linux-kernel - Re: Hardware-accelerated video decoders used through a firmware instead of hardware registers

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] [day] [month] [year] [list]

Message-ID: <1557755257.4442.8.camel@pengutronix.de>
Date:   Mon, 13 May 2019 15:47:37 +0200
From:   Philipp Zabel <p.zabel@...gutronix.de>
To:     Paul Kocialkowski <paul.kocialkowski@...tlin.com>,
        Nicolas Dufresne <nicolas@...fresne.ca>,
        Hans Verkuil <hans.verkuil@...co.com>,
        Sakari Ailus <sakari.ailus@...ux.intel.com>,
        Mauro Carvalho Chehab <mchehab@...nel.org>
Cc:     Laurent Pinchart <laurent.pinchart@...asonboard.com>,
        Maxime Ripard <maxime.ripard@...tlin.com>,
        linux-kernel@...r.kernel.org, linux-media@...r.kernel.org,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        Thierry Reding <thierry.reding@...il.com>,
        Tiffany Lin <tiffany.lin@...iatek.com>,
        Andrew-CT Chen <andrew-ct.chen@...iatek.com>
Subject: Re: Hardware-accelerated video decoders used through a firmware
 instead of hardware registers

Hi,

On Sun, 2019-05-12 at 18:32 +0200, Paul Kocialkowski wrote:
[...]
> I would be curious to know what the situation is on the i.MX6 coda
> block, which also seems pretty obscure.

FWIW, I had started collecting things I learned about the BIT processor
in the CODA IP cores, mostly by looking at the firmware files
distributed by Freescale/NXP: https://github.com/pH5/coda-bits

It is a somewhat strange custom 16-bit DSP architecture. There is a
rudimentary start for a disassembler in there as well, but large parts
of the instruction set are still completely unknown, and I have no idea
how the address generator / DMA units or bitstream accelerators work.

I would be delighted if somebody would like to pick up analyzing the BIT
processor ISA further. I think it could be fruitful to start
systematically throwing instructions at it and see what happens, to
learn more. I haven't had much motivation to play with this, recently.

About the internal connections and available accelerator units, there is
a block diagram in the i.MX53 TRM, but I am not aware of any register
level description for any of these.

regards
Philipp