Commit Graph
47 Commits
Author SHA1 Message Date
Kaustubh RasteandMichael Niedermayer a211626978 avcodec/mips: Improve hevc uni vt and hv mc msa functions
Remove unused macro.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-11-01 02:10:09 +01:00
Kaustubh RasteandMichael Niedermayer e39c0f29a1 avcodec/mips: Improve hevc bi hz and hv mc msa functions
Align the mask buffer.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-11-01 02:10:09 +01:00
Kaustubh RasteandMichael Niedermayer 6854dd7039 avcodec/mips: Improve hevc bi weighted copy, hz and vt mc msa functions
Pack the data to half word before clipping.
Use immediate unsigned saturation for clip to max saving one vector register.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-11-01 02:10:09 +01:00
Kaustubh RasteandMichael Niedermayer 93218c2234 avcodec/mips: Improve avc chroma avg hv mc msa functions
Replace generic with block size specific function.
Load the specific destination bytes instead of MSA load and pack.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-10-30 21:32:26 +01:00
Kaustubh RasteandMichael Niedermayer 1181d93231 avcodec/mips: Improve avc avg mc 22, 11, 31, 13 and 33 msa functions
Remove loops and unroll as block sizes are known.
Load the specific destination bytes instead of MSA load and pack.
Remove unused macro and functions.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-10-30 21:32:26 +01:00
Kaustubh RasteandMichael Niedermayer 736a48901f avcodec/mips: Improve hevc bi weighted hv mc msa functions
Use immediate unsigned saturation for clip to max saving one vector register.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-10-25 21:50:37 +02:00
Kaustubh RasteandMichael Niedermayer ce0a52e9e9 avcodec/mips: Improve avc chroma copy and avg vert mc msa functions
Replace generic with block size specific function.
Load the specific destination bytes instead of MSA load and pack.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-10-25 21:50:37 +02:00
Kaustubh RasteandMichael Niedermayer 2aab7c2dfa avcodec/mips: Improve avc put mc 11, 31, 13 and 33 msa functions
Remove loops and unroll as block sizes are known.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-10-25 21:50:37 +02:00
Kaustubh RasteandMichael Niedermayer e5439e272e avcodec/mips: Improve hevc uni weighted vert mc msa functions
Pack the data to half word before clipping.
Use immediate unsigned saturation for clip to max saving one vector register.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-10-13 02:46:31 +02:00
Kaustubh RasteandMichael Niedermayer 6ca821a3e7 avcodec/mips: Improve hevc uni horiz mc msa functions
Update macros to remove adds.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-10-13 02:46:31 +02:00
Kaustubh RasteandMichael Niedermayer e63758468c avcodec/mips: Improve hevc bi copy mc msa functions
Load the specific destination bytes instead of MSA load and pack.
Use immediate unsigned saturation for clip to max saving one vector register.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-10-13 02:46:31 +02:00
Kaustubh RasteandMichael Niedermayer e549933a27 avcodec/mips: Improve avc put mc 12, 32 and 22 msa functions
Remove loops and unroll as block sizes are known.
Removed unused functions.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-10-13 02:46:31 +02:00
Kaustubh RasteandMichael Niedermayer 27a0a83880 avcodec/mips: Improve avc chroma avg horiz mc msa functions
Replace generic with block size specific function.
Load the specific destination bytes instead of MSA load and pack.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-10-13 02:46:31 +02:00
Kaustubh RasteandMichael Niedermayer ff53f4dc2d avcodec/mips: Improve avc uni copy mc msa functions
Load the specific bytes instead of MSA load.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-10-10 23:58:41 +02:00
Kaustubh RasteandMichael Niedermayer eadb911643 avcodec/mips: Improve hevc uni-w horiz mc msa functions
Load the specific destination bytes instead of MSA load and pack.
Pack the data to half word before clipping.
Use immediate unsigned saturation for clip to max saving one vector register.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-10-10 23:58:41 +02:00
Kaustubh RasteandMichael Niedermayer 662234a9a2 avcodec/mips: Improve avc put mc 21, 23 and 02 msa functions
Remove loops and unroll as block sizes are known.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-10-10 23:58:41 +02:00
Kaustubh RasteandMichael Niedermayer b59323cb72 avcodec/mips: Improve avc chroma hv mc msa functions
Replace generic with block size specific function.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-10-10 23:58:41 +02:00
Kaustubh RasteandMichael Niedermayer af9433b1d6 avcodec/mips: Improve avc bi-weighted mc msa functions
Replace generic with block size specific function.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-10-10 23:58:41 +02:00
Kaustubh RasteandMichael Niedermayer 56822b074b avcodec/mips: preload data in hevc sao edge 135 degree filter msa functions
Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-10-10 23:58:41 +02:00
Kaustubh RasteandMichael Niedermayer 51ebce7d7d avcodec/mips: Cleanup unused functions
Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-10-06 02:42:49 +02:00
Kaustubh RasteandMichael Niedermayer 6796a1dd8c avcodec/mips: Improve avc put mc 20, 01 and 03 msa functions
Remove loops and unroll as block sizes are known.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-09-27 21:15:57 +02:00
Kaustubh RasteandMichael Niedermayer b8854e2439 avcodec/mips: Improve avc chroma vert mc msa functions
Replace generic with block size specific function.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-09-27 21:15:57 +02:00
Kaustubh RasteandMichael Niedermayer 10ab5534e0 avcodec/mips: Improve avc weighted mc msa functions
Replace generic with block size specific function.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-09-27 21:15:57 +02:00
Kaustubh RasteandMichael Niedermayer ed1586b921 avcodec/mips: Removed generic function call in avc intra msa functions
Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-09-27 21:15:57 +02:00
Kaustubh RasteandMichael Niedermayer deeaaba1ab avcodec/mips: preload data in hevc sao edge 45 degree filter msa functions
Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-09-27 21:15:57 +02:00
Kaustubh RasteandMichael Niedermayer 16adbfe60c avcodec/mips: Improve avc chroma horiz mc msa functions
Replace generic with block size specific function.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-09-24 02:33:48 +02:00
Kaustubh RasteandMichael Niedermayer d6737539e7 avcodec/mips: Unrolled loops avc intra msa functions
Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-09-24 02:33:48 +02:00
Kaustubh RasteandMichael Niedermayer 7f8417f226 avcodec/mips: Improve hevc uni-w copy mc msa functions
Load the specific destination bytes instead of MSA load and pack.
