[PATCH] OMAP: config : disable the usb host configuration in omap2plus_defconfig

[PATCH] OMAP: config : disable the usb host configuration in omap2plus_defconfig

[Keshava Munegowda]

The usb host is disabled in the omap2 build; This is because
usb host is causing the retention to break in cpu idle.

Signed-off-by: Keshava Munegowda <keshava_mgowda@ti.com>
---
 arch/arm/configs/omap2plus_defconfig |   11 -----------
 1 file changed, 11 deletions(-)

diff --git a/arch/arm/configs/omap2plus_defconfig b/arch/arm/configs/omap2plus_defconfig
index 9854ff4..7b32da6 100644
--- a/arch/arm/configs/omap2plus_defconfig
+++ b/arch/arm/configs/omap2plus_defconfig
@@ -170,17 +170,6 @@ CONFIG_SND_USB_AUDIO=m
 CONFIG_SND_SOC=m
 CONFIG_SND_OMAP_SOC=m
 CONFIG_SND_OMAP_SOC_OMAP3_PANDORA=m
-CONFIG_USB=y
-CONFIG_USB_DEBUG=y
-CONFIG_USB_ANNOUNCE_NEW_DEVICES=y
-CONFIG_USB_DEVICEFS=y
-CONFIG_USB_SUSPEND=y
-CONFIG_USB_MON=y
-CONFIG_USB_EHCI_HCD=y
-CONFIG_USB_WDM=y
-CONFIG_USB_STORAGE=y
-CONFIG_USB_LIBUSUAL=y
-CONFIG_USB_TEST=y
 CONFIG_USB_GADGET=y
 CONFIG_USB_GADGET_DEBUG=y
 CONFIG_USB_GADGET_DEBUG_FILES=y
-- 
1.7.9.5

Re: [PATCH] OMAP: config : disable the usb host configuration in omap2plus_defconfig

[Munegowda, Keshava]

On Fri, Jul 6, 2012 at 5:19 PM, Keshava Munegowda <keshava_mgowda@ti.com> wrote:
> The usb host is disabled in the omap2 build; This is because
> usb host is causing the retention to break in cpu idle.
>
> Signed-off-by: Keshava Munegowda <keshava_mgowda@ti.com>
> ---
>  arch/arm/configs/omap2plus_defconfig |   11 -----------
>  1 file changed, 11 deletions(-)
>
> diff --git a/arch/arm/configs/omap2plus_defconfig b/arch/arm/configs/omap2plus_defconfig
> index 9854ff4..7b32da6 100644
> --- a/arch/arm/configs/omap2plus_defconfig
> +++ b/arch/arm/configs/omap2plus_defconfig
> @@ -170,17 +170,6 @@ CONFIG_SND_USB_AUDIO=m
>  CONFIG_SND_SOC=m
>  CONFIG_SND_OMAP_SOC=m
>  CONFIG_SND_OMAP_SOC_OMAP3_PANDORA=m
> -CONFIG_USB=y
> -CONFIG_USB_DEBUG=y
> -CONFIG_USB_ANNOUNCE_NEW_DEVICES=y
> -CONFIG_USB_DEVICEFS=y
> -CONFIG_USB_SUSPEND=y
> -CONFIG_USB_MON=y
> -CONFIG_USB_EHCI_HCD=y
> -CONFIG_USB_WDM=y
> -CONFIG_USB_STORAGE=y
> -CONFIG_USB_LIBUSUAL=y
> -CONFIG_USB_TEST=y
>  CONFIG_USB_GADGET=y
>  CONFIG_USB_GADGET_DEBUG=y
>  CONFIG_USB_GADGET_DEBUG_FILES=y
> --
> 1.7.9.5
>

Sameo
    please merge this patch ; this makes the core retention to work in
omap2plus_defconfig

regards
keshava

Re: [PATCH] OMAP: config : disable the usb host configuration in omap2plus_defconfig

[Shilimkar, Santosh]

