Re: [PATCH RESEND number 2] libata: eliminate the home grown dma padding in favour of that provided by the block layer

[Tejun Heo <htejun@gmail.com>]

And, here's working version.  I'll splite and post them tomorrow.

Thanks.

Index: work/block/blk-core.c
===================================================================
--- work.orig/block/blk-core.c
+++ work/block/blk-core.c
@@ -116,6 +116,7 @@ void rq_init(struct request_queue *q, st
 	rq->ref_count = 1;
 	rq->q = q;
 	rq->special = NULL;
+	rq->raw_data_len = 0;
 	rq->data_len = 0;
 	rq->data = NULL;
 	rq->nr_phys_segments = 0;
@@ -1982,6 +1983,7 @@ void blk_rq_bio_prep(struct request_queu
 	rq->hard_cur_sectors = rq->current_nr_sectors;
 	rq->hard_nr_sectors = rq->nr_sectors = bio_sectors(bio);
 	rq->buffer = bio_data(bio);
+	rq->raw_data_len = bio->bi_size;
 	rq->data_len = bio->bi_size;
 
 	rq->bio = rq->biotail = bio;
Index: work/block/blk-map.c
===================================================================
--- work.orig/block/blk-map.c
+++ work/block/blk-map.c
@@ -19,6 +19,7 @@ int blk_rq_append_bio(struct request_que
 		rq->biotail->bi_next = bio;
 		rq->biotail = bio;
 
+		rq->raw_data_len += bio->bi_size;
 		rq->data_len += bio->bi_size;
 	}
 	return 0;
@@ -139,6 +140,25 @@ int blk_rq_map_user(struct request_queue
 		ubuf += ret;
 	}
 
+	/*
+	 * __blk_rq_map_user() copies the buffers if starting address
+	 * or length aren't aligned.  As the copied buffer is always
+	 * page aligned, we know for a fact that there's enough room
+	 * for padding.  Extend the last bio and update rq->data_len
+	 * accordingly.
+	 *
+	 * On unmap, bio_uncopy_user() will use unmodified
+	 * bio_map_data pointed to by bio->bi_private.
+	 */
+	if (len & queue_dma_alignment(q)) {
+		unsigned int pad_len = (queue_dma_alignment(q) & ~len) + 1;
+		struct bio *bio = rq->biotail;
+
+		bio->bi_io_vec[bio->bi_vcnt - 1].bv_len += pad_len;
+		bio->bi_size += pad_len;
+		rq->data_len += pad_len;
+	}
+
 	rq->buffer = rq->data = NULL;
 	return 0;
 unmap_rq:
Index: work/include/linux/blkdev.h
===================================================================
--- work.orig/include/linux/blkdev.h
+++ work/include/linux/blkdev.h
@@ -214,6 +214,7 @@ struct request {
 	unsigned int cmd_len;
 	unsigned char cmd[BLK_MAX_CDB];
 
+	unsigned int raw_data_len;
 	unsigned int data_len;
 	unsigned int sense_len;
 	void *data;
@@ -256,6 +257,7 @@ struct bio_vec;
 typedef int (merge_bvec_fn) (struct request_queue *, struct bio *, struct bio_vec *);
 typedef void (prepare_flush_fn) (struct request_queue *, struct request *);
 typedef void (softirq_done_fn)(struct request *);
+typedef int (dma_drain_needed_fn)(struct request *);
 
 enum blk_queue_state {
 	Queue_down,
@@ -292,6 +294,7 @@ struct request_queue
 	merge_bvec_fn		*merge_bvec_fn;
 	prepare_flush_fn	*prepare_flush_fn;
 	softirq_done_fn		*softirq_done_fn;
+	dma_drain_needed_fn	*dma_drain_needed;
 
 	/*
 	 * Dispatch queue sorting
@@ -696,8 +699,9 @@ extern void blk_queue_max_hw_segments(st
 extern void blk_queue_max_segment_size(struct request_queue *, unsigned int);
 extern void blk_queue_hardsect_size(struct request_queue *, unsigned short);
 extern void blk_queue_stack_limits(struct request_queue *t, struct request_queue *b);
-extern int blk_queue_dma_drain(struct request_queue *q, void *buf,
-			       unsigned int size);
+extern int blk_queue_dma_drain(struct request_queue *q,
+			       dma_drain_needed_fn *dma_drain_needed,
+			       void *buf, unsigned int size);
 extern void blk_queue_segment_boundary(struct request_queue *, unsigned long);
 extern void blk_queue_prep_rq(struct request_queue *, prep_rq_fn *pfn);
 extern void blk_queue_merge_bvec(struct request_queue *, merge_bvec_fn *);
Index: work/block/blk-merge.c
===================================================================
--- work.orig/block/blk-merge.c
+++ work/block/blk-merge.c
@@ -220,7 +220,7 @@ new_segment:
 		bvprv = bvec;
 	} /* segments in rq */
 
