diff --git a/libavfilter/x86/vf_noise.c b/libavfilter/x86/vf_noise.c index 95f97b3b06..3edcbd64e5 100644 --- a/libavfilter/x86/vf_noise.c +++ b/libavfilter/x86/vf_noise.c @@ -39,7 +39,7 @@ static void line_noise_avg_sse2(uint8_t *dst, const uint8_t *src, "movdqu (%1, %%"FF_REG_a"), %%xmm1 \n\t" "movdqu (%2, %%"FF_REG_a"), %%xmm2 \n\t" "movdqu (%3, %%"FF_REG_a"), %%xmm3 \n\t" - "movdqa (%0, %%"FF_REG_a"), %%xmm0 \n\t" + "movdqu (%0, %%"FF_REG_a"), %%xmm0 \n\t" "paddb %%xmm2, %%xmm1 \n\t" "paddb %%xmm3, %%xmm1 \n\t" "movdqa %%xmm4, %%xmm5 \n\t" @@ -59,7 +59,7 @@ static void line_noise_avg_sse2(uint8_t *dst, const uint8_t *src, "psraw $7, %%xmm3 \n\t" "packsswb %%xmm3, %%xmm1 \n\t" "paddb %%xmm6, %%xmm1 \n\t" - "movdqa %%xmm1, (%4, %%"FF_REG_a") \n\t" + "movdqu %%xmm1, (%4, %%"FF_REG_a") \n\t" "add $16, %%"FF_REG_a" \n\t" " js 1b \n\t" :: "r" (src+xmm_len), "r" (shift[0]+xmm_len), "r" (shift[1]+xmm_len), "r" (shift[2]+xmm_len), @@ -88,12 +88,12 @@ static void line_noise_sse2(uint8_t *dst, const uint8_t *src, "packsswb %%xmm2, %%xmm2 \n\t" ".p2align 4 \n\t" "1: \n\t" - "movdqa (%0, %%"FF_REG_a"), %%xmm0 \n\t" + "movdqu (%0, %%"FF_REG_a"), %%xmm0 \n\t" "movdqu (%1, %%"FF_REG_a"), %%xmm1 \n\t" "pxor %%xmm2, %%xmm0 \n\t" "paddsb %%xmm1, %%xmm0 \n\t" "pxor %%xmm2, %%xmm0 \n\t" - "movntdq %%xmm0, (%2, %%"FF_REG_a") \n\t" + "movdqu %%xmm0, (%2, %%"FF_REG_a") \n\t" "add $16, %%"FF_REG_a" \n\t" " js 1b \n\t" :: "r" (src+xmm_len), "r" (noise+xmm_len), "r" (dst+xmm_len), "g" (-xmm_len)