summaryrefslogtreecommitdiffstats
path: root/libavcodec/ppc
Commit message (Collapse)AuthorAgeFilesLines
* lavc: typo fix uncliped -> unclippedGanesh Ajjanagadde2015-12-151-3/+3
| | | | | | Untested due to lack of ppc. Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
* avcodec/ppc/fdctdsp: use more accurate constantsGanesh Ajjanagadde2015-12-011-7/+7
| | | | | | | | Whoever wrote this stuff had a pretty bad libm - digits differ pretty quickly. Reviewed-by: Ronald S. Bultje <rsbultje@gmail.com> Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
* avcodec/ppc/fdctdsp: use M_SQRT2 instead of ad-hoc SQRT2Ganesh Ajjanagadde2015-11-221-11/+10
| | | | | | | | | | | | | This actually fixes an incorrect float literal. It is believed by examining the precision that the literals were all pre-computed as floats, resulting in this needless loss of precision. There is no benefit to keeping such reduced precision: 1. These constants are used for static array computation, hence compile-time. 2. They will be treated as doubles anyway, since f specifier was not present. Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
* blockdsp: reindent after parameter removalChristophe Gisquet2015-10-031-11/+11
| | | | Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
* blockdsp: remove high bitdepth parameterChristophe Gisquet2015-10-021-4/+1
| | | | | | | | | It is only (mis-)used to set the dsp fucntions clear_block(s). But these functions always work on 16bits-wide elements, which make the parameter useless and actually harmful, as it causes all content on more than 8-bits to not use accelerated functions. Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
* Merge commit '7c6eb0a1b7bf1aac7f033a7ec6d8cacc3b5c2615'Michael Niedermayer2015-07-271-1/+1
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * commit '7c6eb0a1b7bf1aac7f033a7ec6d8cacc3b5c2615': lavc: AV-prefix all codec flags Conflicts: doc/examples/muxing.c ffmpeg.c ffmpeg_opt.c ffplay.c libavcodec/aacdec.c libavcodec/aacenc.c libavcodec/ac3dec.c libavcodec/ac3enc_float.c libavcodec/atrac1.c libavcodec/atrac3.c libavcodec/atrac3plusdec.c libavcodec/dcadec.c libavcodec/ffv1enc.c libavcodec/h264.c libavcodec/h264_loopfilter.c libavcodec/h264_mb.c libavcodec/imc.c libavcodec/libmp3lame.c libavcodec/libtheoraenc.c libavcodec/libtwolame.c libavcodec/libvpxenc.c libavcodec/libxavs.c libavcodec/libxvid.c libavcodec/mpeg12dec.c libavcodec/mpeg12enc.c libavcodec/mpegaudiodec_template.c libavcodec/mpegvideo.c libavcodec/mpegvideo_enc.c libavcodec/mpegvideo_motion.c libavcodec/nellymoserdec.c libavcodec/nellymoserenc.c libavcodec/nvenc.c libavcodec/on2avc.c libavcodec/options_table.h libavcodec/opus_celt.c libavcodec/pngenc.c libavcodec/ra288.c libavcodec/ratecontrol.c libavcodec/twinvq.c libavcodec/vc1_block.c libavcodec/vc1_loopfilter.c libavcodec/vc1_mc.c libavcodec/vc1dec.c libavcodec/vorbisdec.c libavcodec/vp3.c libavcodec/wma.c libavcodec/wmaprodec.c libavcodec/x86/hpeldsp_init.c libavcodec/x86/me_cmp_init.c Merged-by: Michael Niedermayer <michael@niedermayer.cc>
* \ Merge commit 'd42191c78befc1983f23b1899b2dda513b72f1ed'Michael Niedermayer2015-07-171-1/+1
|\ \ | |/ | | | | | | | | | | | | | | | | | | | | * commit 'd42191c78befc1983f23b1899b2dda513b72f1ed': configure: Factor out vp8dsp module Conflicts: configure libavcodec/Makefile libavcodec/x86/Makefile Merged-by: Michael Niedermayer <michael@niedermayer.cc>
| * configure: Factor out vp8dsp moduleVittorio Giovara2015-07-171-1/+1
| |
* | doc: avoid incorrect phrase 'allows to'Andreas Cadhalpun2015-06-161-1/+1
| | | | | | | | | | | | | | | | Also fix typo found by Lou Logan: Sacrifying -> Sacrificing Reviewed-by: Lou Logan <lou@lrcd.com> Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
* | avcodec/ppc/pixblockdsp: Fix type of get_pixels_vsx()Michael Niedermayer2015-06-011-1/+1
| | | | | | | | Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* | Merge commit 'd0bf20a4f25ac5de021c860a0c8ad05638ee2078'Michael Niedermayer2015-05-311-0/+44
|\ \ | |/ | | | | | | | | | | * commit 'd0bf20a4f25ac5de021c860a0c8ad05638ee2078': ppc: vsx: Implement diff_pixels and get_pixels Merged-by: Michael Niedermayer <michaelni@gmx.at>
| * ppc: vsx: Implement diff_pixels and get_pixelsLuca Barbato2015-05-311-0/+44
| | | | | | | | Use a macro to abstract the endianness.
