Merge commit 'dmaengine-3.13-v2' of git://git.kernel.org/pub/scm/linux/kernel/git/djbw/dmaengine

Pull dmaengine changes from Dan

1/ Bartlomiej and Dan finalized a rework of the dma address unmap
   implementation.

2/ In the course of testing 1/ a collection of enhancements to dmatest
   fell out.  Notably basic performance statistics, and fixed / enhanced
   test control through new module parameters 'run', 'wait', 'noverify',
   and 'verbose'.  Thanks to Andriy and Linus for their review.

3/ Testing the raid related corner cases of 1/ triggered bugs in the
   recently added 16-source operation support in the ioatdma driver.

4/ Some minor fixes / cleanups to mv_xor and ioatdma.

Conflicts:
	drivers/dma/dmatest.c

Signed-off-by: Vinod Koul <vinod.koul@intel.com>

1 files changed, 351 insertions, 564 deletions

diff --git a/drivers/dma/dmatest.c b/drivers/dma/dmatest.c
index 59e287f56df..20f9a3aaf92 100644
--- a/drivers/dma/dmatest.c
+++ b/drivers/dma/dmatest.c
@@ -8,6 +8,8 @@
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
  */
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
 #include <linux/delay.h>
 #include <linux/dma-mapping.h>
 #include <linux/dmaengine.h>
@@ -19,10 +21,6 @@
 #include <linux/random.h>
 #include <linux/slab.h>
 #include <linux/wait.h>
-#include <linux/ctype.h>
-#include <linux/debugfs.h>
-#include <linux/uaccess.h>
-#include <linux/seq_file.h>
 
 static unsigned int test_buf_size = 16384;
 module_param(test_buf_size, uint, S_IRUGO | S_IWUSR);
@@ -68,92 +66,13 @@ module_param(timeout, uint, S_IRUGO | S_IWUSR);
 MODULE_PARM_DESC(timeout, "Transfer Timeout in msec (default: 3000), "
 		 "Pass -1 for infinite timeout");
 
-/* Maximum amount of mismatched bytes in buffer to print */
-#define MAX_ERROR_COUNT		32
-
-/*
- * Initialization patterns. All bytes in the source buffer has bit 7
- * set, all bytes in the destination buffer has bit 7 cleared.
- *
- * Bit 6 is set for all bytes which are to be copied by the DMA
- * engine. Bit 5 is set for all bytes which are to be overwritten by
- * the DMA engine.
- *
- * The remaining bits are the inverse of a counter which increments by
- * one for each byte address.
- */
-#define PATTERN_SRC		0x80
-#define PATTERN_DST		0x00
-#define PATTERN_COPY		0x40
-#define PATTERN_OVERWRITE	0x20
-#define PATTERN_COUNT_MASK	0x1f
-
-enum dmatest_error_type {
-	DMATEST_ET_OK,
-	DMATEST_ET_MAP_SRC,
-	DMATEST_ET_MAP_DST,
-	DMATEST_ET_PREP,
-	DMATEST_ET_SUBMIT,
-	DMATEST_ET_TIMEOUT,
-	DMATEST_ET_DMA_ERROR,
-	DMATEST_ET_DMA_IN_PROGRESS,
-	DMATEST_ET_VERIFY,
-	DMATEST_ET_VERIFY_BUF,
-};
-
-struct dmatest_verify_buffer {
-	unsigned int	index;
-	u8		expected;
-	u8		actual;
-};
-
-struct dmatest_verify_result {
-	unsigned int			error_count;
-	struct dmatest_verify_buffer	data[MAX_ERROR_COUNT];
-	u8				pattern;
-	bool				is_srcbuf;
-};
-
-struct dmatest_thread_result {
-	struct list_head	node;
-	unsigned int		n;
-	unsigned int		src_off;
-	unsigned int		dst_off;
-	unsigned int		len;
-	enum dmatest_error_type	type;
-	union {
-		unsigned long			data;
-		dma_cookie_t			cookie;
-		enum dma_status			status;
-		int				error;
-		struct dmatest_verify_result	*vr;
-	};
-};
-
-struct dmatest_result {
-	struct list_head	node;
-	char			*name;
-	struct list_head	results;
-};
-
-struct dmatest_info;
-
-struct dmatest_thread {
-	struct list_head	node;
-	struct dmatest_info	*info;
-	struct task_struct	*task;
-	struct dma_chan		*chan;
-	u8			**srcs;
-	u8			**dsts;
-	enum dma_transaction_type type;
-	bool			done;
-};
+static bool noverify;
+module_param(noverify, bool, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(noverify, "Disable random data setup and verification");
 
-struct dmatest_chan {
-	struct list_head	node;
-	struct dma_chan		*chan;
-	struct list_head	threads;
-};
+static bool verbose;
+module_param(verbose, bool, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(verbose, "Enable \"success\" result messages (default: off)");
 
 /**
  * struct dmatest_params - test parameters.
@@ -177,6 +96,7 @@ struct dmatest_params {
 	unsigned int	xor_sources;
 	unsigned int	pq_sources;
 	int		timeout;
+	bool		noverify;
 };
 
 /**
@@ -184,7 +104,7 @@ struct dmatest_params {
  * @params:		test parameters
  * @lock:		access protection to the fields of this structure
  */
-struct dmatest_info {
+static struct dmatest_info {
 	/* Test parameters */
 	struct dmatest_params	params;
 
