linux-kernel - Re: [PATCH v2 02/36] media: chips-media: coda: Remove useless setting of min_buffers

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:   Mon, 04 Dec 2023 14:45:05 +0100
From:   Philipp Zabel <p.zabel@...gutronix.de>
To:     Benjamin Gaignard <benjamin.gaignard@...labora.com>,
        hverkuil@...all.nl, mchehab@...nel.org, tfiga@...omium.org,
        m.szyprowski@...sung.com, matt.ranostay@...sulko.com
Cc:     linux-kernel@...r.kernel.org, linux-media@...r.kernel.org,
        linux-staging@...ts.linux.dev, kernel@...labora.com
Subject: Re: [PATCH v2 02/36] media: chips-media: coda: Remove useless
 setting of min_buffers_needed

Hi Benjamin,

On Mo, 2023-12-04 at 14:22 +0100, Benjamin Gaignard wrote:
> This driver uses min_buffers_needed which vb2 uses to ensure
> start_streaming is called when at least 'min_buffers_needed'
> buffers are queued. However, this driver doesn't need this,
> it can stream fine without any buffers queued.
> Just drop this unnecessary restriction.

Did you test this? I think there currently is some initialization in
start_streaming() that doesn't quite work when starting without any
buffers queued (it returns -EINVAL if called with count < 1).
For example, the CODA960 decoder initialization expects 512 bytes of
bitstream payload to run SEQ_INIT, and for JPEG decoding it tries to
determine chroma subsampling from the JPEG header.

regards
Philipp