dma: mv_xor: Remove multi-slot support

Although the driver supported multiple-slot allocation, only one slot was
ever allocated for each transaction. So, given we have no users of the
multi-slot support, we can remove it and greatly simplify the code.

Signed-off-by: Lior Amsalem <alior@marvell.com>
Signed-off-by: Ezequiel Garcia <ezequiel.garcia@free-electrons.com>
Signed-off-by: Vinod Koul <vinod.koul@intel.com>
diff --git a/drivers/dma/mv_xor.c b/drivers/dma/mv_xor.c
index 1e43f18..a30e221 100644
--- a/drivers/dma/mv_xor.c
+++ b/drivers/dma/mv_xor.c
@@ -45,19 +45,15 @@
 #define mv_chan_to_devp(chan)           \
 	((chan)->dmadev.dev)
 
-static void mv_desc_init(struct mv_xor_desc_slot *desc, unsigned long flags)
+static void mv_desc_init(struct mv_xor_desc_slot *desc,
+			 dma_addr_t addr, u32 byte_count)
 {
 	struct mv_xor_desc *hw_desc = desc->hw_desc;
 
 	hw_desc->status = (1 << 31);
 	hw_desc->phy_next_desc = 0;
 	hw_desc->desc_command = (1 << 31);
-}
-
-static void mv_desc_set_byte_count(struct mv_xor_desc_slot *desc,
-				   u32 byte_count)
-{
-	struct mv_xor_desc *hw_desc = desc->hw_desc;
+	hw_desc->phy_dest_addr = addr;
 	hw_desc->byte_count = byte_count;
 }
 
@@ -75,13 +71,6 @@
 	hw_desc->phy_next_desc = 0;
 }
 
-static void mv_desc_set_dest_addr(struct mv_xor_desc_slot *desc,
-				  dma_addr_t addr)
-{
-	struct mv_xor_desc *hw_desc = desc->hw_desc;
-	hw_desc->phy_dest_addr = addr;
-}
-
 static void mv_desc_set_src_addr(struct mv_xor_desc_slot *desc,
 				 int index, dma_addr_t addr)
 {
@@ -188,11 +177,6 @@
 	return (state == 1) ? 1 : 0;
 }
 
-static int mv_chan_xor_slot_count(size_t len, int src_cnt)
-{
-	return 1;
-}
-
 /**
  * mv_xor_free_slots - flags descriptor slots for reuse
  * @slot: Slot to free
@@ -204,7 +188,7 @@
 	dev_dbg(mv_chan_to_devp(mv_chan), "%s %d slot %p\n",
 		__func__, __LINE__, slot);
 
-	slot->slots_per_op = 0;
+	slot->slot_used = 0;
 
 }
 
@@ -222,7 +206,7 @@
 	/* set the hardware chain */
 	mv_chan_set_next_descriptor(mv_chan, sw_desc->async_tx.phys);
 
-	mv_chan->pending += sw_desc->slot_cnt;
+	mv_chan->pending++;
 	mv_xor_issue_pending(&mv_chan->dmachan);
 }
 
@@ -243,8 +227,6 @@
 				desc->async_tx.callback_param);
 
 		dma_descriptor_unmap(&desc->async_tx);
-		if (desc->group_head)
-			desc->group_head = NULL;
 	}
 
 	/* run dependent operations */
@@ -361,19 +343,16 @@
 }
 
 static struct mv_xor_desc_slot *
-mv_xor_alloc_slots(struct mv_xor_chan *mv_chan, int num_slots,
-		    int slots_per_op)
+mv_xor_alloc_slot(struct mv_xor_chan *mv_chan)
 {
-	struct mv_xor_desc_slot *iter, *_iter, *alloc_start = NULL;
-	LIST_HEAD(chain);
-	int slots_found, retry = 0;
+	struct mv_xor_desc_slot *iter, *_iter;
+	int retry = 0;
 