@@ -192,16 +112,95 @@ struct dmatest_info {
 	struct list_head	channels;
 	unsigned int		nr_channels;
 	struct mutex		lock;
+	bool			did_init;
+} test_info = {
+	.channels = LIST_HEAD_INIT(test_info.channels),
+	.lock = __MUTEX_INITIALIZER(test_info.lock),
+};
+
+static int dmatest_run_set(const char *val, const struct kernel_param *kp);
+static int dmatest_run_get(char *val, const struct kernel_param *kp);
+static struct kernel_param_ops run_ops = {
+	.set = dmatest_run_set,
+	.get = dmatest_run_get,
+};
+static bool dmatest_run;
+module_param_cb(run, &run_ops, &dmatest_run, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(run, "Run the test (default: false)");
+
+/* Maximum amount of mismatched bytes in buffer to print */
+#define MAX_ERROR_COUNT		32
+
+/*
+ * Initialization patterns. All bytes in the source buffer has bit 7
+ * set, all bytes in the destination buffer has bit 7 cleared.
+ *
+ * Bit 6 is set for all bytes which are to be copied by the DMA
+ * engine. Bit 5 is set for all bytes which are to be overwritten by
+ * the DMA engine.
+ *
+ * The remaining bits are the inverse of a counter which increments by
+ * one for each byte address.
+ */
+#define PATTERN_SRC		0x80
+#define PATTERN_DST		0x00
+#define PATTERN_COPY		0x40
+#define PATTERN_OVERWRITE	0x20
+#define PATTERN_COUNT_MASK	0x1f
 
-	/* debugfs related stuff */
-	struct dentry		*root;
+struct dmatest_thread {
+	struct list_head	node;
+	struct dmatest_info	*info;
+	struct task_struct	*task;
+	struct dma_chan		*chan;
+	u8			**srcs;
+	u8			**dsts;
+	enum dma_transaction_type type;
+	bool			done;
+};
 
-	/* Test results */
-	struct list_head	results;
-	struct mutex		results_lock;
+struct dmatest_chan {
+	struct list_head	node;
+	struct dma_chan		*chan;
+	struct list_head	threads;
 };
 
-static struct dmatest_info test_info;
+static DECLARE_WAIT_QUEUE_HEAD(thread_wait);
+static bool wait;
+
+static bool is_threaded_test_run(struct dmatest_info *info)
+{
+	struct dmatest_chan *dtc;
+
+	list_for_each_entry(dtc, &info->channels, node) {
+		struct dmatest_thread *thread;
+
+		list_for_each_entry(thread, &dtc->threads, node) {
+			if (!thread->done)
+				return true;
+		}
+	}
+
+	return false;
+}
+
+static int dmatest_wait_get(char *val, const struct kernel_param *kp)
+{
+	struct dmatest_info *info = &test_info;
+	struct dmatest_params *params = &info->params;
+
+	if (params->iterations)
+		wait_event(thread_wait, !is_threaded_test_run(info));
+	wait = true;
+	return param_get_bool(val, kp);
+}
+
+static struct kernel_param_ops wait_ops = {
+	.get = dmatest_wait_get,
+	.set = param_set_bool,
+};
+module_param_cb(wait, &wait_ops, &wait, S_IRUGO);
+MODULE_PARM_DESC(wait, "Wait for tests to complete (default: false)");
 
 static bool dmatest_match_channel(struct dmatest_params *params,
 		struct dma_chan *chan)
@@ -223,7 +222,7 @@ static unsigned long dmatest_random(void)
 {
 	unsigned long buf;
 
-	get_random_bytes(&buf, sizeof(buf));
+	prandom_bytes(&buf, sizeof(buf));
 	return buf;
 }
 
@@ -262,9 +261,31 @@ static void dmatest_init_dsts(u8 **bufs, unsigned int start, unsigned int len,
 	}
 }
 
-static unsigned int dmatest_verify(struct dmatest_verify_result *vr, u8 **bufs,
-		unsigned int start, unsigned int end, unsigned int counter,
-		u8 pattern, bool is_srcbuf)
+static void dmatest_mismatch(u8 actual, u8 pattern, unsigned int index,
+		unsigned int counter, bool is_srcbuf)
+{
+	u8		diff = actual ^ pattern;
+	u8		expected = pattern | (~counter & PATTERN_COUNT_MASK);
+	const char	*thread_name = current->comm;
+
+	if (is_srcbuf)
+		pr_warn("%s: srcbuf[0x%x] overwritten! Expected %02x, got %02x\n",
+			thread_name, index, expected, actual);
+	else if ((pattern & PATTERN_COPY)
+			&& (diff & (PATTERN_COPY | PATTERN_OVERWRITE)))
+		pr_warn("%s: dstbuf[0x%x] not copied! Expected %02x, got %02x\n",
+			thread_name, index, expected, actual);
+	else if (diff & PATTERN_SRC)
+		pr_warn("%s: dstbuf[0x%x] was copied! Expected %02x, got %02x\n",
+			thread_name, index, expected, actual);
+	else
+		pr_warn("%s: dstbuf[0x%x] mismatch! Expected %02x, got %02x\n",
+			thread_name, index, expected, actual);
+}
+
+static unsigned int dmatest_verify(u8 **bufs, unsigned int start,
+		unsigned int end, unsigned int counter, u8 pattern,
+		bool is_srcbuf)
 {
 	unsigned int i;
 	unsigned int error_count = 0;
@@ -272,7 +293,6 @@ static unsigned int dmatest_verify(struct dmatest_verify_result *vr, u8 **bufs,
 	u8 expected;
 	u8 *buf;
 	unsigned int counter_orig = counter;
-	struct dmatest_verify_buffer *vb;
 
 	for (; (buf = *bufs); bufs++) {
 		counter = counter_orig;
@@ -280,12 +300,9 @@ static unsigned int dmatest_verify(struct dmatest_verify_result *vr, u8 **bufs,
 			actual = buf[i];
 			expected = pattern | (~counter & PATTERN_COUNT_MASK);
 			if (actual != expected) {
-				if (error_count < MAX_ERROR_COUNT && vr) {
-					vb = &vr->data[error_count];
-					vb->index = i;
-					vb->expected = expected;
-					vb->actual = actual;
-				}
+				if (error_count < MAX_ERROR_COUNT)
+					dmatest_mismatch(actual, pattern, i,
+							 counter, is_srcbuf);
 				error_count++;
 			}
 			counter++;
@@ -293,7 +310,7 @@ static unsigned int dmatest_verify(struct dmatest_verify_result *vr, u8 **bufs,
 	}
 
 	if (error_count > MAX_ERROR_COUNT)
-		pr_warning("%s: %u errors suppressed\n",
+		pr_warn("%s: %u errors suppressed\n",
 			current->comm, error_count - MAX_ERROR_COUNT);
 
 	return error_count;
@@ -313,20 +330,6 @@ static void dmatest_callback(void *arg)
 	wake_up_all(done->wait);
 }
 
