📄 i420_rgb_mmx.h
字号:
packuswb %%xmm1, %%xmm1 # R6 R4 R2 R0 / R6 R4 R2 R0 \n\packuswb %%xmm2, %%xmm2 # G6 G4 G2 G0 / G6 G4 G2 G0 \n\ \n\# Limit RGB odd to 0..255 \n\packuswb %%xmm3, %%xmm3 # B7 B5 B3 B1 / B7 B5 B3 B1 \n\packuswb %%xmm4, %%xmm4 # R7 R5 R3 R1 / R7 R5 R3 R1 \n\packuswb %%xmm5, %%xmm5 # G7 G5 G3 G1 / G7 G5 G3 G1 \n\ \n\# Interleave RGB even and odd \n\punpcklbw %%xmm3, %%xmm0 # B7 B6 B5 B4 B3 B2 B1 B0 \n\punpcklbw %%xmm4, %%xmm1 # R7 R6 R5 R4 R3 R2 R1 R0 \n\punpcklbw %%xmm5, %%xmm2 # G7 G6 G5 G4 G3 G2 G1 G0 \n\"#define SSE2_UNPACK_15_ALIGNED " \n\# mask unneeded bits off \n\movl $0xf8f8f8f8, %%eax # \n\movd %%eax, %%xmm5 # \n\pshufd $0, %%xmm5, %%xmm5 # set xmm5 to f8f8 f8f8 ... f8f8 f8f8 \n\pand %%xmm5, %%xmm0 # b7b6b5b4 b3______ b7b6b5b4 b3______ \n\psrlw $3,%%xmm0 # ______b7 b6b5b4b3 ______b7 b6b5b4b3 \n\pand %%xmm5, %%xmm2 # g7g6g5g4 g3______ g7g6g5g4 g3______ \n\pand %%xmm5, %%xmm1 # r7r6r5r4 r3______ r7r6r5r4 r3______ \n\psrlw $1,%%xmm1 # __r7r6r5 r4r3____ __r7r6r5 r4r3____ \n\pxor %%xmm4, %%xmm4 # zero mm4 \n\movdqa %%xmm0, %%xmm5 # Copy B15-B0 \n\movdqa %%xmm2, %%xmm7 # Copy G15-G0 \n\ \n\# convert rgb24 plane to rgb15 pack for pixel 0-7 \n\punpcklbw %%xmm4, %%xmm2 # ________ ________ g7g6g5g4 g3______ \n\punpcklbw %%xmm1, %%xmm0 # r7r6r5r4 r3______ ______b7 b6b5b4b3 \n\psllw $2,%%xmm2 # ________ ____g7g6 g5g4g3__ ________ \n\por %%xmm2, %%xmm0 # r7r6r5r4 r3__g7g6 g5g4g3b7 b6b5b4b3 \n\movntdq %%xmm0, (%3) # store pixel 0-7 \n\ \n\# convert rgb24 plane to rgb15 pack for pixel 8-15 \n\punpckhbw %%xmm4, %%xmm7 # ________ ________ g7g6g5g4 g3______ \n\punpckhbw %%xmm1, %%xmm5 # r7r6r5r4 r3______ ______b7 b6b5b4b3 \n\psllw $2,%%xmm7 # ________ ____g7g6 g5g4g3__ ________ \n\por %%xmm7, %%xmm5 # r7r6r5r4 r3__g7g6 g5g4g3b7 b6b5b4b3 \n\movntdq %%xmm5, 16(%3) # store pixel 4-7 \n\"#define SSE2_UNPACK_15_UNALIGNED " \n\# mask unneeded bits off \n\movl $0xf8f8f8f8, %%eax # \n\movd %%eax, %%xmm5 # \n\pshufd $0, %%xmm5, %%xmm5 # set xmm5 to f8f8 f8f8 ... f8f8 f8f8 \n\pand %%xmm5, %%xmm0 # b7b6b5b4 b3______ b7b6b5b4 b3______ \n\psrlw $3,%%xmm0 # ______b7 b6b5b4b3 ______b7 b6b5b4b3 \n\pand %%xmm5, %%xmm2 # g7g6g5g4 g3______ g7g6g5g4 g3______ \n\pand %%xmm5, %%xmm1 # r7r6r5r4 r3______ r7r6r5r4 r3______ \n\psrlw $1,%%xmm1 # __r7r6r5 r4r3____ __r7r6r5 r4r3____ \n\pxor %%xmm4, %%xmm4 # zero mm4 \n\movdqa %%xmm0, %%xmm5 # Copy B15-B0 \n\movdqa %%xmm2, %%xmm7 # Copy G15-G0 \n\ \n\# convert rgb24 plane to rgb15 pack for pixel 0-7 \n\punpcklbw %%xmm4, %%xmm2 # ________ ________ g7g6g5g4 g3______ \n\punpcklbw %%xmm1, %%xmm0 # r7r6r5r4 r3______ ______b7 b6b5b4b3 \n\psllw $2,%%xmm2 # ________ ____g7g6 g5g4g3__ ________ \n\por %%xmm2, %%xmm0 # r7r6r5r4 r3__g7g6 g5g4g3b7 b6b5b4b3 \n\movdqu %%xmm0, (%3) # store pixel 0-7 \n\ \n\# convert rgb24 plane to rgb15 pack for pixel 8-15 \n\punpckhbw %%xmm4, %%xmm7 # ________ ________ g7g6g5g4 g3______ \n\punpckhbw %%xmm1, %%xmm5 # r7r6r5r4 r3______ ______b7 b6b5b4b3 \n\psllw $2,%%xmm7 # ________ ____g7g6 g5g4g3__ ________ \n\por %%xmm7, %%xmm5 # r7r6r5r4 r3__g7g6 g5g4g3b7 b6b5b4b3 \n\movdqu %%xmm5, 16(%3) # store pixel 4-7 \n\"#define SSE2_UNPACK_16_ALIGNED " \n\# mask unneeded bits off \n\movl $0xf8f8f8f8, %%eax # \n\movd %%eax, %%xmm5 # \n\pshufd $0, %%xmm5, %%xmm5 # set xmm5 to f8f8 f8f8 ... f8f8 f8f8 \n\pand %%xmm5, %%xmm0 # b7b6b5b4 b3______ b7b6b5b4 b3______ \n\pand %%xmm5, %%xmm1 # r7r6r5r4 r3______ r7r6r5r4 r3______ \n\movl $0xfcfcfcfc, %%eax # \n\movd %%eax, %%xmm5 # \n\pshufd $0, %%xmm5, %%xmm5 # set xmm5 to f8f8 f8f8 ... f8f8 f8f8 \n\pand %%xmm5, %%xmm2 # g7g6g5g4 g3g2____ g7g6g5g4 g3g2____ \n\psrlw $3,%%xmm0 # ______b7 b6b5b4b3 ______b7 b6b5b4b3 \n\pxor %%xmm4, %%xmm4 # zero mm4 \n\movdqa %%xmm0, %%xmm5 # Copy B15-B0 \n\movdqa %%xmm2, %%xmm7 # Copy G15-G0 \n\ \n\# convert rgb24 plane to rgb16 pack for pixel 0-7 \n\punpcklbw %%xmm4, %%xmm2 # ________ ________ g7g6g5g4 g3g2____ \n\punpcklbw %%xmm1, %%xmm0 # r7r6r5r4 r3______ ______b7 b6b5b4b3 \n\psllw $3,%%xmm2 # ________ __g7g6g5 g4g3g2__ ________ \n\por %%xmm2, %%xmm0 # r7r6r5r4 r3g7g6g5 g4g3g2b7 b6b5b4b3 \n\movntdq %%xmm0, (%3) # store pixel 0-7 \n\ \n\# convert rgb24 plane to rgb16 pack for pixel 8-15 \n\punpckhbw %%xmm4, %%xmm7 # ________ ________ g7g6g5g4 g3g2____ \n\punpckhbw %%xmm1, %%xmm5 # r7r6r5r4 r3______ ______b7 b6b5b4b3 \n\psllw $3,%%xmm7 # ________ __g7g6g5 g4g3g2__ ________ \n\por %%xmm7, %%xmm5 # r7r6r5r4 r3g7g6g5 g4g3g2b7 b6b5b4b3 \n\movntdq %%xmm5, 16(%3) # store pixel 4-7 \n\"#define SSE2_UNPACK_16_UNALIGNED " \n\# mask unneeded bits off \n\movl $0xf8f8f8f8, %%eax # \n\movd %%eax, %%xmm5 # \n\pshufd $0, %%xmm5, %%xmm5 # set xmm5 to f8f8 f8f8 ... f8f8 f8f8 \n\pand %%xmm5, %%xmm0 # b7b6b5b4 b3______ b7b6b5b4 b3______ \n\pand %%xmm5, %%xmm1 # r7r6r5r4 r3______ r7r6r5r4 r3______ \n\movl $0xfcfcfcfc, %%eax # \n\movd %%eax, %%xmm5 # \n\pshufd $0, %%xmm5, %%xmm5 # set xmm5 to f8f8 f8f8 ... f8f8 f8f8 \n\pand %%xmm5, %%xmm2 # g7g6g5g4 g3g2____ g7g6g5g4 g3g2____ \n\psrlw $3,%%xmm0 # ______b7 b6b5b4b3 ______b7 b6b5b4b3 \n\pxor %%xmm4, %%xmm4 # zero mm4 \n\movdqa %%xmm0, %%xmm5 # Copy B15-B0 \n\movdqa %%xmm2, %%xmm7 # Copy G15-G0 \n\ \n\# convert rgb24 plane to rgb16 pack for pixel 0-7 \n\punpcklbw %%xmm4, %%xmm2 # ________ ________ g7g6g5g4 g3g2____ \n\punpcklbw %%xmm1, %%xmm0 # r7r6r5r4 r3______ ______b7 b6b5b4b3 \n\psllw $3,%%xmm2 # ________ __g7g6g5 g4g3g2__ ________ \n\por %%xmm2, %%xmm0 # r7r6r5r4 r3g7g6g5 g4g3g2b7 b6b5b4b3 \n\movdqu %%xmm0, (%3) # store pixel 0-7 \n\ \n\# convert rgb24 plane to rgb16 pack for pixel 8-15 \n\punpckhbw %%xmm4, %%xmm7 # ________ ________ g7g6g5g4 g3g2____ \n\punpckhbw %%xmm1, %%xmm5 # r7r6r5r4 r3______ ______b7 b6b5b4b3 \n\psllw $3,%%xmm7 # ________ __g7g6g5 g4g3g2__ ________ \n\por %%xmm7, %%xmm5 # r7r6r5r4 r3g7g6g5 g4g3g2b7 b6b5b4b3 \n\movdqu %%xmm5, 16(%3) # store pixel 4-7 \n\"#define SSE2_UNPACK_32_ARGB_ALIGNED " \n\pxor %%xmm3, %%xmm3 # zero xmm3 \n\movdqa %%xmm0, %%xmm4 # B7 B6 B5 B4 B3 B2 B1 B0 \n\punpcklbw %%xmm2, %%xmm4 # G3 B3 G2 B2 G1 B1 G0 B0 \n\movdqa %%xmm1, %%xmm5 # R7 R6 R5 R4 R3 R2 R1 R0 \n\punpcklbw %%xmm3, %%xmm5 # 00 R3 00 R2 00 R1 00 R0 \n\movdqa %%xmm4, %%xmm6 # G3 