On 02/07/14 20:30, Nicholas A. Bellinger wrote:
> All that scsi_debug with NOP'ed REQ_TYPE_FS commands is doing is calling
> scsi_cmd->done() as soon as the descriptor has been dispatched into LLD
> ->queuecommand() code.
>
> It's useful for determining an absolute performance ceiling between
> scsi_mq vs. scsi_request_fn() based codepaths without involving any
> other LLD specific overheads.
Sorry but I'm not convinced that the scsi_debug driver is well suited
for testing scsi-mq. In source file drivers/scsi/scsi_debug.c of kernel
3.14-rc1 I found the following:
static DEF_SCSI_QCMD(scsi_debug_queuecommand)
>From include/scsi/scsi_host.h:
/*
* Temporary #define for host lock push down. Can be removed when all
* drivers have been updated to take advantage of unlocked
* queuecommand.
*
*/
#define DEF_SCSI_QCMD(func_name) \
int func_name(struct Scsi_Host *shost, struct scsi_cmnd *cmd) \
{ \
unsigned long irq_flags; \
int rc; \
spin_lock_irqsave(shost->host_lock, irq_flags); \
scsi_cmd_get_serial(shost, cmd); \
rc = func_name##_lck (cmd, cmd->scsi_done); \
spin_unlock_irqrestore(shost->host_lock, irq_flags); \
return rc; \
}
In other words, all scsi_debug_queuecommand() are serialized. I think
for testing the scsi-mq code properly a SCSI LLD driver is needed that
allows concurrent queuecommand() calls.
Bart.
--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