-static inline void unmap_src(struct device *dev, dma_addr_t *addr, size_t len,
-			     unsigned int count)
-{
-	while (count--)
-		dma_unmap_single(dev, addr[count], len, DMA_TO_DEVICE);
-}
-
-static inline void unmap_dst(struct device *dev, dma_addr_t *addr, size_t len,
-			     unsigned int count)
-{
-	while (count--)
-		dma_unmap_single(dev, addr[count], len, DMA_BIDIRECTIONAL);
-}
-
 static unsigned int min_odd(unsigned int x, unsigned int y)
 {
 	unsigned int val = min(x, y);
@@ -334,172 +337,49 @@ static unsigned int min_odd(unsigned int x, unsigned int y)
 	return val % 2 ? val : val - 1;
 }
 
-static char *verify_result_get_one(struct dmatest_verify_result *vr,
-		unsigned int i)
+static void result(const char *err, unsigned int n, unsigned int src_off,
+		   unsigned int dst_off, unsigned int len, unsigned long data)
 {
-	struct dmatest_verify_buffer *vb = &vr->data[i];
-	u8 diff = vb->actual ^ vr->pattern;
-	static char buf[512];
-	char *msg;
-
-	if (vr->is_srcbuf)
-		msg = "srcbuf overwritten!";
-	else if ((vr->pattern & PATTERN_COPY)
-			&& (diff & (PATTERN_COPY | PATTERN_OVERWRITE)))
-		msg = "dstbuf not copied!";
-	else if (diff & PATTERN_SRC)
-		msg = "dstbuf was copied!";
-	else
-		msg = "dstbuf mismatch!";
-
-	snprintf(buf, sizeof(buf) - 1, "%s [0x%x] Expected %02x, got %02x", msg,
-		 vb->index, vb->expected, vb->actual);
-
-	return buf;
+	pr_info("%s: result #%u: '%s' with src_off=0x%x dst_off=0x%x len=0x%x (%lu)",
+		current->comm, n, err, src_off, dst_off, len, data);
 }
 
-static char *thread_result_get(const char *name,
-		struct dmatest_thread_result *tr)
+static void dbg_result(const char *err, unsigned int n, unsigned int src_off,
+		       unsigned int dst_off, unsigned int len,
+		       unsigned long data)
 {
-	static const char * const messages[] = {
-		[DMATEST_ET_OK]			= "No errors",
-		[DMATEST_ET_MAP_SRC]		= "src mapping error",
-		[DMATEST_ET_MAP_DST]		= "dst mapping error",
-		[DMATEST_ET_PREP]		= "prep error",
-		[DMATEST_ET_SUBMIT]		= "submit error",
-		[DMATEST_ET_TIMEOUT]		= "test timed out",
-		[DMATEST_ET_DMA_ERROR]		=
-			"got completion callback (DMA_ERROR)",
-		[DMATEST_ET_DMA_IN_PROGRESS]	=
-			"got completion callback (DMA_IN_PROGRESS)",
-		[DMATEST_ET_VERIFY]		= "errors",
-		[DMATEST_ET_VERIFY_BUF]		= "verify errors",
-	};
-	static char buf[512];
-
-	snprintf(buf, sizeof(buf) - 1,
-		 "%s: #%u: %s with src_off=0x%x ""dst_off=0x%x len=0x%x (%lu)",
-		 name, tr->n, messages[tr->type], tr->src_off, tr->dst_off,
-		 tr->len, tr->data);
-
-	return buf;
+	pr_debug("%s: result #%u: '%s' with src_off=0x%x dst_off=0x%x len=0x%x (%lu)",
+		   current->comm, n, err, src_off, dst_off, len, data);
 }
 
-static int thread_result_add(struct dmatest_info *info,
-		struct dmatest_result *r, enum dmatest_error_type type,
-		unsigned int n, unsigned int src_off, unsigned int dst_off,
-		unsigned int len, unsigned long data)
-{
-	struct dmatest_thread_result *tr;
-
-	tr = kzalloc(sizeof(*tr), GFP_KERNEL);
-	if (!tr)
-		return -ENOMEM;
-
-	tr->type = type;
-	tr->n = n;
-	tr->src_off = src_off;
-	tr->dst_off = dst_off;
-	tr->len = len;
-	tr->data = data;
+#define verbose_result(err, n, src_off, dst_off, len, data) ({ \
+	if (verbose) \
+		result(err, n, src_off, dst_off, len, data); \
+	else \
+		dbg_result(err, n, src_off, dst_off, len, data); \
+})
 
-	mutex_lock(&info->results_lock);
-	list_add_tail(&tr->node, &r->results);
-	mutex_unlock(&info->results_lock);
-
-	if (tr->type == DMATEST_ET_OK)
-		pr_debug("%s\n", thread_result_get(r->name, tr));
-	else
-		pr_warn("%s\n", thread_result_get(r->name, tr));
-
-	return 0;
-}
-
-static unsigned int verify_result_add(struct dmatest_info *info,
-		struct dmatest_result *r, unsigned int n,
-		unsigned int src_off, unsigned int dst_off, unsigned int len,
-		u8 **bufs, int whence, unsigned int counter, u8 pattern,
-		bool is_srcbuf)
+static unsigned long long dmatest_persec(s64 runtime, unsigned int val)
 {
-	struct dmatest_verify_result *vr;
-	unsigned int error_count;
-	unsigned int buf_off = is_srcbuf ? src_off : dst_off;
-	unsigned int start, end;
-
-	if (whence < 0) {
-		start = 0;
-		end = buf_off;
-	} else if (whence > 0) {
-		start = buf_off + len;
-		end = info->params.buf_size;
-	} else {
-		start = buf_off;
-		end = buf_off + len;
-	}
+	unsigned long long per_sec = 1000000;
 