| * ppc: Restrict some Altivec implementations to Big EndianLuca Barbato2015-05-3123-45/+45
| | | | | | | | | | In Little Endian the vec_ld/vec_st operations work as expected only for byte-vectors.
* | Merge commit '7014b65995b1fe6188fb53447bf61b08e3963355'Michael Niedermayer2015-05-311-23/+20
|\ \ | |/ | | | | | | | | | | * commit '7014b65995b1fe6188fb53447bf61b08e3963355': ppc: pixblockdsp: Use the abriged vector types Merged-by: Michael Niedermayer <michaelni@gmx.at>
| * ppc: pixblockdsp: Use the abriged vector typesLuca Barbato2015-05-311-23/+20
| |
* | avcodec/ppc/h264dsp: POWER LE support in h264_idct_dc_add_internal() fix ↵Rong Yan2015-05-141-11/+15
| | | | | | | | | | | | | | | | | | | | | | | | | | vec_lvsl bug We got defective video when use GCC 4.9.2 instead of GCC 4.9.1 to compile FFMEPG. And further found that GCC 4.8 and 4.9 need patch to fix the lvsl/lvsr bug on POWER LE, and GCC 5.1 contains the correct code since its release. The message on gcc-patches requesting approval for lvsl/lvsr patch is at https://gcc.gnu.org/ml/gcc-patches/2014-10/msg00228.html. The fixed code avoids using lvsl and will not depends on GCC version, also it uses less instructions on POWER LE. Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* | avcodec/ppc/h264qpel_template: Fix alignment assertMichael Niedermayer2015-05-101-1/+1
| | | | | | | | Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* | avcodec/ppc/fdctdsp: POWER LE support in ff_fdct_altivec() delete macros ↵Rong Yan2015-04-271-47/+39
| | | | | | | | | | | | | | | | | | | | VEC_FMERGEH() VEC_FMERGEL(), they where wrong GCC tool had a bug of PPC intrinsic interpret, which has been fixed in GCC 4.9.1. This bug lead to errors in two of our previous patches. We found this when we update our GCC tools to 4.9.1 and by reading the related info on GCC website. We fix our previous error in two separate commits Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* | avcodec/ppc/fdctdsp: POWER LE support in ff_fdct_altivec()Rong Yan2015-04-031-39/+47
| | | | | | | | | | | | add marcos VEC_FMERGEH() VEC_FMERGEL() Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* | Revert "ppc: lavc: use LOCAL_ALIGNED instead of DECLARE_ALIGNED"Michael Niedermayer2015-03-153-31/+31
| | | | | | | | | | | | | | This broke build, note, LOCAL_ALIGNED_16() does not work either Found-by: James Almer <jamrial@gmail.com> This reverts commit 98cccdd91c9aab9e031753f6feb81be0c44a9c8e.
* | ppc: lavc: use LOCAL_ALIGNED instead of DECLARE_ALIGNEDChristophe Gisquet2015-03-143-31/+31
| | | | | | | | | | | | The later may yield incorrect code for on-stack variables. Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* | Merge commit '71f1ad37d858b810b71a4af1c25771beaa50b27b'Michael Niedermayer2015-03-011-2/+1
|\ \ | |/ | | | | | | | | | | | | | | | | | | | | * commit '71f1ad37d858b810b71a4af1c25771beaa50b27b': lavc: do not compile fmtconvert unconditionally Conflicts: configure libavcodec/ppc/Makefile libavcodec/x86/Makefile Merged-by: Michael Niedermayer <michaelni@gmx.at>
| * lavc: do not compile fmtconvert unconditionallyAnton Khirnov2015-02-281-2/+1
| | | | | | | | Only ac3dec and dcadec use it.
