1 files changed, 223 insertions, 230 deletions

diff --git a/crypto/async_tx/async_xor.c b/crypto/async_tx/async_xor.c
index 1c445c7bdab..3c562f5a60b 100644
--- a/crypto/async_tx/async_xor.c
+++ b/crypto/async_tx/async_xor.c
@@ -25,328 +25,321 @@
  */
 #include <linux/kernel.h>
 #include <linux/interrupt.h>
+#include <linux/module.h>
 #include <linux/mm.h>
 #include <linux/dma-mapping.h>
 #include <linux/raid/xor.h>
 #include <linux/async_tx.h>
 
-/* do_async_xor - dma map the pages and perform the xor with an engine.
- * 	This routine is marked __always_inline so it can be compiled away
- * 	when CONFIG_DMA_ENGINE=n
- */
-static __always_inline struct dma_async_tx_descriptor *
-do_async_xor(struct dma_device *device,
-	struct dma_chan *chan, struct page *dest, struct page **src_list,
-	unsigned int offset, unsigned int src_cnt, size_t len,
-	enum async_tx_flags flags, struct dma_async_tx_descriptor *depend_tx,
-	dma_async_tx_callback cb_fn, void *cb_param)
+/* do_async_xor - dma map the pages and perform the xor with an engine */
+static __async_inline struct dma_async_tx_descriptor *
+do_async_xor(struct dma_chan *chan, struct dmaengine_unmap_data *unmap,
+	     struct async_submit_ctl *submit)
 {
-	dma_addr_t dma_dest;
-	dma_addr_t *dma_src = (dma_addr_t *) src_list;
-	struct dma_async_tx_descriptor *tx;
-	int i;
-	unsigned long dma_prep_flags = cb_fn ? DMA_PREP_INTERRUPT : 0;
+	struct dma_device *dma = chan->device;
+	struct dma_async_tx_descriptor *tx = NULL;
+	dma_async_tx_callback cb_fn_orig = submit->cb_fn;
+	void *cb_param_orig = submit->cb_param;
+	enum async_tx_flags flags_orig = submit->flags;
+	enum dma_ctrl_flags dma_flags = 0;
+	int src_cnt = unmap->to_cnt;
+	int xor_src_cnt;
+	dma_addr_t dma_dest = unmap->addr[unmap->to_cnt];
+	dma_addr_t *src_list = unmap->addr;
 
-	pr_debug("%s: len: %zu\n", __func__, len);
+	while (src_cnt) {
+		dma_addr_t tmp;
 
-	dma_dest = dma_map_page(device->dev, dest, offset, len,
-				DMA_FROM_DEVICE);
+		submit->flags = flags_orig;
+		xor_src_cnt = min(src_cnt, (int)dma->max_xor);
+		/* if we are submitting additional xors, leave the chain open
+		 * and clear the callback parameters
+		 */
+		if (src_cnt > xor_src_cnt) {
+			submit->flags &= ~ASYNC_TX_ACK;
+			submit->flags |= ASYNC_TX_FENCE;
+			submit->cb_fn = NULL;
+			submit->cb_param = NULL;
+		} else {
+			submit->cb_fn = cb_fn_orig;
+			submit->cb_param = cb_param_orig;
+		}
+		if (submit->cb_fn)
+			dma_flags |= DMA_PREP_INTERRUPT;
+		if (submit->flags & ASYNC_TX_FENCE)
+			dma_flags |= DMA_PREP_FENCE;
 
