diff mbox

ARM: imx: temporarily remove CONFIG_SOC_FSL from LS1021A

Message ID 1727581.y25NQPsRIm@wuerfel
State New
Headers show

Commit Message

Arnd Bergmann Nov. 11, 2014, 4:03 p.m. UTC 
The newly introduced LS1021A SoC selects CONFIG_SOC_FSL, which
is originally symbol used for the PowerPC based platforms
and guards lots of code that does not build on ARM.

This breaks allmodconfig, so let's remove it for now, until
either all those drivers are fixed or they use a dependency
on IMX instead.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>

---
On Friday 31 October 2014 17:01:12 Jingchang Lu wrote:
> From: Jingchang Lu <b35083@freescale.com>
> 
> The LS1021A SoC is a dual-core Cortex-A7 based processor,
> this adds the initial support for it.
> 
> Signed-off-by: Jingchang Lu <b35083@freescale.com>

>  
> +config FSL_SOC
> +       bool
> +
> +config SOC_LS1021A
> +       bool "Freescale LS1021A support"
> +       select ARM_GIC
> +       select HAVE_ARM_ARCH_TIMER
> +       select FSL_SOC
>


This broke allmodconfig builds in linux-next, please apply this
partial revert for the moment.

	Arnd

---

Comments

Shawn Guo Nov. 12, 2014, 10:42 a.m. UTC | #1 
On Tue, Nov 11, 2014 at 05:03:25PM +0100, Arnd Bergmann wrote:
> The newly introduced LS1021A SoC selects CONFIG_SOC_FSL, which
> is originally symbol used for the PowerPC based platforms
> and guards lots of code that does not build on ARM.
> 
> This breaks allmodconfig, so let's remove it for now, until
> either all those drivers are fixed or they use a dependency
> on IMX instead.
> 
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> 
> ---
> On Friday 31 October 2014 17:01:12 Jingchang Lu wrote:
> > From: Jingchang Lu <b35083@freescale.com>
> > 
> > The LS1021A SoC is a dual-core Cortex-A7 based processor,
> > this adds the initial support for it.
> > 
> > Signed-off-by: Jingchang Lu <b35083@freescale.com>
> 
> >  
> > +config FSL_SOC
> > +       bool
> > +
> > +config SOC_LS1021A
> > +       bool "Freescale LS1021A support"
> > +       select ARM_GIC
> > +       select HAVE_ARM_ARCH_TIMER
> > +       select FSL_SOC
> >
> 
> 
> This broke allmodconfig builds in linux-next, please apply this
> partial revert for the moment.

I noticed the issue from 0-DAY testing and was trying to drop the patch.
Jingchang said someone is trying to fixing the issues from driver code.
But it seems to take some time.  So I agree with this partial reverting.

Patch applied.  Thanks Arnd.

Shawn

> diff --git a/arch/arm/mach-imx/Kconfig b/arch/arm/mach-imx/Kconfig
> index ef549ef1cd6d..e8627e04e1e6 100644
> --- a/arch/arm/mach-imx/Kconfig
> +++ b/arch/arm/mach-imx/Kconfig
> @@ -658,14 +658,10 @@ choice
>  
>  endchoice
>  
> -config FSL_SOC
> -	bool
> -
>  config SOC_LS1021A
>  	bool "Freescale LS1021A support"
>  	select ARM_GIC
>  	select HAVE_ARM_ARCH_TIMER
> -	select FSL_SOC
>  	select PCI_DOMAINS if PCI
>  	select ZONE_DMA if ARM_LPAE
>  
>
Arnd Bergmann Nov. 12, 2014, 11:04 a.m. UTC | #2 
On Wednesday 12 November 2014 18:42:41 Shawn Guo wrote:
> 
> I noticed the issue from 0-DAY testing and was trying to drop the patch.
> Jingchang said someone is trying to fixing the issues from driver code.
> But it seems to take some time.  So I agree with this partial reverting.
> 
> Patch applied.  Thanks Arnd.

Thanks a lot!

Jingchang, can you say which of the drivers that are currently depending
on FSL_SOC are reused on LS1021A? I see these ones in the source tree:

drivers/ata/sata_fsl.c
drivers/crypto/caam/
drivers/crypto/talitos.c
drivers/dma/fsldma.c
drivers/memory/fsl_ifc.c
drivers/misc/carma/
drivers/mtd/nand/fsl_ifc_nand.c
drivers/net/ethernet/freescale/fsl_pq_mdio.c
drivers/net/ethernet/freescale/gianfar*
drivers/net/ethernet/freescale/xgmac_mdio.c
drivers/spi/spi-fsl-cpm.c
drivers/spi/spi-fsl-dspi.c
drivers/spi/spi-fsl-espi.c
drivers/usb/host/ehci-fsl.c
drivers/video/fbdev/fsl-diu-fb.c
drivers/virt/fsl_hypervisor.c
sound/soc/fsl/mpc*

I suspect that you only want a small subset of these, but I have no idea
which.

