[4/4] ARM: multi_v7_defconfig: Enable DW USB2 support and rockchip phy
diff mbox

Message ID 1441570566-14001-5-git-send-email-sjoerd.simons@collabora.co.uk
State New
Headers show

Commit Message

Sjoerd Simons Sept. 6, 2015, 8:16 p.m. UTC
The DesignWare HS OTG USB 2.0 is used in various SoC families among wich
the various rockchip SoCs, enable the driver and the PHY glue for
rockchip

Signed-off-by: Sjoerd Simons <sjoerd.simons@collabora.co.uk>

---

 arch/arm/configs/multi_v7_defconfig | 2 ++
 1 file changed, 2 insertions(+)

Comments

Krzysztof Kozlowski Sept. 7, 2015, 12:02 a.m. UTC | #1
On 07.09.2015 05:16, Sjoerd Simons wrote:
> The DesignWare HS OTG USB 2.0 is used in various SoC families among wich
> the various rockchip SoCs, enable the driver and the PHY glue for
> rockchip
> 
> Signed-off-by: Sjoerd Simons <sjoerd.simons@collabora.co.uk>
> 
> ---
> 
>  arch/arm/configs/multi_v7_defconfig | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/arch/arm/configs/multi_v7_defconfig b/arch/arm/configs/multi_v7_defconfig
> index fb673e2..49c733a 100644
> --- a/arch/arm/configs/multi_v7_defconfig
> +++ b/arch/arm/configs/multi_v7_defconfig
> @@ -514,6 +514,7 @@ CONFIG_USB_STORAGE=y
>  CONFIG_USB_DWC3=y
>  CONFIG_USB_CHIPIDEA=y
>  CONFIG_USB_CHIPIDEA_HOST=y
> +CONFIG_USB_DWC2=y

+Cc Marek and Kukjin,

This would conflict with Marek's:
[PATCH v2] ARM: multi_v7_defconfig: Enable DWC2 USB driver and USB
ethernet gadget
http://www.spinics.net/lists/linux-samsung-soc/msg46743.html

Probably your series and Marek's change should go together to the same
tree. How about rebasing your work on top of it?

Best regards,
Krzysztof

>  CONFIG_AB8500_USB=y
>  CONFIG_KEYSTONE_USB_PHY=y
>  CONFIG_OMAP_USB3=y
> @@ -658,6 +659,7 @@ CONFIG_PHY_HIX5HD2_SATA=y
>  CONFIG_PWM_STI=m
>  CONFIG_OMAP_USB2=y
>  CONFIG_TI_PIPE3=y
> +CONFIG_PHY_ROCKCHIP_USB=y
>  CONFIG_PHY_MIPHY28LP=y
>  CONFIG_PHY_MIPHY365X=y
>  CONFIG_PHY_RCAR_GEN2=m
>
Sjoerd Simons Sept. 7, 2015, 7:10 a.m. UTC | #2
Hey Krzysztof,

On Mon, 2015-09-07 at 09:02 +0900, Krzysztof Kozlowski wrote:
> On 07.09.2015 05:16, Sjoerd Simons wrote:
> > The DesignWare HS OTG USB 2.0 is used in various SoC families among
> > wich
> > the various rockchip SoCs, enable the driver and the PHY glue for
> > rockchip
> +Cc Marek and Kukjin,
> 
> This would conflict with Marek's:
> [PATCH v2] ARM: multi_v7_defconfig: Enable DWC2 USB driver and USB
> ethernet gadget
> http://www.spinics.net/lists/linux-samsung-soc/msg46743.html
> 
> Probably your series and Marek's change should go together to the
> same
> tree. How about rebasing your work on top of it?

Thanks for the pointer! I'm entirely happy to rebase on Mareks patch or
this one dropped for now until Mareks patch lands. 