 	/* start search from the last allocated descrtiptor
 	 * if a contiguous allocation can not be found start searching
 	 * from the beginning of the list
 	 */
 retry:
-	slots_found = 0;
 	if (retry == 0)
 		iter = mv_chan->last_used;
 	else
@@ -383,55 +362,29 @@
 
 	list_for_each_entry_safe_continue(
 		iter, _iter, &mv_chan->all_slots, slot_node) {
+
 		prefetch(_iter);
 		prefetch(&_iter->async_tx);
-		if (iter->slots_per_op) {
+		if (iter->slot_used) {
 			/* give up after finding the first busy slot
 			 * on the second pass through the list
 			 */
 			if (retry)
 				break;
-
-			slots_found = 0;
 			continue;
 		}
 
-		/* start the allocation if the slot is correctly aligned */
-		if (!slots_found++)
-			alloc_start = iter;
+		/* pre-ack descriptor */
+		async_tx_ack(&iter->async_tx);
 
-		if (slots_found == num_slots) {
-			struct mv_xor_desc_slot *alloc_tail = NULL;
-			struct mv_xor_desc_slot *last_used = NULL;
-			iter = alloc_start;
-			while (num_slots) {
-				int i;
+		iter->slot_used = 1;
+		INIT_LIST_HEAD(&iter->chain_node);
+		iter->async_tx.cookie = -EBUSY;
+		mv_chan->last_used = iter;
+		mv_desc_clear_next_desc(iter);
 
-				/* pre-ack all but the last descriptor */
-				async_tx_ack(&iter->async_tx);
+		return iter;
 
-				list_add_tail(&iter->chain_node, &chain);
-				alloc_tail = iter;
-				iter->async_tx.cookie = 0;
-				iter->slot_cnt = num_slots;
-				iter->xor_check_result = NULL;
-				for (i = 0; i < slots_per_op; i++) {
-					iter->slots_per_op = slots_per_op - i;
-					last_used = iter;
-					iter = list_entry(iter->slot_node.next,
-						struct mv_xor_desc_slot,
-						slot_node);
-				}
-				num_slots -= slots_per_op;
-			}
-			alloc_tail->group_head = alloc_start;
-			alloc_tail->async_tx.cookie = -EBUSY;
-			list_splice(&chain, &alloc_tail->tx_list);
-			mv_chan->last_used = last_used;
-			mv_desc_clear_next_desc(alloc_start);
-			mv_desc_clear_next_desc(alloc_tail);
-			return alloc_tail;
-		}
 	}
 	if (!retry++)
 		goto retry;
@@ -448,7 +401,7 @@
 {
 	struct mv_xor_desc_slot *sw_desc = to_mv_xor_slot(tx);
 	struct mv_xor_chan *mv_chan = to_mv_xor_chan(tx->chan);
-	struct mv_xor_desc_slot *grp_start, *old_chain_tail;
+	struct mv_xor_desc_slot *old_chain_tail;
 	dma_cookie_t cookie;
 	int new_hw_chain = 1;
 
@@ -456,27 +409,24 @@
 		"%s sw_desc %p: async_tx %p\n",
 		__func__, sw_desc, &sw_desc->async_tx);
 
-	grp_start = sw_desc->group_head;
-
 	spin_lock_bh(&mv_chan->lock);
 	cookie = dma_cookie_assign(tx);
 
 	if (list_empty(&mv_chan->chain))
-		list_splice_init(&sw_desc->tx_list, &mv_chan->chain);
+		list_add_tail(&sw_desc->chain_node, &mv_chan->chain);
 	else {
 		new_hw_chain = 0;
 
 		old_chain_tail = list_entry(mv_chan->chain.prev,
 					    struct mv_xor_desc_slot,
 					    chain_node);
-		list_splice_init(&grp_start->tx_list,
-				 &old_chain_tail->chain_node);
+		list_add_tail(&sw_desc->chain_node, &mv_chan->chain);
 
 		dev_dbg(mv_chan_to_devp(mv_chan), "Append to last desc %pa\n",
 			&old_chain_tail->async_tx.phys);
 
 		/* fix up the hardware chain */
-		mv_desc_set_next_desc(old_chain_tail, grp_start->async_tx.phys);
+		mv_desc_set_next_desc(old_chain_tail, sw_desc->async_tx.phys);
 
