by Bjorn Andersson

[permalink] [raw]

Subject: Re: [PATCH v10 03/11] remoteproc: Update Kconfig setup to 'depends on REMOTEPROC'

On Sat 08 Oct 05:52 PDT 2016, Peter Griffin wrote:

> Make REMOTEPROC core a selectable kconfig option, and update
> remoteproc client drivers to 'depends on' the core. This avoids
> some nasty Kconfig recursive dependency issues. Also when using
> menuconfig client drivers will be hidden until the core has been
> enabled.
>
> Documentation/kbuild/kconfig-language.txt:
>
> Note:
> select should be used with care. select will force
> a symbol to a value without visiting the dependencies.
> By abusing select you are able to select a symbol FOO even
> if FOO depends on BAR that is not set.
> In general use select only for non-visible symbols
> (no prompts anywhere) and for symbols with no dependencies.
> That will limit the usefulness but on the other hand avoid
> the illegal configurations all over.
>
> Signed-off-by: Peter Griffin <[email protected]>

Sorry, I missed this patch in the set - but spotted it in linux-next.

I still don't like the change, but remoteproc has dependencies so I
guess I have to pick it until we fix that.

It's however not okay to take this patch through the DMA tree, as it
effectively stops me from introducing any changes in the rproc tree.
Further more, it's not based on v4.9, so it currently introduces another
Kconfig dependency problem - that I can't fix in my tree without
conflicting with Vinod's.

So, Vinod, can you please drop this patch from your tree? I'll pick it
up for now.

Regards,
Bjorn

> ---
> drivers/remoteproc/Kconfig | 21 ++++++++++++---------
> 1 file changed, 12 insertions(+), 9 deletions(-)
>
> diff --git a/drivers/remoteproc/Kconfig b/drivers/remoteproc/Kconfig
> index a7bedc6..decdcbe 100644
> --- a/drivers/remoteproc/Kconfig
> +++ b/drivers/remoteproc/Kconfig
> @@ -1,20 +1,21 @@
> menu "Remoteproc drivers"
>
> -# REMOTEPROC gets selected by whoever wants it
> config REMOTEPROC
> - tristate
> + tristate "Support for Remote Processor subsystem"
> depends on HAS_DMA
> select CRC32
> select FW_LOADER
> select VIRTIO
> select VIRTUALIZATION
>
> +if REMOTEPROC
> +
> config OMAP_REMOTEPROC
> tristate "OMAP remoteproc support"
> depends on HAS_DMA
> depends on ARCH_OMAP4 || SOC_OMAP5
> depends on OMAP_IOMMU
> - select REMOTEPROC
> + depends on REMOTEPROC
> select MAILBOX
> select OMAP2PLUS_MBOX
> select RPMSG
> @@ -34,7 +35,7 @@ config OMAP_REMOTEPROC
> config STE_MODEM_RPROC
> tristate "STE-Modem remoteproc support"
> depends on HAS_DMA
> - select REMOTEPROC
> + depends on REMOTEPROC
> default n
> help
> Say y or m here to support STE-Modem shared memory driver.
> @@ -44,7 +45,7 @@ config STE_MODEM_RPROC
> config WKUP_M3_RPROC
> tristate "AMx3xx Wakeup M3 remoteproc support"
> depends on SOC_AM33XX || SOC_AM43XX
> - select REMOTEPROC
> + depends on REMOTEPROC
> help
> Say y here to support Wakeup M3 remote processor on TI AM33xx
> and AM43xx family of SoCs.
> @@ -57,8 +58,8 @@ config WKUP_M3_RPROC
> config DA8XX_REMOTEPROC
> tristate "DA8xx/OMAP-L13x remoteproc support"
> depends on ARCH_DAVINCI_DA8XX
> + depends on REMOTEPROC
> select CMA if MMU
> - select REMOTEPROC
> select RPMSG
> help
> Say y here to support DA8xx/OMAP-L13x remote processors via the
> @@ -84,9 +85,9 @@ config QCOM_Q6V5_PIL
> tristate "Qualcomm Hexagon V5 Peripherial Image Loader"
> depends on OF && ARCH_QCOM
> depends on QCOM_SMEM
> + depends on REMOTEPROC
> select MFD_SYSCON
> select QCOM_MDT_LOADER
> - select REMOTEPROC
> help
> Say y here to support the Qualcomm Peripherial Image Loader for the
> Hexagon V5 based remote processors.
> @@ -94,7 +95,7 @@ config QCOM_Q6V5_PIL
> config ST_REMOTEPROC
> tristate "ST remoteproc support"
> depends on ARCH_STI
> - select REMOTEPROC
> + depends on REMOTEPROC
> help
> Say y here to support ST's adjunct processors via the remote
> processor framework.
> @@ -102,6 +103,8 @@ config ST_REMOTEPROC
>
> config ST_SLIM_REMOTEPROC
> tristate
> - select REMOTEPROC
> + depends on REMOTEPROC
> +
> +endif # REMOTEPROC
>
> endmenu
> --
> 1.9.1
>

2016-11-14 05:03:14

by Vinod Koul

[permalink] [raw]

Subject: Re: [PATCH v10 03/11] remoteproc: Update Kconfig setup to 'depends on REMOTEPROC'

On Thu, Nov 03, 2016 at 02:24:39PM -0700, Bjorn Andersson wrote:
> On Sat 08 Oct 05:52 PDT 2016, Peter Griffin wrote:
>
> > Make REMOTEPROC core a selectable kconfig option, and update
> > remoteproc client drivers to 'depends on' the core. This avoids
> > some nasty Kconfig recursive dependency issues. Also when using
> > menuconfig client drivers will be hidden until the core has been
> > enabled.
> >
> > Documentation/kbuild/kconfig-language.txt:
> >
> > Note:
> > select should be used with care. select will force
> > a symbol to a value without visiting the dependencies.
> > By abusing select you are able to select a symbol FOO even
> > if FOO depends on BAR that is not set.
> > In general use select only for non-visible symbols
> > (no prompts anywhere) and for symbols with no dependencies.
> > That will limit the usefulness but on the other hand avoid
> > the illegal configurations all over.
> >
> > Signed-off-by: Peter Griffin <[email protected]>
>
> Sorry, I missed this patch in the set - but spotted it in linux-next.
>
> I still don't like the change, but remoteproc has dependencies so I
> guess I have to pick it until we fix that.
>
> It's however not okay to take this patch through the DMA tree, as it
> effectively stops me from introducing any changes in the rproc tree.
> Further more, it's not based on v4.9, so it currently introduces another
> Kconfig dependency problem - that I can't fix in my tree without
> conflicting with Vinod's.
>
>
> So, Vinod, can you please drop this patch from your tree? I'll pick it
> up for now.

Sorry for the delay, b/w KS/LPC and travel, was slow on email.

I have dropped this one now..

--
~Vinod