On Tue, 11 Oct 2022, Harry Wentland <harry.wentland@xxxxxxx> wrote: > On 2022-10-11 09:49, Jani Nikula wrote: >> The connector->override_edid flag is strictly for EDID override debugfs >> management, and drivers have no business using it. >> >> Cc: Alex Deucher <alexander.deucher@xxxxxxx> >> Cc: Christian König <christian.koenig@xxxxxxx> >> Cc: Xinhui Pan <Xinhui.Pan@xxxxxxx> >> Cc: amd-gfx@xxxxxxxxxxxxxxxxxxxxx >> Signed-off-by: Jani Nikula <jani.nikula@xxxxxxxxx> >> >> --- >> >> I really have no idea what the functional impact here is. I can only >> guess that the intention is to abuse ->override_edid to block EDID >> property updates. In any case, this use needs to go. >> >> It also seems really curious we get here via connector .get_modes hook! >> --- >> drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 3 --- >> 1 file changed, 3 deletions(-) >> >> diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c >> index 4c73727e0b7d..d96877196a7f 100644 >> --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c >> +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c >> @@ -6109,7 +6109,6 @@ static void create_eml_sink(struct amdgpu_dm_connector *aconnector) >> aconnector->base.name); >> >> aconnector->base.force = DRM_FORCE_OFF; >> - aconnector->base.override_edid = false; > > I'm not even sure the DRM_FORCE_OFF business is right. > > Either way, I don't think amdgpu should be messing with > override_edid, so this is > Reviewed-by: Harry Wentland <harry.wentland@xxxxxxx> Thanks for the swift review; is it okay to merge this via drm-misc-next along with the rest (once they've been reviewed, of course)? BR, Jani. > > Harry > >> return; >> } >> >> @@ -6144,8 +6143,6 @@ static void handle_edid_mgmt(struct amdgpu_dm_connector *aconnector) >> link->verified_link_cap.link_rate = LINK_RATE_HIGH2; >> } >> >> - >> - aconnector->base.override_edid = true; >> create_eml_sink(aconnector); >> } >> > -- Jani Nikula, Intel Open Source Graphics Center
