Re: [PATCH v2 6/7] drm/msm: Drop unnecessary mach include

Re: [PATCH v2 6/7] drm/msm: Drop unnecessary mach include

[Rob Clark]

On Mon, Dec 30, 2013 at 4:15 PM, Stephen Boyd <sboyd@codeaurora.org> wrote:
> This file doesn't use the clk_reset() API that is exposed in
> mach-msm's mach/clk.h file. Remove the include so that this
> driver can be compiled as part of the multi-platform kernel.

Thanks!

Signed-off-by: Rob Clark <robdclark@gmail.com>


> Cc: Rob Clark <robdclark@gmail.com>
> Cc: Dave Airlie <airlied@redhat.com>
> ---
>  drivers/gpu/drm/msm/mdp4/mdp4_dtv_encoder.c | 2 --
>  1 file changed, 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/msm/mdp4/mdp4_dtv_encoder.c b/drivers/gpu/drm/msm/mdp4/mdp4_dtv_encoder.c
> index 5e0dcae..3799ccc 100644
> --- a/drivers/gpu/drm/msm/mdp4/mdp4_dtv_encoder.c
> +++ b/drivers/gpu/drm/msm/mdp4/mdp4_dtv_encoder.c
> @@ -15,8 +15,6 @@
>   * this program.  If not, see <http://www.gnu.org/licenses/>.
>   */
>
> -#include <mach/clk.h>
> -
>  #include "mdp4_kms.h"
>
>  #include "drm_crtc.h"
> --
> The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
> hosted by The Linux Foundation
>

Re: [PATCH v2 6/7] drm/msm: Drop unnecessary mach include

[Rob Clark]

On Mon, Dec 30, 2013 at 8:12 PM, Rob Clark <robdclark@gmail.com> wrote:
> On Mon, Dec 30, 2013 at 4:15 PM, Stephen Boyd <sboyd@codeaurora.org> wrote:
>> This file doesn't use the clk_reset() API that is exposed in
>> mach-msm's mach/clk.h file. Remove the include so that this
>> driver can be compiled as part of the multi-platform kernel.
>
> Thanks!
>
> Signed-off-by: Rob Clark <robdclark@gmail.com>

oh, fwiw, things are a bit re-arranged on msm-next-staging (to add
8074 and 8060a support).. currently I have:

------------------
#if defined(CONFIG_ARCH_MSM)
#  include <mach/clk.h>
#elif defined(CONFIG_COMPILE_TEST)
/* stubs we need for compile-test: */
static inline struct device *msm_iommu_get_ctx(const char *ctx_name)
{
        return NULL;
}
#endif
------------------

Not really sure what to do about msm_iommu_get_ctx(), or whether I
still need that?

BR,
-R


>
>
>> Cc: Rob Clark <robdclark@gmail.com>
>> Cc: Dave Airlie <airlied@redhat.com>
>> ---
>>  drivers/gpu/drm/msm/mdp4/mdp4_dtv_encoder.c | 2 --
>>  1 file changed, 2 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/msm/mdp4/mdp4_dtv_encoder.c b/drivers/gpu/drm/msm/mdp4/mdp4_dtv_encoder.c
>> index 5e0dcae..3799ccc 100644
>> --- a/drivers/gpu/drm/msm/mdp4/mdp4_dtv_encoder.c
>> +++ b/drivers/gpu/drm/msm/mdp4/mdp4_dtv_encoder.c
>> @@ -15,8 +15,6 @@
>>   * this program.  If not, see <http://www.gnu.org/licenses/>.
>>   */
>>
>> -#include <mach/clk.h>
>> -
>>  #include "mdp4_kms.h"
>>
>>  #include "drm_crtc.h"
>> --
>> The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
>> hosted by The Linux Foundation
>>

Re: [PATCH v2 6/7] drm/msm: Drop unnecessary mach include

[Stephen Boyd]

On 12/30, Rob Clark wrote:
> On Mon, Dec 30, 2013 at 8:12 PM, Rob Clark <robdclark@gmail.com> wrote:
> > On Mon, Dec 30, 2013 at 4:15 PM, Stephen Boyd <sboyd@codeaurora.org> wrote:
> >> This file doesn't use the clk_reset() API that is exposed in
> >> mach-msm's mach/clk.h file. Remove the include so that this
> >> driver can be compiled as part of the multi-platform kernel.
> >
> > Thanks!
> >
> > Signed-off-by: Rob Clark <robdclark@gmail.com>

I think you mean Acked-by? Signed-off-by usually means you're
sending the patch along.