-	for (i = 0; i < src_cnt; i++)
-		dma_src[i] = dma_map_page(device->dev, src_list[i], offset,
-					  len, DMA_TO_DEVICE);
-
-	/* Since we have clobbered the src_list we are committed
-	 * to doing this asynchronously.  Drivers force forward progress
-	 * in case they can not provide a descriptor
-	 */
-	tx = device->device_prep_dma_xor(chan, dma_dest, dma_src, src_cnt, len,
-					 dma_prep_flags);
-	if (!tx) {
-		if (depend_tx)
-			dma_wait_for_async_tx(depend_tx);
-
-		while (!tx)
-			tx = device->device_prep_dma_xor(chan, dma_dest,
-							 dma_src, src_cnt, len,
-							 dma_prep_flags);
-	}
+		/* Drivers force forward progress in case they can not provide a
+		 * descriptor
+		 */
+		tmp = src_list[0];
+		if (src_list > unmap->addr)
+			src_list[0] = dma_dest;
+		tx = dma->device_prep_dma_xor(chan, dma_dest, src_list,
+					      xor_src_cnt, unmap->len,
+					      dma_flags);
+		src_list[0] = tmp;
+
+
+		if (unlikely(!tx))
+			async_tx_quiesce(&submit->depend_tx);
+
+		/* spin wait for the preceding transactions to complete */
+		while (unlikely(!tx)) {
+			dma_async_issue_pending(chan);
+			tx = dma->device_prep_dma_xor(chan, dma_dest,
+						      src_list,
+						      xor_src_cnt, unmap->len,
+						      dma_flags);
+		}
 
-	async_tx_submit(chan, tx, flags, depend_tx, cb_fn, cb_param);
+		dma_set_unmap(tx, unmap);
+		async_tx_submit(chan, tx, submit);
+		submit->depend_tx = tx;
+
+		if (src_cnt > xor_src_cnt) {
+			/* drop completed sources */
+			src_cnt -= xor_src_cnt;
+			/* use the intermediate result a source */
+			src_cnt++;
+			src_list += xor_src_cnt - 1;
+		} else
+			break;
+	}
 
 	return tx;
 }
 
 static void
 do_sync_xor(struct page *dest, struct page **src_list, unsigned int offset,
-	unsigned int src_cnt, size_t len, enum async_tx_flags flags,
-	struct dma_async_tx_descriptor *depend_tx,
-	dma_async_tx_callback cb_fn, void *cb_param)
+	    int src_cnt, size_t len, struct async_submit_ctl *submit)
 {
-	void *_dest;
 	int i;
+	int xor_src_cnt = 0;
+	int src_off = 0;
+	void *dest_buf;
+	void **srcs;
 
-	pr_debug("%s: len: %zu\n", __func__, len);
+	if (submit->scribble)
+		srcs = submit->scribble;
+	else
+		srcs = (void **) src_list;
 
-	/* reuse the 'src_list' array to convert to buffer pointers */
+	/* convert to buffer pointers */
 	for (i = 0; i < src_cnt; i++)
-		src_list[i] = (struct page *)
-			(page_address(src_list[i]) + offset);
-
+		if (src_list[i])
+			srcs[xor_src_cnt++] = page_address(src_list[i]) + offset;
+	src_cnt = xor_src_cnt;
 	/* set destination address */
-	_dest = page_address(dest) + offset;
+	dest_buf = page_address(dest) + offset;
 
-	if (flags & ASYNC_TX_XOR_ZERO_DST)
-		memset(_dest, 0, len);
+	if (submit->flags & ASYNC_TX_XOR_ZERO_DST)
+		memset(dest_buf, 0, len);
 
-	xor_blocks(src_cnt, len, _dest,
-		(void **) src_list);
+	while (src_cnt > 0) {
+		/* process up to 'MAX_XOR_BLOCKS' sources */
+		xor_src_cnt = min(src_cnt, MAX_XOR_BLOCKS);
+		xor_blocks(xor_src_cnt, len, dest_buf, &srcs[src_off]);
 
-	async_tx_sync_epilog(flags, depend_tx, cb_fn, cb_param);
+		/* drop completed sources */
+		src_cnt -= xor_src_cnt;
+		src_off += xor_src_cnt;
+	}
+
+	async_tx_sync_epilog(submit);
 }
 