These patches were based on current linux-next, so i guess Mareks pathc
didn't land just yet? Tbh, i don't really know what tries multi_v7
tends to go through (especially for non-platform specific changes)

> Best regards,
> Krzysztof
> 
> >  CONFIG_AB8500_USB=y
> >  CONFIG_KEYSTONE_USB_PHY=y
> >  CONFIG_OMAP_USB3=y
> > @@ -658,6 +659,7 @@ CONFIG_PHY_HIX5HD2_SATA=y
> >  CONFIG_PWM_STI=m
> >  CONFIG_OMAP_USB2=y
> >  CONFIG_TI_PIPE3=y
> > +CONFIG_PHY_ROCKCHIP_USB=y
> >  CONFIG_PHY_MIPHY28LP=y
> >  CONFIG_PHY_MIPHY365X=y
> >  CONFIG_PHY_RCAR_GEN2=m
> > 
>
Krzysztof Kozlowski Sept. 7, 2015, 7:24 a.m. UTC | #3
On 07.09.2015 16:10, Sjoerd Simons wrote:
> Hey Krzysztof,
> 
> On Mon, 2015-09-07 at 09:02 +0900, Krzysztof Kozlowski wrote:
>> On 07.09.2015 05:16, Sjoerd Simons wrote:
>>> The DesignWare HS OTG USB 2.0 is used in various SoC families among
>>> wich
>>> the various rockchip SoCs, enable the driver and the PHY glue for
>>> rockchip
>> +Cc Marek and Kukjin,
>>
>> This would conflict with Marek's:
>> [PATCH v2] ARM: multi_v7_defconfig: Enable DWC2 USB driver and USB
>> ethernet gadget
>> http://www.spinics.net/lists/linux-samsung-soc/msg46743.html
>>
>> Probably your series and Marek's change should go together to the
>> same
>> tree. How about rebasing your work on top of it?
> 
> Thanks for the pointer! I'm entirely happy to rebase on Mareks patch or
> this one dropped for now until Mareks patch lands. 
> 
> These patches were based on current linux-next, so i guess Mareks pathc
> didn't land just yet? Tbh, i don't really know what tries multi_v7
> tends to go through (especially for non-platform specific changes)

It was too late for Marek's patch to go for v4.3 merge window, so I
applied it but not pushed to my for-next branch. AFAIU, Stephen does not
want v4.4 material during merge window because of mangling with statistics.


If you want to rebase (or cherry-pick), you can find Marek's patch here:
branch: defconfig-for-next
repo: git://github.com/krzk/linux.git

However this *branch will be rebased on v4.3-rc1* (because this is
material for v4.4). After v4.3-rc1 I can prepare a tag for other
interested parties.


As for multi_v7_defconfig policy, I got impression from the
Arnd/Olof/Kevin comments on LKML, that platform specific changes should
go through platform tree. However this is multi-platform change so maybe
we can push both patches through arm-soc.


Arnd, Olof, Kevin (BTW, you missed to CC him),

How do you wish to handle this case (USB_DWC2 and two patches: Marek's
and Sjoerd's)?

Best regards,
Krzysztof
Heiko Stübner Sept. 7, 2015, 7:53 a.m. UTC | #4
Am Montag, 7. September 2015, 16:24:53 schrieb Krzysztof Kozlowski:
> On 07.09.2015 16:10, Sjoerd Simons wrote:
> > Hey Krzysztof,
> > 
> > On Mon, 2015-09-07 at 09:02 +0900, Krzysztof Kozlowski wrote:
> >> On 07.09.2015 05:16, Sjoerd Simons wrote:
> >>> The DesignWare HS OTG USB 2.0 is used in various SoC families among
> >>> wich
> >>> the various rockchip SoCs, enable the driver and the PHY glue for
> >>> rockchip
> >> 
> >> +Cc Marek and Kukjin,
> >> 
> >> This would conflict with Marek's:
> >> [PATCH v2] ARM: multi_v7_defconfig: Enable DWC2 USB driver and USB
> >> ethernet gadget
> >> http://www.spinics.net/lists/linux-samsung-soc/msg46743.html
> >> 
> >> Probably your series and Marek's change should go together to the
> >> same
> >> tree. How about rebasing your work on top of it?
> > 
> > Thanks for the pointer! I'm entirely happy to rebase on Mareks patch or
> > this one dropped for now until Mareks patch lands.
> > 
> > These patches were based on current linux-next, so i guess Mareks pathc
> > didn't land just yet? Tbh, i don't really know what tries multi_v7
> > tends to go through (especially for non-platform specific changes)
> 
> It was too late for Marek's patch to go for v4.3 merge window, so I
> applied it but not pushed to my for-next branch. AFAIU, Stephen does not
> want v4.4 material during merge window because of mangling with statistics.
> 
> 
> If you want to rebase (or cherry-pick), you can find Marek's patch here:
> branch: defconfig-for-next
> repo: git://github.com/krzk/linux.git
> 
> However this *branch will be rebased on v4.3-rc1* (because this is
> material for v4.4). After v4.3-rc1 I can prepare a tag for other
> interested parties.
> 
> 
> As for multi_v7_defconfig policy, I got impression from the
> Arnd/Olof/Kevin comments on LKML, that platform specific changes should
> go through platform tree. However this is multi-platform change so maybe
> we can push both patches through arm-soc.
> 
> 
> Arnd, Olof, Kevin (BTW, you missed to CC him),
> 
> How do you wish to handle this case (USB_DWC2 and two patches: Marek's
> and Sjoerd's)?

