linux-kernel - Re: [RFC PATCH v10 6/9] media: tegra: Add Tegra210 Video input driver

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:   Sun, 26 Apr 2020 05:10:51 +0300
From:   Dmitry Osipenko <digetx@...il.com>
To:     Sowjanya Komatineni <skomatineni@...dia.com>,
        Hans Verkuil <hverkuil@...all.nl>, thierry.reding@...il.com,
        jonathanh@...dia.com, frankc@...dia.com, sakari.ailus@....fi,
        helen.koike@...labora.com
Cc:     sboyd@...nel.org, linux-media@...r.kernel.org,
        devicetree@...r.kernel.org, linux-clk@...r.kernel.org,
        linux-tegra@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [RFC PATCH v10 6/9] media: tegra: Add Tegra210 Video input driver

26.04.2020 04:43, Sowjanya Komatineni пишет:
...
>> It looks to me that at least all those hardcoded HW format IDs do not
>> match the older SoCs.
> 
> TPG hard coded formats are supported on prior Tegra.
> 
> Other supported formats are SoC dependent and  part of soc data in the
> driver already.

But I don't see where that SoC-dependent definition is made in
terga210.c. That tegra_image_format enum looks T210-specific, isn't it?

...
>> The driver will need to have a bit better separation if it's supposed to
>> have a common core for all SoCs. Each incompatible VI/CSI hardware
>> version should have its own kernel module.
> 
> currently other Tegra host1x driver (drm) also does similar. Single
> module for all Tegra SoCs.

DRM driver has a proper separation of the sub-drivers where sub-driver
won't load on unsupported hardware. The tegra-video driver should do the
same, i.e. VI and CSI should be individual drivers (and not OPS). There
could be a some common core, but for now it's not obvious to me what
that core should be, maybe just the video.c.

> With current tegra-video, all the v4l2 related common part of
> implementation is same for all tegra's and only
> tegra210.c/tegra186.c/tegra194.c will have corresponding tegra soc
> specific vi/csi programming sequence.

This code shouldn't be shared within the same driver module, IMO.


>> The tegra-video should be okay, although the "video" part sounds a bit
>> too broad since video could mean a lot of things. I think downstream
>> kernel uses (or at least used) the tegra-camera name for the driver,
>> perhaps it could be a reasonable variant as well.
> prior feedback suggests not to use camera variant instead to use video

Alright, then the tegra-video should be fine.