From 9ec626726d1cc30db54390ee81f52cfe53ebeedf Mon Sep 17 00:00:00 2001 From: Rong Yan Date: Fri, 7 Nov 2014 09:09:09 +0000 Subject: [PATCH 4/4] libswscale/ppc/swscale_altivec.c : fix hScale_altivec_real() yuv2planeX_16_altivec() yuv2planeX_8() for POWER LE --- libswscale/ppc/swscale_altivec.c | 142 ++++++++++++++------------------------- 1 file changed, 50 insertions(+), 92 deletions(-) diff --git a/libswscale/ppc/swscale_altivec.c b/libswscale/ppc/swscale_altivec.c index 86f40ab..1673b8b 100644 --- a/libswscale/ppc/swscale_altivec.c +++ b/libswscale/ppc/swscale_altivec.c @@ -29,20 +29,30 @@ #include "libavutil/attributes.h" #include "libavutil/cpu.h" #include "yuv2rgb_altivec.h" +#include "libavutil/ppc/util_altivec.h" #if HAVE_ALTIVEC #define vzero vec_splat_s32(0) -#define yuv2planeX_8(d1, d2, l1, src, x, perm, filter) do { \ - vector signed short l2 = vec_ld(((x) << 1) + 16, src); \ - vector signed short ls = vec_perm(l1, l2, perm); \ - vector signed int i1 = vec_mule(filter, ls); \ - vector signed int i2 = vec_mulo(filter, ls); \ - vector signed int vf1 = vec_mergeh(i1, i2); \ - vector signed int vf2 = vec_mergel(i1, i2); \ - d1 = vec_add(d1, vf1); \ - d2 = vec_add(d2, vf2); \ - l1 = l2; \ +#if HAVE_BIGENDIAN +#define GET_VF(a, b) {\ + a = vec_mergeh(i1, i2);\ + b = vec_mergel(i1, i2);\ + } +#else +#define GET_VF(a, b) {\ + a = vec_mergel(i2, i1);\ + b = vec_mergeh(i2, i1);\ + } +#endif + +#define yuv2planeX_8(d1, d2, l1, src, x, filter) do {\ + vector signed int i1 = vec_mule(filter, l1);\ + vector signed int i2 = vec_mulo(filter, l1);\ + vector signed int vf1, vf2;\ + GET_VF(vf1, vf2);\ + d1 = vec_add(d1, vf1);\ + d2 = vec_add(d2, vf2);\ } while (0) static void yuv2planeX_16_altivec(const int16_t *filter, int filterSize, @@ -66,16 +76,13 @@ static void yuv2planeX_16_altivec(const int16_t *filter, int filterSize, vo4 = vec_ld(48, val); for (j = 0; j < filterSize; j++) { - vector signed short l1, vLumFilter = vec_ld(j << 1, filter); - vector unsigned char perm, perm0 = vec_lvsl(j << 1, filter); - vLumFilter = vec_perm(vLumFilter, vLumFilter, perm0); + vector signed short l1, l2; + vector signed short vLumFilter = (vector signed short)unaligned_load(j << 1, filter); vLumFilter = vec_splat(vLumFilter, 0); // lumFilter[j] is loaded 8 times in vLumFilter - - perm = vec_lvsl(x << 1, src[j]); - l1 = vec_ld(x << 1, src[j]); - - yuv2planeX_8(vo1, vo2, l1, src[j], x, perm, vLumFilter); - yuv2planeX_8(vo3, vo4, l1, src[j], x + 8, perm, vLumFilter); + l1 = (vector signed short)unaligned_load(x << 1, src[j]); + l2 = (vector signed short)unaligned_load(((x) << 1) + 16, src[j]); + yuv2planeX_8(vo1, vo2, l1, src[j], x, vLumFilter); + yuv2planeX_8(vo3, vo4, l2, src[j], x + 8, vLumFilter); } vo1 = vec_sra(vo1, altivec_vectorShiftInt19); @@ -85,7 +92,7 @@ static void yuv2planeX_16_altivec(const int16_t *filter, int filterSize, vs1 = vec_packsu(vo1, vo2); vs2 = vec_packsu(vo3, vo4); vf = vec_packsu(vs1, vs2); - vec_st(vf, 0, dest); + VEC_ST(vf, 0, dest); } static inline void yuv2planeX_u(const int16_t *filter, int filterSize, @@ -140,31 +147,15 @@ static void hScale_altivec_real(SwsContext *c, int16_t *dst, int dstW, for (i = 0; i < dstW; i++) { register int srcPos = filterPos[i]; - vector unsigned char src_v0 = vec_ld(srcPos, src); - vector unsigned char src_v1, src_vF; + vector unsigned char src_vF = unaligned_load(srcPos, src); vector signed short src_v, filter_v; vector signed int val_vEven, val_s; - if ((((uintptr_t)src + srcPos) % 16) > 12) { - src_v1 = vec_ld(srcPos + 16, src); - } - src_vF = vec_perm(src_v0, src_v1, vec_lvsl(srcPos, src)); - src_v = // vec_unpackh sign-extends... - (vector signed short)(vec_mergeh((vector unsigned char)vzero, src_vF)); + (vector signed short)(VEC_MERGEH((vector unsigned char)vzero, src_vF)); // now put our elements in the even slots src_v = vec_mergeh(src_v, (vector signed short)vzero); - - filter_v = vec_ld(i << 3, filter); - // The 3 above is 2 (filterSize == 4) + 1 (sizeof(short) == 2). - - // The neat trick: We only care for half the elements, - // high or low depending on (i<<3)%16 (it's 0 or 8 here), - // and we're going to use vec_mule, so we choose - // carefully how to "unpack" the elements into the even slots. - if ((i << 3) % 16) - filter_v = vec_mergel(filter_v, (vector signed short)vzero); - else - filter_v = vec_mergeh(filter_v, (vector signed short)vzero); + filter_v = (vector signed short)unaligned_load(i << 3, filter); + filter_v = vec_mergeh(filter_v, (vector signed short)vzero); val_vEven = vec_mule(src_v, filter_v); val_s = vec_sums(val_vEven, vzero); @@ -177,19 +168,12 @@ static void hScale_altivec_real(SwsContext *c, int16_t *dst, int dstW, for (i = 0; i < dstW; i++) { register int srcPos = filterPos[i]; - vector unsigned char src_v0 = vec_ld(srcPos, src); - vector unsigned char src_v1, src_vF; + vector unsigned char src_vF = unaligned_load(srcPos, src); vector signed short src_v, filter_v; vector signed int val_v, val_s; - if ((((uintptr_t)src + srcPos) % 16) > 8) { - src_v1 = vec_ld(srcPos + 16, src); - } - src_vF = vec_perm(src_v0, src_v1, vec_lvsl(srcPos, src)); - src_v = // vec_unpackh sign-extends... - (vector signed short)(vec_mergeh((vector unsigned char)vzero, src_vF)); - filter_v = vec_ld(i << 4, filter); - // the 4 above is 3 (filterSize == 8) + 1 (sizeof(short) == 2) + (vector signed short)(VEC_MERGEH((vector unsigned char)vzero, src_vF)); + filter_v = (vector signed short)unaligned_load(i << 4, filter); val_v = vec_msums(src_v, filter_v, (vector signed int)vzero); val_s = vec_sums(val_v, vzero); @@ -202,25 +186,21 @@ static void hScale_altivec_real(SwsContext *c, int16_t *dst, int dstW, for (i = 0; i < dstW; i++) { register int srcPos = filterPos[i]; - vector unsigned char src_v0 = vec_ld(srcPos, src); - vector unsigned char src_v1 = vec_ld(srcPos + 16, src); - vector unsigned char src_vF = vec_perm(src_v0, src_v1, vec_lvsl(srcPos, src)); - + vector unsigned char src_vF = unaligned_load(srcPos, src); vector signed short src_vA = // vec_unpackh sign-extends... - (vector signed short)(vec_mergeh((vector unsigned char)vzero, src_vF)); + (vector signed short)(VEC_MERGEH((vector unsigned char)vzero, src_vF)); vector signed short src_vB = // vec_unpackh sign-extends... - (vector signed short)(vec_mergel((vector unsigned char)vzero, src_vF)); + (vector signed short)(VEC_MERGEL((vector unsigned char)vzero, src_vF)); - vector signed short filter_v0 = vec_ld(i << 5, filter); - vector signed short filter_v1 = vec_ld((i << 5) + 16, filter); - // the 5 above are 4 (filterSize == 16) + 1 (sizeof(short) == 2) + vector signed short filter_v0 = (vector signed short)unaligned_load(i << 5, filter); + vector signed short filter_v1 = (vector signed