2021-05-09 13:59:28

by Vinod Koul

[permalink] [raw]
Subject: Re: [PATCH v2 14/17] dma: qcom: bam_dma: Create a new header file for BAM DMA driver

On 06-05-21, 03:07, Bhupesh Sharma wrote:
> Create a new header file for BAM DMA driver to make sure
> that it can be included in the follow-up patch to defer probing
> drivers which require BAM DMA driver to be first probed successfully.
>
> Cc: Thara Gopinath <[email protected]>
> Cc: Bjorn Andersson <[email protected]>
> Cc: Rob Herring <[email protected]>
> Cc: Andy Gross <[email protected]>
> Cc: Herbert Xu <[email protected]>
> Cc: David S. Miller <[email protected]>
> Cc: Stephen Boyd <[email protected]>
> Cc: Michael Turquette <[email protected]>
> Cc: Vinod Koul <[email protected]>
> Cc: [email protected]
> Cc: [email protected]
> Cc: [email protected]
> Cc: [email protected]
> Cc: [email protected]
> Cc: [email protected]
> Signed-off-by: Bhupesh Sharma <[email protected]>
> ---
> drivers/dma/qcom/bam_dma.c | 283 +-----------------------------------
> include/soc/qcom/bam_dma.h | 290 +++++++++++++++++++++++++++++++++++++

1. Please use -M with move patches...

2. susbsytem is dmaengine

3. Why move..? These things are internal to the driver and I dont think
it is wise for clients to use everything here... If the client needs to
know defer probe, it should request a channel and check the status
returned for EPROBE_DEFER

Thanks

--
~Vinod


2021-05-09 19:21:27

by Bhupesh Sharma

[permalink] [raw]
Subject: Re: [PATCH v2 14/17] dma: qcom: bam_dma: Create a new header file for BAM DMA driver

Hi Vinod,

Thanks for the review.

On Sun, 9 May 2021 at 19:28, Vinod Koul <[email protected]> wrote:
>
> On 06-05-21, 03:07, Bhupesh Sharma wrote:
> > Create a new header file for BAM DMA driver to make sure
> > that it can be included in the follow-up patch to defer probing
> > drivers which require BAM DMA driver to be first probed successfully.
> >
> > Cc: Thara Gopinath <[email protected]>
> > Cc: Bjorn Andersson <[email protected]>
> > Cc: Rob Herring <[email protected]>
> > Cc: Andy Gross <[email protected]>
> > Cc: Herbert Xu <[email protected]>
> > Cc: David S. Miller <[email protected]>
> > Cc: Stephen Boyd <[email protected]>
> > Cc: Michael Turquette <[email protected]>
> > Cc: Vinod Koul <[email protected]>
> > Cc: [email protected]
> > Cc: [email protected]
> > Cc: [email protected]
> > Cc: [email protected]
> > Cc: [email protected]
> > Cc: [email protected]
> > Signed-off-by: Bhupesh Sharma <[email protected]>
> > ---
> > drivers/dma/qcom/bam_dma.c | 283 +-----------------------------------
> > include/soc/qcom/bam_dma.h | 290 +++++++++++++++++++++++++++++++++++++
>
> 1. Please use -M with move patches...

Oops, will do.

> 2. susbsytem is dmaengine
>
> 3. Why move..? These things are internal to the driver and I dont think
> it is wise for clients to use everything here... If the client needs to
> know defer probe, it should request a channel and check the status
> returned for EPROBE_DEFER

Yes, the main intent is to defer the probe of the calling client driver in
case the BAM DMA is not probed() yet.

Sure, I will make the suggested change in v3,

Regards,
Bhupesh