-	if (q->dma_drain_size) {
+	if (q->dma_drain_size && q->dma_drain_needed(rq)) {
 		sg->page_link &= ~0x02;
 		sg = sg_next(sg);
 		sg_set_page(sg, virt_to_page(q->dma_drain_buffer),
@@ -228,6 +228,7 @@ new_segment:
 			    ((unsigned long)q->dma_drain_buffer) &
 			    (PAGE_SIZE - 1));
 		nsegs++;
+		rq->data_len += q->dma_drain_size;
 	}
 
 	if (sg)
Index: work/block/bsg.c
===================================================================
--- work.orig/block/bsg.c
+++ work/block/bsg.c
@@ -437,14 +437,14 @@ static int blk_complete_sgv4_hdr_rq(stru
 	}
 
 	if (rq->next_rq) {
-		hdr->dout_resid = rq->data_len;
-		hdr->din_resid = rq->next_rq->data_len;
+		hdr->dout_resid = rq->raw_data_len;
+		hdr->din_resid = rq->next_rq->raw_data_len;
 		blk_rq_unmap_user(bidi_bio);
 		blk_put_request(rq->next_rq);
 	} else if (rq_data_dir(rq) == READ)
-		hdr->din_resid = rq->data_len;
+		hdr->din_resid = rq->raw_data_len;
 	else
-		hdr->dout_resid = rq->data_len;
+		hdr->dout_resid = rq->raw_data_len;
 
 	/*
 	 * If the request generated a negative error number, return it
Index: work/block/scsi_ioctl.c
===================================================================
--- work.orig/block/scsi_ioctl.c
+++ work/block/scsi_ioctl.c
@@ -266,7 +266,7 @@ static int blk_complete_sghdr_rq(struct 
 	hdr->info = 0;
 	if (hdr->masked_status || hdr->host_status || hdr->driver_status)
 		hdr->info |= SG_INFO_CHECK;
-	hdr->resid = rq->data_len;
+	hdr->resid = rq->raw_data_len;
 	hdr->sb_len_wr = 0;
 
 	if (rq->sense_len && hdr->sbp) {
@@ -528,6 +528,7 @@ static int __blk_send_generic(struct req
 	rq = blk_get_request(q, WRITE, __GFP_WAIT);
 	rq->cmd_type = REQ_TYPE_BLOCK_PC;
 	rq->data = NULL;
+	rq->raw_data_len = 0;
 	rq->data_len = 0;
 	rq->timeout = BLK_DEFAULT_SG_TIMEOUT;
 	memset(rq->cmd, 0, sizeof(rq->cmd));
Index: work/drivers/scsi/scsi_lib.c
===================================================================
--- work.orig/drivers/scsi/scsi_lib.c
+++ work/drivers/scsi/scsi_lib.c
@@ -1015,10 +1015,6 @@ static int scsi_init_sgtable(struct requ
 	}
 
 	req->buffer = NULL;
-	if (blk_pc_request(req))
-		sdb->length = req->data_len;
-	else
-		sdb->length = req->nr_sectors << 9;
 
 	/* 
 	 * Next, walk the list, and fill in the addresses and sizes of
@@ -1027,6 +1023,10 @@ static int scsi_init_sgtable(struct requ
 	count = blk_rq_map_sg(req->q, req, sdb->table.sgl);
 	BUG_ON(count > sdb->table.nents);
 	sdb->table.nents = count;
+	if (blk_pc_request(req))
+		sdb->length = req->data_len;
+	else
+		sdb->length = req->nr_sectors << 9;
 	return BLKPREP_OK;
 }
 
