On Thu, 2020-12-10 at 14:35 -0600, Don Brace wrote:
> * Enhance performance by adding sequential stream detection.
> for R5/R6 sequential write requests.
> * Reduce stripe lock contention with full-stripe write
> operations.
I suppose that "stripe lock" is used by the firmware? Could you
elaborate a bit more how this technique improves performance?
>
> Reviewed-by: Scott Benesh <scott.benesh@xxxxxxxxxxxxx>
> Reviewed-by: Scott Teel <scott.teel@xxxxxxxxxxxxx>
> Signed-off-by: Don Brace <don.brace@xxxxxxxxxxxxx>
> ---
> drivers/scsi/smartpqi/smartpqi.h | 8 +++
> drivers/scsi/smartpqi/smartpqi_init.c | 87
> +++++++++++++++++++++++++++++++--
> 2 files changed, 89 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/scsi/smartpqi/smartpqi.h
> b/drivers/scsi/smartpqi/smartpqi.h
> index a5e271dd2742..343f06e44220 100644
> --- a/drivers/scsi/smartpqi/smartpqi.h
> +++ b/drivers/scsi/smartpqi/smartpqi.h
> @@ -1042,6 +1042,12 @@ struct pqi_scsi_dev_raid_map_data {
>
> #define RAID_CTLR_LUNID "\0\0\0\0\0\0\0\0"
>
> +#define NUM_STREAMS_PER_LUN 8
> +
> +struct pqi_stream_data {
> + u64 next_lba;
> + u32 last_accessed;
> +};
>
> struct pqi_scsi_dev {
> int devtype; /* as reported by INQUIRY
> commmand */
> @@ -1097,6 +1103,7 @@ struct pqi_scsi_dev {
> struct list_head add_list_entry;
> struct list_head delete_list_entry;
>
> + struct pqi_stream_data stream_data[NUM_STREAMS_PER_LUN];
> atomic_t scsi_cmds_outstanding;
> atomic_t raid_bypass_cnt;
> };
> @@ -1296,6 +1303,7 @@ struct pqi_ctrl_info {
> u8 enable_r5_writes : 1;
> u8 enable_r6_writes : 1;
> u8 lv_drive_type_mix_valid : 1;
> + u8 enable_stream_detection : 1;
>
> u8 ciss_report_log_flags;
> u32 max_transfer_encrypted_sas_sata;
> diff --git a/drivers/scsi/smartpqi/smartpqi_init.c
> b/drivers/scsi/smartpqi/smartpqi_init.c
> index fc8fafab480d..96383d047a88 100644
> --- a/drivers/scsi/smartpqi/smartpqi_init.c
> +++ b/drivers/scsi/smartpqi/smartpqi_init.c
> @@ -5721,8 +5721,82 @@ void pqi_prep_for_scsi_done(struct scsi_cmnd
> *scmd)
> atomic_dec(&device->scsi_cmds_outstanding);
> }
>
> -static int pqi_scsi_queue_command(struct Scsi_Host *shost,
> +static bool pqi_is_parity_write_stream(struct pqi_ctrl_info
> *ctrl_info,
> struct scsi_cmnd *scmd)
> +{
> + u32 oldest_jiffies;
> + u8 lru_index;
> + int i;
> + int rc;
> + struct pqi_scsi_dev *device;
> + struct pqi_stream_data *pqi_stream_data;
> + struct pqi_scsi_dev_raid_map_data rmd;
> +
> + if (!ctrl_info->enable_stream_detection)
> + return false;
> +
> + rc = pqi_get_aio_lba_and_block_count(scmd, &rmd);
> + if (rc)
> + return false;
> +
> + /* Check writes only. */
> + if (!rmd.is_write)
> + return false;
> +
> + device = scmd->device->hostdata;
> +
> + /* Check for RAID 5/6 streams. */
> + if (device->raid_level != SA_RAID_5 && device->raid_level !=
> SA_RAID_6)
> + return false;
> +
> + /*
> + * If controller does not support AIO RAID{5,6} writes, need
> to send
> + * requests down non-AIO path.
> + */
> + if ((device->raid_level == SA_RAID_5 && !ctrl_info-
> >enable_r5_writes) ||
> + (device->raid_level == SA_RAID_6 && !ctrl_info-
> >enable_r6_writes))
> + return true;
> +
> + lru_index = 0;
> + oldest_jiffies = INT_MAX;
> + for (i = 0; i < NUM_STREAMS_PER_LUN; i++) {
> + pqi_stream_data = &device->stream_data[i];
> + /*
> + * Check for adjacent request or request is within
> + * the previous request.
> + */
> + if ((pqi_stream_data->next_lba &&
> + rmd.first_block >= pqi_stream_data->next_lba)
> &&
> + rmd.first_block <= pqi_stream_data->next_lba
> +
> + rmd.block_cnt) {
Here you seem to assume that the previous write had the same block_cnt.
What's the justification for that?
> + pqi_stream_data->next_lba = rmd.first_block +
> + rmd.block_cnt;
> + pqi_stream_data->last_accessed = jiffies;
> + return true;
> + }
> +
> + /* unused entry */
> + if (pqi_stream_data->last_accessed == 0) {
> + lru_index = i;
> + break;
> + }
> +
> + /* Find entry with oldest last accessed time. */
> + if (pqi_stream_data->last_accessed <= oldest_jiffies)
> {
> + oldest_jiffies = pqi_stream_data-
> >last_accessed;
> + lru_index = i;
> + }
> + }
> +
> + /* Set LRU entry. */
> + pqi_stream_data = &device->stream_data[lru_index];
> + pqi_stream_data->last_accessed = jiffies;
> + pqi_stream_data->next_lba = rmd.first_block + rmd.block_cnt;
> +
> + return false;
> +}
> +
> +static int pqi_scsi_queue_command(struct Scsi_Host *shost, struct
> scsi_cmnd *scmd)
> {
> int rc;
> struct pqi_ctrl_info *ctrl_info;
> @@ -5768,11 +5842,12 @@ static int pqi_scsi_queue_command(struct
> Scsi_Host *shost,
> raid_bypassed = false;
> if (device->raid_bypass_enabled &&
> !blk_rq_is_passthrough(scmd->request)) {
> - rc =
> pqi_raid_bypass_submit_scsi_cmd(ctrl_info, device,
> - scmd, queue_group);
> - if (rc == 0 || rc == SCSI_MLQUEUE_HOST_BUSY)
> {
> - raid_bypassed = true;
> - atomic_inc(&device->raid_bypass_cnt);
> + if (!pqi_is_parity_write_stream(ctrl_info,
> scmd)) {
> + rc =
> pqi_raid_bypass_submit_scsi_cmd(ctrl_info, device, scmd,
> queue_group);
> + if (rc == 0 || rc ==
> SCSI_MLQUEUE_HOST_BUSY) {
> + raid_bypassed = true;
> + atomic_inc(&device-
> >raid_bypass_cnt);
> + }
> }
> }
> if (!raid_bypassed)
>
