Re: [RFC PATCH v1 0/3] Introducing ARM SIMD Support
Lucas Clemente Vella <[email protected]> Wed, 1 Oct 2014 13:03:07 -0300
| Newsgroups | gmane.comp.multimedia.ogg.vorbis.devel |
|---|---|
| Message-ID | <CAGCathwKhi15b1YXZM3S1LNEB8EZU=xbagLCYMmGD9xgoJpz3g@mail.gmail.com> |
This list is almost dead. I am not sure the vorbis base libs maintainers are watching this (suposing they are maintained right now), but I find it very nice to see ARM SIMD support. 2014-10-01 12:05 GMT-03:00 Viswanath Puttagunta < [email protected]>: > Gentlemen! > > Am I at the right place? I posted these patches almost a month ago to " > [email protected]" > > I haven't seen any response/reviews about this patch. Neither do I see any > other emails from this mailing list. > > I can see that they got posted > http://lists.xiph.org/pipermail/vorbis-dev/2014-September/thread.html > > Please advise. > > Regards, > Vish (Viswanath Puttagunta) > Cell: 972-342-0205 > Technical Program Manager > Member Services, Linaro > > On 23 September 2014 15:50, Viswanath Puttagunta < > [email protected]> wrote: > >> Hello Vorbis Developers / Monty, >> >> Am I at the right place? I posted these patches about 2 weeks ago to " >> [email protected]" >> >> I haven't seen any response/reviews about this patch. Neither do I see >> any other emails from this mailing list. >> >> Please advise. >> >> Regards, >> Vish (Viswanath Puttagunta) >> Cell: 972-342-0205 >> Technical Program Manager >> Member Services, Linaro >> >> On 10 September 2014 14:15, Viswanath Puttagunta < >> [email protected]> wrote: >> >>> libvorbis does not currently have any simd/vectorization. >>> >>> Following patches add generic framework for simd/vectorization >>> and on top, add ARM-NEON simd vectorization using intrinsics. >>> >>> I was able to get over 34% performance improvement on my >>> Beaglebone Black which is single Cortex-A8 based CPU. >>> >>> You can find more information on metrics and procedure I used >>> to measure at >>> >>> https://wiki.linaro.org/WorkingGroups/Middleware/Graphics/MediaLibs/libvorbis >>> >>> As described in above link/doc, you can see my work-in-progress >>> tree at >>> https://git.linaro.org/people/viswanath.puttagunta/vorbis.git >>> >>> Also what standard should one follow when submitting patches >>> to libvorbis? I'm pretty sure linux kernel checkpatch.pl will >>> fail all patches for libvorbis. Please advise. >>> >>> Viswanath Puttagunta (3): >>> lib/simd: Introduce vectorization framework for libvorbis >>> lib/block.c: Use optimized routine for wave_operation >>> mdct: implement arm simd implementation for mdct >>> >>> configure.ac | 5 ++ >>> lib/Makefile.am | 6 +- >>> lib/block.c | 14 ++--- >>> lib/mdct.c | 102 +------------------------------- >>> lib/simd/Makefile.am | 9 +++ >>> lib/simd/neon_simd.c | 158 >>> ++++++++++++++++++++++++++++++++++++++++++++++++++ >>> lib/simd/no_simd.c | 129 +++++++++++++++++++++++++++++++++++++++++ >>> lib/simd/simd.h | 40 +++++++++++++ >>> 8 files changed, 351 insertions(+), 112 deletions(-) >>> create mode 100644 lib/simd/Makefile.am >>> create mode 100644 lib/simd/neon_simd.c >>> create mode 100644 lib/simd/no_simd.c >>> create mode 100644 lib/simd/simd.h >>> >>> -- >>> 1.7.9.5 >>> >>> >> > > _______________________________________________ > Vorbis-dev mailing list > [email protected] > http://lists.xiph.org/mailman/listinfo/vorbis-dev > > -- Lucas Clemente Vella [email protected] _______________________________________________ Vorbis-dev mailing list [email protected] http://lists.xiph.org/mailman/listinfo/vorbis-dev