diff options
Diffstat (limited to 'libavcodec/ppc/mpegvideoencdsp.c')
| -rw-r--r-- | libavcodec/ppc/mpegvideoencdsp.c | 72 |
1 files changed, 66 insertions, 6 deletions
diff --git a/libavcodec/ppc/mpegvideoencdsp.c b/libavcodec/ppc/mpegvideoencdsp.c index 7354816208..3e6765ce15 100644 --- a/libavcodec/ppc/mpegvideoencdsp.c +++ b/libavcodec/ppc/mpegvideoencdsp.c @@ -1,18 +1,18 @@ /* - * This file is part of Libav. + * This file is part of FFmpeg. * - * Libav is free software; you can redistribute it and/or + * FFmpeg is free software; you can redistribute it and/or * modify it under the terms of the GNU Lesser General Public * License as published by the Free Software Foundation; either * version 2.1 of the License, or (at your option) any later version. * - * Libav is distributed in the hope that it will be useful, + * FFmpeg is distributed in the hope that it will be useful, * but WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU * Lesser General Public License for more details. * * You should have received a copy of the GNU Lesser General Public - * License along with Libav; if not, write to the Free Software + * License along with FFmpeg; if not, write to the Free Software * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA */ @@ -29,8 +29,36 @@ #include "libavutil/ppc/util_altivec.h" #include "libavcodec/mpegvideoencdsp.h" -#if HAVE_ALTIVEC && HAVE_BIGENDIAN +#if HAVE_ALTIVEC +#if HAVE_VSX +static int pix_norm1_altivec(uint8_t *pix, int line_size) +{ + int i, s = 0; + const vector unsigned int zero = + (const vector unsigned int) vec_splat_u32(0); + vector unsigned int sv = (vector unsigned int) vec_splat_u32(0); + vector signed int sum; + + for (i = 0; i < 16; i++) { + /* Read the potentially unaligned pixels. */ + //vector unsigned char pixl = vec_ld(0, pix); + //vector unsigned char pixr = vec_ld(15, pix); + //vector unsigned char pixv = vec_perm(pixl, pixr, perm); + vector unsigned char pixv = vec_vsx_ld(0, pix); + + /* Square the values, and add them to our sum. */ + sv = vec_msum(pixv, pixv, sv); + + pix += line_size; + } + /* Sum up the four partial sums, and put the result into s. */ + sum = vec_sums((vector signed int) sv, (vector signed int) zero); + sum = vec_splat(sum, 3); + vec_ste(sum, 0, &s); + return s; +} +#else static int pix_norm1_altivec(uint8_t *pix, int line_size) { int i, s = 0; @@ -58,7 +86,37 @@ static int pix_norm1_altivec(uint8_t *pix, int line_size) return s; } +#endif /* HAVE_VSX */ + +#if HAVE_VSX +static int pix_sum_altivec(uint8_t *pix, int line_size) +{ + int i, s; + const vector unsigned int zero = + (const vector unsigned int) vec_splat_u32(0); + vector unsigned int sad = (vector unsigned int) vec_splat_u32(0); + vector signed int sumdiffs; + + for (i = 0; i < 16; i++) { + /* Read the potentially unaligned 16 pixels into t1. */ + //vector unsigned char pixl = vec_ld(0, pix); + //vector unsigned char pixr = vec_ld(15, pix); + //vector unsigned char t1 = vec_perm(pixl, pixr, perm); + vector unsigned char t1 = vec_vsx_ld(0, pix); + /* Add each 4 pixel group together and put 4 results into sad. */ + sad = vec_sum4s(t1, sad); + + pix += line_size; + } + + /* Sum up the four partial sums, and put the result into s. */ + sumdiffs = vec_sums((vector signed int) sad, (vector signed int) zero); + sumdiffs = vec_splat(sumdiffs, 3); + vec_ste(sumdiffs, 0, &s); + return s; +} +#else static int pix_sum_altivec(uint8_t *pix, int line_size) { int i, s; @@ -88,12 +146,14 @@ static int pix_sum_altivec(uint8_t *pix, int line_size) return s; } +#endif /* HAVE_VSX */ + #endif /* HAVE_ALTIVEC */ av_cold void ff_mpegvideoencdsp_init_ppc(MpegvideoEncDSPContext *c, AVCodecContext *avctx) { -#if HAVE_ALTIVEC && HAVE_BIGENDIAN +#if HAVE_ALTIVEC if (!PPC_ALTIVEC(av_get_cpu_flags())) return; |