-	vr = kmalloc(sizeof(*vr), GFP_KERNEL);
-	if (!vr) {
-		pr_warn("dmatest: No memory to store verify result\n");
-		return dmatest_verify(NULL, bufs, start, end, counter, pattern,
-				      is_srcbuf);
-	}
-
-	vr->pattern = pattern;
-	vr->is_srcbuf = is_srcbuf;
-
-	error_count = dmatest_verify(vr, bufs, start, end, counter, pattern,
-				     is_srcbuf);
-	if (error_count) {
-		vr->error_count = error_count;
-		thread_result_add(info, r, DMATEST_ET_VERIFY_BUF, n, src_off,
-				  dst_off, len, (unsigned long)vr);
-		return error_count;
-	}
-
-	kfree(vr);
-	return 0;
-}
-
-static void result_free(struct dmatest_info *info, const char *name)
-{
-	struct dmatest_result *r, *_r;
-
-	mutex_lock(&info->results_lock);
-	list_for_each_entry_safe(r, _r, &info->results, node) {
-		struct dmatest_thread_result *tr, *_tr;
-
-		if (name && strcmp(r->name, name))
-			continue;
-
-		list_for_each_entry_safe(tr, _tr, &r->results, node) {
-			if (tr->type == DMATEST_ET_VERIFY_BUF)
-				kfree(tr->vr);
-			list_del(&tr->node);
-			kfree(tr);
-		}
+	if (runtime <= 0)
+		return 0;
 
-		kfree(r->name);
-		list_del(&r->node);
-		kfree(r);
+	/* drop precision until runtime is 32-bits */
+	while (runtime > UINT_MAX) {
+		runtime >>= 1;
+		per_sec <<= 1;
 	}
 
-	mutex_unlock(&info->results_lock);
+	per_sec *= val;
+	do_div(per_sec, runtime);
+	return per_sec;
 }
 
-static struct dmatest_result *result_init(struct dmatest_info *info,
-		const char *name)
+static unsigned long long dmatest_KBs(s64 runtime, unsigned long long len)
 {
-	struct dmatest_result *r;
-
-	r = kzalloc(sizeof(*r), GFP_KERNEL);
-	if (r) {
-		r->name = kstrdup(name, GFP_KERNEL);
-		INIT_LIST_HEAD(&r->results);
-		mutex_lock(&info->results_lock);
-		list_add_tail(&r->node, &info->results);
-		mutex_unlock(&info->results_lock);
-	}
-	return r;
+	return dmatest_persec(runtime, len >> 10);
 }
 
 /*
@@ -525,7 +405,6 @@ static int dmatest_func(void *data)
 	struct dmatest_params	*params;
 	struct dma_chan		*chan;
 	struct dma_device	*dev;
-	const char		*thread_name;
 	unsigned int		src_off, dst_off, len;
 	unsigned int		error_count;
 	unsigned int		failed_tests = 0;
@@ -538,9 +417,10 @@ static int dmatest_func(void *data)
 	int			src_cnt;
 	int			dst_cnt;
 	int			i;
-	struct dmatest_result	*result;
+	ktime_t			ktime;
+	s64			runtime = 0;
+	unsigned long long	total_len = 0;
 
-	thread_name = current->comm;
 	set_freezable();
 
 	ret = -ENOMEM;
@@ -570,10 +450,6 @@ static int dmatest_func(void *data)
 	} else
 		goto err_thread_type;
 
-	result = result_init(info, thread_name);
-	if (!result)
-		goto err_srcs;
-
 	thread->srcs = kcalloc(src_cnt+1, sizeof(u8 *), GFP_KERNEL);
 	if (!thread->srcs)
 		goto err_srcs;
@@ -597,17 +473,17 @@ static int dmatest_func(void *data)
 	set_user_nice(current, 10);
 
 	/*
-	 * src buffers are freed by the DMAEngine code with dma_unmap_single()
-	 * dst buffers are freed by ourselves below
+	 * src and dst buffers are freed by ourselves below
 	 */
-	flags = DMA_CTRL_ACK | DMA_PREP_INTERRUPT
-	      | DMA_COMPL_SKIP_DEST_UNMAP | DMA_COMPL_SRC_UNMAP_SINGLE;
+	flags = DMA_CTRL_ACK | DMA_PREP_INTERRUPT;
 
+	ktime = ktime_get();
 	while (!kthread_should_stop()
 	       && !(params->iterations && total_tests >= params->iterations)) {
 		struct dma_async_tx_descriptor *tx = NULL;
-		dma_addr_t dma_srcs[src_cnt];
-		dma_addr_t dma_dsts[dst_cnt];
+		struct dmaengine_unmap_data *um;
+		dma_addr_t srcs[src_cnt];
+		dma_addr_t *dsts;
 		u8 align = 0;
 
 		total_tests++;
@@ -626,81 +502,103 @@ static int dmatest_func(void *data)
 			break;
 		}
 
-		len = dmatest_random() % params->buf_size + 1;
+		if (params->noverify) {
+			len = params->buf_size;
+			src_off = 0;
+			dst_off = 0;
+		} else {
+			len = dmatest_random() % params->buf_size + 1;
+			len = (len >> align) << align;
+			if (!len)
+				len = 1 << align;
+			src_off = dmatest_random() % (params->buf_size - len + 1);
+			dst_off = dmatest_random() % (params->buf_size - len + 1);
+
+			src_off = (src_off >> align) << align;
+			dst_off = (dst_off >> align) << align;
+
+			dmatest_init_srcs(thread->srcs, src_off, len,
+					  params->buf_size);
+			dmatest_init_dsts(thread->dsts, dst_off, len,
+					  params->buf_size);
+		}
+
 		len = (len >> align) << align;
 		if (!len)
 			len = 1 << align;
-		src_off = dmatest_random() % (params->buf_size - len + 1);
-		dst_off = dmatest_random() % (params->buf_size - len + 1);
+		total_len += len;
 