 /**
  * async_xor - attempt to xor a set of blocks with a dma engine.
- *	xor_blocks always uses the dest as a source so the ASYNC_TX_XOR_ZERO_DST
- *	flag must be set to not include dest data in the calculation.  The
- *	assumption with dma eninges is that they only use the destination
- *	buffer as a source when it is explicity specified in the source list.
  * @dest: destination page
- * @src_list: array of source pages (if the dest is also a source it must be
- *	at index zero).  The contents of this array may be overwritten.
- * @offset: offset in pages to start transaction
+ * @src_list: array of source pages
+ * @offset: common src/dst offset to start transaction
  * @src_cnt: number of source pages
  * @len: length in bytes
- * @flags: ASYNC_TX_XOR_ZERO_DST, ASYNC_TX_XOR_DROP_DEST,
- *	ASYNC_TX_ACK, ASYNC_TX_DEP_ACK
- * @depend_tx: xor depends on the result of this transaction.
- * @cb_fn: function to call when the xor completes
- * @cb_param: parameter to pass to the callback routine
+ * @submit: submission / completion modifiers
+ *
+ * honored flags: ASYNC_TX_ACK, ASYNC_TX_XOR_ZERO_DST, ASYNC_TX_XOR_DROP_DST
+ *
+ * xor_blocks always uses the dest as a source so the
+ * ASYNC_TX_XOR_ZERO_DST flag must be set to not include dest data in
+ * the calculation.  The assumption with dma eninges is that they only
+ * use the destination buffer as a source when it is explicity specified
+ * in the source list.
+ *
+ * src_list note: if the dest is also a source it must be at index zero.
+ * The contents of this array will be overwritten if a scribble region
+ * is not specified.
  */
 struct dma_async_tx_descriptor *
 async_xor(struct page *dest, struct page **src_list, unsigned int offset,
-	int src_cnt, size_t len, enum async_tx_flags flags,
-	struct dma_async_tx_descriptor *depend_tx,
-	dma_async_tx_callback cb_fn, void *cb_param)
+	  int src_cnt, size_t len, struct async_submit_ctl *submit)
 {
-	struct dma_chan *chan = async_tx_find_channel(depend_tx, DMA_XOR,
+	struct dma_chan *chan = async_tx_find_channel(submit, DMA_XOR,
 						      &dest, 1, src_list,
 						      src_cnt, len);
 	struct dma_device *device = chan ? chan->device : NULL;
-	struct dma_async_tx_descriptor *tx = NULL;
-	dma_async_tx_callback _cb_fn;
-	void *_cb_param;
-	unsigned long local_flags;
-	int xor_src_cnt;
-	int i = 0, src_off = 0;
+	struct dmaengine_unmap_data *unmap = NULL;
 
 	BUG_ON(src_cnt <= 1);
 
-	while (src_cnt) {
-		local_flags = flags;
-		if (device) { /* run the xor asynchronously */
-			xor_src_cnt = min(src_cnt, device->max_xor);
-			/* if we are submitting additional xors
-			 * only set the callback on the last transaction
-			 */
-			if (src_cnt > xor_src_cnt) {
-				local_flags &= ~ASYNC_TX_ACK;
-				_cb_fn = NULL;
-				_cb_param = NULL;
-			} else {
-				_cb_fn = cb_fn;
-				_cb_param = cb_param;
-			}
-
-			tx = do_async_xor(device, chan, dest,
-					  &src_list[src_off], offset,
-					  xor_src_cnt, len, local_flags,
-					  depend_tx, _cb_fn, _cb_param);
-		} else { /* run the xor synchronously */
-			/* in the sync case the dest is an implied source
-			 * (assumes the dest is at the src_off index)
-			 */
-			if (flags & ASYNC_TX_XOR_DROP_DST) {
-				src_cnt--;
-				src_off++;
-			}
+	if (device)
+		unmap = dmaengine_get_unmap_data(device->dev, src_cnt+1, GFP_NOIO);
 
