The rx path of the 8250_dma user in the RX-timeout case:
- it starts the RX transfer
- if the rx-timeout interrupt occurs, it dmaengine_pause() the transfer
- step two is dmaengine_terminate_all() on this channel.
- the rx interrupt occurs, it does not start the RX-transfer because
  according to dmaengine_tx_status() the status of the current transfer
  is still DMA_IN_PROGRESS because the earlier dmaengine_terminate_all()
  did not reset this.

To fix this problem, the patch adds dma_cookie_complete() "complete" the
current cookie.

Cc: Russell King <redacted>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
---
 drivers/dma/omap-dma.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/dma/omap-dma.c b/drivers/dma/omap-dma.c
index b19f04f..81ede01 100644
--- a/drivers/dma/omap-dma.c
+++ b/drivers/dma/omap-dma.c

@@ -970,6 +970,12 @@ static int omap_dma_terminate_all(struct omap_chan *c)
 
 	/* Prevent this channel being scheduled */
 	spin_lock(&d->lock);
+	if (!c->desc && !list_empty(&c->node)) {
+		struct virt_dma_desc *vd = vchan_next_desc(&c->vc);
+
+		if (vd)
+			dma_cookie_complete(&vd->tx);
+	}
 	list_del_init(&c->node);
 	spin_unlock(&d->lock);
 

@@ -979,6 +985,7 @@ static int omap_dma_terminate_all(struct omap_chan *c)
 	 * c->desc is NULL and exit.)
 	 */
 	if (c->desc) {
+		dma_cookie_complete(&c->desc->vd.tx);
 		c->desc = NULL;
 		/* Avoid stopping the dma twice */
 		if (!c->paused)

-- 
2.0.1