 		/* if the channel is not busy */
 		if (!mv_chan_is_busy(mv_chan)) {
@@ -491,7 +441,7 @@
 	}
 
 	if (new_hw_chain)
-		mv_xor_start_new_chain(mv_chan, grp_start);
+		mv_xor_start_new_chain(mv_chan, sw_desc);
 
 	spin_unlock_bh(&mv_chan->lock);
 
@@ -525,7 +475,6 @@
 		slot->async_tx.tx_submit = mv_xor_tx_submit;
 		INIT_LIST_HEAD(&slot->chain_node);
 		INIT_LIST_HEAD(&slot->slot_node);
-		INIT_LIST_HEAD(&slot->tx_list);
 		dma_desc = mv_chan->dma_desc_pool;
 		slot->async_tx.phys = dma_desc + idx * MV_XOR_SLOT_SIZE;
 		slot->idx = idx++;
@@ -553,8 +502,7 @@
 		    unsigned int src_cnt, size_t len, unsigned long flags)
 {
 	struct mv_xor_chan *mv_chan = to_mv_xor_chan(chan);
-	struct mv_xor_desc_slot *sw_desc, *grp_start;
-	int slot_cnt;
+	struct mv_xor_desc_slot *sw_desc;
 
 	if (unlikely(len < MV_XOR_MIN_BYTE_COUNT))
 		return NULL;
@@ -566,19 +514,15 @@
 		__func__, src_cnt, len, &dest, flags);
 
 	spin_lock_bh(&mv_chan->lock);
-	slot_cnt = mv_chan_xor_slot_count(len, src_cnt);
-	sw_desc = mv_xor_alloc_slots(mv_chan, slot_cnt, 1);
+	sw_desc = mv_xor_alloc_slot(mv_chan);
 	if (sw_desc) {
 		sw_desc->type = DMA_XOR;
 		sw_desc->async_tx.flags = flags;
-		grp_start = sw_desc->group_head;
-		mv_desc_init(grp_start, flags);
-		mv_desc_set_byte_count(grp_start, len);
-		mv_desc_set_dest_addr(sw_desc->group_head, dest);
+		mv_desc_init(sw_desc, dest, len);
 		sw_desc->unmap_src_cnt = src_cnt;
 		sw_desc->unmap_len = len;
 		while (src_cnt--)
-			mv_desc_set_src_addr(grp_start, src_cnt, src[src_cnt]);
+			mv_desc_set_src_addr(sw_desc, src_cnt, src[src_cnt]);
 	}
 	spin_unlock_bh(&mv_chan->lock);
 	dev_dbg(mv_chan_to_devp(mv_chan),
diff --git a/drivers/dma/mv_xor.h b/drivers/dma/mv_xor.h
index d074922..e03021e 100644
--- a/drivers/dma/mv_xor.h
+++ b/drivers/dma/mv_xor.h
@@ -110,9 +110,7 @@
  * @completed_node: node on the mv_xor_chan.completed_slots list
  * @hw_desc: virtual address of the hardware descriptor chain
  * @phys: hardware address of the hardware descriptor chain
- * @group_head: first operation in a transaction
- * @slot_cnt: total slots used in an transaction (group of operations)
- * @slots_per_op: number of slots per operation
+ * @slot_used: slot in use or not
  * @idx: pool index
  * @unmap_src_cnt: number of xor sources
  * @unmap_len: transaction bytecount
@@ -127,14 +125,11 @@
 	struct list_head	completed_node;
 	enum dma_transaction_type	type;
 	void			*hw_desc;
-	struct mv_xor_desc_slot	*group_head;
-	u16			slot_cnt;
-	u16			slots_per_op;
+	u16			slot_used;
 	u16			idx;
 	u16			unmap_src_cnt;
 	u32			value;
 	size_t			unmap_len;
-	struct list_head	tx_list;
 	struct dma_async_tx_descriptor	async_tx;
 	union {
 		u32		*xor_check_result;