B3 G2 B2 G1 B1 G0 B0 \n\punpcklwd %%xmm5, %%xmm4 # 00 R1 B1 G1 00 R0 B0 G0 \n\movntdq %%xmm4, (%3) # Store ARGB3 ARGB2 ARGB1 ARGB0 \n\punpckhwd %%xmm5, %%xmm6 # 00 R3 B3 G3 00 R2 B2 G2 \n\movntdq %%xmm6, 16(%3) # Store ARGB7 ARGB6 ARGB5 ARGB4 \n\punpckhbw %%xmm2, %%xmm0 # G7 B7 G6 B6 G5 B5 G4 B4 \n\punpckhbw %%xmm3, %%xmm1 # 00 R7 00 R6 00 R5 00 R4 \n\movdqa %%xmm0, %%xmm5 # G7 B7 G6 B6 G5 B5 G4 B4 \n\punpcklwd %%xmm1, %%xmm5 # 00 R5 B5 G5 00 R4 B4 G4 \n\movntdq %%xmm5, 32(%3) # Store ARGB11 ARGB10 ARGB9 ARGB8 \n\punpckhwd %%xmm1, %%xmm0 # 00 R7 B7 G7 00 R6 B6 G6 \n\movntdq %%xmm0, 48(%3) # Store ARGB15 ARGB14 ARGB13 ARGB12 \n\"#define SSE2_UNPACK_32_ARGB_UNALIGNED " \n\pxor %%xmm3, %%xmm3 # zero xmm3 \n\movdqa %%xmm0, %%xmm4 # B7 B6 B5 B4 B3 B2 B1 B0 \n\punpcklbw %%xmm2, %%xmm4 # G3 B3 G2 B2 G1 B1 G0 B0 \n\movdqa %%xmm1, %%xmm5 # R7 R6 R5 R4 R3 R2 R1 R0 \n\punpcklbw %%xmm3, %%xmm5 # 00 R3 00 R2 00 R1 00 R0 \n\movdqa %%xmm4, %%xmm6 # G3 B3 G2 B2 G1 B1 G0 B0 \n\punpcklwd %%xmm5, %%xmm4 # 00 R1 B1 G1 00 R0 B0 G0 \n\movdqu %%xmm4, (%3) # Store ARGB3 ARGB2 ARGB1 ARGB0 \n\punpckhwd %%xmm5, %%xmm6 # 00 R3 B3 G3 00 R2 B2 G2 \n\movdqu %%xmm6, 16(%3) # Store ARGB7 ARGB6 ARGB5 ARGB4 \n\punpckhbw %%xmm2, %%xmm0 # G7 B7 G6 B6 G5 B5 G4 B4 \n\punpckhbw %%xmm3, %%xmm1 # 00 R7 00 R6 00 R5 00 R4 \n\movdqa %%xmm0, %%xmm5 # G7 B7 G6 B6 G5 B5 G4 B4 \n\punpcklwd %%xmm1, %%xmm5 # 00 R5 B5 G5 00 R4 B4 G4 \n\movdqu %%xmm5, 32(%3) # Store ARGB11 ARGB10 ARGB9 ARGB8 \n\punpckhwd %%xmm1, %%xmm0 # 00 R7 B7 G7 00 R6 B6 G6 \n\movdqu %%xmm0, 48(%3) # Store ARGB15 ARGB14 ARGB13 ARGB12 \n\"#define SSE2_UNPACK_32_RGBA_ALIGNED " \n\pxor %%xmm3, %%xmm3 # zero mm3 \n\movdqa %%xmm2, %%xmm4 # G7 G6 G5 G4 G3 G2 G1 G0 \n\punpcklbw %%xmm1, %%xmm4 # R3 G3 R2 G2 R1 G1 R0 G0 \n\punpcklbw %%xmm0, %%xmm3 # B3 00 B2 00 B1 00 B0 00 \n\movdqa %%xmm3, %%xmm5 # R3 00 R2 00 R1 00 R0 00 \n\punpcklwd %%xmm4, %%xmm3 # R1 G1 B1 00 R0 B0 G0 00 \n\movntdq %%xmm3, (%3) # Store RGBA3 RGBA2 RGBA1 RGBA0 \n\punpckhwd %%xmm4, %%xmm5 # R3 G3 B3 00 R2 G2 B2 00 \n\movntdq %%xmm5, 