diff options
Diffstat (limited to 'libavcodec/x86/dsputil_mmx.c')
| -rw-r--r-- | libavcodec/x86/dsputil_mmx.c | 68 |
1 files changed, 3 insertions, 65 deletions
diff --git a/libavcodec/x86/dsputil_mmx.c b/libavcodec/x86/dsputil_mmx.c index 70b2249efd..097cab8285 100644 --- a/libavcodec/x86/dsputil_mmx.c +++ b/libavcodec/x86/dsputil_mmx.c @@ -253,68 +253,6 @@ void ff_add_pixels_clamped_mmx(const int16_t *block, uint8_t *pixels, } while (--i); } -static void put_pixels8_mmx(uint8_t *block, const uint8_t *pixels, - ptrdiff_t line_size, int h) -{ - __asm__ volatile ( - "lea (%3, %3), %%"REG_a" \n\t" - ".p2align 3 \n\t" - "1: \n\t" - "movq (%1 ), %%mm0 \n\t" - "movq (%1, %3), %%mm1 \n\t" - "movq %%mm0, (%2) \n\t" - "movq %%mm1, (%2, %3) \n\t" - "add %%"REG_a", %1 \n\t" - "add %%"REG_a", %2 \n\t" - "movq (%1 ), %%mm0 \n\t" - "movq (%1, %3), %%mm1 \n\t" - "movq %%mm0, (%2) \n\t" - "movq %%mm1, (%2, %3) \n\t" - "add %%"REG_a", %1 \n\t" - "add %%"REG_a", %2 \n\t" - "subl $4, %0 \n\t" - "jnz 1b \n\t" - : "+g"(h), "+r"(pixels), "+r"(block) - : "r"((x86_reg)line_size) - : "%"REG_a, "memory" - ); -} - -static void put_pixels16_mmx(uint8_t *block, const uint8_t *pixels, - ptrdiff_t line_size, int h) -{ - __asm__ volatile ( - "lea (%3, %3), %%"REG_a" \n\t" - ".p2align 3 \n\t" - "1: \n\t" - "movq (%1 ), %%mm0 \n\t" - "movq 8(%1 ), %%mm4 \n\t" - "movq (%1, %3), %%mm1 \n\t" - "movq 8(%1, %3), %%mm5 \n\t" - "movq %%mm0, (%2) \n\t" - "movq %%mm4, 8(%2) \n\t" - "movq %%mm1, (%2, %3) \n\t" - "movq %%mm5, 8(%2, %3) \n\t" - "add %%"REG_a", %1 \n\t" - "add %%"REG_a", %2 \n\t" - "movq (%1 ), %%mm0 \n\t" - "movq 8(%1 ), %%mm4 \n\t" - "movq (%1, %3), %%mm1 \n\t" - "movq 8(%1, %3), %%mm5 \n\t" - "movq %%mm0, (%2) \n\t" - "movq %%mm4, 8(%2) \n\t" - "movq %%mm1, (%2, %3) \n\t" - "movq %%mm5, 8(%2, %3) \n\t" - "add %%"REG_a", %1 \n\t" - "add %%"REG_a", %2 \n\t" - "subl $4, %0 \n\t" - "jnz 1b \n\t" - : "+g"(h), "+r"(pixels), "+r"(block) - : "r"((x86_reg)line_size) - : "%"REG_a, "memory" - ); -} - #define CLEAR_BLOCKS(name, n) \ static void name(int16_t *blocks) \ { \ @@ -1075,7 +1013,7 @@ static void gmc_mmx(uint8_t *dst, uint8_t *src, /* CAVS-specific */ void ff_put_cavs_qpel8_mc00_mmx(uint8_t *dst, uint8_t *src, ptrdiff_t stride) { - put_pixels8_mmx(dst, src, stride, 8); + ff_put_pixels8_mmx(dst, src, stride, 8); } void ff_avg_cavs_qpel8_mc00_mmx(uint8_t *dst, uint8_t *src, ptrdiff_t stride) @@ -1085,7 +1023,7 @@ void ff_avg_cavs_qpel8_mc00_mmx(uint8_t *dst, uint8_t *src, ptrdiff_t stride) void ff_put_cavs_qpel16_mc00_mmx(uint8_t *dst, uint8_t *src, ptrdiff_t stride) { - put_pixels16_mmx(dst, src, stride, 16); + ff_put_pixels16_mmx(dst, src, stride, 16); } void ff_avg_cavs_qpel16_mc00_mmx(uint8_t *dst, uint8_t *src, ptrdiff_t stride) @@ -1097,7 +1035,7 @@ void ff_avg_cavs_qpel16_mc00_mmx(uint8_t *dst, uint8_t *src, ptrdiff_t stride) void ff_put_vc1_mspel_mc00_mmx(uint8_t *dst, const uint8_t *src, ptrdiff_t stride, int rnd) { - put_pixels8_mmx(dst, src, stride, 8); + ff_put_pixels8_mmx(dst, src, stride, 8); } static void vector_clipf_sse(float *dst, const float *src, |