On 2018-07-03 06:13 PM, Felix Kuehling wrote: > On 2018-07-03 10:19 AM, Tom St Denis wrote: >> Hi all, >> >> This block >> >> #if defined(CONFIG_DRM_AMD_DC) >>        else if (amdgpu_device_has_dc_support(adev)) >>            amdgpu_device_ip_block_add(adev, &dm_ip_block); >> #else >> #   warning "Enable CONFIG_DRM_AMD_DC for display support on SOC15." >> #endif >> >> >> in soc15_set_ip_blocks() should probably be ported to a runtime >> drm_warn() no? This means a kernel with no DC support will just >> silently fail to light up on SOC15. >> >> I can write a patch for this if nobody objects. > > Maybe do both. So someone building a kernel without DC gets a warning as > well as someone running it. Yep, we already do both for CONFIG_MTRR & CONFIG_X86_PAT. BTW, the build warning should also be guarded by #ifndef CONFIG_COMPILE_TEST, per commit 31bb90f1cd084 "drm/amdgpu: shut up #warning for compile testing". OTOH, do we even need the capability to build the driver without DC anymore? -- Earthling Michel Dänzer | http://www.amd.com Libre software enthusiast | Mesa and X developer
