[lkml]   [2016]   [Jan]   [11]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [RFC PATCH v0] Add tw5864 driver
Hi Andrey,

On 01/03/2016 02:41 AM, Andrey Utkin wrote:
> (Disclaimer up to scissors mark)
> Please be so kind to take a look at a new driver.
> We aim to follow high standards of kernel development and possibly to get this driver in mainline kernel.
> The device is multichannel video and audio capture and compression chip TW5864 of Intersil/Techwell.
> The code is in repo , branch "tw5864", subdir drivers/staging/media/tw5864 .
> This branch is regularly rebased during development, so that there's a single commit adding this driver on top of current linux-next.
> Direct link to subdir:
> Implementation status
> * H.264 streaming work stable, including concurrent work of multiple channels on same chip;
> * Audio streaming functionality is not implemented, but is considered for future;
> * The chip has motion detection capability, but of same sensitivity level for whole frame; this was considered quite limiting for our needs and we have implemented per-grid-cell sensitivity with a bit of heuristic processing of motion vector data exposed by hardware. Datasheet-suggested mechanism is not used currently.
> Testing status
> * Runtime tests on my test machine show that video streaming works stable. Multichannel streaming was working, but I haven't test this with latest revision lately yet.
> * Runtime performance will be tested later also on few early-adopters' CCTV machines.
> * -f on files reports no warnings, errors or style issues;
> * on patch reports no warnings, errors or style issues;
> * sparse reports nothing;
> * compilation shows no warnings (gcc 4.9.3);
> * compilation with EXTRA_CFLAGS=-W shows a lot of warnings from included headers (over 9000 lines of output). Seems this technique from Documentation/SubmitChecklist is not practical in this case
> * Other Documentation/SubmitChecklist advises weren't thoroughly worked out.

Did you also test with v4l2-compliance? Before I accept the driver I want to see the
output of 'v4l2-compliance' and 'v4l2-compliance -s'. Basically there shouldn't be
any failures.

I did a quick scan over the source and I saw nothing big. When you post the new
version I will go over it more carefully and do a proper review.



 \ /
  Last update: 2016-01-11 12:41    [W:0.078 / U:0.800 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site