> 
> oh, fwiw, things are a bit re-arranged on msm-next-staging (to add
> 8074 and 8060a support).. currently I have:
> 
> ------------------
> #if defined(CONFIG_ARCH_MSM)
> #  include <mach/clk.h>
> #elif defined(CONFIG_COMPILE_TEST)
> /* stubs we need for compile-test: */
> static inline struct device *msm_iommu_get_ctx(const char *ctx_name)
> {
>         return NULL;
> }
> #endif
> ------------------
> 
> Not really sure what to do about msm_iommu_get_ctx(), or whether I
> still need that?
> 

I think the drm Kconfig should be selecting the MSM iommu Kconfig
symbol. At least for now, it's a direct compile time dependency
of this driver. I still don't see a use of mach/clk.h though, so
we should be able to drop it completely.

Is your msm-next-staging tree going into v3.14? I don't see
anything in linux-next so far.

-- 
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
hosted by The Linux Foundation

Re: [PATCH v2 6/7] drm/msm: Drop unnecessary mach include

[Olof Johansson]

On Mon, Dec 30, 2013 at 09:15:58PM -0800, Stephen Boyd wrote:
> On 12/30, Rob Clark wrote:
> > On Mon, Dec 30, 2013 at 8:12 PM, Rob Clark <robdclark@gmail.com> wrote:
> > > On Mon, Dec 30, 2013 at 4:15 PM, Stephen Boyd <sboyd@codeaurora.org> wrote:
> > >> This file doesn't use the clk_reset() API that is exposed in
> > >> mach-msm's mach/clk.h file. Remove the include so that this
> > >> driver can be compiled as part of the multi-platform kernel.
> > >
> > > Thanks!
> > >
> > > Signed-off-by: Rob Clark <robdclark@gmail.com>
> 
> I think you mean Acked-by? Signed-off-by usually means you're
> sending the patch along.

I've held off applying this, but I have applied the others -- please
merge this through the DRM tree since it will need to go on top of
changes there.

That also means that enabling the MSM DTM driver in a multiplatform
kernel will cause build breakages until the equivalent patch has been
picked up. Since it's not enabled by any defconfigs I'm not too worried,
but it will show up on randconfigs.


-Olof

Re: [PATCH v2 6/7] drm/msm: Drop unnecessary mach include

[Stephen Boyd]

On 01/09/14 00:13, Olof Johansson wrote:
> On Mon, Dec 30, 2013 at 09:15:58PM -0800, Stephen Boyd wrote:
>> On 12/30, Rob Clark wrote:
>>> On Mon, Dec 30, 2013 at 8:12 PM, Rob Clark <robdclark@gmail.com> wrote:
>>>> On Mon, Dec 30, 2013 at 4:15 PM, Stephen Boyd <sboyd@codeaurora.org> wrote:
>>>>> This file doesn't use the clk_reset() API that is exposed in
>>>>> mach-msm's mach/clk.h file. Remove the include so that this
>>>>> driver can be compiled as part of the multi-platform kernel.
>>>> Thanks!
>>>>
>>>> Signed-off-by: Rob Clark <robdclark@gmail.com>
>> I think you mean Acked-by? Signed-off-by usually means you're
>> sending the patch along.
> I've held off applying this, but I have applied the others -- please
> merge this through the DRM tree since it will need to go on top of
> changes there.
>
> That also means that enabling the MSM DTM driver in a multiplatform
> kernel will cause build breakages until the equivalent patch has been
> picked up. Since it's not enabled by any defconfigs I'm not too worried,
> but it will show up on randconfigs.
>

Ok. I think Rob's still planning to send his patch through the DRM tree
so we should be ok on the other end of the merge window.

-- 
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
hosted by The Linux Foundation