I guess the easiest option would be to just drop the dwc2 from here, only 
keeping the phy enablement and let everything (Marek's and Sjoerd's patch) 
come together in the armsoc defconfig branch? As the rest of both patches don't 
seem to touch common areas, they should be easily mergable from different trees 
as well.


Heiko
Krzysztof Kozlowski Sept. 7, 2015, 7:56 a.m. UTC | #5
On 07.09.2015 16:53, Heiko Stübner wrote:
> Am Montag, 7. September 2015, 16:24:53 schrieb Krzysztof Kozlowski:
>> On 07.09.2015 16:10, Sjoerd Simons wrote:
>>> Hey Krzysztof,
>>>
>>> On Mon, 2015-09-07 at 09:02 +0900, Krzysztof Kozlowski wrote:
>>>> On 07.09.2015 05:16, Sjoerd Simons wrote:
>>>>> The DesignWare HS OTG USB 2.0 is used in various SoC families among
>>>>> wich
>>>>> the various rockchip SoCs, enable the driver and the PHY glue for
>>>>> rockchip
>>>>
>>>> +Cc Marek and Kukjin,
>>>>
>>>> This would conflict with Marek's:
>>>> [PATCH v2] ARM: multi_v7_defconfig: Enable DWC2 USB driver and USB
>>>> ethernet gadget
>>>> http://www.spinics.net/lists/linux-samsung-soc/msg46743.html
>>>>
>>>> Probably your series and Marek's change should go together to the
>>>> same
>>>> tree. How about rebasing your work on top of it?
>>>
>>> Thanks for the pointer! I'm entirely happy to rebase on Mareks patch or
>>> this one dropped for now until Mareks patch lands.
>>>
>>> These patches were based on current linux-next, so i guess Mareks pathc
>>> didn't land just yet? Tbh, i don't really know what tries multi_v7
>>> tends to go through (especially for non-platform specific changes)
>>
>> It was too late for Marek's patch to go for v4.3 merge window, so I
>> applied it but not pushed to my for-next branch. AFAIU, Stephen does not
>> want v4.4 material during merge window because of mangling with statistics.
>>
>>
>> If you want to rebase (or cherry-pick), you can find Marek's patch here:
>> branch: defconfig-for-next
>> repo: git://github.com/krzk/linux.git
>>
>> However this *branch will be rebased on v4.3-rc1* (because this is
>> material for v4.4). After v4.3-rc1 I can prepare a tag for other
>> interested parties.
>>
>>
>> As for multi_v7_defconfig policy, I got impression from the
>> Arnd/Olof/Kevin comments on LKML, that platform specific changes should
>> go through platform tree. However this is multi-platform change so maybe
>> we can push both patches through arm-soc.
>>
>>
>> Arnd, Olof, Kevin (BTW, you missed to CC him),
>>
>> How do you wish to handle this case (USB_DWC2 and two patches: Marek's
>> and Sjoerd's)?
> 
> I guess the easiest option would be to just drop the dwc2 from here, only 
> keeping the phy enablement and let everything (Marek's and Sjoerd's patch) 
> come together in the armsoc defconfig branch? As the rest of both patches don't 
> seem to touch common areas, they should be easily mergable from different trees 
> as well.

Yes, I agree. The config PHY_ROCKCHIP_USB does not depend on it directly
so this should work fine.

