[PATCH] misc: Kconfig: add a new dependency for MARVELL_CN10K_DPI

[PATCH] misc: Kconfig: add a new dependency for MARVELL_CN10K_DPI

[Vamsi Attunuru]

DPI hardware is an on-chip PCIe device on Marvell's arm64 SoC
platforms. As Arnd suggested, CN10K belongs to ARCH_THUNDER
lineage.

Patch makes mrvl_cn10k_dpi driver dependent on CONFIG_ARCH_THUNDER.

Signed-off-by: Vamsi Attunuru <vattunuru@marvell.com>
---
 drivers/misc/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/misc/Kconfig b/drivers/misc/Kconfig
index 64fcca9e44d7..f3bb75384627 100644
--- a/drivers/misc/Kconfig
+++ b/drivers/misc/Kconfig
@@ -588,6 +588,7 @@ config NSM
 config MARVELL_CN10K_DPI
 	tristate "Octeon CN10K DPI driver"
 	depends on PCI
+	depends on ARCH_THUNDER || COMPILE_TEST
 	help
 	  Enables Octeon CN10K DMA packet interface (DPI) driver which
 	  intializes DPI hardware's physical function (PF) device's
-- 
2.25.1

Re: [PATCH] misc: Kconfig: add a new dependency for MARVELL_CN10K_DPI

[Nathan Chancellor]

Hi Vamsi,

On Thu, Jul 11, 2024 at 05:01:15AM -0700, Vamsi Attunuru wrote:
> DPI hardware is an on-chip PCIe device on Marvell's arm64 SoC
> platforms. As Arnd suggested, CN10K belongs to ARCH_THUNDER
> lineage.
> 
> Patch makes mrvl_cn10k_dpi driver dependent on CONFIG_ARCH_THUNDER.
> 
> Signed-off-by: Vamsi Attunuru <vattunuru@marvell.com>
> ---
>  drivers/misc/Kconfig | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/drivers/misc/Kconfig b/drivers/misc/Kconfig
> index 64fcca9e44d7..f3bb75384627 100644
> --- a/drivers/misc/Kconfig
> +++ b/drivers/misc/Kconfig
> @@ -588,6 +588,7 @@ config NSM
>  config MARVELL_CN10K_DPI
>  	tristate "Octeon CN10K DPI driver"
>  	depends on PCI
> +	depends on ARCH_THUNDER || COMPILE_TEST
>  	help
>  	  Enables Octeon CN10K DMA packet interface (DPI) driver which
>  	  intializes DPI hardware's physical function (PF) device's
> -- 
> 2.25.1
> 

After this change, ARCH=arm allmodconfig fails with:

  drivers/misc/mrvl_cn10k_dpi.c: In function 'dpi_reg_write':
  drivers/misc/mrvl_cn10k_dpi.c:190:9: error: implicit declaration of function 'writeq'; did you mean 'writeb'? [-Wimplicit-function-declaration]
    190 |         writeq(val, dpi->reg_base + offset);
        |         ^~~~~~
        |         writeb
  drivers/misc/mrvl_cn10k_dpi.c: In function 'dpi_reg_read':
  drivers/misc/mrvl_cn10k_dpi.c:195:16: error: implicit declaration of function 'readq'; did you mean 'readb'? [-Wimplicit-function-declaration]
    195 |         return readq(dpi->reg_base + offset);
        |                ^~~~~
        |                readb

Including one of the io-64-nonatomic headers would resolve this but I am
not sure which one would be appropriate (or perhaps the dependency
should be tightened to requiring 64BIT, as some other drivers have
done).

Cheers,
Nathan

Re: [PATCH] misc: Kconfig: add a new dependency for MARVELL_CN10K_DPI

[Arnd Bergmann]

On Tue, Jul 16, 2024, at 15:26, Nathan Chancellor wrote:
> On Thu, Jul 11, 2024 at 05:01:15AM -0700, Vamsi Attunuru wrote:
>> 
>
> After this change, ARCH=arm allmodconfig fails with:
>
>   drivers/misc/mrvl_cn10k_dpi.c: In function 'dpi_reg_write':
>   drivers/misc/mrvl_cn10k_dpi.c:190:9: error: implicit declaration of 
> function 'writeq'; did you mean 'writeb'? 
> [-Wimplicit-function-declaration]
>     190 |         writeq(val, dpi->reg_base + offset);
>         |         ^~~~~~
>         |         writeb
>   drivers/misc/mrvl_cn10k_dpi.c: In function 'dpi_reg_read':
>   drivers/misc/mrvl_cn10k_dpi.c:195:16: error: implicit declaration of 
> function 'readq'; did you mean 'readb'? 
> [-Wimplicit-function-declaration]
>     195 |         return readq(dpi->reg_base + offset);
>         |                ^~~~~
>         |                readb
>
> Including one of the io-64-nonatomic headers would resolve this but I am
> not sure which one would be appropriate (or perhaps the dependency
> should be tightened to requiring 64BIT, as some other drivers have
> done).

Right, a dependency on 64BIT makes sense here. The alternative is
to include linux/io-64-nonatomic-hi-lo.h or linux/io-64-nonatomic-lo-hi.h
in order to have a replacement readq/writeq implementation that
works on 32-bit architectures. However, doing this requires
understanding whether what the side-effects of accessing the
64-bit registers are and whether they require writing the upper
or lower half of the register last.