On Fri, Jul 6, 2012 at 5:25 PM, Munegowda, Keshava
<keshava_mgowda@ti.com> wrote:
> On Fri, Jul 6, 2012 at 5:19 PM, Keshava Munegowda <keshava_mgowda@ti.com> wrote:
>> The usb host is disabled in the omap2 build; This is because
>> usb host is causing the retention to break in cpu idle.
>>
>> Signed-off-by: Keshava Munegowda <keshava_mgowda@ti.com>
>> ---
>>  arch/arm/configs/omap2plus_defconfig |   11 -----------
>>  1 file changed, 11 deletions(-)
>>
>> diff --git a/arch/arm/configs/omap2plus_defconfig b/arch/arm/configs/omap2plus_defconfig
>> index 9854ff4..7b32da6 100644
>> --- a/arch/arm/configs/omap2plus_defconfig
>> +++ b/arch/arm/configs/omap2plus_defconfig
>> @@ -170,17 +170,6 @@ CONFIG_SND_USB_AUDIO=m
>>  CONFIG_SND_SOC=m
>>  CONFIG_SND_OMAP_SOC=m
>>  CONFIG_SND_OMAP_SOC_OMAP3_PANDORA=m
>> -CONFIG_USB=y
>> -CONFIG_USB_DEBUG=y
>> -CONFIG_USB_ANNOUNCE_NEW_DEVICES=y
>> -CONFIG_USB_DEVICEFS=y
>> -CONFIG_USB_SUSPEND=y
>> -CONFIG_USB_MON=y
>> -CONFIG_USB_EHCI_HCD=y
>> -CONFIG_USB_WDM=y
>> -CONFIG_USB_STORAGE=y
>> -CONFIG_USB_LIBUSUAL=y
>> -CONFIG_USB_TEST=y
>>  CONFIG_USB_GADGET=y
>>  CONFIG_USB_GADGET_DEBUG=y
>>  CONFIG_USB_GADGET_DEBUG_FILES=y
>> --
>> 1.7.9.5
>>
>
> Sameo
>     please merge this patch ; this makes the core retention to work in
> omap2plus_defconfig
>
If at all this patch has to be merged, it should be via Tony's tree,
otherwise you
might have merge conflicts.

Regards
Santosh

Re: [PATCH] OMAP: config : disable the usb host configuration in omap2plus_defconfig

[Munegowda, Keshava]

On Fri, Jul 6, 2012 at 5:29 PM, Shilimkar, Santosh
<santosh.shilimkar@ti.com> wrote:
> On Fri, Jul 6, 2012 at 5:25 PM, Munegowda, Keshava
> <keshava_mgowda@ti.com> wrote:
>> On Fri, Jul 6, 2012 at 5:19 PM, Keshava Munegowda <keshava_mgowda@ti.com> wrote:
>>> The usb host is disabled in the omap2 build; This is because
>>> usb host is causing the retention to break in cpu idle.
>>>
>>> Signed-off-by: Keshava Munegowda <keshava_mgowda@ti.com>
>>> ---
>>>  arch/arm/configs/omap2plus_defconfig |   11 -----------
>>>  1 file changed, 11 deletions(-)
>>>
>>> diff --git a/arch/arm/configs/omap2plus_defconfig b/arch/arm/configs/omap2plus_defconfig
>>> index 9854ff4..7b32da6 100644
>>> --- a/arch/arm/configs/omap2plus_defconfig
>>> +++ b/arch/arm/configs/omap2plus_defconfig
>>> @@ -170,17 +170,6 @@ CONFIG_SND_USB_AUDIO=m
>>>  CONFIG_SND_SOC=m
>>>  CONFIG_SND_OMAP_SOC=m
>>>  CONFIG_SND_OMAP_SOC_OMAP3_PANDORA=m
>>> -CONFIG_USB=y
>>> -CONFIG_USB_DEBUG=y
>>> -CONFIG_USB_ANNOUNCE_NEW_DEVICES=y
>>> -CONFIG_USB_DEVICEFS=y
>>> -CONFIG_USB_SUSPEND=y
>>> -CONFIG_USB_MON=y
>>> -CONFIG_USB_EHCI_HCD=y
>>> -CONFIG_USB_WDM=y
>>> -CONFIG_USB_STORAGE=y
>>> -CONFIG_USB_LIBUSUAL=y
>>> -CONFIG_USB_TEST=y
>>>  CONFIG_USB_GADGET=y
>>>  CONFIG_USB_GADGET_DEBUG=y
>>>  CONFIG_USB_GADGET_DEBUG_FILES=y
>>> --
>>> 1.7.9.5
>>>
>>
>> Sameo
>>     please merge this patch ; this makes the core retention to work in
>> omap2plus_defconfig
>>
> If at all this patch has to be merged, it should be via Tony's tree,
> otherwise you
> might have merge conflicts.
>
> Regards
> Santosh

