NVMe: avoid kmalloc/kfree for smaller IO

Currently we allocate an nvme_iod for each IO, which holds the
sg list, prps, and other IO related info. Set a threshold of
2 pages and/or 8KB of data, below which we can just embed this
in the per-command pdu in blk-mq. For any IO at or below
NVME_INT_PAGES and NVME_INT_BYTES, we save a kmalloc and kfree.

For higher IOPS, this saves up to 1% of CPU time.

Signed-off-by: Jens Axboe <axboe@fb.com>
Reviewed-by: Keith Busch <keith.busch@intel.com>

include/linux/nvme.h

@@ -132,13 +132,12 @@ struct nvme_ns {
  * allocated to store the PRP list.
  */
 struct nvme_iod {
-	void *private;		/* For the use of the submitter of the I/O */
+	unsigned long private;	/* For the use of the submitter of the I/O */
 	int npages;		/* In the PRP list. 0 means small pool in use */
 	int offset;		/* Of PRP list */
 	int nents;		/* Used in scatterlist */
 	int length;		/* Of data, in bytes */
 	dma_addr_t first_dma;
-	struct list_head node;
 	struct scatterlist sg[0];
 };