Re: [PATCH v2 -next] drm/display: Fix build error without CONFIG_OF

[Dmitry Baryshkov <dmitry.baryshkov@linaro.org>]

On 23/05/2022 15:54, Linus Walleij wrote:
> On Mon, May 23, 2022 at 2:46 PM Linus Walleij <linus.walleij@linaro.org> wrote:
>> On Fri, May 6, 2022 at 2:33 PM YueHaibing <yuehaibing@huawei.com> wrote:
>>
>>> While CONFIG_OF is n but COMPILE_TEST is y, we got this:
>>>
>>> WARNING: unmet direct dependencies detected for DRM_DP_AUX_BUS
>>>    Depends on [n]: HAS_IOMEM [=y] && DRM [=y] && OF [=n]
>>>    Selected by [y]:
>>>    - DRM_MSM [=y] && HAS_IOMEM [=y] && DRM [=y] && (ARCH_QCOM || SOC_IMX5 || COMPILE_TEST [=y]) && COMMON_CLK [=y] && IOMMU_SUPPORT [=y] && (QCOM_OCMEM [=n] || QCOM_OCMEM [=n]=n) && (QCOM_LLCC [=y] || QCOM_LLCC [=y]=n) && (QCOM_COMMAND_DB [=n] || QCOM_COMMAND_DB [=n]=n)
>>>
>>> Make DRM_DP_AUX_BUS depends on OF || COMPILE_TEST to fix this warning.
>>>
>>> Fixes: f5d01644921b ("drm/msm: select DRM_DP_AUX_BUS for the AUX bus support")
>>> Signed-off-by: YueHaibing <yuehaibing@huawei.com>
>>
>> Patch applied to the DRM tree.
> 
> Nope, failed:
> 
> $ dim push-branch drm-misc-next
> dim: ac890b9eeb9b ("drm/display: Fix build error without CONFIG_OF"):
> Fixes: SHA1 in not pointing at an ancestor:
> dim:     f5d01644921b ("drm/msm: select DRM_DP_AUX_BUS for the AUX bus support")
> dim: ERROR: issues in commits detected, aborting
> 
> I don't know what to do with this, sorry. The other committers are maybe better
> with this kind of situations. I think it is designed to stop me from
> shooting myself
> in the foot.

It 'Fixes' a commit in the drm/msm tree, however a fix is not 
msm-specific. We can push it out through the drm/msm, but we'd need an 
ack from drm/core.

-- 
With best wishes
Dmitry