scsi: pm8001: Expose hardware queues for pm80xx

In commit 7266bb90ee0d ("scsi: pm80xx: Increase number of supported
queues"), support for 80xx chip was improved by enabling multiple HW
queues.

In this, like other SCSI MQ HBA drivers at the time, the HW queues were not
exposed to upper layer, and instead the driver managed the queues
internally.

However, this management duplicates blk-mq code. In addition, the HW queue
management is sub-optimal for a system where the number of CPUs exceeds the
HW queues - this is because queues are selected in a round-robin fashion,
when it would be better to make adjacent CPUs submit on the same queue. And
finally, the affinity of the completion queue interrupts is not set to
mirror the cpu<->HQ queue mapping, which is suboptimal.

As such, for when MSIX is supported, expose HW queues to upper layer. We
always use queue index #0 for "internal" commands, i.e. anything which does
not come from the block layer, so omit this from the affinity spreading.

Link: https://lore.kernel.org/r/1654879602-33497-5-git-send-email-john.garry@huawei.com
Tested-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Signed-off-by: John Garry <john.garry@huawei.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

diff --git a/drivers/scsi/pm8001/pm8001_init.c b/drivers/scsi/pm8001/pm8001_init.c
index 171f0a2cf2d64ed6151be2f4a2aed2fe191ee8d9..a3a972c2e532370d395c6956deb52ba794efa199 100644 (file)
--- a/drivers/scsi/pm8001/pm8001_init.c
+++ b/drivers/scsi/pm8001/pm8001_init.c
@@ -81,6 +81,18 @@ LIST_HEAD(hba_list);
 
 struct workqueue_struct *pm8001_wq;
 
+static int pm8001_map_queues(struct Scsi_Host *shost)
+{
+       struct sas_ha_struct *sha = SHOST_TO_SAS_HA(shost);
+       struct pm8001_hba_info *pm8001_ha = sha->lldd_ha;
+       struct blk_mq_queue_map *qmap = &shost->tag_set.map[HCTX_TYPE_DEFAULT];
+
+       if (pm8001_ha->number_of_intr > 1)
+               blk_mq_pci_map_queues(qmap, pm8001_ha->pdev, 1);
+
+       return blk_mq_map_queues(qmap);
+}
+
 /*
  * The main structure which LLDD must register for scsi core.
  */
@@ -110,6 +122,7 @@ static struct scsi_host_template pm8001_sht = {
        .shost_groups           = pm8001_host_groups,
        .track_queue_depth      = 1,
        .cmd_per_lun            = 32,
+       .map_queues             = pm8001_map_queues,
 };
 
 /*
@@ -928,31 +941,35 @@ static int pm8001_configure_phy_settings(struct pm8001_hba_info *pm8001_ha)
  */
 static u32 pm8001_setup_msix(struct pm8001_hba_info *pm8001_ha)
 {
-       u32 number_of_intr;
-       int rc, cpu_online_count;
        unsigned int allocated_irq_vectors;
+       int rc;
 
        /* SPCv controllers supports 64 msi-x */
        if (pm8001_ha->chip_id == chip_8001) {
-               number_of_intr = 1;
+               rc = pci_alloc_irq_vectors(pm8001_ha->pdev, 1, 1,
+                                          PCI_IRQ_MSIX);
        } else {
-               number_of_intr = PM8001_MAX_MSIX_VEC;
+               /*
+                * Queue index #0 is used always for housekeeping, so don't
+                * include in the affinity spreading.
+                */
+               struct irq_affinity desc = {
+                       .pre_vectors = 1,
+               };
+               rc = pci_alloc_irq_vectors_affinity(
+                               pm8001_ha->pdev, 2, PM8001_MAX_MSIX_VEC,
+                               PCI_IRQ_MSIX | PCI_IRQ_AFFINITY, &desc);
        }
 
-       cpu_online_count = num_online_cpus();
-       number_of_intr = min_t(int, cpu_online_count, number_of_intr);
-       rc = pci_alloc_irq_vectors(pm8001_ha->pdev, number_of_intr,
-                       number_of_intr, PCI_IRQ_MSIX);
        allocated_irq_vectors = rc;
        if (rc < 0)
                return rc;
 
        /* Assigns the number of interrupts */
-       number_of_intr = min_t(int, allocated_irq_vectors, number_of_intr);
-       pm8001_ha->number_of_intr = number_of_intr;
+       pm8001_ha->number_of_intr = allocated_irq_vectors;
 
        /* Maximum queue number updating in HBA structure */
-       pm8001_ha->max_q_num = number_of_intr;
+       pm8001_ha->max_q_num = allocated_irq_vectors;
 
        pm8001_dbg(pm8001_ha, INIT,
                   "pci_alloc_irq_vectors request ret:%d no of intr %d\n",
@@ -1123,8 +1140,19 @@ static int pm8001_pci_probe(struct pci_dev *pdev,
        if (rc)
                goto err_out_enable;
 
+
        PM8001_CHIP_DISP->chip_post_init(pm8001_ha);
 
