Yes, exactly. The logic in two successive if checks of "(update_type != UPDATE_TYPE_FAST) " is incorrect still.
Possibly we can update code by dropping one if check as below. Anyway, this needs confirm from display team @Lee, Alvin @Lei, Jun.
if (update_type != UPDATE_TYPE_FAST) {
dc->hwss.post_unlock_program_front_end(dc, context);
if (dc->hwss.commit_subvp_config)
dc->hwss.commit_subvp_config(dc, context);
}
Regards,
Guchun
-----Original Message-----
From: Koenig, Christian <Christian.Koenig@xxxxxxx>
Sent: Thursday, September 1, 2022 7:07 PM
To: Song, Asher <Asher.Song@xxxxxxx>; amd-gfx@xxxxxxxxxxxxxxxxxxxxx; Lee, Alvin <Alvin.Lee2@xxxxxxx>; Lei, Jun <Jun.Lei@xxxxxxx>; Chen, Guchun <Guchun.Chen@xxxxxxx>
Subject: Re: [PATCH] drm/display: guard if clause
Well please adjust the subject line, that should read something like "drm/amd/display:..." or "drm/amdgpu:...".
Am 01.09.22 um 11:11 schrieb Asher Song:
> To eliminate the following compiling error, guard if clause.
>
> drivers/gpu/drm/amd/amdgpu/../display/dc/core/dc.c: In function 'commit_planes_for_stream':
> drivers/gpu/drm/amd/amdgpu/../display/dc/core/dc.c:3521:2: error: this
> 'if' clause does not guard... [-Werror=misleading-indentation]
> drivers/gpu/drm/amd/amdgpu/../display/dc/core/dc.c:3523:3: note: ...this statement, but the latter is misleadingly indented as if it were guarded by the 'if'
> if (update_type != UPDATE_TYPE_FAST)
> ^~
>
> Signed-off-by: Asher Song <Asher.Song@xxxxxxx>
> ---
> drivers/gpu/drm/amd/display/dc/core/dc.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c
> b/drivers/gpu/drm/amd/display/dc/core/dc.c
> index b49237390cce..66072ac1bb4f 100644
> --- a/drivers/gpu/drm/amd/display/dc/core/dc.c
> +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
> @@ -3505,11 +3505,12 @@ static void commit_planes_for_stream(struct dc *dc,
> top_pipe_to_program->stream_res.tg);
> }
>
> - if (update_type != UPDATE_TYPE_FAST)
> + if (update_type != UPDATE_TYPE_FAST){
> dc->hwss.post_unlock_program_front_end(dc, context);
> if (update_type != UPDATE_TYPE_FAST)
> if (dc->hwss.commit_subvp_config)
> dc->hwss.commit_subvp_config(dc, context);
> + }
That looks like a step into the right directly, but please re-read the code:
if (update_type != UPDATE_TYPE_FAST) {
....
if (update_type != UPDATE_TYPE_FAST)
....
}
That's certainly still not correct.
Regards,
Christian.
>
> /* Since phantom pipe programming is moved to post_unlock_program_front_end,
> * move the SubVP lock to after the phantom pipes have been setup
