Re: [linux-next:master 7437/7719] drivers/gpu/drm/rockchip/rockchip_vop2_reg.c:833:28: error: implicit declaration of function 'FIELD_PREP'

[Heiko Stuebner <heiko@sntech.de>]

Hi Andy,

Am Dienstag, 4. März 2025, 04:40:47 MEZ schrieb Andy Yan:
> On 3/4/25 01:20, kernel test robot wrote:
> > tree:   https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git master
> > head:   cd3215bbcb9d4321def93fea6cfad4d5b42b9d1d
> > commit: 328e6885996ca2c6eb8b07d3c9bb1439fdcb088f [7437/7719] drm/rockchip: vop2: Add platform specific callback
> > config: alpha-allyesconfig (https://download.01.org/0day-ci/archive/20250304/202503040135.fgoyWdLB-lkp@intel.com/config)
> > compiler: alpha-linux-gcc (GCC) 14.2.0
> > reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20250304/202503040135.fgoyWdLB-lkp@intel.com/reproduce)
> 
> I can't reproduce this error on arm64 defconfig, so this let me wonder that should we
> let drm/rockchip depends on ARCH_ROCKCHIP, so as to reduce the testing overhead on other
> unrelated platforms?

Hmm, depends.

Limiting Rockchip-drm to ARCH_ROCKCHIP || COMPILE_TEST could make sense,
as as you say, completely different architectures have absolute no use
for this driver.

But the error was real nevertheless. We should never rely on headers
getting included by accident. the vop2-reg.c change does introduces the use
of FIELD_PREP, so we should include the header ourself - as my change from
yesterday did. So the test-robot showed us an actual problem :-)


Heiko