linux-kernel - Re: [PATCH] drm: mediatek: mtk_dsi: Fix NO_EOT

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]

Message-ID: <07c93d61-c5fd-f074-abb2-73fdaa81fd65@collabora.com>
Date:   Wed, 19 Jul 2023 09:39:45 +0200
From:   AngeloGioacchino Del Regno 
        <angelogioacchino.delregno@...labora.com>
To:     chunkuang.hu@...nel.org
Cc:     p.zabel@...gutronix.de, airlied@...il.com, daniel@...ll.ch,
        matthias.bgg@...il.com, jitao.shi@...iatek.com, ck.hu@...iatek.com,
        shaoming.chen@...iatek.com, yt.shen@...iatek.com,
        dri-devel@...ts.freedesktop.org,
        linux-mediatek@...ts.infradead.org, linux-kernel@...r.kernel.org,
        linux-arm-kernel@...ts.infradead.org,
        Alexandre Mergnat <amergnat@...libre.com>
Subject: Re: [PATCH] drm: mediatek: mtk_dsi: Fix NO_EOT_PACKET
 settings/handling

Il 23/06/23 13:29, Alexandre Mergnat ha scritto:
> 
> 
> On 23/06/2023 11:47, AngeloGioacchino Del Regno wrote:
>> Il 23/05/23 12:42, AngeloGioacchino Del Regno ha scritto:
>>> Due to the initial confusion about MIPI_DSI_MODE_EOT_PACKET, properly
>>> renamed to MIPI_DSI_MODE_NO_EOT_PACKET, reflecting its actual meaning,
>>> both the DSI_TXRX_CON register setting for bit (HSTX_)DIS_EOT and the
>>> later calculation for horizontal sync-active (HSA), back (HBP) and
>>> front (HFP) porches got incorrect due to the logic being inverted.
>>>
>>> This means that a number of settings were wrong because....:
>>>   - DSI_TXRX_CON register setting: bit (HSTX_)DIS_EOT should be
>>>     set in order to disable the End of Transmission packet;
>>>   - Horizontal Sync and Back/Front porches: The delta used to
>>>     calculate all of HSA, HBP and HFP should account for the
>>>     additional EOT packet.
>>>
>>> Before this change...
>>>   - Bit (HSTX_)DIS_EOT was being set when EOT packet was enabled;
>>>   - For HSA/HBP/HFP delta... all three were wrong, as words were
>>>     added when EOT disabled, instead of when EOT packet enabled!
>>>
>>> Invert the logic around flag MIPI_DSI_MODE_NO_EOT_PACKET in the
>>> MediaTek DSI driver to fix the aforementioned issues.
>>>
>>> Fixes: 8b2b99fd7931 ("drm/mediatek: dsi: Fine tune the line time caused by EOTp")
>>> Fixes: 2d52bfba09d1 ("drm/mediatek: add non-continuous clock mode and EOT packet 
>>> control")
>>> Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@...labora.com>
>>
>> Gentle ping for an important fix.
>>
>> Regards,
>> Angelo
>>
>>> ---
>>>   drivers/gpu/drm/mediatek/mtk_dsi.c | 4 ++--
>>>   1 file changed, 2 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/drivers/gpu/drm/mediatek/mtk_dsi.c 
>>> b/drivers/gpu/drm/mediatek/mtk_dsi.c
>>> index 7d5250351193..b0ab38e59db9 100644
>>> --- a/drivers/gpu/drm/mediatek/mtk_dsi.c
>>> +++ b/drivers/gpu/drm/mediatek/mtk_dsi.c
>>> @@ -407,7 +407,7 @@ static void mtk_dsi_rxtx_control(struct mtk_dsi *dsi)
>>>       if (dsi->mode_flags & MIPI_DSI_CLOCK_NON_CONTINUOUS)
>>>           tmp_reg |= HSTX_CKLP_EN;
>>> -    if (!(dsi->mode_flags & MIPI_DSI_MODE_NO_EOT_PACKET))
>>> +    if (dsi->mode_flags & MIPI_DSI_MODE_NO_EOT_PACKET)
>>>           tmp_reg |= DIS_EOT;
>>>       writel(tmp_reg, dsi->regs + DSI_TXRX_CTRL);
>>> @@ -484,7 +484,7 @@ static void mtk_dsi_config_vdo_timing(struct mtk_dsi *dsi)
>>>                 timing->da_hs_zero + timing->da_hs_exit + 3;
>>>       delta = dsi->mode_flags & MIPI_DSI_MODE_VIDEO_BURST ? 18 : 12;
>>> -    delta += dsi->mode_flags & MIPI_DSI_MODE_NO_EOT_PACKET ? 2 : 0;
>>> +    delta += dsi->mode_flags & MIPI_DSI_MODE_NO_EOT_PACKET ? 0 : 2;
>>>       horizontal_frontporch_byte = vm->hfront_porch * dsi_tmp_buf_bpp;
>>>       horizontal_front_back_byte = horizontal_frontporch_byte + 
>>> horizontal_backporch_byte;
>>
>>
> 
> Sounds logic
> 
> Reviewed-by: Alexandre Mergnat <amergnat@...libre.com>
> 

Ping again.

Regards,
Angelo