* | Merge commit 'd74a8cb7e42f703be5796eeb485f06af710ae8ca'Michael Niedermayer2015-02-281-111/+0
|\ \ | |/ | | | | | | | | | | | | | | | | | | | | * commit 'd74a8cb7e42f703be5796eeb485f06af710ae8ca': fmtconvert: drop unused functions Conflicts: libavcodec/arm/fmtconvert_vfp_armv6.S libavcodec/x86/fmtconvert.asm libavcodec/x86/fmtconvert_init.c Merged-by: Michael Niedermayer <michaelni@gmx.at>
| * fmtconvert: drop unused functionsAnton Khirnov2015-02-281-111/+0
| |
* | avcodec/ppc/fdctdsp: fix const warningMichael Niedermayer2015-02-281-1/+1
| | | | | | | | Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* | avcodec/ppc/idctdsp.c: POWER LE support in idct_add_altivec()Rong Yan2015-02-031-3/+13
| | | | | | | | | | | | also add GET_TMP2() macro Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* | avcodec/ppc/h264qpel_template: Change if DEBUG assert() to av_assert2()Michael Niedermayer2015-01-301-5/+2
| | | | | | | | Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* | ppc/mpegvideo_altivec: Change assert to av_assert2()Michael Niedermayer2015-01-301-2/+1
| | | | | | | | | | | | And move it to where the variable is used Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* | libavcodec/ppc/mpegvideoencdsp.c: fix stack smashing in pix_norm1_altivec() ↵Andreas Cadhalpun2015-01-211-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | and pix_sum_altivec() The vec_ste calls were mistakenly changed to vec_vsx_st in c5ca76a, which caused stack smashing. Changing them back fixes crashes on ppc64el, when configured with --toolchain=hardened. Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* | avcodec/ppc/vp3dsp_altivec: POWER LE support to vp3_idct_add_altivec()Rong Yan2014-12-081-2/+16
| | | | | | | | | | | | add GET_VDST16() macro Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* | avcodec/ppc/vp8dsp_altivec.c: POWER LE support put_vp8_epel_h_altivec_core() ↵Rong Yan2014-12-041-37/+49
| | | | | | | | | | | | put_vp8_epel_v_altivec_core() put_vp8_pixels16_altivec() add marcos GET_PIXHL() GET_OUTER() LOAD_HL() Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* | avcodec/ppc/vc1dsp_altivec: add POWER LE support to vc1_inv_trans_8x4_altivec()Rong Yan2014-12-041-3/+10
| | | | | | | | Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* | avcodec/ppc/lossless_audiodsp_altivec: POWER LE support for ↵Rong Yan2014-12-021-10/+21
| | | | | | | | | | | | | | | | scalarproduct_and_madd_int16_altivec() adds macro GET_T() Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* | avcodec/ppc/h264qpel_template: protect unistd.h by #if HAVE_UNISTD_HMichael Niedermayer2014-11-281-0/+4
| | | | | | | | Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* | avcodec/ppc/h264qpel: POWER LE support for put_pixels16_l2_altivec() and ↵Rong Yan2014-11-281-58/+51
| | | | | | | | | | | | | | | | avg_pixels16_l2_altivec() add put_unligned_store() and avg_unligned_store() macros Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* | avcodec/ppc/h264qpel_template: POWER LE support for ↵Rong Yan2014-11-281-199/+165
| | | | | | | | | | | | | | | | PREFIX_h264_qpel16_h_lowpass_altivec() PREFIX_h264_qpel16_v_lowpass_altivec() PREFIX_h264_qpel16_hv_lowpass_altivec() add load_alignment() macro Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* | avcodec/ppc/h264chroma_template: POWER LE support for ↵Rong Yan2014-11-281-135/+88