Best regards,
Krzysztof
Sjoerd Simons Sept. 7, 2015, 8:49 a.m. UTC | #6
On Mon, 2015-09-07 at 16:56 +0900, Krzysztof Kozlowski wrote:
> On 07.09.2015 16:53, Heiko Stübner wrote:
> > Am Montag, 7. September 2015, 16:24:53 schrieb Krzysztof Kozlowski:
> > I guess the easiest option would be to just drop the dwc2 from
> > here, only 
> > keeping the phy enablement and let everything (Marek's and Sjoerd's
> > patch) 
> > come together in the armsoc defconfig branch? As the rest of both
> > patches don't 
> > seem to touch common areas, they should be easily mergable from
> > different trees 
> > as well.
> 
> Yes, I agree. The config PHY_ROCKCHIP_USB does not depend on it
> directly
> so this should work fine.

Sounds good, I'll wait a bit longer for more feedback and then send a
V2 with patch 4 adjusted to just enable the rockchip PHY.

Thanks for the feedback!
Javier Martinez Canillas Sept. 7, 2015, 9:33 a.m. UTC | #7
Hello Sjoerd,

On 09/06/2015 10:16 PM, Sjoerd Simons wrote:
> The DesignWare HS OTG USB 2.0 is used in various SoC families among wich
> the various rockchip SoCs, enable the driver and the PHY glue for
> rockchip
> 
> Signed-off-by: Sjoerd Simons <sjoerd.simons@collabora.co.uk>
> 
> ---
> 
>  arch/arm/configs/multi_v7_defconfig | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/arch/arm/configs/multi_v7_defconfig b/arch/arm/configs/multi_v7_defconfig
> index fb673e2..49c733a 100644
> --- a/arch/arm/configs/multi_v7_defconfig
> +++ b/arch/arm/configs/multi_v7_defconfig
> @@ -514,6 +514,7 @@ CONFIG_USB_STORAGE=y
>  CONFIG_USB_DWC3=y
>  CONFIG_USB_CHIPIDEA=y
>  CONFIG_USB_CHIPIDEA_HOST=y
> +CONFIG_USB_DWC2=y
>  CONFIG_AB8500_USB=y
>  CONFIG_KEYSTONE_USB_PHY=y
>  CONFIG_OMAP_USB3=y
> @@ -658,6 +659,7 @@ CONFIG_PHY_HIX5HD2_SATA=y
>  CONFIG_PWM_STI=m
>  CONFIG_OMAP_USB2=y
>  CONFIG_TI_PIPE3=y
> +CONFIG_PHY_ROCKCHIP_USB=y
>  CONFIG_PHY_MIPHY28LP=y
>  CONFIG_PHY_MIPHY365X=y
>  CONFIG_PHY_RCAR_GEN2=m
>

After dropping the DWC2 and keeping only the PHY as suggested
by Heiko and Krzysztof:

Reviewed-by: Javier Martinez Canillas <javier@osg.samsung.com>

Best regards,

Patch
diff mbox

diff --git a/arch/arm/configs/multi_v7_defconfig b/arch/arm/configs/multi_v7_defconfig
index fb673e2..49c733a 100644
--- a/arch/arm/configs/multi_v7_defconfig
+++ b/arch/arm/configs/multi_v7_defconfig
@@ -514,6 +514,7 @@  CONFIG_USB_STORAGE=y
 CONFIG_USB_DWC3=y
 CONFIG_USB_CHIPIDEA=y
 CONFIG_USB_CHIPIDEA_HOST=y
+CONFIG_USB_DWC2=y
 CONFIG_AB8500_USB=y
 CONFIG_KEYSTONE_USB_PHY=y
 CONFIG_OMAP_USB3=y
@@ -658,6 +659,7 @@  CONFIG_PHY_HIX5HD2_SATA=y
 CONFIG_PWM_STI=m
 CONFIG_OMAP_USB2=y
 CONFIG_TI_PIPE3=y
+CONFIG_PHY_ROCKCHIP_USB=y
 CONFIG_PHY_MIPHY28LP=y
 CONFIG_PHY_MIPHY365X=y
 CONFIG_PHY_RCAR_GEN2=m