-			/* process up to 'MAX_XOR_BLOCKS' sources */
-			xor_src_cnt = min(src_cnt, MAX_XOR_BLOCKS);
-
-			/* if we are submitting additional xors
-			 * only set the callback on the last transaction
-			 */
-			if (src_cnt > xor_src_cnt) {
-				local_flags &= ~ASYNC_TX_ACK;
-				_cb_fn = NULL;
-				_cb_param = NULL;
-			} else {
-				_cb_fn = cb_fn;
-				_cb_param = cb_param;
-			}
+	if (unmap && is_dma_xor_aligned(device, offset, 0, len)) {
+		struct dma_async_tx_descriptor *tx;
+		int i, j;
 
-			/* wait for any prerequisite operations */
-			if (depend_tx) {
-				/* if ack is already set then we cannot be sure
-				 * we are referring to the correct operation
-				 */
-				BUG_ON(depend_tx->ack);
-				if (dma_wait_for_async_tx(depend_tx) ==
-					DMA_ERROR)
-					panic("%s: DMA_ERROR waiting for "
-						"depend_tx\n",
-						__func__);
-			}
+		/* run the xor asynchronously */
+		pr_debug("%s (async): len: %zu\n", __func__, len);
 
-			do_sync_xor(dest, &src_list[src_off], offset,
-				xor_src_cnt, len, local_flags, depend_tx,
-				_cb_fn, _cb_param);
+		unmap->len = len;
+		for (i = 0, j = 0; i < src_cnt; i++) {
+			if (!src_list[i])
+				continue;
+			unmap->to_cnt++;
+			unmap->addr[j++] = dma_map_page(device->dev, src_list[i],
+							offset, len, DMA_TO_DEVICE);
 		}
 
-		/* the previous tx is hidden from the client,
-		 * so ack it
-		 */
-		if (i && depend_tx)
-			async_tx_ack(depend_tx);
+		/* map it bidirectional as it may be re-used as a source */
+		unmap->addr[j] = dma_map_page(device->dev, dest, offset, len,
+					      DMA_BIDIRECTIONAL);
+		unmap->bidi_cnt = 1;
 
-		depend_tx = tx;
+		tx = do_async_xor(chan, unmap, submit);
+		dmaengine_unmap_put(unmap);
+		return tx;
+	} else {
+		dmaengine_unmap_put(unmap);
+		/* run the xor synchronously */
+		pr_debug("%s (sync): len: %zu\n", __func__, len);
+		WARN_ONCE(chan, "%s: no space for dma address conversion\n",
+			  __func__);
+
+		/* in the sync case the dest is an implied source
+		 * (assumes the dest is the first source)
+		 */
+		if (submit->flags & ASYNC_TX_XOR_DROP_DST) {
+			src_cnt--;
+			src_list++;
+		}
 
-		if (src_cnt > xor_src_cnt) {
-			/* drop completed sources */
-			src_cnt -= xor_src_cnt;
-			src_off += xor_src_cnt;
+		/* wait for any prerequisite operations */
+		async_tx_quiesce(&submit->depend_tx);
 
-			/* unconditionally preserve the destination */
-			flags &= ~ASYNC_TX_XOR_ZERO_DST;
+		do_sync_xor(dest, src_list, offset, src_cnt, len, submit);
 
