scsi: lib/sg_pool.c: improve APIs for allocating sg pool

sg_alloc_table_chained() currently allows the caller to provide one
preallocated SGL and returns if the requested number isn't bigger than
size of that SGL. This is used to inline an SGL for an IO request.

However, scattergather code only allows that size of the 1st preallocated
SGL to be SG_CHUNK_SIZE(128). This means a substantial amount of memory
(4KB) is claimed for the SGL for each IO request. If the I/O is small, it
would be prudent to allocate a smaller SGL.

Introduce an extra parameter to sg_alloc_table_chained() and
sg_free_table_chained() for specifying size of the preallocated SGL.

Both __sg_free_table() and __sg_alloc_table() assume that each SGL has the
same size except for the last one.  Change the code to allow both functions
to accept a variable size for the 1st preallocated SGL.

[mkp: attempted to clarify commit desc]

Cc: Christoph Hellwig <hch@lst.de>
Cc: Bart Van Assche <bvanassche@acm.org>
Cc: Ewan D. Milne <emilne@redhat.com>
Cc: Hannes Reinecke <hare@suse.com>
Cc: Sagi Grimberg <sagi@grimberg.me>
Cc: Chuck Lever <chuck.lever@oracle.com>
Cc: netdev@vger.kernel.org
Cc: linux-nvme@lists.infradead.org
Suggested-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

3 files changed, 10 insertions, 8 deletions

diff --git a/drivers/nvme/host/fc.c b/drivers/nvme/host/fc.c
index dd8169bbf0d2..46811caac9d2 100644
--- a/drivers/nvme/host/fc.c
+++ b/drivers/nvme/host/fc.c
@@ -2112,7 +2112,8 @@ nvme_fc_map_data(struct nvme_fc_ctrl *ctrl, struct request *rq,
 
 	freq->sg_table.sgl = freq->first_sgl;
 	ret = sg_alloc_table_chained(&freq->sg_table,
-			blk_rq_nr_phys_segments(rq), freq->sg_table.sgl);
+			blk_rq_nr_phys_segments(rq), freq->sg_table.sgl,
+			SG_CHUNK_SIZE);
 	if (ret)
 		return -ENOMEM;
 
@@ -2122,7 +2123,7 @@ nvme_fc_map_data(struct nvme_fc_ctrl *ctrl, struct request *rq,
 	freq->sg_cnt = fc_dma_map_sg(ctrl->lport->dev, freq->sg_table.sgl,
 				op->nents, dir);
 	if (unlikely(freq->sg_cnt <= 0)) {
-		sg_free_table_chained(&freq->sg_table, true);
+		sg_free_table_chained(&freq->sg_table, SG_CHUNK_SIZE);
 		freq->sg_cnt = 0;
 		return -EFAULT;
 	}
@@ -2148,7 +2149,7 @@ nvme_fc_unmap_data(struct nvme_fc_ctrl *ctrl, struct request *rq,
 
 	nvme_cleanup_cmd(rq);
 
-	sg_free_table_chained(&freq->sg_table, true);
+	sg_free_table_chained(&freq->sg_table, SG_CHUNK_SIZE);
 
 	freq->sg_cnt = 0;
 }
diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c
index f383146e7d0f..f7ea19b45798 100644
--- a/drivers/nvme/host/rdma.c
+++ b/drivers/nvme/host/rdma.c
@@ -1133,7 +1133,7 @@ static void nvme_rdma_unmap_data(struct nvme_rdma_queue *queue,
 				    WRITE ? DMA_TO_DEVICE : DMA_FROM_DEVICE);
 
 	nvme_cleanup_cmd(rq);
-	sg_free_table_chained(&req->sg_table, true);
+	sg_free_table_chained(&req->sg_table, SG_CHUNK_SIZE);
 }
 
 static int nvme_rdma_set_sg_null(struct nvme_command *c)
@@ -1248,7 +1248,8 @@ static int nvme_rdma_map_data(struct nvme_rdma_queue *queue,
 
 	req->sg_table.sgl = req->first_sgl;
 	ret = sg_alloc_table_chained(&req->sg_table,
-			blk_rq_nr_phys_segments(rq), req->sg_table.sgl);
+			blk_rq_nr_phys_segments(rq), req->sg_table.sgl,
+			SG_CHUNK_SIZE);
 	if (ret)
 		return -ENOMEM;
 
@@ -1288,7 +1289,7 @@ out_unmap_sg:
 			req->nents, rq_data_dir(rq) ==
 			WRITE ? DMA_TO_DEVICE : DMA_FROM_DEVICE);
 out_free_table:
-	sg_free_table_chained(&req->sg_table, true);
+	sg_free_table_chained(&req->sg_table, SG_CHUNK_SIZE);
 	return ret;
 }
 
diff --git a/drivers/nvme/target/loop.c b/drivers/nvme/target/loop.c
index 9e211ad6bdd3..b16dc3981c69 100644
--- a/drivers/nvme/target/loop.c
+++ b/drivers/nvme/target/loop.c
@@ -77,7 +77,7 @@ static void nvme_loop_complete_rq(struct request *req)
 	struct nvme_loop_iod *iod = blk_mq_rq_to_pdu(req);
 
 	nvme_cleanup_cmd(req);
-	sg_free_table_chained(&iod->sg_table, true);
+	sg_free_table_chained(&iod->sg_table, SG_CHUNK_SIZE);
 	nvme_complete_rq(req);
 }
 
@@ -157,7 +157,7 @@ static blk_status_t nvme_loop_queue_rq(struct blk_mq_hw_ctx *hctx,
 		iod->sg_table.sgl = iod->first_sgl;
 		if (sg_alloc_table_chained(&iod->sg_table,
 				blk_rq_nr_phys_segments(req),
-				iod->sg_table.sgl))
+				iod->sg_table.sgl, SG_CHUNK_SIZE))
 			return BLK_STS_RESOURCE;
 
 		iod->req.sg = iod->sg_table.sgl;