Index: work/block/blk-settings.c
===================================================================
--- work.orig/block/blk-settings.c
+++ work/block/blk-settings.c
@@ -296,6 +296,7 @@ EXPORT_SYMBOL(blk_queue_stack_limits);
  * blk_queue_dma_drain - Set up a drain buffer for excess dma.
  *
  * @q:  the request queue for the device
+ * @dma_drain_needed: fn which returns non-zero if drain is necessary
  * @buf:	physically contiguous buffer
  * @size:	size of the buffer in bytes
  *
@@ -315,14 +316,16 @@ EXPORT_SYMBOL(blk_queue_stack_limits);
  * device can support otherwise there won't be room for the drain
  * buffer.
  */
-int blk_queue_dma_drain(struct request_queue *q, void *buf,
-				unsigned int size)
+extern int blk_queue_dma_drain(struct request_queue *q,
+			       dma_drain_needed_fn *dma_drain_needed,
+			       void *buf, unsigned int size)
 {
 	if (q->max_hw_segments < 2 || q->max_phys_segments < 2)
 		return -EINVAL;
 	/* make room for appending the drain */
 	--q->max_hw_segments;
 	--q->max_phys_segments;
+	q->dma_drain_needed = dma_drain_needed;
 	q->dma_drain_buffer = buf;
 	q->dma_drain_size = size;
 
Index: work/drivers/scsi/sr.c
===================================================================
--- work.orig/drivers/scsi/sr.c
+++ work/drivers/scsi/sr.c
@@ -557,11 +557,30 @@ static void sr_release(struct cdrom_devi
 
 }
 
+static int sr_drain_needed(struct request *rq)
+{
+	if (likely(!blk_pc_request(rq)))
+		return 0;
+
+	switch (rq->cmd[0]) {
+	case GPCMD_READ_10:
+	case GPCMD_READ_12:
+	case GPCMD_WRITE_10:
+	case GPCMD_WRITE_12:
+	case GPCMD_WRITE_AND_VERIFY_10:
+		return 0;
+	}
+
+	return 1;
+}
+
 static int sr_probe(struct device *dev)
 {
 	struct scsi_device *sdev = to_scsi_device(dev);
+	struct request_queue *queue = sdev->request_queue;
 	struct gendisk *disk;
-	struct scsi_cd *cd;
+	struct scsi_cd *cd = NULL;
+	void *drain_buf = NULL;
 	int minor, error;
 
 	error = -ENODEV;
@@ -573,11 +592,15 @@ static int sr_probe(struct device *dev)
 	if (!cd)
 		goto fail;
 
+	drain_buf = kmalloc(SR_DRAIN_SIZE, queue->bounce_gfp | GFP_KERNEL);
+	if (!drain_buf)
+		goto fail;
+
 	kref_init(&cd->kref);
 
 	disk = alloc_disk(1);
 	if (!disk)
-		goto fail_free;
+		goto fail;
 
 	spin_lock(&sr_index_lock);
 	minor = find_first_zero_bit(sr_index_bits, SR_DISKS);
@@ -615,13 +638,14 @@ static int sr_probe(struct device *dev)
 
 	/* FIXME: need to handle a get_capabilities failure properly ?? */
 	get_capabilities(cd);
-	blk_queue_prep_rq(sdev->request_queue, sr_prep_fn);
+	blk_queue_prep_rq(queue, sr_prep_fn);
+	blk_queue_dma_drain(queue, sr_drain_needed, drain_buf, SR_DRAIN_SIZE);
 	sr_vendor_init(cd);
 
 	disk->driverfs_dev = &sdev->sdev_gendev;
 	set_capacity(disk, cd->capacity);
 	disk->private_data = &cd->driver;
-	disk->queue = sdev->request_queue;
+	disk->queue = queue;
 	cd->cdi.disk = disk;
 
 	if (register_cdrom(&cd->cdi))
@@ -637,9 +661,9 @@ static int sr_probe(struct device *dev)
 
 fail_put:
 	put_disk(disk);
-fail_free:
-	kfree(cd);
 fail:
+	kfree(cd);
+	kfree(drain_buf);
 	return error;
 }
 
@@ -894,6 +918,12 @@ static void sr_kref_release(struct kref 
 static int sr_remove(struct device *dev)
 {
 	struct scsi_cd *cd = dev_get_drvdata(dev);
+	struct scsi_device *sdev = to_scsi_device(dev);
+	struct request_queue *queue = sdev->request_queue;
+
+	kfree(queue->dma_drain_buffer);
+	queue->dma_drain_buffer = NULL;
+	queue->dma_drain_size = 0;
 
 	del_gendisk(cd->disk);
 
Index: work/drivers/scsi/sr.h
===================================================================
--- work.orig/drivers/scsi/sr.h
+++ work/drivers/scsi/sr.h
@@ -22,6 +22,7 @@
 