Thanks Santosh
           looping Tony........

regards
keshava

Re: [PATCH] OMAP: config : disable the usb host configuration in omap2plus_defconfig

[Tony Lindgren]

* Munegowda, Keshava <keshava_mgowda@ti.com> [120706 05:06]:
> On Fri, Jul 6, 2012 at 5:29 PM, Shilimkar, Santosh
> <santosh.shilimkar@ti.com> wrote:
> > On Fri, Jul 6, 2012 at 5:25 PM, Munegowda, Keshava
> > <keshava_mgowda@ti.com> wrote:
> >> On Fri, Jul 6, 2012 at 5:19 PM, Keshava Munegowda <keshava_mgowda@ti.com> wrote:
> >>> The usb host is disabled in the omap2 build; This is because
> >>> usb host is causing the retention to break in cpu idle.
...
> >> Sameo
> >>     please merge this patch ; this makes the core retention to work in
> >> omap2plus_defconfig
> >>
> > If at all this patch has to be merged, it should be via Tony's tree,
> > otherwise you
> > might have merge conflicts.

I think it's best to fix things properly, even if it means that it will
take longer and go into next kernel version.

Tony

Re: [PATCH] OMAP: config : disable the usb host configuration in omap2plus_defconfig

[Munegowda, Keshava]

On Fri, Jul 6, 2012 at 5:37 PM, Tony Lindgren <tony@atomide.com> wrote:
> * Munegowda, Keshava <keshava_mgowda@ti.com> [120706 05:06]:
>> On Fri, Jul 6, 2012 at 5:29 PM, Shilimkar, Santosh
>> <santosh.shilimkar@ti.com> wrote:
>> > On Fri, Jul 6, 2012 at 5:25 PM, Munegowda, Keshava
>> > <keshava_mgowda@ti.com> wrote:
>> >> On Fri, Jul 6, 2012 at 5:19 PM, Keshava Munegowda <keshava_mgowda@ti.com> wrote:
>> >>> The usb host is disabled in the omap2 build; This is because
>> >>> usb host is causing the retention to break in cpu idle.
> ...
>> >> Sameo
>> >>     please merge this patch ; this makes the core retention to work in
>> >> omap2plus_defconfig
>> >>
>> > If at all this patch has to be merged, it should be via Tony's tree,
>> > otherwise you
>> > might have merge conflicts.
>
> I think it's best to fix things properly, even if it means that it will
> take longer and go into next kernel version.
>
> Tony

Hi tony

    This requires ehci remote wakeup feature to be implemented, which
will take at least to 3 months to
get reviewed in opensource.
hence for now  kevin is also agreed to disable it by default;


regards
keshava

Re: [PATCH] OMAP: config : disable the usb host configuration in omap2plus_defconfig

[Tony Lindgren]

* Munegowda, Keshava <keshava_mgowda@ti.com> [120706 05:30]:
> On Fri, Jul 6, 2012 at 5:37 PM, Tony Lindgren <tony@atomide.com> wrote:
> >
> > I think it's best to fix things properly, even if it means that it will
> > take longer and go into next kernel version.
> 
>     This requires ehci remote wakeup feature to be implemented, which
> will take at least to 3 months to
> get reviewed in opensource.
> hence for now  kevin is also agreed to disable it by default;

Why does echi need a "remote wakeup" feature? If you're talking about
echi wake-up events, I don't see how that would affect core retention.

Tony

Re: [PATCH] OMAP: config : disable the usb host configuration in omap2plus_defconfig

[Munegowda, Keshava]

On Fri, Jul 6, 2012 at 6:09 PM, Tony Lindgren <tony@atomide.com> wrote:
> * Munegowda, Keshava <keshava_mgowda@ti.com> [120706 05:30]:
>> On Fri, Jul 6, 2012 at 5:37 PM, Tony Lindgren <tony@atomide.com> wrote:
>> >
>> > I think it's best to fix things properly, even if it means that it will
>> > take longer and go into next kernel version.
>>
>>     This requires ehci remote wakeup feature to be implemented, which
>> will take at least to 3 months to
>> get reviewed in opensource.
>> hence for now  kevin is also agreed to disable it by default;
>
> Why does echi need a "remote wakeup" feature? If you're talking about
> echi wake-up events, I don't see how that would affect core retention.