-		src_off = (src_off >> align) << align;
-		dst_off = (dst_off >> align) << align;
-
-		dmatest_init_srcs(thread->srcs, src_off, len, params->buf_size);
-		dmatest_init_dsts(thread->dsts, dst_off, len, params->buf_size);
+		um = dmaengine_get_unmap_data(dev->dev, src_cnt+dst_cnt,
+					      GFP_KERNEL);
+		if (!um) {
+			failed_tests++;
+			result("unmap data NULL", total_tests,
+			       src_off, dst_off, len, ret);
+			continue;
+		}
 
+		um->len = params->buf_size;
 		for (i = 0; i < src_cnt; i++) {
-			u8 *buf = thread->srcs[i] + src_off;
-
-			dma_srcs[i] = dma_map_single(dev->dev, buf, len,
-						     DMA_TO_DEVICE);
-			ret = dma_mapping_error(dev->dev, dma_srcs[i]);
+			unsigned long buf = (unsigned long) thread->srcs[i];
+			struct page *pg = virt_to_page(buf);
+			unsigned pg_off = buf & ~PAGE_MASK;
+
+			um->addr[i] = dma_map_page(dev->dev, pg, pg_off,
+						   um->len, DMA_TO_DEVICE);
+			srcs[i] = um->addr[i] + src_off;
+			ret = dma_mapping_error(dev->dev, um->addr[i]);
 			if (ret) {
-				unmap_src(dev->dev, dma_srcs, len, i);
-				thread_result_add(info, result,
-						  DMATEST_ET_MAP_SRC,
-						  total_tests, src_off, dst_off,
-						  len, ret);
+				dmaengine_unmap_put(um);
+				result("src mapping error", total_tests,
+				       src_off, dst_off, len, ret);
 				failed_tests++;
 				continue;
 			}
+			um->to_cnt++;
 		}
 		/* map with DMA_BIDIRECTIONAL to force writeback/invalidate */
+		dsts = &um->addr[src_cnt];
 		for (i = 0; i < dst_cnt; i++) {
-			dma_dsts[i] = dma_map_single(dev->dev, thread->dsts[i],
-						     params->buf_size,
-						     DMA_BIDIRECTIONAL);
-			ret = dma_mapping_error(dev->dev, dma_dsts[i]);
+			unsigned long buf = (unsigned long) thread->dsts[i];
+			struct page *pg = virt_to_page(buf);
+			unsigned pg_off = buf & ~PAGE_MASK;
+
+			dsts[i] = dma_map_page(dev->dev, pg, pg_off, um->len,
+					       DMA_BIDIRECTIONAL);
+			ret = dma_mapping_error(dev->dev, dsts[i]);
 			if (ret) {
-				unmap_src(dev->dev, dma_srcs, len, src_cnt);
-				unmap_dst(dev->dev, dma_dsts, params->buf_size,
-					  i);
-				thread_result_add(info, result,
-						  DMATEST_ET_MAP_DST,
-						  total_tests, src_off, dst_off,
-						  len, ret);
+				dmaengine_unmap_put(um);
+				result("dst mapping error", total_tests,
+				       src_off, dst_off, len, ret);
 				failed_tests++;
 				continue;
 			}
+			um->bidi_cnt++;
 		}
 
 		if (thread->type == DMA_MEMCPY)
 			tx = dev->device_prep_dma_memcpy(chan,
-							 dma_dsts[0] + dst_off,
-							 dma_srcs[0], len,
-							 flags);
+							 dsts[0] + dst_off,
+							 srcs[0], len, flags);
 		else if (thread->type == DMA_XOR)
 			tx = dev->device_prep_dma_xor(chan,
-						      dma_dsts[0] + dst_off,
-						      dma_srcs, src_cnt,
+						      dsts[0] + dst_off,
+						      srcs, src_cnt,
 						      len, flags);
 		else if (thread->type == DMA_PQ) {
 			dma_addr_t dma_pq[dst_cnt];
 
 			for (i = 0; i < dst_cnt; i++)
-				dma_pq[i] = dma_dsts[i] + dst_off;
-			tx = dev->device_prep_dma_pq(chan, dma_pq, dma_srcs,
+				dma_pq[i] = dsts[i] + dst_off;
+			tx = dev->device_prep_dma_pq(chan, dma_pq, srcs,
 						     src_cnt, pq_coefs,
 						     len, flags);
 		}
 
 		if (!tx) {
-			unmap_src(dev->dev, dma_srcs, len, src_cnt);
-			unmap_dst(dev->dev, dma_dsts, params->buf_size,
-				  dst_cnt);
-			thread_result_add(info, result, DMATEST_ET_PREP,
-					  total_tests, src_off, dst_off,
-					  len, 0);
+			dmaengine_unmap_put(um);
+			result("prep error", total_tests, src_off,
+			       dst_off, len, ret);
 			msleep(100);
 			failed_tests++;
 			continue;
@@ -712,9 +610,9 @@ static int dmatest_func(void *data)
 		cookie = tx->tx_submit(tx);
 
 		if (dma_submit_error(cookie)) {
-			thread_result_add(info, result, DMATEST_ET_SUBMIT,
-					  total_tests, src_off, dst_off,
-					  len, cookie);
+			dmaengine_unmap_put(um);
+			result("submit error", total_tests, src_off,
+			       dst_off, len, ret);
 			msleep(100);
 			failed_tests++;
 			continue;
@@ -735,59 +633,59 @@ static int dmatest_func(void *data)
 			 * free it this time?" dancing.  For now, just
 			 * leave it dangling.
 			 */
-			thread_result_add(info, result, DMATEST_ET_TIMEOUT,
-					  total_tests, src_off, dst_off,
-					  len, 0);
+			dmaengine_unmap_put(um);
+			result("test timed out", total_tests, src_off, dst_off,
+			       len, 0);
 			failed_tests++;
 			continue;
 		} else if (status != DMA_COMPLETE) {
-			enum dmatest_error_type type = (status == DMA_ERROR) ?
-				DMATEST_ET_DMA_ERROR : DMATEST_ET_DMA_IN_PROGRESS;
-			thread_result_add(info, result, type,
-					  total_tests, src_off, dst_off,
-					  len, status);
+			dmaengine_unmap_put(um);
+			result(status == DMA_ERROR ?
+			       "completion error status" :
+			       "completion busy status", total_tests, src_off,
+			       dst_off, len, ret);
 			failed_tests++;
 			continue;
 		}
 