short)unaligned_load((i << 5) + 16, filter); vector signed int val_acc = vec_msums(src_vA, filter_v0, (vector signed int)vzero); vector signed int val_v = vec_msums(src_vB, filter_v1, val_acc); vector signed int val_s = vec_sums(val_v, vzero); - vec_st(val_s, 0, tempo); + VEC_ST(val_s, 0, tempo); dst[i] = FFMIN(tempo[3] >> 7, (1 << 15) - 1); } break; @@ -231,56 +211,34 @@ static void hScale_altivec_real(SwsContext *c, int16_t *dst, int dstW, register int srcPos = filterPos[i]; vector signed int val_s, val_v = (vector signed int)vzero; - vector signed short filter_v0R = vec_ld(i * 2 * filterSize, filter); - vector unsigned char permF = vec_lvsl((i * 2 * filterSize), filter); - - vector unsigned char src_v0 = vec_ld(srcPos, src); - vector unsigned char permS = vec_lvsl(srcPos, src); for (j = 0; j < filterSize - 15; j += 16) { - vector unsigned char src_v1 = vec_ld(srcPos + j + 16, src); - vector unsigned char src_vF = vec_perm(src_v0, src_v1, permS); - + vector unsigned char src_vF = unaligned_load(srcPos + j, src); vector signed short src_vA = // vec_unpackh sign-extends... - (vector signed short)(vec_mergeh((vector unsigned char)vzero, src_vF)); + (vector signed short)(VEC_MERGEH((vector unsigned char)vzero, src_vF)); vector signed short src_vB = // vec_unpackh sign-extends... - (vector signed short)(vec_mergel((vector unsigned char)vzero, src_vF)); - - vector signed short filter_v1R = vec_ld((i * 2 * filterSize) + (j * 2) + 16, filter); - vector signed short filter_v2R = vec_ld((i * 2 * filterSize) + (j * 2) + 32, filter); - vector signed short filter_v0 = vec_perm(filter_v0R, filter_v1R, permF); - vector signed short filter_v1 = vec_perm(filter_v1R, filter_v2R, permF); + (vector signed short)(VEC_MERGEL((vector unsigned char)vzero, src_vF)); + vector signed short filter_v0 = (vector signed short)unaligned_load((i * 2 * filterSize) + (j * 2), filter); + vector signed short filter_v1 = (vector signed short)unaligned_load((i * 2 * filterSize) + (j * 2) + 16, filter); vector signed int val_acc = vec_msums(src_vA, filter_v0, val_v); val_v = vec_msums(src_vB, filter_v1, val_acc); - - filter_v0R = filter_v2R; - src_v0 = src_v1; } if (j < filterSize - 7) { // loading src_v0 is useless, it's already done above - // vector unsigned char src_v0 = vec_ld(srcPos + j, src); - vector unsigned char src_v1, src_vF; - vector signed short src_v, filter_v1R, filter_v; - if ((((uintptr_t)src + srcPos) % 16) > 8) { - src_v1 = vec_ld(srcPos + j + 16, src); - } - src_vF = vec_perm(src_v0, src_v1, permS); + vector unsigned char src_vF = unaligned_load(srcPos + j, src); + vector signed short src_v, filter_v; src_v = // vec_unpackh sign-extends... - (vector signed short)(vec_mergeh((vector unsigned char)vzero, src_vF)); - // loading filter_v0R is useless, it's already done above - // vector signed short filter_v0R = vec_ld((i * 2 * filterSize) + j, filter); - filter_v1R = vec_ld((i * 2 * filterSize) + (j * 2) + 16, filter); - filter_v = vec_perm(filter_v0R, filter_v1R, permF); - + (vector signed short)(VEC_MERGEH((vector unsigned char)vzero, src_vF)); + filter_v = (vector signed short)unaligned_load((i * 2 * filterSize) + (j * 2), filter); val_v = vec_msums(src_v, filter_v, val_v); } val_s = vec_sums(val_v, vzero); - vec_st(val_s, 0, tempo); + VEC_ST(val_s, 0, tempo); dst[i] = FFMIN(tempo[3] >> 7, (1 << 15) - 1); } } -- 1.9.1