+       if (pm8001_ha->number_of_intr > 1) {
+               shost->nr_hw_queues = pm8001_ha->number_of_intr - 1;
+               /*
+                * For now, ensure we're not sent too many commands by setting
+                * host_tagset. This is also required if we start using request
+                * tag.
+                */
+               shost->host_tagset = 1;
+       }
+
        rc = scsi_add_host(shost, &pdev->dev);
        if (rc)
                goto err_out_ha_free;

diff --git a/drivers/scsi/pm8001/pm8001_sas.h b/drivers/scsi/pm8001/pm8001_sas.h
index c761a2fd25e50d15894f7c4e05d37bad3918207e..c5e3f380a01cd1390496d3723b979e8e0826d1af 100644 (file)
--- a/drivers/scsi/pm8001/pm8001_sas.h
+++ b/drivers/scsi/pm8001/pm8001_sas.h
@@ -55,6 +55,8 @@
 #include <scsi/scsi_tcq.h>
 #include <scsi/sas_ata.h>
 #include <linux/atomic.h>
+#include <linux/blk-mq.h>
+#include <linux/blk-mq-pci.h>
 #include "pm8001_defs.h"
 
 #define DRV_NAME               "pm80xx"

diff --git a/drivers/scsi/pm8001/pm80xx_hwi.c b/drivers/scsi/pm8001/pm80xx_hwi.c
index d5cc726c034e05ca5a049246fe24cc8957c3d062..3893f704602ef8e66875cce6ebcf7e3008ee7eae 100644 (file)
--- a/drivers/scsi/pm8001/pm80xx_hwi.c
+++ b/drivers/scsi/pm8001/pm80xx_hwi.c
@@ -4351,6 +4351,29 @@ static int check_enc_sat_cmd(struct sas_task *task)
        return ret;
 }
 
+static u32 pm80xx_chip_get_q_index(struct sas_task *task)
+{
+       struct scsi_cmnd *scmd = NULL;
+       u32 blk_tag;
+
+       if (task->uldd_task) {
+               struct ata_queued_cmd *qc;
+
+               if (dev_is_sata(task->dev)) {
+                       qc = task->uldd_task;
+                       scmd = qc->scsicmd;
+               } else {
+                       scmd = task->uldd_task;
+               }
+       }
+
+       if (!scmd)
+               return 0;
+
+       blk_tag = blk_mq_unique_tag(scsi_cmd_to_rq(scmd));
+       return blk_mq_unique_tag_to_hwq(blk_tag);
+}
+
 /**
  * pm80xx_chip_ssp_io_req - send an SSP task to FW
  * @pm8001_ha: our hba card information.
@@ -4366,7 +4389,7 @@ static int pm80xx_chip_ssp_io_req(struct pm8001_hba_info *pm8001_ha,
        u32 tag = ccb->ccb_tag;
        u64 phys_addr, end_addr;
        u32 end_addr_high, end_addr_low;
-       u32 q_index, cpu_id;
+       u32 q_index;
        u32 opc = OPC_INB_SSPINIIOSTART;
 
        memset(&ssp_cmd, 0, sizeof(ssp_cmd));
@@ -4387,8 +4410,7 @@ static int pm80xx_chip_ssp_io_req(struct pm8001_hba_info *pm8001_ha,
        ssp_cmd.ssp_iu.efb_prio_attr |= (task->ssp_task.task_attr & 7);
        memcpy(ssp_cmd.ssp_iu.cdb, task->ssp_task.cmd->cmnd,
                       task->ssp_task.cmd->cmd_len);
-       cpu_id = smp_processor_id();
-       q_index = (u32) (cpu_id) % (pm8001_ha->max_q_num);
+       q_index = pm80xx_chip_get_q_index(task);
 
        /* Check if encryption is set */
        if (pm8001_ha->chip->encrypt &&
@@ -4517,8 +4539,7 @@ static int pm80xx_chip_sata_req(struct pm8001_hba_info *pm8001_ha,
        struct domain_device *dev = task->dev;
        struct pm8001_device *pm8001_ha_dev = dev->lldd_dev;
        struct ata_queued_cmd *qc = task->uldd_task;
-       u32 tag = ccb->ccb_tag;
-       u32 q_index, cpu_id;
+       u32 tag = ccb->ccb_tag, q_index;
        struct sata_start_req sata_cmd;
        u32 hdr_tag, ncg_tag = 0;
        u64 phys_addr, end_addr;
@@ -4528,8 +4549,8 @@ static int pm80xx_chip_sata_req(struct pm8001_hba_info *pm8001_ha,
        unsigned long flags;
        u32 opc = OPC_INB_SATA_HOST_OPSTART;
        memset(&sata_cmd, 0, sizeof(sata_cmd));
-       cpu_id = smp_processor_id();
-       q_index = (u32) (cpu_id) % (pm8001_ha->max_q_num);
+
+       q_index = pm80xx_chip_get_q_index(task);
 
        if (task->data_dir == DMA_NONE && !task->ata_task.use_ncq) {
                ATAP = 0x04; /* no data*/