scsi: core: add support for request batching

This allows a list of requests to be issued, with the LLD only writing the
hardware doorbell when necessary, after the last request was prepared.
This is more efficient if we have lists of requests to issue, particularly
on virtualized hardware, where writing the doorbell is more expensive than
on real hardware.

The use case for this is plugged IO, where blk-mq flushes a batch of
requests all at once.

The API is the same as for blk-mq, just with blk-mq concepts tweaked to
fit the SCSI subsystem API: the "last" flag in blk_mq_queue_data becomes a
flag in scsi_cmnd, while the queue_num in the commit_rqs callback is
extracted from the hctx and passed as a parameter.

The only complication is that blk-mq uses different plugging heuristics
depending on whether commit_rqs is present or not.  So we have two
different sets of blk_mq_ops and pick one depending on whether the
scsi_host template uses commit_rqs or not.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

1 files changed, 14 insertions, 2 deletions

diff --git a/include/scsi/scsi_host.h b/include/scsi/scsi_host.h
index cc139dbd71e5..31e0d6ca1eba 100644
--- a/include/scsi/scsi_host.h
+++ b/include/scsi/scsi_host.h
@@ -80,8 +80,10 @@ struct scsi_host_template {
 	 * command block to the LLDD.  When the driver finished
 	 * processing the command the done callback is invoked.
 	 *
-	 * If queuecommand returns 0, then the HBA has accepted the
-	 * command.  The done() function must be called on the command
+	 * If queuecommand returns 0, then the driver has accepted the
+	 * command.  It must also push it to the HBA if the scsi_cmnd
+	 * flag SCMD_LAST is set, or if the driver does not implement
+	 * commit_rqs.  The done() function must be called on the command
 	 * when the driver has finished with it. (you may call done on the
 	 * command before queuecommand returns, but in this case you
 	 * *must* return 0 from queuecommand).
@@ -110,6 +112,16 @@ struct scsi_host_template {
 	int (* queuecommand)(struct Scsi_Host *, struct scsi_cmnd *);
 
 	/*
+	 * The commit_rqs function is used to trigger a hardware
+	 * doorbell after some requests have been queued with
+	 * queuecommand, when an error is encountered before sending
+	 * the request with SCMD_LAST set.
+	 *
+	 * STATUS: OPTIONAL
+	 */
+	void (*commit_rqs)(struct Scsi_Host *, u16);
+
+	/*
 	 * This is an error handling strategy routine.  You don't need to
 	 * define one of these if you don't want to - there is a default
 	 * routine that is present that should work in most cases.  For those