Pack the data to half word before clipping.
Use immediate unsigned saturation for clip to max saving one vector register.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-09-24 02:33:48 +02:00
Kaustubh RasteandMichael Niedermayer f160a63bad avcodec/mips: Remove generic func use in hevc non-uni copy mc msa functions
Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-09-22 12:12:59 +02:00
Kaustubh RasteandMichael Niedermayer 2b15626997 avcodec/mips: preload data in hevc sao edge 90 degree filter msa functions
Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-09-22 12:12:59 +02:00
Kaustubh RasteandMichael Niedermayer bba9c1c6bb avcodec/mips: Reduced conditional cases in avc inter lpf msa functions
Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-09-21 03:02:24 +02:00
Kaustubh RasteandMichael Niedermayer b5da07d434 avcodec/mips: Unrolled loops and expanded functions in avc put mc 10 & 30 msa functions
Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-09-21 03:02:24 +02:00
Kaustubh RasteandMichael Niedermayer e428e5ded6 avcodec/mips: preload data in hevc sao edge 0 degree filter msa functions
Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-09-20 02:56:20 +02:00
Kaustubh RasteandMichael Niedermayer f4ba85dc82 avcodec/mips: Fixed rnd_val variable to 6 in hevc uni mc msa functions
Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-09-20 02:56:20 +02:00
Kaustubh RasteandMichael Niedermayer 1a85fb7e1e avcodec/mips: Improve hevc sao band filter msa functions
Preload data in band filter 0-8 for better pipeline parallelization.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-09-15 22:36:42 +02:00
Kaustubh RasteandMichael Niedermayer 0105ed551c avcodec/mips: Improve avc mc copy msa functions
Remove loops and unroll as block sizes are known.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-09-15 22:19:07 +02:00
Kaustubh RasteandMichael Niedermayer e5a650e141 avcodec/mips: Improve avc lpf msa functions
Optimize luma intra case by reducing conditional cases.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-09-15 22:19:07 +02:00
Kaustubh RasteandMichael Niedermayer c6314cd750 avcodec/mips: Improve hevc idct msa functions
Align the buffers. Remove reduandant constant array.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-09-15 01:47:14 +02:00
Kaustubh RasteandMichael Niedermayer f692e55aab avcodec/mips: Improve hevc lpf msa functions
Seperate the filter processing in all strong, all weak and strong + weak cases.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-09-15 01:47:14 +02:00
Kaustubh RasteandMichael Niedermayer 9b2c3c406f avcodec/mips: Improve vp9 mc msa functions
Load the specific destination bytes instead of MSA load and pack.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-09-08 13:48:40 +02:00
Kaustubh RasteandMichael Niedermayer c75b23cbea avcodec/mips: Improve vp9 idct msa functions
Removed memset calls.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-09-08 13:48:40 +02:00
Kaustubh RasteandMichael Niedermayer 2e79813a8e avcodec/mips: Improve vp9 lpf msa functions
Updated VP9_LPF_FILTER4_4W macro to process on 8 bit data.
Replaced VP9_LPF_FILTER4_8W with VP9_LPF_FILTER4_4W.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-09-08 13:48:40 +02:00
Kaustubh RasteandMichael Niedermayer fa805df060 libavcodec/mips: Improve avc idct8 msa function
Replace memset call with msa stores.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-09-05 22:16:03 +02:00
Kaustubh RasteandMichael Niedermayer 36ea41de37 libavcodec/mips: Improve avc dequant-idct luma dc msa function
Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-07-29 03:47:50 +02:00
Kaustubh RasteandMichael Niedermayer a776cb2074 libavcodec/mips: Optimize avc idct 4x4 for msa
Removed memset call and improved performance.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-07-25 22:04:34 +02:00
Kaustubh RasteandMichael Niedermayer ef1b4bdf44 libavutil/mips: Updated msa generic macros
Reduced msa load-store code.
Removed inline asm of GP load-store for 64 bit.
Updated variable names in GP load-store macros for naming consistency.
Corrected macro descriptions.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-07-21 17:37:05 +02:00
Kaustubh RasteandMichael Niedermayer df806605f7 avcodec: Add prefetch for mips
Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>
Reviewed-by: Manojkumar Bhosale <Manojkumar.Bhosale@imgtec.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2017-07-14 02:30:37 +02:00