-		/* Unmap by myself (see DMA_COMPL_SKIP_DEST_UNMAP above) */
-		unmap_dst(dev->dev, dma_dsts, params->buf_size, dst_cnt);
+		dmaengine_unmap_put(um);
 
-		error_count = 0;
+		if (params->noverify) {
+			verbose_result("test passed", total_tests, src_off,
+				       dst_off, len, 0);
+			continue;
+		}
 
-		pr_debug("%s: verifying source buffer...\n", thread_name);
-		error_count += verify_result_add(info, result, total_tests,
-				src_off, dst_off, len, thread->srcs, -1,
+		pr_debug("%s: verifying source buffer...\n", current->comm);
+		error_count = dmatest_verify(thread->srcs, 0, src_off,
 				0, PATTERN_SRC, true);
-		error_count += verify_result_add(info, result, total_tests,
-				src_off, dst_off, len, thread->srcs, 0,
-				src_off, PATTERN_SRC | PATTERN_COPY, true);
-		error_count += verify_result_add(info, result, total_tests,
-				src_off, dst_off, len, thread->srcs, 1,
-				src_off + len, PATTERN_SRC, true);
-
-		pr_debug("%s: verifying dest buffer...\n", thread_name);
-		error_count += verify_result_add(info, result, total_tests,
-				src_off, dst_off, len, thread->dsts, -1,
+		error_count += dmatest_verify(thread->srcs, src_off,
+				src_off + len, src_off,
+				PATTERN_SRC | PATTERN_COPY, true);
+		error_count += dmatest_verify(thread->srcs, src_off + len,
+				params->buf_size, src_off + len,
+				PATTERN_SRC, true);
+
+		pr_debug("%s: verifying dest buffer...\n", current->comm);
+		error_count += dmatest_verify(thread->dsts, 0, dst_off,
 				0, PATTERN_DST, false);
-		error_count += verify_result_add(info, result, total_tests,
-				src_off, dst_off, len, thread->dsts, 0,
-				src_off, PATTERN_SRC | PATTERN_COPY, false);
-		error_count += verify_result_add(info, result, total_tests,
-				src_off, dst_off, len, thread->dsts, 1,
-				dst_off + len, PATTERN_DST, false);
+		error_count += dmatest_verify(thread->dsts, dst_off,
+				dst_off + len, src_off,
+				PATTERN_SRC | PATTERN_COPY, false);
+		error_count += dmatest_verify(thread->dsts, dst_off + len,
+				params->buf_size, dst_off + len,
+				PATTERN_DST, false);
 
 		if (error_count) {
-			thread_result_add(info, result, DMATEST_ET_VERIFY,
-					  total_tests, src_off, dst_off,
-					  len, error_count);
+			result("data error", total_tests, src_off, dst_off,
+			       len, error_count);
 			failed_tests++;
 		} else {
-			thread_result_add(info, result, DMATEST_ET_OK,
-					  total_tests, src_off, dst_off,
-					  len, 0);
+			verbose_result("test passed", total_tests, src_off,
+				       dst_off, len, 0);
 		}
 	}
+	runtime = ktime_us_delta(ktime_get(), ktime);
 
 	ret = 0;
 	for (i = 0; thread->dsts[i]; i++)
@@ -802,20 +700,17 @@ err_srcbuf:
 err_srcs:
 	kfree(pq_coefs);
 err_thread_type:
-	pr_notice("%s: terminating after %u tests, %u failures (status %d)\n",
-			thread_name, total_tests, failed_tests, ret);
+	pr_info("%s: summary %u tests, %u failures %llu iops %llu KB/s (%d)\n",
+		current->comm, total_tests, failed_tests,
+		dmatest_persec(runtime, total_tests),
+		dmatest_KBs(runtime, total_len), ret);
 
 	/* terminate all transfers on specified channels */
 	if (ret)
 		dmaengine_terminate_all(chan);
 
 	thread->done = true;
-
-	if (params->iterations > 0)
-		while (!kthread_should_stop()) {
-			DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wait_dmatest_exit);
-			interruptible_sleep_on(&wait_dmatest_exit);
-		}
+	wake_up(&thread_wait);
 
 	return ret;
 }
@@ -828,9 +723,10 @@ static void dmatest_cleanup_channel(struct dmatest_chan *dtc)
 
 	list_for_each_entry_safe(thread, _thread, &dtc->threads, node) {
 		ret = kthread_stop(thread->task);
-		pr_debug("dmatest: thread %s exited with status %d\n",
-				thread->task->comm, ret);
+		pr_debug("thread %s exited with status %d\n",
+			 thread->task->comm, ret);
 		list_del(&thread->node);
+		put_task_struct(thread->task);
 		kfree(thread);
 	}
 
@@ -861,27 +757,27 @@ static int dmatest_add_threads(struct dmatest_info *info,
 	for (i = 0; i < params->threads_per_chan; i++) {
 		thread = kzalloc(sizeof(struct dmatest_thread), GFP_KERNEL);
 		if (!thread) {
-			pr_warning("dmatest: No memory for %s-%s%u\n",
-				   dma_chan_name(chan), op, i);
-
+			pr_warn("No memory for %s-%s%u\n",
+				dma_chan_name(chan), op, i);
 			break;
 		}
 		thread->info = info;
 		thread->chan = dtc->chan;
 		thread->type = type;
 		smp_wmb();
-		thread->task = kthread_run(dmatest_func, thread, "%s-%s%u",
+		thread->task = kthread_create(dmatest_func, thread, "%s-%s%u",
 				dma_chan_name(chan), op, i);
 		if (IS_ERR(thread->task)) {
-			pr_warning("dmatest: Failed to run thread %s-%s%u\n",
-					dma_chan_name(chan), op, i);
+			pr_warn("Failed to create thread %s-%s%u\n",
+				dma_chan_name(chan), op, i);
 			kfree(thread);
 			break;
 		}
 
