Re: [PATCH] drm/sti: load HQVDP firmware the first time HQVDP's plane is used

From: Vincent ABRIOU <vincent.abriou@st.com>
To: Emil Velikov <emil.l.velikov@gmail.com>
Cc: Benjamin Gaignard <benjamin.gaignard@linaro.org>,
	Vincent ABRIOU <vincent.abriou@st.com>,
	Fabien DESSENNE <fabien.dessenne@st.com>,
	ML dri-devel <dri-devel@lists.freedesktop.org>
Subject: Re: [PATCH] drm/sti: load HQVDP firmware the first time HQVDP's plane is used
Date: Wed, 4 Nov 2015 10:17:26 +0100	[thread overview]
Message-ID: <5639CD26.7000609@st.com> (raw)
In-Reply-To: <CACvgo53hui00oHxvH6m1_qvMS7z+Twpk1bdCL6gwqSjn7V16wA@mail.gmail.com>

Hi Emil,

On 11/03/2015 03:55 PM, Emil Velikov wrote:
> Hi Vincent,
>
> On 2 November 2015 at 10:05, Vincent Abriou <vincent.abriou@st.com> wrote:
>> @@ -754,6 +901,9 @@ static void sti_hqvdp_atomic_update(struct drm_plane *drm_plane,
>>          sti_hqvdp_update_hvsrc(HVSRC_VERT, scale_v, &cmd->hvsrc);
>>
>>          if (first_prepare) {
>> +               /* Start HQVDP XP70 coprocessor */
>> +               sti_hqvdp_start_xp70(hqvdp);
>> +
> Shouldn't one check for things such as this (and
> sti_hqvdp_check_hw_scaling, amongst other things) in atomic_check() ?
> Mostl/all other drivers seems to do so. After all drm core check()s
> and if everything/state is ok then calls update().
>

Thanks to highlight me this point.
Atomic has been newly introduced in our sti driver and we updated our 
driver in a straightforward way to make it work with atomic leaving 
aside such point.

I will trig a rework for this for the next update of our driver taking 
into account atomic_check for planes management.

Regards
Vincent

> Regards,
> Emil
>
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dri-devel