-			/* use the intermediate result a source, but remember
-			 * it's dropped, because it's implied, in the sync case
-			 */
-			src_list[--src_off] = dest;
-			src_cnt++;
-			flags |= ASYNC_TX_XOR_DROP_DST;
-		} else
-			src_cnt = 0;
-		i++;
+		return NULL;
 	}
-
-	return tx;
 }
 EXPORT_SYMBOL_GPL(async_xor);
 
 static int page_is_zero(struct page *p, unsigned int offset, size_t len)
 {
-	char *a = page_address(p) + offset;
-	return ((*(u32 *) a) == 0 &&
-		memcmp(a, a + 4, len - 4) == 0);
+	return !memchr_inv(page_address(p) + offset, 0, len);
+}
+
+static inline struct dma_chan *
+xor_val_chan(struct async_submit_ctl *submit, struct page *dest,
+		 struct page **src_list, int src_cnt, size_t len)
+{
+	#ifdef CONFIG_ASYNC_TX_DISABLE_XOR_VAL_DMA
+	return NULL;
+	#endif
+	return async_tx_find_channel(submit, DMA_XOR_VAL, &dest, 1, src_list,
+				     src_cnt, len);
 }
 
 /**
- * async_xor_zero_sum - attempt a xor parity check with a dma engine.
+ * async_xor_val - attempt a xor parity check with a dma engine.
  * @dest: destination page used if the xor is performed synchronously
- * @src_list: array of source pages.  The dest page must be listed as a source
- * 	at index zero.  The contents of this array may be overwritten.
+ * @src_list: array of source pages
  * @offset: offset in pages to start transaction
  * @src_cnt: number of source pages
  * @len: length in bytes
  * @result: 0 if sum == 0 else non-zero
- * @flags: ASYNC_TX_ACK, ASYNC_TX_DEP_ACK
- * @depend_tx: xor depends on the result of this transaction.
- * @cb_fn: function to call when the xor completes
- * @cb_param: parameter to pass to the callback routine
+ * @submit: submission / completion modifiers
+ *
+ * honored flags: ASYNC_TX_ACK
+ *
+ * src_list note: if the dest is also a source it must be at index zero.
+ * The contents of this array will be overwritten if a scribble region
+ * is not specified.
  */
 struct dma_async_tx_descriptor *
-async_xor_zero_sum(struct page *dest, struct page **src_list,
-	unsigned int offset, int src_cnt, size_t len,
-	u32 *result, enum async_tx_flags flags,
-	struct dma_async_tx_descriptor *depend_tx,
-	dma_async_tx_callback cb_fn, void *cb_param)
+async_xor_val(struct page *dest, struct page **src_list, unsigned int offset,
+	      int src_cnt, size_t len, enum sum_check_flags *result,
+	      struct async_submit_ctl *submit)
 {
-	struct dma_chan *chan = async_tx_find_channel(depend_tx, DMA_ZERO_SUM,
-						      &dest, 1, src_list,
-						      src_cnt, len);
+	struct dma_chan *chan = xor_val_chan(submit, dest, src_list, src_cnt, len);
 	struct dma_device *device = chan ? chan->device : NULL;
 	struct dma_async_tx_descriptor *tx = NULL;
+	struct dmaengine_unmap_data *unmap = NULL;
 
 	BUG_ON(src_cnt <= 1);
 
-	if (device && src_cnt <= device->max_xor) {
-		dma_addr_t *dma_src = (dma_addr_t *) src_list;
-		unsigned long dma_prep_flags = cb_fn ? DMA_PREP_INTERRUPT : 0;
+	if (device)
+		unmap = dmaengine_get_unmap_data(device->dev, src_cnt, GFP_NOIO);
+
+	if (unmap && src_cnt <= device->max_xor &&
+	    is_dma_xor_aligned(device, offset, 0, len)) {
+		unsigned long dma_prep_flags = 0;
 		int i;
 
 		pr_debug("%s: (async) len: %zu\n", __func__, len);
 
-		for (i = 0; i < src_cnt; i++)
-			dma_src[i] = dma_map_page(device->dev, src_list[i],
-						  offset, len, DMA_TO_DEVICE);
-
-		tx = device->device_prep_dma_zero_sum(chan, dma_src, src_cnt,
-						      len, result,
-						      dma_prep_flags);
-		if (!tx) {
-			if (depend_tx)
-				dma_wait_for_async_tx(depend_tx);
+		if (submit->cb_fn)
+			dma_prep_flags |= DMA_PREP_INTERRUPT;
+		if (submit->flags & ASYNC_TX_FENCE)
+			dma_prep_flags |= DMA_PREP_FENCE;
 