 		/* srcbuf and dstbuf are allocated by the thread itself */
-
+		get_task_struct(thread->task);
 		list_add_tail(&thread->node, &dtc->threads);
+		wake_up_process(thread->task);
 	}
 
 	return i;
@@ -897,7 +793,7 @@ static int dmatest_add_channel(struct dmatest_info *info,
 
 	dtc = kmalloc(sizeof(struct dmatest_chan), GFP_KERNEL);
 	if (!dtc) {
-		pr_warning("dmatest: No memory for %s\n", dma_chan_name(chan));
+		pr_warn("No memory for %s\n", dma_chan_name(chan));
 		return -ENOMEM;
 	}
 
@@ -917,7 +813,7 @@ static int dmatest_add_channel(struct dmatest_info *info,
 		thread_count += cnt > 0 ? cnt : 0;
 	}
 
-	pr_info("dmatest: Started %u threads using %s\n",
+	pr_info("Started %u threads using %s\n",
 		thread_count, dma_chan_name(chan));
 
 	list_add_tail(&dtc->node, &info->channels);
@@ -937,20 +833,20 @@ static bool filter(struct dma_chan *chan, void *param)
 		return true;
 }
 
-static int __run_threaded_test(struct dmatest_info *info)
+static void request_channels(struct dmatest_info *info,
+			     enum dma_transaction_type type)
 {
 	dma_cap_mask_t mask;
-	struct dma_chan *chan;
-	struct dmatest_params *params = &info->params;
-	int err = 0;
 
 	dma_cap_zero(mask);
-	dma_cap_set(DMA_MEMCPY, mask);
+	dma_cap_set(type, mask);
 	for (;;) {
+		struct dmatest_params *params = &info->params;
+		struct dma_chan *chan;
+
 		chan = dma_request_channel(mask, filter, params);
 		if (chan) {
-			err = dmatest_add_channel(info, chan);
-			if (err) {
+			if (dmatest_add_channel(info, chan)) {
 				dma_release_channel(chan);
 				break; /* add_channel failed, punt */
 			}
@@ -960,22 +856,30 @@ static int __run_threaded_test(struct dmatest_info *info)
 		    info->nr_channels >= params->max_channels)
 			break; /* we have all we need */
 	}
-	return err;
 }
 
-#ifndef MODULE
-static int run_threaded_test(struct dmatest_info *info)
+static void run_threaded_test(struct dmatest_info *info)
 {
-	int ret;
+	struct dmatest_params *params = &info->params;
 
-	mutex_lock(&info->lock);
-	ret = __run_threaded_test(info);
-	mutex_unlock(&info->lock);
-	return ret;
+	/* Copy test parameters */
+	params->buf_size = test_buf_size;
+	strlcpy(params->channel, strim(test_channel), sizeof(params->channel));
+	strlcpy(params->device, strim(test_device), sizeof(params->device));
+	params->threads_per_chan = threads_per_chan;
+	params->max_channels = max_channels;
+	params->iterations = iterations;
+	params->xor_sources = xor_sources;
+	params->pq_sources = pq_sources;
+	params->timeout = timeout;
+	params->noverify = noverify;
+
+	request_channels(info, DMA_MEMCPY);
+	request_channels(info, DMA_XOR);
+	request_channels(info, DMA_PQ);
 }
-#endif
 
-static void __stop_threaded_test(struct dmatest_info *info)
+static void stop_threaded_test(struct dmatest_info *info)
 {
 	struct dmatest_chan *dtc, *_dtc;
 	struct dma_chan *chan;
@@ -984,203 +888,86 @@ static void __stop_threaded_test(struct dmatest_info *info)
 		list_del(&dtc->node);
 		chan = dtc->chan;
 		dmatest_cleanup_channel(dtc);
-		pr_debug("dmatest: dropped channel %s\n", dma_chan_name(chan));
+		pr_debug("dropped channel %s\n", dma_chan_name(chan));
 		dma_release_channel(chan);
 	}
 
 	info->nr_channels = 0;
 }
 
-static void stop_threaded_test(struct dmatest_info *info)
+static void restart_threaded_test(struct dmatest_info *info, bool run)
 {
-	mutex_lock(&info->lock);
-	__stop_threaded_test(info);
-	mutex_unlock(&info->lock);
-}
-
-static int __restart_threaded_test(struct dmatest_info *info, bool run)
-{
-	struct dmatest_params *params = &info->params;
+	/* we might be called early to set run=, defer running until all
+	 * parameters have been evaluated
+	 */
+	if (!info->did_init)
+		return;
 
 	/* Stop any running test first */
-	__stop_threaded_test(info);
-
-	if (run == false)
-		return 0;
-
-	/* Clear results from previous run */
-	result_free(info, NULL);
-
-	/* Copy test parameters */
-	params->buf_size = test_buf_size;
-	strlcpy(params->channel, strim(test_channel), sizeof(params->channel));
-	strlcpy(params->device, strim(test_device), sizeof(params->device));
-	params->threads_per_chan = threads_per_chan;
-	params->max_channels = max_channels;
-	params->iterations = iterations;
-	params->xor_sources = xor_sources;
-	params->pq_sources = pq_sources;
-	params->timeout = timeout;
+	stop_threaded_test(info);
 
 	/* Run test with new parameters */
-	return __run_threaded_test(info);
-}
-
-static bool __is_threaded_test_run(struct dmatest_info *info)
-{
-	struct dmatest_chan *dtc;
-
-	list_for_each_entry(dtc, &info->channels, node) {
-		struct dmatest_thread *thread;
-
-		list_for_each_entry(thread, &dtc->threads, node) {
-			if (!thread->done)
-				return true;
-		}
-	}
-
-	return false;
+	run_threaded_test(info);
 }
 