 #define MAX_RETRIES	3
 #define SR_TIMEOUT	(30 * HZ)
+#define SR_DRAIN_SIZE	PAGE_SIZE
 
 struct scsi_device;
 
Index: work/drivers/ata/libata-core.c
===================================================================
--- work.orig/drivers/ata/libata-core.c
+++ work/drivers/ata/libata-core.c
@@ -4476,30 +4476,13 @@ void ata_sg_clean(struct ata_queued_cmd 
 	struct ata_port *ap = qc->ap;
 	struct scatterlist *sg = qc->sg;
 	int dir = qc->dma_dir;
-	void *pad_buf = NULL;
 
 	WARN_ON(sg == NULL);
 
-	VPRINTK("unmapping %u sg elements\n", qc->mapped_n_elem);
+	VPRINTK("unmapping %u sg elements\n", qc->n_elem);
 
-	/* if we padded the buffer out to 32-bit bound, and data
-	 * xfer direction is from-device, we must copy from the
-	 * pad buffer back into the supplied buffer
-	 */
-	if (qc->pad_len && !(qc->tf.flags & ATA_TFLAG_WRITE))
-		pad_buf = ap->pad + (qc->tag * ATA_DMA_PAD_SZ);
-
-	if (qc->mapped_n_elem)
-		dma_unmap_sg(ap->dev, sg, qc->mapped_n_elem, dir);
-	/* restore last sg */
-	if (qc->last_sg)
-		*qc->last_sg = qc->saved_last_sg;
-	if (pad_buf) {
-		struct scatterlist *psg = &qc->extra_sg[1];
-		void *addr = kmap_atomic(sg_page(psg), KM_IRQ0);
-		memcpy(addr + psg->offset, pad_buf, qc->pad_len);
-		kunmap_atomic(addr, KM_IRQ0);
-	}
+	if (qc->n_elem)
+		dma_unmap_sg(ap->dev, sg, qc->n_elem, dir);
 
 	qc->flags &= ~ATA_QCFLAG_DMAMAP;
 	qc->sg = NULL;
@@ -4765,97 +4748,6 @@ void ata_sg_init(struct ata_queued_cmd *
 	qc->cursg = qc->sg;
 }
 
-static unsigned int ata_sg_setup_extra(struct ata_queued_cmd *qc,
-				       unsigned int *n_elem_extra,
-				       unsigned int *nbytes_extra)
-{
-	struct ata_port *ap = qc->ap;
-	unsigned int n_elem = qc->n_elem;
-	struct scatterlist *lsg, *copy_lsg = NULL, *tsg = NULL, *esg = NULL;
-
-	*n_elem_extra = 0;
-	*nbytes_extra = 0;
-
-	/* needs padding? */
-	qc->pad_len = qc->nbytes & 3;
-
-	if (likely(!qc->pad_len))
-		return n_elem;
-
-	/* locate last sg and save it */
-	lsg = sg_last(qc->sg, n_elem);
-	qc->last_sg = lsg;
-	qc->saved_last_sg = *lsg;
-
-	sg_init_table(qc->extra_sg, ARRAY_SIZE(qc->extra_sg));
-
-	if (qc->pad_len) {
-		struct scatterlist *psg = &qc->extra_sg[1];
-		void *pad_buf = ap->pad + (qc->tag * ATA_DMA_PAD_SZ);
-		unsigned int offset;
-
-		WARN_ON(qc->dev->class != ATA_DEV_ATAPI);
-
-		memset(pad_buf, 0, ATA_DMA_PAD_SZ);
-
-		/* psg->page/offset are used to copy to-be-written
-		 * data in this function or read data in ata_sg_clean.
-		 */
-		offset = lsg->offset + lsg->length - qc->pad_len;
-		sg_set_page(psg, nth_page(sg_page(lsg), offset >> PAGE_SHIFT),
-			    qc->pad_len, offset_in_page(offset));
-
-		if (qc->tf.flags & ATA_TFLAG_WRITE) {
-			void *addr = kmap_atomic(sg_page(psg), KM_IRQ0);
-			memcpy(pad_buf, addr + psg->offset, qc->pad_len);
-			kunmap_atomic(addr, KM_IRQ0);
-		}
-
-		sg_dma_address(psg) = ap->pad_dma + (qc->tag * ATA_DMA_PAD_SZ);
-		sg_dma_len(psg) = ATA_DMA_PAD_SZ;
-
-		/* Trim the last sg entry and chain the original and
-		 * padding sg lists.
-		 *
-		 * Because chaining consumes one sg entry, one extra
-		 * sg entry is allocated and the last sg entry is
-		 * copied to it if the length isn't zero after padded
-		 * amount is removed.
-		 *
-		 * If the last sg entry is completely replaced by
-		 * padding sg entry, the first sg entry is skipped
-		 * while chaining.
-		 */
-		lsg->length -= qc->pad_len;
-		if (lsg->length) {
-			copy_lsg = &qc->extra_sg[0];
-			tsg = &qc->extra_sg[0];
-		} else {
-			n_elem--;
-			tsg = &qc->extra_sg[1];
-		}
-
-		esg = &qc->extra_sg[1];
-
-		(*n_elem_extra)++;
-		(*nbytes_extra) += 4 - qc->pad_len;
-	}
-
-	if (copy_lsg)
-		sg_set_page(copy_lsg, sg_page(lsg), lsg->length, lsg->offset);
-
-	sg_chain(lsg, 1, tsg);
-	sg_mark_end(esg);
-
-	/* sglist can't start with chaining sg entry, fast forward */
-	if (qc->sg == lsg) {
-		qc->sg = tsg;
-		qc->cursg = tsg;
-	}
-
-	return n_elem;
-}
-
 /**
  *	ata_sg_setup - DMA-map the scatter-gather table associated with a command.
  *	@qc: Command with scatter-gather table to be mapped.
@@ -4872,26 +4764,27 @@ static unsigned int ata_sg_setup_extra(s
 static int ata_sg_setup(struct ata_queued_cmd *qc)
 {
 	struct ata_port *ap = qc->ap;
-	unsigned int n_elem, n_elem_extra, nbytes_extra;
 
