diff mbox

[v3,4/6] dma: rcar-dma: use result of updated get_residue in tx_status

Message ID 1443559488-2416-5-git-send-email-hamzahfrq.sub@gmail.com
State Under Review
Delegated to: Geert Uytterhoeven
Headers show

Commit Message

hamzahfrq.sub@gmail.com Sept. 29, 2015, 8:44 p.m. UTC
From: Muhammad Hamza Farooq <mfarooq@visteon.com>

Signed-off-by: Muhammad Hamza Farooq <mfarooq@visteon.com>
---
 drivers/dma/sh/rcar-dmac.c | 4 ++++
 1 file changed, 4 insertions(+)

Comments

Vinod Koul Oct. 12, 2015, 3:12 p.m. UTC | #1
On Tue, Sep 29, 2015 at 10:44:46PM +0200, hamzahfrq.sub@gmail.com wrote:
> From: Muhammad Hamza Farooq <mfarooq@visteon.com>
> 
> Signed-off-by: Muhammad Hamza Farooq <mfarooq@visteon.com>
> ---
>  drivers/dma/sh/rcar-dmac.c | 4 ++++
>  1 file changed, 4 insertions(+)
> 
> diff --git a/drivers/dma/sh/rcar-dmac.c b/drivers/dma/sh/rcar-dmac.c
> index 6c806d0..0b5a367 100644
> --- a/drivers/dma/sh/rcar-dmac.c
> +++ b/drivers/dma/sh/rcar-dmac.c
> @@ -1258,6 +1258,10 @@ static enum dma_status rcar_dmac_tx_status(struct dma_chan *chan,
>  	residue = rcar_dmac_chan_get_residue(rchan, cookie);
>  	spin_unlock_irqrestore(&rchan->lock, flags);
>  
> +	/* if there's no residue, the cookie is complete */
> +	if (!residue)
> +		return DMA_COMPLETE;
> +

why would we get this, you should not be here if the status is complete and
exited before you checked the status
Laurent Pinchart Oct. 15, 2015, 4:49 p.m. UTC | #2
Hi Vinod,

On Monday 12 October 2015 20:42:32 Vinod Koul wrote:
> On Tue, Sep 29, 2015 at 10:44:46PM +0200, hamzahfrq.sub@gmail.com wrote:
> > From: Muhammad Hamza Farooq <mfarooq@visteon.com>
> > 
> > Signed-off-by: Muhammad Hamza Farooq <mfarooq@visteon.com>
> > ---
> > 
> >  drivers/dma/sh/rcar-dmac.c | 4 ++++
> >  1 file changed, 4 insertions(+)
> > 
> > diff --git a/drivers/dma/sh/rcar-dmac.c b/drivers/dma/sh/rcar-dmac.c
> > index 6c806d0..0b5a367 100644
> > --- a/drivers/dma/sh/rcar-dmac.c
> > +++ b/drivers/dma/sh/rcar-dmac.c
> > @@ -1258,6 +1258,10 @@ static enum dma_status rcar_dmac_tx_status(struct
> > dma_chan *chan,> 
> >  	residue = rcar_dmac_chan_get_residue(rchan, cookie);
> >  	spin_unlock_irqrestore(&rchan->lock, flags);
> > 
> > +	/* if there's no residue, the cookie is complete */
> > +	if (!residue)
> > +		return DMA_COMPLETE;
> > +
> 
> why would we get this, you should not be here if the status is complete and
> exited before you checked the status

The status check is a software check, while the residue check reads hardware 
registers. The hardware might have complete the transfer but the interrupt 
handler might not have had a chance to run yet.

Is it valid from a DMA engine API point of view to return a DMA_IN_PROGRESS 
status with a 0 residue ?
diff mbox

Patch

diff --git a/drivers/dma/sh/rcar-dmac.c b/drivers/dma/sh/rcar-dmac.c
index 6c806d0..0b5a367 100644
--- a/drivers/dma/sh/rcar-dmac.c
+++ b/drivers/dma/sh/rcar-dmac.c
@@ -1258,6 +1258,10 @@  static enum dma_status rcar_dmac_tx_status(struct dma_chan *chan,
 	residue = rcar_dmac_chan_get_residue(rchan, cookie);
 	spin_unlock_irqrestore(&rchan->lock, flags);
 
+	/* if there's no residue, the cookie is complete */
+	if (!residue)
+		return DMA_COMPLETE;
+
 	dma_set_residue(txstate, residue);
 
 	return status;