     Arnd

RE: [EXTERNAL] Re: [PATCH] misc: Kconfig: add a new dependency for MARVELL_CN10K_DPI

[Vamsi Krishna Attunuru]

>-----Original Message-----
>From: Arnd Bergmann <arnd@arndb.de>
>Sent: Tuesday, July 16, 2024 7:21 PM
>To: Nathan Chancellor <nathan@kernel.org>; Vamsi Krishna Attunuru
><vattunuru@marvell.com>
>Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>; linux-
>kernel@vger.kernel.org
>Subject: [EXTERNAL] Re: [PATCH] misc: Kconfig: add a new dependency for
>MARVELL_CN10K_DPI
>
>On Tue, Jul 16, 2024, at 15: 26, Nathan Chancellor wrote: > On Thu, Jul 11, 2024
>at 05: 01: 15AM -0700, Vamsi Attunuru wrote: >> > > After this change,
>ARCH=arm allmodconfig fails with: > > drivers/misc/mrvl_cn10k_dpi. c: In
>
>On Tue, Jul 16, 2024, at 15:26, Nathan Chancellor wrote:
>> On Thu, Jul 11, 2024 at 05:01:15AM -0700, Vamsi Attunuru wrote:
>>>
>>
>> After this change, ARCH=arm allmodconfig fails with:
>>
>>   drivers/misc/mrvl_cn10k_dpi.c: In function 'dpi_reg_write':
>>   drivers/misc/mrvl_cn10k_dpi.c:190:9: error: implicit declaration of
>> function 'writeq'; did you mean 'writeb'?
>> [-Wimplicit-function-declaration]
>>     190 |         writeq(val, dpi->reg_base + offset);
>>         |         ^~~~~~
>>         |         writeb
>>   drivers/misc/mrvl_cn10k_dpi.c: In function 'dpi_reg_read':
>>   drivers/misc/mrvl_cn10k_dpi.c:195:16: error: implicit declaration of
>> function 'readq'; did you mean 'readb'?
>> [-Wimplicit-function-declaration]
>>     195 |         return readq(dpi->reg_base + offset);
>>         |                ^~~~~
>>         |                readb
>>
>> Including one of the io-64-nonatomic headers would resolve this but I
>> am not sure which one would be appropriate (or perhaps the dependency
>> should be tightened to requiring 64BIT, as some other drivers have
>> done).
>
>Right, a dependency on 64BIT makes sense here. The alternative is to include
>linux/io-64-nonatomic-hi-lo.h or linux/io-64-nonatomic-lo-hi.h in order to
>have a replacement readq/writeq implementation that works on 32-bit
>architectures. However, doing this requires understanding whether what the
>side-effects of accessing the 64-bit registers are and whether they require
>writing the upper or lower half of the register last.
>

Yes Arnd, I am checking the functionality using lo-hi calls that you suggested.
If it has any implications, I will fix it with the 64BIT dependency.

Thanks, Nathan, for reaching out. Could you please advise on the branch where
I should implement the fix.



>     Arnd

Re: [EXTERNAL] Re: [PATCH] misc: Kconfig: add a new dependency for MARVELL_CN10K_DPI

[Arnd Bergmann]

On Tue, Jul 16, 2024, at 17:33, Vamsi Krishna Attunuru wrote:
>>> Including one of the io-64-nonatomic headers would resolve this but I
>>> am not sure which one would be appropriate (or perhaps the dependency
>>> should be tightened to requiring 64BIT, as some other drivers have
>>> done).
>>
>>Right, a dependency on 64BIT makes sense here. The alternative is to include
>>linux/io-64-nonatomic-hi-lo.h or linux/io-64-nonatomic-lo-hi.h in order to
>>have a replacement readq/writeq implementation that works on 32-bit
>>architectures. However, doing this requires understanding whether what the
>>side-effects of accessing the 64-bit registers are and whether they require
>>writing the upper or lower half of the register last.
>>
>
> Yes Arnd, I am checking the functionality using lo-hi calls that you suggested.
> If it has any implications, I will fix it with the 64BIT dependency.
>

If the datasheet does not give you any insight about how it works,
you can try it out on a 64-bit kernel by changing all the
readq/writeq into lo_hi_readq()/lo_hi_writeq() or
hi_lo_readq()/hi_lo_writeq() after including the corresponding header.

Usually only one of the two works, or possibly neither if the registers
require 64-bit access. If you figure out which one works, you
can change the calls back to readq/writeq and leave the correct
header included that will provide fallback implementations on 32-bit
architectures but use the native 64-bit access on 64-bit architectures.

In practice this is not really important because nobody will ever
need to use this driver on a 32-bit kernel, given that this SoC
only supports 32-bit usermode but not kernel.

       Arnd

Re: [EXTERNAL] Re: [PATCH] misc: Kconfig: add a new dependency for MARVELL_CN10K_DPI

[Nathan Chancellor]

On Tue, Jul 16, 2024 at 03:33:49PM +0000, Vamsi Krishna Attunuru wrote:
> Thanks, Nathan, for reaching out. Could you please advise on the branch where
> I should implement the fix.

I assume char-misc-next, as that is where the broken changes are.

https://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc.git/log/?h=char-misc-next

Cheers,
Nathan