| | | | | | | | | | | | | | | | PREFIX_h264_chroma_mc8_altivec() PREFIX_no_rnd_vc1_chroma_mc8_altivec() CHROMA_MC8_ALTIVEC_CORE_SIMPLE() CHROMA_MC8_ALTIVEC_CORE() add marcos GET_VSRC1() GET_VSRC() Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* | avcodec/ppc/h264dsp: POWER LE support for h264_idct8_add_altivec() ↵Rong Yan2014-11-281-27/+62
| | | | | | | | | | | | | | | | h264_idct_dc_add_internal() h264_loop_filter_luma_altivec() write16x4() VEC_1D_DCT() weight_h264_W_altivec() biweight_h264_W_altivec() VEC_LOAD_U8_ADD_S16_STORE_U8() ALTIVEC_STORE_SUM_CLIP() add marcos GET_2PERM() dstv_load() vdst_load() dest_unligned_store() Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* | Merge commit '9c12c6ff9539e926df0b2a2299e915ae71872600'Michael Niedermayer2014-11-241-28/+28
|\ \ | |/ | | | | | | | | | | | | | | | | | | | | | | * commit '9c12c6ff9539e926df0b2a2299e915ae71872600': motion_est: convert stride to ptrdiff_t Conflicts: libavcodec/me_cmp.c libavcodec/ppc/me_cmp.c libavcodec/x86/me_cmp_init.c See: 9c669672c7fd45ef1cad782ab551be438ceac6cd Merged-by: Michael Niedermayer <michaelni@gmx.at>
| * motion_est: convert stride to ptrdiff_tVittorio Giovara2014-11-241-28/+28
| | | | | | | | | | CC: libav-stable@libav.org Bug-Id: CID 700556 / CID 700557 / CID 700558
* | libavcodec/ppc/me_cmp.c : support little endian in sse8_altivec(), ↵Rong Yan2014-11-111-37/+21
| | | | | | | | | | | | hadamard8_diff8x8_altivec() and hadamard8_diff16x8_altivec() Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* | libavcodec/ppc/me_cmp.c : factorize little and big endian codeRong Yan2014-11-111-351/+55
| | | | | | | | | | | | add marcos GET_PERM() LOAD_PIX() for POWER LE Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* | libavcodec/ppc/hpeldsp_altivec.c : fix ff_put_pixels16_altivec() ↵Rong Yan2014-11-071-184/+71
| | | | | | | | | | | | put_no_rnd_pixels16_xy2_altivec() put_no_rnd_pixels8_xy2_altivec() avg_pixels8_altivec() avg_pixels8_xy2_altivec() put_pixels16_xy2_altivec() put_pixels8_xy2_altivec() ff_avg_pixels16_altivec() for POWER LE Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* | libavcodec/ppc/hpeldsp_altivec.c : fix ff_put_pixels16_altivec() for POWER LERong Yan2014-10-101-0/+34
| | | | | | | | Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* | libavcodec/ppc/me_cmp.c : fix sad16_altivec() sse16_altivec() ↵Rong Yan2014-10-101-0/+294
| | | | | | | | | | | | sad16_xy2_altivec() sad16_x2_altivec() sad16_y2_altivec() sad8_altivec() for POWER LE Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* | libavcodec/ppc/mpegvideoencdsp.c : fix pix_norm1_altivec() and ↵Rong Yan2014-10-101-0/+60
| | | | | | | | | | | | pix_sum_altivec() for POWER LE Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* | libavcodec/ppc/pixblockdsp.c : fix get_pixels_altivec() and ↵Rong Yan2014-10-101-0/+95
| | | | | | | | | | | | diff_pixels_altivec() for POWER LE Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* | Add missing "const" all over the place.Reimar Döffinger2014-08-291-1/+1
| | | | | | | | | | | | Only "./configure --enable-gpl" on x86 was tested. Signed-off-by: Reimar Döffinger <Reimar.Doeffinger@gmx.de>
* | Merge commit 'efd26bedec9a345a5960dbfcbaec888418f2d4e6'Michael Niedermayer2014-08-151-0/+2
|\ \ | |/ | | | | | | | | | | | | | | | | | | * commit 'efd26bedec9a345a5960dbfcbaec888418f2d4e6': build: Add explanatory comments to (optimization) blocks in the Makefiles Conflicts: libavcodec/ppc/Makefile libavcodec/x86/Makefile Merged-by: Michael Niedermayer <michaelni@gmx.at>
OpenPOWER on IntegriCloud