[FFmpeg-devel] [PATCH 2/3] VP8: split out declarations to new header

Jason Garrett-Glaser jason
Fri Feb 18 00:29:13 CET 2011


On Wed, Feb 16, 2011 at 10:42 AM, Ronald S. Bultje <rsbultje at gmail.com> wrote:
> Hi,
>
> On Wed, Feb 16, 2011 at 1:38 PM, Jason Garrett-Glaser <jason at x264.com> wrote:
>> On Wed, Feb 16, 2011 at 10:37 AM, Ronald S. Bultje <rsbultje at gmail.com> wrote:
>>> On Wed, Feb 16, 2011 at 1:33 PM, Jason Garrett-Glaser <jason at x264.com> wrote:
>>>> On Wed, Feb 16, 2011 at 10:30 AM, Ronald S. Bultje <rsbultje at gmail.com> wrote:
>>>>> On Wed, Feb 16, 2011 at 1:22 PM, Jason Garrett-Glaser <jason at x264.com> wrote:
>>>>>> ---
>>>>>> ?libavcodec/vp8.c ? ? | ?177 +------------------------------------
>>>>>> ?libavcodec/vp8.h ? ? | ?242 ++++++++++++++++++++++++++++++++++++++++++++++++++
>>>>>> ?libavcodec/vp8data.h | ? 44 ++--------
>>>>>> ?3 files changed, 249 insertions(+), 214 deletions(-)
>>>>>> ?create mode 100644 libavcodec/vp8.h
>>>>>
>>>>> Why? You want to use VP8Context or parts of it in x86/vp8*.[ch]?
>>>>
>>>> Well, I wanted to write an mmx version of clamp_mv as x264 has for
>>>> median, but now I realize that we don't have any inline mmx anywhere
>>>> in libavcodec, so this would be a new thing...
>>>
>>> Shall we delay pushing this patch until that's done?
>>
>> Well, this patch cleans things up anyways, for example the fact that
>> vp8data was missing include guards... I think this way fits more with
>> how h264.c/h works for example.
>
> I guess it doesn't hurt - so sure, go for it.
>
> Ronald
>

Applied.

Jason



More information about the ffmpeg-devel mailing list