options: set minimum for "threads" to zero

[ffmpeg] / libavcodec / ppc / vp8dsp_altivec.c

diff --git a/libavcodec/ppc/vp8dsp_altivec.c b/libavcodec/ppc/vp8dsp_altivec.c
index b294154c6c90349069c642e24ac92fe1a2a855a2..06874b8b878d50f911633aab80e5e366458855f1 100644 (file)
--- a/libavcodec/ppc/vp8dsp_altivec.c
+++ b/libavcodec/ppc/vp8dsp_altivec.c
@@ -3,23 +3,24 @@
  *
  * Copyright (C) 2010 David Conrad
  *
- * This file is part of FFmpeg.
+ * This file is part of Libav.
  *
- * FFmpeg is free software; you can redistribute it and/or
+ * Libav 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.
  *
- * FFmpeg is distributed in the hope that it will be useful,
+ * Libav 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 FFmpeg; if not, write to the Free Software
+ * License along with Libav; if not, write to the Free Software
  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
  */
 
+#include "libavutil/cpu.h"
 #include "libavcodec/vp8dsp.h"
 #include "dsputil_altivec.h"
 #include "types_altivec.h"
@@ -55,8 +56,8 @@ static const vec_s8 h_subpel_filters_outer[3] =
     vec_s8 filter_outerl = vec_sld(filter_outerh, filter_outerh, 2)
 
 #define FILTER_H(dstv, off) \
-    a = vec_ld((off)-2,    src); \
-    b = vec_ld((off)-2+15, src); \
+    a = vec_ld((off)-is6tap-1,    src); \
+    b = vec_ld((off)-is6tap-1+15, src); \
 \
     pixh  = vec_perm(a, b, permh##off); \
     pixl  = vec_perm(a, b, perml##off); \
@@ -85,13 +86,15 @@ void put_vp8_epel_h_altivec_core(uint8_t *dst, int dst_stride,
     vec_s16 f16h, f16l;
     vec_s32 filth, filtl;
 
-    vec_u8 perm_inner = { 1,2,3,4, 2,3,4,5, 3,4,5,6, 4,5,6,7 };
+    vec_u8 perm_inner6 = { 1,2,3,4, 2,3,4,5, 3,4,5,6, 4,5,6,7 };
+    vec_u8 perm_inner4 = { 0,1,2,3, 1,2,3,4, 2,3,4,5, 3,4,5,6 };
+    vec_u8 perm_inner  = is6tap ? perm_inner6 : perm_inner4;
     vec_u8 perm_outer = { 4,9, 0,5, 5,10, 1,6, 6,11, 2,7, 7,12, 3,8 };
     vec_s32 c64 = vec_sl(vec_splat_s32(1), vec_splat_u32(6));
     vec_u16 c7  = vec_splat_u16(7);
 
-    align_vec0 = vec_lvsl( -2, src);
-    align_vec8 = vec_lvsl(8-2, src);
+    align_vec0 = vec_lvsl( -is6tap-1, src);
+    align_vec8 = vec_lvsl(8-is6tap-1, src);
 
     permh0     = vec_perm(align_vec0, align_vec0, perm_inner);
     permh8     = vec_perm(align_vec8, align_vec8, perm_inner);
@@ -238,8 +241,13 @@ void put_vp8_epel ## WIDTH ## _v ## TAPS ## _altivec(uint8_t *dst, int dst_strid
 static void put_vp8_epel ## WIDTH ## _h ## HTAPS ## v ## VTAPS ## _altivec(uint8_t *dst, int stride, uint8_t *src, int s, int h, int mx, int my) \
 { \
     DECLARE_ALIGNED(16, uint8_t, tmp)[(2*WIDTH+5)*16]; \
-    put_vp8_epel ## WIDTH ## _h ## HTAPS ## _altivec(tmp, 16,     src-2*stride, stride, h+5, mx, my); \
-    put_vp8_epel ## WIDTH ## _v ## VTAPS ## _altivec(dst, stride, tmp+2*16,     16,     h,   mx, my); \
+    if (VTAPS == 6) { \
+        put_vp8_epel ## WIDTH ## _h ## HTAPS ## _altivec(tmp, 16,     src-2*stride, stride, h+5, mx, my); \
+        put_vp8_epel ## WIDTH ## _v ## VTAPS ## _altivec(dst, stride, tmp+2*16,     16,     h,   mx, my); \
+    } else { \
+        put_vp8_epel ## WIDTH ## _h ## HTAPS ## _altivec(tmp, 16,     src-stride, stride, h+4, mx, my); \
+        put_vp8_epel ## WIDTH ## _v ## VTAPS ## _altivec(dst, stride, tmp+16,     16,     h,   mx, my); \
+    } \
 }
 
 EPEL_FUNCS(16,6)
@@ -265,7 +273,7 @@ static void put_vp8_pixels16_altivec(uint8_t *dst, int stride, uint8_t *src, int
 
 av_cold void ff_vp8dsp_init_altivec(VP8DSPContext *c)
 {
-    if (!(mm_support() & AV_CPU_FLAG_ALTIVEC))
+    if (!(av_get_cpu_flags() & AV_CPU_FLAG_ALTIVEC))
         return;
 
     c->put_vp8_epel_pixels_tab[0][0][0] = put_vp8_pixels16_altivec;