	Arnd
Jingchang Lu Nov. 14, 2014, 9:31 a.m. UTC | #3 
>-----Original Message-----
>From: Arnd Bergmann [mailto:arnd@arndb.de]
>Sent: Wednesday, November 12, 2014 7:05 PM
>To: linux-arm-kernel@lists.infradead.org
>Cc: Shawn Guo; mark.rutland@arm.com; devicetree@vger.kernel.org; Lu
>Jingchang-B35083
>Subject: Re: [PATCH] ARM: imx: temporarily remove CONFIG_SOC_FSL from
>LS1021A
>
>On Wednesday 12 November 2014 18:42:41 Shawn Guo wrote:
>>
>> I noticed the issue from 0-DAY testing and was trying to drop the patch.
>> Jingchang said someone is trying to fixing the issues from driver code.
>> But it seems to take some time.  So I agree with this partial reverting.
>>
>> Patch applied.  Thanks Arnd.
>
>Thanks a lot!
>
>Jingchang, can you say which of the drivers that are currently depending
>on FSL_SOC are reused on LS1021A? I see these ones in the source tree:
>
>drivers/ata/sata_fsl.c
>drivers/crypto/caam/
>drivers/crypto/talitos.c
>drivers/dma/fsldma.c
>drivers/memory/fsl_ifc.c
>drivers/misc/carma/
>drivers/mtd/nand/fsl_ifc_nand.c
>drivers/net/ethernet/freescale/fsl_pq_mdio.c
>drivers/net/ethernet/freescale/gianfar*
>drivers/net/ethernet/freescale/xgmac_mdio.c
>drivers/spi/spi-fsl-cpm.c
>drivers/spi/spi-fsl-dspi.c
>drivers/spi/spi-fsl-espi.c
>drivers/usb/host/ehci-fsl.c
>drivers/video/fbdev/fsl-diu-fb.c
>drivers/virt/fsl_hypervisor.c
>sound/soc/fsl/mpc*
>
>I suspect that you only want a small subset of these, but I have no idea
>which.
>
>	Arnd

Hi, Arnd and Shawn,

  Many thanks for your work on resolving the FSL_SOC breaking allmodconfig.

  Formerly many drivers on PPC depend on a FSL_SOC configuration, when they
are shared on our ARM platform, I add this configuration internally for
quick reference and also some internal developers' requirement, thus I
also add it into my upstream code.

 Yes, only a subset of these drivers are shared between our ARM and PPC
platforms. Since there will be allmodconfig breaking fault due to the
FSL_SOC configuration, we will try to fix this or select another dependency,
I will also inform our driver owners about this.
 
Below drivers are shared on LS1021A platform currently:

drivers/crypto/caam/
drivers/memory/fsl_ifc.c
drivers/mtd/nand/fsl_ifc_nand.c
drivers/net/ethernet/freescale/fsl_pq_mdio.c
drivers/net/ethernet/freescale/gianfar*
drivers/spi/spi-fsl-dspi.c


Thanks and Best Regards,
Jingchang
Arnd Bergmann Nov. 14, 2014, 10:56 a.m. UTC | #4 
On Friday 14 November 2014 09:31:42 Jingchang Lu wrote:
>  Yes, only a subset of these drivers are shared between our ARM and PPC
> platforms. Since there will be allmodconfig breaking fault due to the
> FSL_SOC configuration, we will try to fix this or select another dependency,
> I will also inform our driver owners about this.
>  
> Below drivers are shared on LS1021A platform currently:
> 
> drivers/crypto/caam/
> drivers/memory/fsl_ifc.c
> drivers/mtd/nand/fsl_ifc_nand.c
> drivers/net/ethernet/freescale/fsl_pq_mdio.c
> drivers/net/ethernet/freescale/gianfar*
> drivers/spi/spi-fsl-dspi.c

Ok, since this is only a small subset of the FSL_SOC drivers, I'd
suggest you make these depend on (FSL_SOC || ARCH_IMX) after
verifying that they build without warnings on ARM.

	Arnd
diff mbox

Patch

diff --git a/arch/arm/mach-imx/Kconfig b/arch/arm/mach-imx/Kconfig
index ef549ef1cd6d..e8627e04e1e6 100644
--- a/arch/arm/mach-imx/Kconfig
+++ b/arch/arm/mach-imx/Kconfig
@@ -658,14 +658,10 @@  choice
 
 endchoice
 
-config FSL_SOC
-	bool
-
 config SOC_LS1021A
 	bool "Freescale LS1021A support"
 	select ARM_GIC
 	select HAVE_ARM_ARCH_TIMER
-	select FSL_SOC
 	select PCI_DOMAINS if PCI
 	select ZONE_DMA if ARM_LPAE