On Thu, Jun 23, 2011 at 3:30 AM, Christian Robottom Reis
<k...@linaro.org> wrote:
> Hi there,
>
>    I took a look at the AOSP libjpeg code which is included in
>
>         git://android.git.kernel.org/platform/external/jpeg
>
> during my flight back home (which incidentally had been diverted and
> landed me in Rio de Janeiro; not sure if I celebrate or cry) and noted
> the following things:
>
>    - There is a v6 implementation of the fast IDCT algorithm which
>      lives in armv6_idct.S.
>
>    - The commit which adds this implementation was added October 2010,
>      and there haven't been any changes since.
>
>    - The code that selects the decoder IDCT implementation in
>      jddctmgr.c always uses that implementation if ANDROID_ARMV6_IDCT
>      is defined.
>
>    - Google have an "ashmem" backing store implementation, and have
>      code to enable tile-based mode. It's a fairly non-intrusive change
>      to use ashmem since it just replaces jpeg_open_backing_store.
>
>    - The code is pretty much standard libjpeg without any structural
>      changes to it.
>
>    - There isn't any NEON code in this branch.
>
>    - Mans has an optimized version here:
>
>        http://git.mansr.com/?p=libjpeg;a=summary
>
>      I don't know if he's submitted this to AOSP or not.
>
> This suggests to me that a simple drop-in of libjpeg-turbo might be
> actually easy to do, and that there is probably a significant
> performance benefit to be achieved. One thing to keep in mind is that
> this code still supports armv6, so we'd probably want to preserve that.

sounds good. Seems like a great starter project for MM WG to get their
hands dirty on android.

They could have three builds:

 1. plain LEB
 2. LEB with mans optimized libjpeg
 3. LEB with libjpeg-turbo

Then benchmark all three and keep AOSP community posted on our intent,
progress and results.

-- 

 - Alexander

_______________________________________________
linaro-dev mailing list
linaro-dev@lists.linaro.org
http://lists.linaro.org/mailman/listinfo/linaro-dev

Reply via email to