-static ssize_t dtf_read_run(struct file *file, char __user *user_buf,
-		size_t count, loff_t *ppos)
+static int dmatest_run_get(char *val, const struct kernel_param *kp)
 {
-	struct dmatest_info *info = file->private_data;
-	char buf[3];
+	struct dmatest_info *info = &test_info;
 
 	mutex_lock(&info->lock);
-
-	if (__is_threaded_test_run(info)) {
-		buf[0] = 'Y';
+	if (is_threaded_test_run(info)) {
+		dmatest_run = true;
 	} else {
-		__stop_threaded_test(info);
-		buf[0] = 'N';
+		stop_threaded_test(info);
+		dmatest_run = false;
 	}
-
 	mutex_unlock(&info->lock);
-	buf[1] = '\n';
-	buf[2] = 0x00;
-	return simple_read_from_buffer(user_buf, count, ppos, buf, 2);
-}
-
-static ssize_t dtf_write_run(struct file *file, const char __user *user_buf,
-		size_t count, loff_t *ppos)
-{
-	struct dmatest_info *info = file->private_data;
-	char buf[16];
-	bool bv;
-	int ret = 0;
 
-	if (copy_from_user(buf, user_buf, min(count, (sizeof(buf) - 1))))
-		return -EFAULT;
-
-	if (strtobool(buf, &bv) == 0) {
-		mutex_lock(&info->lock);
-
-		if (__is_threaded_test_run(info))
-			ret = -EBUSY;
-		else
-			ret = __restart_threaded_test(info, bv);
-
-		mutex_unlock(&info->lock);
-	}
-
-	return ret ? ret : count;
+	return param_get_bool(val, kp);
 }
 
-static const struct file_operations dtf_run_fops = {
-	.read	= dtf_read_run,
-	.write	= dtf_write_run,
-	.open	= simple_open,
-	.llseek	= default_llseek,
-};
-
-static int dtf_results_show(struct seq_file *sf, void *data)
+static int dmatest_run_set(const char *val, const struct kernel_param *kp)
 {
-	struct dmatest_info *info = sf->private;
-	struct dmatest_result *result;
-	struct dmatest_thread_result *tr;
-	unsigned int i;
+	struct dmatest_info *info = &test_info;
+	int ret;
 
-	mutex_lock(&info->results_lock);
-	list_for_each_entry(result, &info->results, node) {
-		list_for_each_entry(tr, &result->results, node) {
-			seq_printf(sf, "%s\n",
-				thread_result_get(result->name, tr));
-			if (tr->type == DMATEST_ET_VERIFY_BUF) {
-				for (i = 0; i < tr->vr->error_count; i++) {
-					seq_printf(sf, "\t%s\n",
-						verify_result_get_one(tr->vr, i));
-				}
-			}
-		}
+	mutex_lock(&info->lock);
+	ret = param_set_bool(val, kp);
+	if (ret) {
+		mutex_unlock(&info->lock);
+		return ret;
 	}
 
-	mutex_unlock(&info->results_lock);
-	return 0;
-}
-
-static int dtf_results_open(struct inode *inode, struct file *file)
-{
-	return single_open(file, dtf_results_show, inode->i_private);
-}
-
-static const struct file_operations dtf_results_fops = {
-	.open		= dtf_results_open,
-	.read		= seq_read,
-	.llseek		= seq_lseek,
-	.release	= single_release,
-};
-
-static int dmatest_register_dbgfs(struct dmatest_info *info)
-{
-	struct dentry *d;
-
-	d = debugfs_create_dir("dmatest", NULL);
-	if (IS_ERR(d))
-		return PTR_ERR(d);
-	if (!d)
-		goto err_root;
+	if (is_threaded_test_run(info))
+		ret = -EBUSY;
+	else if (dmatest_run)
+		restart_threaded_test(info, dmatest_run);
 
-	info->root = d;
-
-	/* Run or stop threaded test */
-	debugfs_create_file("run", S_IWUSR | S_IRUGO, info->root, info,
-			    &dtf_run_fops);
-
-	/* Results of test in progress */
-	debugfs_create_file("results", S_IRUGO, info->root, info,
-			    &dtf_results_fops);
-
-	return 0;
+	mutex_unlock(&info->lock);
 
-err_root:
-	pr_err("dmatest: Failed to initialize debugfs\n");
-	return -ENOMEM;
+	return ret;
 }
 
 static int __init dmatest_init(void)
 {
 	struct dmatest_info *info = &test_info;
-	int ret;
-
-	memset(info, 0, sizeof(*info));
+	struct dmatest_params *params = &info->params;
 
-	mutex_init(&info->lock);
-	INIT_LIST_HEAD(&info->channels);
+	if (dmatest_run) {
+		mutex_lock(&info->lock);
+		run_threaded_test(info);
+		mutex_unlock(&info->lock);
+	}
 
-	mutex_init(&info->results_lock);
-	INIT_LIST_HEAD(&info->results);
+	if (params->iterations && wait)
+		wait_event(thread_wait, !is_threaded_test_run(info));
 
-	ret = dmatest_register_dbgfs(info);
-	if (ret)
-		return ret;
+	/* module parameters are stable, inittime tests are started,
+	 * let userspace take over 'run' control
+	 */
+	info->did_init = true;
 
-#ifdef MODULE
 	return 0;
-#else
-	return run_threaded_test(info);
-#endif
 }
 /* when compiled-in wait for drivers to load first */
 late_initcall(dmatest_init);
@@ -1189,9 +976,9 @@ static void __exit dmatest_exit(void)
 {
 	struct dmatest_info *info = &test_info;
 
-	debugfs_remove_recursive(info->root);
+	mutex_lock(&info->lock);
 	stop_threaded_test(info);
-	result_free(info, NULL);
+	mutex_unlock(&info->lock);
 }
 module_exit(dmatest_exit);