16(%3) # Store RGBA7 RGBA6 RGBA5 RGBA4 \n\pxor %%xmm6, %%xmm6 # zero mm6 \n\punpckhbw %%xmm1, %%xmm2 # R7 G7 R6 G6 R5 G5 R4 G4 \n\punpckhbw %%xmm0, %%xmm6 # B7 00 B6 00 B5 00 B4 00 \n\movdqa %%xmm6, %%xmm0 # B7 00 B6 00 B5 00 B4 00 \n\punpcklwd %%xmm2, %%xmm6 # R5 G5 B5 00 R4 G4 B4 00 \n\movntdq %%xmm6, 32(%3) # Store BGRA11 BGRA10 BGRA9 RGBA8 \n\punpckhwd %%xmm2, %%xmm0 # R7 G7 B7 00 R6 G6 B6 00 \n\movntdq %%xmm0, 48(%3) # Store RGBA15 RGBA14 RGBA13 RGBA12 \n\"#define SSE2_UNPACK_32_RGBA_UNALIGNED " \n\pxor %%xmm3, %%xmm3 # zero mm3 \n\movdqa %%xmm2, %%xmm4 # G7 G6 G5 G4 G3 G2 G1 G0 \n\punpcklbw %%xmm1, %%xmm4 # R3 G3 R2 G2 R1 G1 R0 G0 \n\punpcklbw %%xmm0, %%xmm3 # B3 00 B2 00 B1 00 B0 00 \n\movdqa %%xmm3, %%xmm5 # R3 00 R2 00 R1 00 R0 00 \n\punpcklwd %%xmm4, %%xmm3 # R1 G1 B1 00 R0 B0 G0 00 \n\movdqu %%xmm3, (%3) # Store RGBA3 RGBA2 RGBA1 RGBA0 \n\punpckhwd %%xmm4, %%xmm5 # R3 G3 B3 00 R2 G2 B2 00 \n\movdqu %%xmm5, 16(%3) # Store RGBA7 RGBA6 RGBA5 RGBA4 \n\pxor %%xmm6, %%xmm6 # zero mm6 \n\punpckhbw %%xmm1, %%xmm2 # R7 G7 R6 G6 R5 G5 R4 G4 \n\punpckhbw %%xmm0, %%xmm6 # B7 00 B6 00 B5 00 B4 00 \n\movdqa %%xmm6, %%xmm0 # B7 00 B6 00 B5 00 B4 00 \n\punpcklwd %%xmm2, %%xmm6 # R5 G5 B5 00 R4 G4 B4 00 \n\movdqu %%xmm6, 32(%3) # Store RGBA11 RGBA10 RGBA9 RGBA8 \n\punpckhwd %%xmm2, %%xmm0 # R7 G7 B7 00 R6 G6 B6 00 \n\movdqu %%xmm0, 48(%3) # Store RGBA15 RGBA14 RGBA13 RGBA12 \n\"#define SSE2_UNPACK_32_BGRA_ALIGNED " \n\pxor %%xmm3, %%xmm3 # zero mm3 \n\movdqa %%xmm2, %%xmm4 # G7 G6 G5 G4 G3 G2 G1 G0 \n\punpcklbw %%xmm0, %%xmm4 # B3 G3 B2 G2 B1 G1 B0 G0 \n\punpcklbw %%xmm1, %%xmm3 # R3 00 R2 00 R1 00 R0 00 \n\movdqa %%xmm3, %%xmm5 # R3 00 R2 00 R1 00 R0 00 \n\punpcklwd %%xmm4, %%xmm3 # B1 G1 R1 00 B0 G0 R0 00 \n\movntdq %%xmm3, (%3) # Store BGRA3 BGRA2 BGRA1 BGRA0 \n\punpckhwd %%xmm4, %%xmm5 # B3 G3 R3 00 B2 G2 R2 00 \n\movntdq %%xmm5, 16(%3) # Store BGRA7 BGRA6 BGRA5 BGRA4 \n\pxor %%xmm6, %%xmm6 # zero mm6 \n\punpckhbw %%xmm0, %%xmm2 # B7 G7 B6 G6 B5 G5 B4 G4 \n\⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -