From: Kory Maincent <[email protected]>
Only the local interruption was configured, remote interrupt was left
behind. This patch fix it by setting stop and abort remote interrupts when
the DW_EDMA_CHIP_LOCAL flag is not set.
Signed-off-by: Kory Maincent <[email protected]>
Fixes: e74c39573d35 ("dmaengine: dw-edma: Add support for native HDMA")
---
This patch is fixing a commit which is only in dmaengine tree and not
merged mainline.
---
drivers/dma/dw-edma/dw-hdma-v0-core.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/dma/dw-edma/dw-hdma-v0-core.c b/drivers/dma/dw-edma/dw-hdma-v0-core.c
index da8663f45fdb..7bd1a0f742be 100644
--- a/drivers/dma/dw-edma/dw-hdma-v0-core.c
+++ b/drivers/dma/dw-edma/dw-hdma-v0-core.c
@@ -232,6 +232,8 @@ static void dw_hdma_v0_core_start(struct dw_edma_chunk *chunk, bool first)
tmp = GET_CH_32(dw, chan->dir, chan->id, int_setup) |
HDMA_V0_STOP_INT_MASK | HDMA_V0_ABORT_INT_MASK |
HDMA_V0_LOCAL_STOP_INT_EN | HDMA_V0_LOCAL_ABORT_INT_EN;
+ if (!(dw->chip->flags & DW_EDMA_CHIP_LOCAL))
+ tmp |= HDMA_V0_REMOTE_STOP_INT_EN | HDMA_V0_REMOTE_ABORT_INT_EN;
SET_CH_32(dw, chan->dir, chan->id, int_setup, tmp);
/* Channel control */
SET_CH_32(dw, chan->dir, chan->id, control1, HDMA_V0_LINKLIST_EN);
--
2.25.1
On Fri, Jun 09, 2023 at 10:16:48AM +0200, K?ry Maincent wrote:
> From: Kory Maincent <[email protected]>
>
> Only the local interruption was configured, remote interrupt was left
> behind. This patch fix it by setting stop and abort remote interrupts when
> the DW_EDMA_CHIP_LOCAL flag is not set.
>
> Signed-off-by: Kory Maincent <[email protected]>
> Fixes: e74c39573d35 ("dmaengine: dw-edma: Add support for native HDMA")
> ---
>
> This patch is fixing a commit which is only in dmaengine tree and not
> merged mainline.
> ---
> drivers/dma/dw-edma/dw-hdma-v0-core.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/drivers/dma/dw-edma/dw-hdma-v0-core.c b/drivers/dma/dw-edma/dw-hdma-v0-core.c
> index da8663f45fdb..7bd1a0f742be 100644
> --- a/drivers/dma/dw-edma/dw-hdma-v0-core.c
> +++ b/drivers/dma/dw-edma/dw-hdma-v0-core.c
> @@ -232,6 +232,8 @@ static void dw_hdma_v0_core_start(struct dw_edma_chunk *chunk, bool first)
> tmp = GET_CH_32(dw, chan->dir, chan->id, int_setup) |
> HDMA_V0_STOP_INT_MASK | HDMA_V0_ABORT_INT_MASK |
> HDMA_V0_LOCAL_STOP_INT_EN | HDMA_V0_LOCAL_ABORT_INT_EN;
> + if (!(dw->chip->flags & DW_EDMA_CHIP_LOCAL))
> + tmp |= HDMA_V0_REMOTE_STOP_INT_EN | HDMA_V0_REMOTE_ABORT_INT_EN;
Seems reasonable especially seeing there is a code with a similar
semantic in the dw_hdma_v0_core_write_chunk() method.
Reviewed-by: Serge Semin <[email protected]>
Just curious whether we really need to have the local IRQs left
enabled for the remote device setup... The only case I have in mind is
that it would be useful to signal a remote end-point host of such
event in some application-specific environment. It sounds exotic but
still possible.
-Serge(y)
> SET_CH_32(dw, chan->dir, chan->id, int_setup, tmp);
> /* Channel control */
> SET_CH_32(dw, chan->dir, chan->id, control1, HDMA_V0_LINKLIST_EN);
> --
> 2.25.1
>
On Mon, 19 Jun 2023 00:48:00 +0300
Serge Semin <[email protected]> wrote:
> Seems reasonable especially seeing there is a code with a similar
> semantic in the dw_hdma_v0_core_write_chunk() method.
>
> Reviewed-by: Serge Semin <[email protected]>
>
> Just curious whether we really need to have the local IRQs left
> enabled for the remote device setup... The only case I have in mind is
> that it would be useful to signal a remote end-point host of such
> event in some application-specific environment. It sounds exotic but
> still possible.
Thanks for your review.
Yes, we do need to let local IRQs enabled. I have tested to remove them and it
prevent the remote setup to function correctly on my board. Maybe it needs to be
set to know internally when the transfer is done, but it seems weird. I haven't
a full explanation for now.
On Mon, Jun 19, 2023 at 08:16:47PM +0200, K?ry Maincent wrote:
> On Mon, 19 Jun 2023 00:48:00 +0300
> Serge Semin <[email protected]> wrote:
>
> > Seems reasonable especially seeing there is a code with a similar
> > semantic in the dw_hdma_v0_core_write_chunk() method.
> >
> > Reviewed-by: Serge Semin <[email protected]>
> >
> > Just curious whether we really need to have the local IRQs left
> > enabled for the remote device setup... The only case I have in mind is
> > that it would be useful to signal a remote end-point host of such
> > event in some application-specific environment. It sounds exotic but
> > still possible.
>
> Thanks for your review.
> Yes, we do need to let local IRQs enabled. I have tested to remove them and it
> prevent the remote setup to function correctly on my board. Maybe it needs to be
> set to know internally when the transfer is done, but it seems weird. I haven't
> a full explanation for now.
Ok. Thanks for checking it out.
-Serge(y)