 	VPRINTK("ENTER, ata%u\n", ap->print_id);
 
-	n_elem = ata_sg_setup_extra(qc, &n_elem_extra, &nbytes_extra);
+	if (ata_is_atapi(qc->tf.protocol)) {
+		struct scatterlist *sg;
+		int i;
 
-	if (n_elem) {
-		n_elem = dma_map_sg(ap->dev, qc->sg, n_elem, qc->dma_dir);
-		if (n_elem < 1) {
-			/* restore last sg */
-			if (qc->last_sg)
-				*qc->last_sg = qc->saved_last_sg;
-			return -1;
-		}
-		DPRINTK("%d sg elements mapped\n", n_elem);
+		ata_dev_printk(qc->dev, KERN_INFO, "XXX cmd=%02x n_elem=%u nbytes=%u dma_dir=%d\n",
+			       qc->cdb[0], qc->n_elem, qc->nbytes, qc->dma_dir);
+
+		for_each_sg(qc->sg, sg, qc->n_elem, i)
+			ata_dev_printk(qc->dev, KERN_INFO, "YYY pfn=%lu offset=%u length=%u\n",
+				       page_to_pfn(sg_page(sg)), sg->offset, sg->length);
 	}
 
-	qc->n_elem = qc->mapped_n_elem = n_elem;
-	qc->n_elem += n_elem_extra;
-	qc->nbytes += nbytes_extra;
+	qc->n_elem = dma_map_sg(ap->dev, qc->sg, qc->n_elem, qc->dma_dir);
+	if (qc->n_elem < 1)
+		return -1;
+
+	DPRINTK("%d sg elements mapped\n", n_elem);
+
 	qc->flags |= ATA_QCFLAG_DMAMAP;
 
 	return 0;
@@ -5955,9 +5848,6 @@ void ata_qc_issue(struct ata_queued_cmd 
 	 */
 	BUG_ON(ata_is_data(prot) && (!qc->sg || !qc->n_elem || !qc->nbytes));
 
-	/* ata_sg_setup() may update nbytes */
-	qc->raw_nbytes = qc->nbytes;
-
 	if (ata_is_dma(prot) || (ata_is_pio(prot) &&
 				 (ap->flags & ATA_FLAG_PIO_DMA)))
 		if (ata_sg_setup(qc))
@@ -6566,19 +6456,12 @@ void ata_host_resume(struct ata_host *ho
 int ata_port_start(struct ata_port *ap)
 {
 	struct device *dev = ap->dev;
-	int rc;
 
 	ap->prd = dmam_alloc_coherent(dev, ATA_PRD_TBL_SZ, &ap->prd_dma,
 				      GFP_KERNEL);
 	if (!ap->prd)
 		return -ENOMEM;
 
