Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Subject: Re: [PATCH][next] drm/rockchip: Remove redundant assignment of
 pointer connector
From:   Alex Bee <knaerzche@gmail.com>
To:     =?UTF-8?Q?Heiko_St=c3=bcbner?= <heiko@sntech.de>,
        Colin King <colin.king@canonical.com>,
        Sandy Huang <hjc@rock-chips.com>,
        David Airlie <airlied@linux.ie>,
        Daniel Vetter <daniel@ffwll.ch>,
        dri-devel@lists.freedesktop.org,
        linux-arm-kernel@lists.infradead.org,
        linux-rockchip@lists.infradead.org
Cc:     kernel-janitors@vger.kernel.org, linux-kernel@vger.kernel.org
References: <20210922112416.182134-1-colin.king@canonical.com>
 <27c79f7a-8e4c-fad8-c6cf-a89793f2e3c6@gmail.com> <22365175.EbdSka62eY@diego>
 <6d18a1a6-37e3-41f9-ddd1-1dae33864d23@gmail.com>
Message-ID: <4a166499-1e83-99c8-8859-9d01a301412d@gmail.com>
Date:   Wed, 22 Sep 2021 21:20:10 +0200
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101
 Thunderbird/78.13.0
MIME-Version: 1.0
In-Reply-To: <6d18a1a6-37e3-41f9-ddd1-1dae33864d23@gmail.com>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: 8bit
Content-Language: en-US
Precedence: bulk

Hi all,

Am 22.09.21 um 19:31 schrieb Alex Bee:
> Hi Heiko,
>
> Am 22.09.21 um 18:45 schrieb Heiko Stübner:
>> Hi Alex,
>>
>> Am Mittwoch, 22. September 2021, 18:35:38 CEST schrieb Alex Bee:
>>> Hi Colin,
>>> Am 22.09.21 um 13:24 schrieb Colin King:
>>>> From: Colin Ian King <colin.king@canonical.com>
>>>>
>>>> The pointer connector is being assigned a value that is never
>>>> read, it is being updated immediately afterwards. The assignment
>>>> is redundant and can be removed.
>>> The pointer to the connector is used in rockchip_rgb_fini for
>>> drm_connector_cleanup.
>>> It's pretty much the same for the encoder, btw.
>> I think the issue is more the two lines
>>
>>     connector = &rgb->connector;
>>       connector = drm_bridge_connector_init(rgb->drm_dev, encoder);
>>
>> hence the connector = &rgb->connector being overwritten immediately 
>> after
>>
>> Now that I look at it again, the whole approach looks strange.
>> drm_bridge_connector_init() creates the connector structure and
>> returns a pointer to it.
>
> Totally agreed.
>
> The main reason I was doing it that way, was the way it was done 
> already in rockchip_lvds.c, where the connector was already existent 
> in the struct rockchip_lvds (and was already used in the panel-case - 
> all places where it is used accept pointers also, btw) and is *no* 
> pointer - and is done already this very strange way.
>
> I wanted to re-use it for the bridge-case and didn't want to differ in 
> coding in rockchip-rgb to much.
>
> The only reason I can think of, why it was done that way is, that we 
> might need a pointer to a fully initialized struct drm_connector for 
> some reason (drm_connector_cleanup ?), what we wouldn't have if have 
> just a pointer and something goes wrong before drm_connector_init 
> respectivly drm_bridge_connector_init.
>
> Alex
>
>
>> So the first line below sets the connector pointer to point to the
>> &rgb->connector element and the second line then set a completely
>> different address into it.
>>
>> So the connector element in rockchip_lvds and rockchip_rgb should 
>> actually
>> become a pointer itself to hold the connector element returned from
>> drm_bridge_connector_init() .
It turns out, nothing bad happens (i.e. rockchip_rgb_fini, the only 
place where the connector is also used, isn't called if 
rockchip_rgb_init fails) - so it will be OK if we make the connector a 
pointer in struct rockchip_rgb. But we'll need to keep it a "full" 
struct drm_connector in struct rockchip_lvds, since in case it's a 
panel  it gets properties assigend before drm_connector_init is called 
and should for that reason initialized before.

I'll send a patch soon.

Alex

>>
>>
>> Heiko
>>
>>> Regards,
>>>
>>> Alex
>>>> Addresses-Coverity: ("Unused value")
>>>> Signed-off-by: Colin Ian King <colin.king@canonical.com>
>>>> ---
>>>>    drivers/gpu/drm/rockchip/rockchip_rgb.c | 1 -
>>>>    1 file changed, 1 deletion(-)
>>>>
>>>> diff --git a/drivers/gpu/drm/rockchip/rockchip_rgb.c 
>>>> b/drivers/gpu/drm/rockchip/rockchip_rgb.c
>>>> index 09be9678f2bd..18fb84068a64 100644
>>>> --- a/drivers/gpu/drm/rockchip/rockchip_rgb.c
>>>> +++ b/drivers/gpu/drm/rockchip/rockchip_rgb.c
>>>> @@ -150,7 +150,6 @@ struct rockchip_rgb *rockchip_rgb_init(struct 
>>>> device *dev,
>>>>        if (ret)
>>>>            goto err_free_encoder;
>>>>    -    connector = &rgb->connector;
>>>>        connector = drm_bridge_connector_init(rgb->drm_dev, encoder);
>>>>        if (IS_ERR(connector)) {
>>>>            DRM_DEV_ERROR(drm_dev->dev,
>>>>
>>>
>>
>>
>>