Re: [dmaengine] Cyclic DMA regression.
From: javier Martin
Date: Mon Apr 09 2012 - 03:09:43 EST
On 5 April 2012 08:08, Vinod Koul <vinod.koul@xxxxxxxxxxxxxxx> wrote:
> this is not right fix. The problem is that we shouldn't mark the cyclic
> descriptor as complete. So for all drivers using cyclic API they should
> do something like this. Can you test if this fixes your issue.
> -----------------
> diff --git a/drivers/dma/imx-dma.c b/drivers/dma/imx-dma.c
> index a45b5d2..0c7362a 100644
> --- a/drivers/dma/imx-dma.c
> +++ b/drivers/dma/imx-dma.c
> @@ -571,11 +571,14 @@ static void imxdma_tasklet(unsigned long data)
> if (desc->desc.callback)
> desc->desc.callback(desc->desc.callback_param);
>
> - dma_cookie_complete(&desc->desc);
> -
> - /* If we are dealing with a cyclic descriptor keep it on ld_active */
> + /* If we are dealing with a cyclic descriptor keep it on ld_active
> + * and don't mark the descriptor as complete.
> + * Only in non-cyclic cases it would be marked as complete
> + */
> if (imxdma_chan_is_doing_cyclic(imxdmac))
> goto out;
> + else
> + dma_cookie_complete(&desc->desc);
>
> /* Free 2D slot if it was an interleaved transfer */
> if (imxdmac->enabled_2d) {
>
>
> --
> ~Vinod
>
Hi Vinod,
thank you for your patch.
It works fine for me:
Tested-by: Javier Martin <javier.martin@xxxxxxxxxxxxxxxxx>
Could you apply it in your tree as a fix for 3.4?
Regards.
--
Javier Martin
Vista Silicon S.L.
CDTUC - FASE C - Oficina S-345
Avda de los Castros s/n
39005- Santander. Cantabria. Spain
+34 942 25 32 60
www.vista-silicon.com
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/