-	rc = ata_pad_alloc(ap, dev);
-	if (rc)
-		return rc;
-
-	DPRINTK("prd alloc, virt %p, dma %llx\n", ap->prd,
-		(unsigned long long)ap->prd_dma);
 	return 0;
 }
 
Index: work/drivers/ata/libata-scsi.c
===================================================================
--- work.orig/drivers/ata/libata-scsi.c
+++ work/drivers/ata/libata-scsi.c
@@ -832,18 +832,11 @@ static void ata_scsi_dev_config(struct s
 	/* configure max sectors */
 	blk_queue_max_sectors(sdev->request_queue, dev->max_sectors);
 
-	/* SATA DMA transfers must be multiples of 4 byte, so
-	 * we need to pad ATAPI transfers using an extra sg.
-	 * Decrement max hw segments accordingly.
-	 */
-	if (dev->class == ATA_DEV_ATAPI) {
-		struct request_queue *q = sdev->request_queue;
-		blk_queue_max_hw_segments(q, q->max_hw_segments - 1);
-
+	if (dev->class == ATA_DEV_ATAPI)
 		/* set the min alignment */
 		blk_queue_update_dma_alignment(sdev->request_queue,
 					       ATA_DMA_PAD_SZ - 1);
-	} else
+	else
 		/* ATA devices must be sector aligned */
 		blk_queue_update_dma_alignment(sdev->request_queue,
 					       ATA_SECT_SIZE - 1);
@@ -2500,7 +2493,9 @@ static unsigned int atapi_xlat(struct at
 	 * want to set it properly, and for DMA where it is
 	 * effectively meaningless.
 	 */
-	nbytes = min(qc->nbytes, (unsigned int)63 * 1024);
+	nbytes = min(scmd->request->raw_data_len, (unsigned int)63 * 1024);
+	ata_dev_printk(qc->dev, KERN_INFO, "XXX raw_data_len=%u\n",
+		       scmd->request->raw_data_len);
 
 	/* Most ATAPI devices which honor transfer chunk size don't
 	 * behave according to the spec when odd chunk size which
@@ -3564,7 +3559,7 @@ EXPORT_SYMBOL_GPL(ata_sas_port_alloc);
  */
 int ata_sas_port_start(struct ata_port *ap)
 {
-	return ata_pad_alloc(ap, ap->dev);
+	return 0;
 }
 EXPORT_SYMBOL_GPL(ata_sas_port_start);
 
@@ -3582,7 +3577,6 @@ EXPORT_SYMBOL_GPL(ata_sas_port_start);
 
 void ata_sas_port_stop(struct ata_port *ap)
 {
-	ata_pad_free(ap, ap->dev);
 }
 EXPORT_SYMBOL_GPL(ata_sas_port_stop);
 
Index: work/drivers/ata/sata_fsl.c
===================================================================
--- work.orig/drivers/ata/sata_fsl.c
+++ work/drivers/ata/sata_fsl.c
@@ -601,21 +601,9 @@ static int sata_fsl_port_start(struct at
 	if (!pp)
 		return -ENOMEM;
 
-	/*
-	 * allocate per command dma alignment pad buffer, which is used
-	 * internally by libATA to ensure that all transfers ending on
-	 * unaligned boundaries are padded, to align on Dword boundaries
-	 */
-	retval = ata_pad_alloc(ap, dev);
-	if (retval) {
-		kfree(pp);
-		return retval;
-	}
-
 	mem = dma_alloc_coherent(dev, SATA_FSL_PORT_PRIV_DMA_SZ, &mem_dma,
 				 GFP_KERNEL);
 	if (!mem) {
-		ata_pad_free(ap, dev);
 		kfree(pp);
 		return -ENOMEM;
 	}
@@ -694,7 +682,6 @@ static void sata_fsl_port_stop(struct at
 	dma_free_coherent(dev, SATA_FSL_PORT_PRIV_DMA_SZ,
 			  pp->cmdslot, pp->cmdslot_paddr);
 
-	ata_pad_free(ap, dev);
 	kfree(pp);
 }
 
Index: work/drivers/ata/sata_mv.c
===================================================================
--- work.orig/drivers/ata/sata_mv.c
+++ work/drivers/ata/sata_mv.c
@@ -1157,17 +1157,13 @@ static int mv_port_start(struct ata_port
 	struct mv_port_priv *pp;
 	void __iomem *port_mmio = mv_ap_base(ap);
 	unsigned long flags;
-	int tag, rc;
+	int tag;
 
 	pp = devm_kzalloc(dev, sizeof(*pp), GFP_KERNEL);
 	if (!pp)
 		return -ENOMEM;
 	ap->private_data = pp;
 
