Setup DMA Mask
---
I realised while (re)creating the test branches (sent in reply to the cover
letter), I missed to squash this hunk in
"staging: vc04_services: bcm2835-isp: Add a more complex ISP processing component"
before sending out the series.

Sending it as a fixup! for now. 
---
 .../staging/vc04_services/bcm2835-isp/bcm2835-v4l2-isp.c    | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/staging/vc04_services/bcm2835-isp/bcm2835-v4l2-isp.c b/drivers/staging/vc04_services/bcm2835-isp/bcm2835-v4l2-isp.c
index 316d35d5f19d..4a23f9feb6e4 100644
--- a/drivers/staging/vc04_services/bcm2835-isp/bcm2835-v4l2-isp.c
+++ b/drivers/staging/vc04_services/bcm2835-isp/bcm2835-v4l2-isp.c
@@ -1774,6 +1774,12 @@ static int bcm2835_isp_probe(struct vchiq_device *device)
 	unsigned int i;
 	int ret;
 
+	ret = dma_set_mask_and_coherent(&device->dev, DMA_BIT_MASK(32));
+	if (ret) {
+		dev_err(&device->dev, "dma_set_mask_and_coherent failed: %d\n", ret);
+		return ret;
+	}
+
 	bcm2835_isp_instances = devm_kzalloc(&device->dev,
 					     sizeof(bcm2835_isp_instances) *
 						      BCM2835_ISP_NUM_INSTANCES,
-- 
2.41.0

On Mon, Nov 13, 2023 at 09:07:55AM -0500, Umang Jain wrote:
> Setup DMA Mask
> ---
> I realised while (re)creating the test branches (sent in reply to the cover
> letter), I missed to squash this hunk in
> "staging: vc04_services: bcm2835-isp: Add a more complex ISP processing component"
> before sending out the series.
> 
> Sending it as a fixup! for now. 

What does that mean?  What can I do with this?

confused,

greg k-h

On Thu, Nov 23, 2023 at 12:56:07PM +0000, Greg KH wrote:
> On Mon, Nov 13, 2023 at 09:07:55AM -0500, Umang Jain wrote:
> > Setup DMA Mask
> > ---
> > I realised while (re)creating the test branches (sent in reply to the cover
> > letter), I missed to squash this hunk in
> > "staging: vc04_services: bcm2835-isp: Add a more complex ISP processing component"
> > before sending out the series.
> > 
> > Sending it as a fixup! for now. 
> 
> What does that mean?  What can I do with this?

Umang, we don't do that for upstream kernel development. If a patch
needs a new version, you should send a new version. In some subsystems
maintainers will prefer if you only resubmit the offending patch when
the series is large (as a "vN.1" instead of a vN+1), while in other
cases resending the whole series is preferred.

Greg, I don't think Umang wanted you to do anything with this patch :-)
This was meant to tell reviewer that the issue was known already.

-- 
Regards,

Laurent Pinchart

Hi Umang,

Am 13.11.23 um 15:07 schrieb Umang Jain:
> Setup DMA Mask
> ---
> I realised while (re)creating the test branches (sent in reply to the cover
> letter), I missed to squash this hunk in
> "staging: vc04_services: bcm2835-isp: Add a more complex ISP processing component"
> before sending out the series.
>
> Sending it as a fixup! for now.
sorry, but i lost track of all your vchiq patch series. Usually a
commiter shouldn't send multiple series to a list at the same time. This
is very hard to review / test.

An expected review time for a patch series should be 2 weeks, especially
such big series like bcm2835-isp.

For this specific case, it would be better to reply to your own patch
and mention this mistake and fix. It's clear that we need a V3 of the
bcm2835-isp, so in this series the fix should be finally included.

I didn't had the time for the big series, but sending more patches
doesn't make me faster ...

Regards
Stefan
> ---
>   .../staging/vc04_services/bcm2835-isp/bcm2835-v4l2-isp.c    | 6 ++++++
>   1 file changed, 6 insertions(+)
>
> diff --git a/drivers/staging/vc04_services/bcm2835-isp/bcm2835-v4l2-isp.c b/drivers/staging/vc04_services/bcm2835-isp/bcm2835-v4l2-isp.c
> index 316d35d5f19d..4a23f9feb6e4 100644
> --- a/drivers/staging/vc04_services/bcm2835-isp/bcm2835-v4l2-isp.c
> +++ b/drivers/staging/vc04_services/bcm2835-isp/bcm2835-v4l2-isp.c
> @@ -1774,6 +1774,12 @@ static int bcm2835_isp_probe(struct vchiq_device *device)
>   	unsigned int i;
>   	int ret;
>
> +	ret = dma_set_mask_and_coherent(&device->dev, DMA_BIT_MASK(32));
> +	if (ret) {
> +		dev_err(&device->dev, "dma_set_mask_and_coherent failed: %d\n", ret);
> +		return ret;
> +	}
> +
>   	bcm2835_isp_instances = devm_kzalloc(&device->dev,
>   					     sizeof(bcm2835_isp_instances) *
>   						      BCM2835_ISP_NUM_INSTANCES,