5936,6 @@ lpfc_sli4_driver_resource_setup(struct lpfc_hba *phba) spin_lock_init(&phba->sli4_hba.abts_nvme_buf_list_lock); INIT_LIST_HEAD(&phba->sli4_hba.lpfc_abts_nvme_buf_list); INIT_LIST_HEAD(&phba->sli4_hba.lpfc_abts_nvmet_ctx_list); - INIT_LIST_HEAD(&phba->sli4_hba.lpfc_nvmet_ctx_get_list); - INIT_LIST_HEAD(&phba->sli4_hba.lpfc_nvmet_ctx_put_list); INIT_LIST_HEAD(&phba->sli4_hba.lpfc_nvmet_io_wait_list); /* Fast-path XRI aborted CQ Event work queue list */ @@ -5947,8 +5944,6 @@ lpfc_sli4_driver_resource_setup(struct lpfc_hba *phba) /* This abort list used by worker thread */ spin_lock_init(&phba->sli4_hba.sgl_list_lock); - spin_lock_init(&phba->sli4_hba.nvmet_ctx_get_lock); - spin_lock_init(&phba->sli4_hba.nvmet_ctx_put_lock); spin_lock_init(&phba->sli4_hba.nvmet_io_wait_lock); /* |