-	rc = ata_pad_alloc(ap, dev);
-	if (rc)
-		return rc;
-
 	pp->crqb = dma_pool_alloc(hpriv->crqb_pool, GFP_KERNEL, &pp->crqb_dma);
 	if (!pp->crqb)
 		return -ENOMEM;
Index: work/drivers/ata/sata_sil24.c
===================================================================
--- work.orig/drivers/ata/sata_sil24.c
+++ work/drivers/ata/sata_sil24.c
@@ -1234,7 +1234,6 @@ static int sil24_port_start(struct ata_p
 	union sil24_cmd_block *cb;
 	size_t cb_size = sizeof(*cb) * SIL24_MAX_CMDS;
 	dma_addr_t cb_dma;
-	int rc;
 
 	pp = devm_kzalloc(dev, sizeof(*pp), GFP_KERNEL);
 	if (!pp)
@@ -1247,10 +1246,6 @@ static int sil24_port_start(struct ata_p
 		return -ENOMEM;
 	memset(cb, 0, cb_size);
 
-	rc = ata_pad_alloc(ap, dev);
-	if (rc)
-		return rc;
-
 	pp->cmd_block = cb;
 	pp->cmd_block_dma = cb_dma;
 
Index: work/include/linux/libata.h
===================================================================
--- work.orig/include/linux/libata.h
+++ work/include/linux/libata.h
@@ -278,7 +278,6 @@ enum {
 
 	/* size of buffer to pad xfers ending on unaligned boundaries */
 	ATA_DMA_PAD_SZ		= 4,
-	ATA_DMA_PAD_BUF_SZ	= ATA_DMA_PAD_SZ * ATA_MAX_QUEUE,
 
 	/* ering size */
 	ATA_ERING_SIZE		= 32,
@@ -458,24 +457,19 @@ struct ata_queued_cmd {
 	unsigned int		tag;
 	unsigned int		n_elem;
 	unsigned int		n_iter;
-	unsigned int		mapped_n_elem;
 
 	int			dma_dir;
 
-	unsigned int		pad_len;
 	unsigned int		sect_size;
 
 	unsigned int		nbytes;
-	unsigned int		raw_nbytes;
 	unsigned int		curbytes;
 
 	struct scatterlist	*cursg;
 	unsigned int		cursg_ofs;
 
 	struct scatterlist	*last_sg;
-	struct scatterlist	saved_last_sg;
 	struct scatterlist	sgent;
-	struct scatterlist	extra_sg[2];
 
 	struct scatterlist	*sg;
 
@@ -620,9 +614,6 @@ struct ata_port {
 	struct ata_prd		*prd;	 /* our SG list */
 	dma_addr_t		prd_dma; /* and its DMA mapping */
 
-	void			*pad;	/* array of DMA pad buffers */
-	dma_addr_t		pad_dma;
-
 	struct ata_ioports	ioaddr;	/* ATA cmd/ctl/dma register blocks */
 
 	u8			ctl;	/* cache of ATA control register */
@@ -1364,12 +1355,10 @@ static inline void ata_qc_reinit(struct 
 	qc->flags = 0;
 	qc->cursg = NULL;
 	qc->cursg_ofs = 0;
-	qc->nbytes = qc->raw_nbytes = qc->curbytes = 0;
+	qc->nbytes = qc->curbytes = 0;
 	qc->n_elem = 0;
-	qc->mapped_n_elem = 0;
 	qc->n_iter = 0;
 	qc->err_mask = 0;
-	qc->pad_len = 0;
 	qc->last_sg = NULL;
 	qc->sect_size = ATA_SECT_SIZE;
 
@@ -1425,19 +1414,6 @@ static inline unsigned int __ac_err_mask
 	return mask;
 }
 