Hi Tony
     the cpu should hit the retention in idle; in this case,  up on
usb bus suspend ( no device connected) , if I cut the clock, then later
device detection happen through I/O wakeup, so implementing the
ehci remote wakeup using I/O chain handler ensure the retention in idle too.
now, with the existing ehci driver in mainline, whether device connected or not
the clocks are not cut and prevents the retention in idle.
so, I need to implement the ehci clock gating mechanism through I/O interrupts
in bus spend /resume . This activity will take some time.

regards
keshava

Re: [PATCH] OMAP: config : disable the usb host configuration in omap2plus_defconfig

[Kevin Hilman]

Keshava Munegowda <keshava_mgowda@ti.com> writes:

> The usb host is disabled in the omap2 build; This is because
> usb host is causing the retention to break in cpu idle.

... and causes warnings during boot, and hangs in suspend, can't suspend
using NFSroot, etc. etc.

Thank you for disabling this by default, But I don't think this fix
is targetted enough (more below.)

Tony, when v3.5 came out, I asked them to fix this properly, but nothing
was done.  So I've resorted to requesting that this feature be disabled
by default since it introduced PM regressions.  Of course, I'd like to
see this fixed properly instead, but we've run out of time for v3.5 and
IMO, this PM regression needs to be fixed for v3.5.  See my more
targetted patches below.

> Signed-off-by: Keshava Munegowda <keshava_mgowda@ti.com>
> ---
>  arch/arm/configs/omap2plus_defconfig |   11 -----------
>  1 file changed, 11 deletions(-)
>
> diff --git a/arch/arm/configs/omap2plus_defconfig b/arch/arm/configs/omap2plus_defconfig
> index 9854ff4..7b32da6 100644
> --- a/arch/arm/configs/omap2plus_defconfig
> +++ b/arch/arm/configs/omap2plus_defconfig
> @@ -170,17 +170,6 @@ CONFIG_SND_USB_AUDIO=m
>  CONFIG_SND_SOC=m
>  CONFIG_SND_OMAP_SOC=m
>  CONFIG_SND_OMAP_SOC_OMAP3_PANDORA=m
> -CONFIG_USB=y
> -CONFIG_USB_DEBUG=y
> -CONFIG_USB_ANNOUNCE_NEW_DEVICES=y
> -CONFIG_USB_DEVICEFS=y
> -CONFIG_USB_SUSPEND=y
> -CONFIG_USB_MON=y
> -CONFIG_USB_EHCI_HCD=y
> -CONFIG_USB_WDM=y
> -CONFIG_USB_STORAGE=y
> -CONFIG_USB_LIBUSUAL=y
> -CONFIG_USB_TEST=y

Why disable all of these?  With that, it's not terribly clear what is
going on, or why.

Instead, what you need is a targetted fix to just disable the problem
driver by default:

Either this:

diff --git a/arch/arm/configs/omap2plus_defconfig b/arch/arm/configs/omap2plus_defconfig
index 9854ff4..11828e6 100644
--- a/arch/arm/configs/omap2plus_defconfig
+++ b/arch/arm/configs/omap2plus_defconfig
@@ -176,7 +176,6 @@ CONFIG_USB_ANNOUNCE_NEW_DEVICES=y
 CONFIG_USB_DEVICEFS=y
 CONFIG_USB_SUSPEND=y
 CONFIG_USB_MON=y
-CONFIG_USB_EHCI_HCD=y
 CONFIG_USB_WDM=y
 CONFIG_USB_STORAGE=y
 CONFIG_USB_LIBUSUAL=y

or this:

diff --git a/drivers/mfd/Kconfig b/drivers/mfd/Kconfig
index e129c82..3747826 100644
--- a/drivers/mfd/Kconfig
+++ b/drivers/mfd/Kconfig
@@ -825,7 +825,7 @@ config MFD_WL1273_CORE
 config MFD_OMAP_USB_HOST
 	bool "Support OMAP USBHS core driver"
 	depends on USB_EHCI_HCD_OMAP || USB_OHCI_HCD_OMAP3
-	default y
+	default n
 	help
 	  This is the core driver for the OAMP EHCI and OHCI drivers.
 	  This MFD driver does the required setup functionalities for

Will have the inteded affect with the benefit of being much clearer
about what is actually being disabled, and why.