-			while (!tx)
-				tx = device->device_prep_dma_zero_sum(chan,
-					dma_src, src_cnt, len, result,
+		for (i = 0; i < src_cnt; i++) {
+			unmap->addr[i] = dma_map_page(device->dev, src_list[i],
+						      offset, len, DMA_TO_DEVICE);
+			unmap->to_cnt++;
+		}
+		unmap->len = len;
+
+		tx = device->device_prep_dma_xor_val(chan, unmap->addr, src_cnt,
+						     len, result,
+						     dma_prep_flags);
+		if (unlikely(!tx)) {
+			async_tx_quiesce(&submit->depend_tx);
+
+			while (!tx) {
+				dma_async_issue_pending(chan);
+				tx = device->device_prep_dma_xor_val(chan,
+					unmap->addr, src_cnt, len, result,
 					dma_prep_flags);
+			}
 		}
-
-		async_tx_submit(chan, tx, flags, depend_tx, cb_fn, cb_param);
+		dma_set_unmap(tx, unmap);
+		async_tx_submit(chan, tx, submit);
 	} else {
-		unsigned long xor_flags = flags;
+		enum async_tx_flags flags_orig = submit->flags;
 
 		pr_debug("%s: (sync) len: %zu\n", __func__, len);
+		WARN_ONCE(device && src_cnt <= device->max_xor,
+			  "%s: no space for dma address conversion\n",
+			  __func__);
 
-		xor_flags |= ASYNC_TX_XOR_DROP_DST;
-		xor_flags &= ~ASYNC_TX_ACK;
-
-		tx = async_xor(dest, src_list, offset, src_cnt, len, xor_flags,
-			depend_tx, NULL, NULL);
+		submit->flags |= ASYNC_TX_XOR_DROP_DST;
+		submit->flags &= ~ASYNC_TX_ACK;
 
-		if (tx) {
-			if (dma_wait_for_async_tx(tx) == DMA_ERROR)
-				panic("%s: DMA_ERROR waiting for tx\n",
-					__func__);
-			async_tx_ack(tx);
-		}
+		tx = async_xor(dest, src_list, offset, src_cnt, len, submit);
 
-		*result = page_is_zero(dest, offset, len) ? 0 : 1;
+		async_tx_quiesce(&tx);
 
-		tx = NULL;
+		*result = !page_is_zero(dest, offset, len) << SUM_CHECK_P;
 
-		async_tx_sync_epilog(flags, depend_tx, cb_fn, cb_param);
+		async_tx_sync_epilog(submit);
+		submit->flags = flags_orig;
 	}
+	dmaengine_unmap_put(unmap);
 
 	return tx;
 }
-EXPORT_SYMBOL_GPL(async_xor_zero_sum);
-
-static int __init async_xor_init(void)
-{
-	#ifdef CONFIG_DMA_ENGINE
-	/* To conserve stack space the input src_list (array of page pointers)
-	 * is reused to hold the array of dma addresses passed to the driver.
-	 * This conversion is only possible when dma_addr_t is less than the
-	 * the size of a pointer.  HIGHMEM64G is known to violate this
-	 * assumption.
-	 */
-	BUILD_BUG_ON(sizeof(dma_addr_t) > sizeof(struct page *));
-	#endif
-
-	return 0;
-}
-
-static void __exit async_xor_exit(void)
-{
-	do { } while (0);
-}
-
-module_init(async_xor_init);
-module_exit(async_xor_exit);
+EXPORT_SYMBOL_GPL(async_xor_val);
 
 MODULE_AUTHOR("Intel Corporation");
 MODULE_DESCRIPTION("asynchronous xor/xor-zero-sum api");