linux-kernel - Re: [PATCH] drm/bridge: analogix-anx6345: fix set of link bandwidth

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Sat, 22 Feb 2020 10:43:02 +0800
From:   Icenowy Zheng <icenowy@...c.io>
To:     Torsten Duwe <duwe@....de>
CC:     Andrzej Hajda <a.hajda@...sung.com>,
        Neil Armstrong <narmstrong@...libre.com>,
        Laurent Pinchart <Laurent.pinchart@...asonboard.com>,
        Jonas Karlman <jonas@...boo.se>,
        Jernej Skrabec <jernej.skrabec@...l.net>,
        Maxime Ripard <maxime@...no.tech>,
        Vasily Khoruzhick <anarsoul@...il.com>,
        dri-devel@...ts.freedesktop.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] drm/bridge: analogix-anx6345: fix set of link bandwidth



于 2020年2月22日 GMT+08:00 上午1:13:28, Torsten Duwe <duwe@....de> 写到:
>On Sat, Feb 22, 2020 at 12:51:27AM +0800, Icenowy Zheng wrote:
>> Current code tries to store the link rate (in bps, which is a big
>> number) in a u8, which surely overflow. Then it's converted back to
>> bandwidth code (which is thus 0) and written to the chip.
>> 
>> The code sometimes works because the chip will automatically fallback
>to
>> the lowest possible DP link rate (1.62Gbps) when get the invalid
>value.
>> However, on the eDP panel of Olimex TERES-I, which wants 2.7Gbps
>link,
>> it failed.
>> 
>> As we had already read the link bandwidth as bandwidth code in
>earlier
>> code (to check whether it is supported), use it when setting
>bandwidth,
>> instead of converting it to link rate and then converting back.
>> 
>> Fixes: e1cff82c1097 ("drm/bridge: fix anx6345 compilation for v5.5")
>> Signed-off-by: Icenowy Zheng <icenowy@...c.io>
>> ---
>>  drivers/gpu/drm/bridge/analogix/analogix-anx6345.c | 3 +--
>>  1 file changed, 1 insertion(+), 2 deletions(-)
>> 
>> diff --git a/drivers/gpu/drm/bridge/analogix/analogix-anx6345.c
>b/drivers/gpu/drm/bridge/analogix/analogix-anx6345.c
>> index 56f55c53abfd..2dfa2fd2a23b 100644
>> --- a/drivers/gpu/drm/bridge/analogix/analogix-anx6345.c
>> +++ b/drivers/gpu/drm/bridge/analogix/analogix-anx6345.c
>> @@ -210,8 +210,7 @@ static int anx6345_dp_link_training(struct
>anx6345 *anx6345)
>>  	if (err)
>>  		return err;
>>  
>> -	dpcd[0] = drm_dp_max_link_rate(anx6345->dpcd);
>> -	dpcd[0] = drm_dp_link_rate_to_bw_code(dpcd[0]);
>> +	dpcd[0] = dp_bw;
>
>Why do you make this assignment and not use dp_bw directly in the call?

Because the dpcd array is then written as a continous array
back to DPCD.

>
>>  	err = regmap_write(anx6345->map[I2C_IDX_DPTX],
>>  			   SP_DP_MAIN_LINK_BW_SET_REG, dpcd[0]);
>                                                       ^^^^^^
>>  	if (err)
>> -- 
>> 2.24.1
>
>BTW, my version is only a bit more verbose:
>
>https://patchwork.freedesktop.org/patch/354344/
>
>	Torsten

-- 
使用 K-9 Mail 发送自我的Android设备。