Kevin

Re: [PATCH] OMAP: config : disable the usb host configuration in omap2plus_defconfig

[Kevin Hilman]

Kevin Hilman <khilman-l0cyMroinI0@public.gmane.org> writes:

> Keshava Munegowda <keshava_mgowda-l0cyMroinI0@public.gmane.org> writes:
>
>> The usb host is disabled in the omap2 build; This is because
>> usb host is causing the retention to break in cpu idle.
>
> ... and causes warnings during boot, and hangs in suspend, can't suspend
> using NFSroot, etc. etc.
>
> Thank you for disabling this by default, But I don't think this fix
> is targetted enough (more below.)
>
> Tony, when v3.5 came out, I asked them to fix this properly, but nothing
> was done.  So I've resorted to requesting that this feature be disabled
> by default since it introduced PM regressions.  Of course, I'd like to
> see this fixed properly instead, but we've run out of time for v3.5 and
> IMO, this PM regression needs to be fixed for v3.5.  See my more
> targetted patches below.
>
>> Signed-off-by: Keshava Munegowda <keshava_mgowda-l0cyMroinI0@public.gmane.org>
>> ---
>>  arch/arm/configs/omap2plus_defconfig |   11 -----------
>>  1 file changed, 11 deletions(-)
>>
>> diff --git a/arch/arm/configs/omap2plus_defconfig b/arch/arm/configs/omap2plus_defconfig
>> index 9854ff4..7b32da6 100644
>> --- a/arch/arm/configs/omap2plus_defconfig
>> +++ b/arch/arm/configs/omap2plus_defconfig
>> @@ -170,17 +170,6 @@ CONFIG_SND_USB_AUDIO=m
>>  CONFIG_SND_SOC=m
>>  CONFIG_SND_OMAP_SOC=m
>>  CONFIG_SND_OMAP_SOC_OMAP3_PANDORA=m
>> -CONFIG_USB=y
>> -CONFIG_USB_DEBUG=y
>> -CONFIG_USB_ANNOUNCE_NEW_DEVICES=y
>> -CONFIG_USB_DEVICEFS=y
>> -CONFIG_USB_SUSPEND=y
>> -CONFIG_USB_MON=y
>> -CONFIG_USB_EHCI_HCD=y
>> -CONFIG_USB_WDM=y
>> -CONFIG_USB_STORAGE=y
>> -CONFIG_USB_LIBUSUAL=y
>> -CONFIG_USB_TEST=y
>
> Why disable all of these?  With that, it's not terribly clear what is
> going on, or why.
>
> Instead, what you need is a targetted fix to just disable the problem
> driver by default:
>
> Either this:
>
> diff --git a/arch/arm/configs/omap2plus_defconfig b/arch/arm/configs/omap2plus_defconfig
> index 9854ff4..11828e6 100644
> --- a/arch/arm/configs/omap2plus_defconfig
> +++ b/arch/arm/configs/omap2plus_defconfig
> @@ -176,7 +176,6 @@ CONFIG_USB_ANNOUNCE_NEW_DEVICES=y
>  CONFIG_USB_DEVICEFS=y
>  CONFIG_USB_SUSPEND=y
>  CONFIG_USB_MON=y
> -CONFIG_USB_EHCI_HCD=y
>  CONFIG_USB_WDM=y
>  CONFIG_USB_STORAGE=y
>  CONFIG_USB_LIBUSUAL=y
>
> or this:
>
> diff --git a/drivers/mfd/Kconfig b/drivers/mfd/Kconfig
> index e129c82..3747826 100644
> --- a/drivers/mfd/Kconfig
> +++ b/drivers/mfd/Kconfig
> @@ -825,7 +825,7 @@ config MFD_WL1273_CORE
>  config MFD_OMAP_USB_HOST
>  	bool "Support OMAP USBHS core driver"
>  	depends on USB_EHCI_HCD_OMAP || USB_OHCI_HCD_OMAP3
> -	default y
> +	default n
>  	help
>  	  This is the core driver for the OAMP EHCI and OHCI drivers.
>  	  This MFD driver does the required setup functionalities for
>
> Will have the inteded affect with the benefit of being much clearer
> about what is actually being disabled, and why.

Since I'm primarily concerned about automated PM testing of
omap2plus_defconfig, I think the first one is better since it just
affects the default omap2plus_defconfig and allows custom defconfigs to
continue to work as usual.

I'll send a patch.

Kevin
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html