-static inline int ata_pad_alloc(struct ata_port *ap, struct device *dev)
-{
-	ap->pad_dma = 0;
-	ap->pad = dmam_alloc_coherent(dev, ATA_DMA_PAD_BUF_SZ,
-				      &ap->pad_dma, GFP_KERNEL);
-	return (ap->pad == NULL) ? -ENOMEM : 0;
-}
-
-static inline void ata_pad_free(struct ata_port *ap, struct device *dev)
-{
-	dmam_free_coherent(dev, ATA_DMA_PAD_BUF_SZ, ap->pad, ap->pad_dma);
-}
-
 static inline struct ata_port *ata_shost_to_port(struct Scsi_Host *host)
 {
 	return *(struct ata_port **)&host->hostdata[0];
Index: work/drivers/scsi/libsas/sas_ata.c
===================================================================
--- work.orig/drivers/scsi/libsas/sas_ata.c
+++ work/drivers/scsi/libsas/sas_ata.c
@@ -178,8 +178,8 @@ static unsigned int sas_ata_qc_issue(str
 	task->uldd_task = qc;
 	if (ata_is_atapi(qc->tf.protocol)) {
 		memcpy(task->ata_task.atapi_packet, qc->cdb, qc->dev->cdb_len);
-		task->total_xfer_len = qc->nbytes + qc->pad_len;
-		task->num_scatter = qc->pad_len ? qc->n_elem + 1 : qc->n_elem;
+		task->total_xfer_len = qc->nbytes;
+		task->num_scatter = qc->n_elem;
 	} else {
 		for_each_sg(qc->sg, sg, qc->n_elem, si)
 			xfer += sg->length;
Index: work/drivers/scsi/ipr.c
===================================================================
--- work.orig/drivers/scsi/ipr.c
+++ work/drivers/scsi/ipr.c
@@ -5140,7 +5140,7 @@ static void ipr_build_ata_ioadl(struct i
 	struct ipr_ioarcb *ioarcb = &ipr_cmd->ioarcb;
 	struct ipr_ioadl_desc *ioadl = ipr_cmd->ioadl;
 	struct ipr_ioadl_desc *last_ioadl = NULL;
-	int len = qc->nbytes + qc->pad_len;
+	int len = qc->nbytes;
 	struct scatterlist *sg;
 	unsigned int si;
 
@@ -5206,7 +5206,7 @@ static unsigned int ipr_qc_issue(struct 
 	ioarcb->cmd_pkt.request_type = IPR_RQTYPE_ATA_PASSTHRU;
 	ioarcb->cmd_pkt.flags_hi |= IPR_FLAGS_HI_NO_LINK_DESC;
 	ioarcb->cmd_pkt.flags_hi |= IPR_FLAGS_HI_NO_ULEN_CHK;
-	ipr_cmd->dma_use_sg = qc->pad_len ? qc->n_elem + 1 : qc->n_elem;
+	ipr_cmd->dma_use_sg = qc->n_elem;
 
 	ipr_build_ata_ioadl(ipr_cmd, qc);
 	regs->flags |= IPR_ATA_FLAG_STATUS_ON_GOOD_COMPLETION;
Index: work/drivers/ata/ahci.c
===================================================================
--- work.orig/drivers/ata/ahci.c
+++ work/drivers/ata/ahci.c
@@ -1979,16 +1979,11 @@ static int ahci_port_start(struct ata_po
 	struct ahci_port_priv *pp;
 	void *mem;
 	dma_addr_t mem_dma;
-	int rc;
 
 	pp = devm_kzalloc(dev, sizeof(*pp), GFP_KERNEL);
 	if (!pp)
 		return -ENOMEM;
 
-	rc = ata_pad_alloc(ap, dev);
-	if (rc)
-		return rc;
-
 	mem = dmam_alloc_coherent(dev, AHCI_PORT_PRIV_DMA_SZ, &mem_dma,
 				  GFP_KERNEL);
 	if (!mem)
Index: work/drivers/ata/pata_icside.c
===================================================================
--- work.orig/drivers/ata/pata_icside.c
+++ work/drivers/ata/pata_icside.c
@@ -304,12 +304,6 @@ static int icside_dma_init(struct pata_i
 }
 
 
-static int pata_icside_port_start(struct ata_port *ap)
-{
-	/* No PRD to alloc */
-	return ata_pad_alloc(ap, ap->dev);
-}
-
 static struct scsi_host_template pata_icside_sht = {
 	.module			= THIS_MODULE,
 	.name			= DRV_NAME,
@@ -389,8 +383,6 @@ static struct ata_port_operations pata_i
 	.irq_clear		= ata_dummy_noret,
 	.irq_on			= ata_irq_on,
 
-	.port_start		= pata_icside_port_start,
-
 	.bmdma_stop		= pata_icside_bmdma_stop,
 	.bmdma_status		= pata_icside_bmdma_status,
 };