michael@0: ; michael@0: ; jiss2fst.asm - fast integer IDCT (SSE2) michael@0: ; michael@0: ; Copyright 2009 Pierre Ossman for Cendio AB michael@0: ; michael@0: ; Based on michael@0: ; x86 SIMD extension for IJG JPEG library michael@0: ; Copyright (C) 1999-2006, MIYASAKA Masaru. michael@0: ; For conditions of distribution and use, see copyright notice in jsimdext.inc michael@0: ; michael@0: ; This file should be assembled with NASM (Netwide Assembler), michael@0: ; can *not* be assembled with Microsoft's MASM or any compatible michael@0: ; assembler (including Borland's Turbo Assembler). michael@0: ; NASM is available from http://nasm.sourceforge.net/ or michael@0: ; http://sourceforge.net/project/showfiles.php?group_id=6208 michael@0: ; michael@0: ; This file contains a fast, not so accurate integer implementation of michael@0: ; the inverse DCT (Discrete Cosine Transform). The following code is michael@0: ; based directly on the IJG's original jidctfst.c; see the jidctfst.c michael@0: ; for more details. michael@0: ; michael@0: ; [TAB8] michael@0: michael@0: %include "jsimdext.inc" michael@0: %include "jdct.inc" michael@0: michael@0: ; -------------------------------------------------------------------------- michael@0: michael@0: %define CONST_BITS 8 ; 14 is also OK. michael@0: %define PASS1_BITS 2 michael@0: michael@0: %if IFAST_SCALE_BITS != PASS1_BITS michael@0: %error "'IFAST_SCALE_BITS' must be equal to 'PASS1_BITS'." michael@0: %endif michael@0: michael@0: %if CONST_BITS == 8 michael@0: F_1_082 equ 277 ; FIX(1.082392200) michael@0: F_1_414 equ 362 ; FIX(1.414213562) michael@0: F_1_847 equ 473 ; FIX(1.847759065) michael@0: F_2_613 equ 669 ; FIX(2.613125930) michael@0: F_1_613 equ (F_2_613 - 256) ; FIX(2.613125930) - FIX(1) michael@0: %else michael@0: ; NASM cannot do compile-time arithmetic on floating-point constants. michael@0: %define DESCALE(x,n) (((x)+(1<<((n)-1)))>>(n)) michael@0: F_1_082 equ DESCALE(1162209775,30-CONST_BITS) ; FIX(1.082392200) michael@0: F_1_414 equ DESCALE(1518500249,30-CONST_BITS) ; FIX(1.414213562) michael@0: F_1_847 equ DESCALE(1984016188,30-CONST_BITS) ; FIX(1.847759065) michael@0: F_2_613 equ DESCALE(2805822602,30-CONST_BITS) ; FIX(2.613125930) michael@0: F_1_613 equ (F_2_613 - (1 << CONST_BITS)) ; FIX(2.613125930) - FIX(1) michael@0: %endif michael@0: michael@0: ; -------------------------------------------------------------------------- michael@0: SECTION SEG_CONST michael@0: michael@0: ; PRE_MULTIPLY_SCALE_BITS <= 2 (to avoid overflow) michael@0: ; CONST_BITS + CONST_SHIFT + PRE_MULTIPLY_SCALE_BITS == 16 (for pmulhw) michael@0: michael@0: %define PRE_MULTIPLY_SCALE_BITS 2 michael@0: %define CONST_SHIFT (16 - PRE_MULTIPLY_SCALE_BITS - CONST_BITS) michael@0: michael@0: alignz 16 michael@0: global EXTN(jconst_idct_ifast_sse2) michael@0: michael@0: EXTN(jconst_idct_ifast_sse2): michael@0: michael@0: PW_F1414 times 8 dw F_1_414 << CONST_SHIFT michael@0: PW_F1847 times 8 dw F_1_847 << CONST_SHIFT michael@0: PW_MF1613 times 8 dw -F_1_613 << CONST_SHIFT michael@0: PW_F1082 times 8 dw F_1_082 << CONST_SHIFT michael@0: PB_CENTERJSAMP times 16 db CENTERJSAMPLE michael@0: michael@0: alignz 16 michael@0: michael@0: ; -------------------------------------------------------------------------- michael@0: SECTION SEG_TEXT michael@0: BITS 32 michael@0: ; michael@0: ; Perform dequantization and inverse DCT on one block of coefficients. michael@0: ; michael@0: ; GLOBAL(void) michael@0: ; jsimd_idct_ifast_sse2 (void * dct_table, JCOEFPTR coef_block, michael@0: ; JSAMPARRAY output_buf, JDIMENSION output_col) michael@0: ; michael@0: michael@0: %define dct_table(b) (b)+8 ; jpeg_component_info * compptr michael@0: %define coef_block(b) (b)+12 ; JCOEFPTR coef_block michael@0: %define output_buf(b) (b)+16 ; JSAMPARRAY output_buf michael@0: %define output_col(b) (b)+20 ; JDIMENSION output_col michael@0: michael@0: %define original_ebp ebp+0 michael@0: %define wk(i) ebp-(WK_NUM-(i))*SIZEOF_XMMWORD ; xmmword wk[WK_NUM] michael@0: %define WK_NUM 2 michael@0: michael@0: align 16 michael@0: global EXTN(jsimd_idct_ifast_sse2) michael@0: michael@0: EXTN(jsimd_idct_ifast_sse2): michael@0: push ebp michael@0: mov eax,esp ; eax = original ebp michael@0: sub esp, byte 4 michael@0: and esp, byte (-SIZEOF_XMMWORD) ; align to 128 bits michael@0: mov [esp],eax michael@0: mov ebp,esp ; ebp = aligned ebp michael@0: lea esp, [wk(0)] michael@0: pushpic ebx michael@0: ; push ecx ; unused michael@0: ; push edx ; need not be preserved michael@0: push esi michael@0: push edi michael@0: michael@0: get_GOT ebx ; get GOT address michael@0: michael@0: ; ---- Pass 1: process columns from input. michael@0: michael@0: ; mov eax, [original_ebp] michael@0: mov edx, POINTER [dct_table(eax)] ; quantptr michael@0: mov esi, JCOEFPTR [coef_block(eax)] ; inptr michael@0: michael@0: %ifndef NO_ZERO_COLUMN_TEST_IFAST_SSE2 michael@0: mov eax, DWORD [DWBLOCK(1,0,esi,SIZEOF_JCOEF)] michael@0: or eax, DWORD [DWBLOCK(2,0,esi,SIZEOF_JCOEF)] michael@0: jnz near .columnDCT michael@0: michael@0: movdqa xmm0, XMMWORD [XMMBLOCK(1,0,esi,SIZEOF_JCOEF)] michael@0: movdqa xmm1, XMMWORD [XMMBLOCK(2,0,esi,SIZEOF_JCOEF)] michael@0: por xmm0, XMMWORD [XMMBLOCK(3,0,esi,SIZEOF_JCOEF)] michael@0: por xmm1, XMMWORD [XMMBLOCK(4,0,esi,SIZEOF_JCOEF)] michael@0: por xmm0, XMMWORD [XMMBLOCK(5,0,esi,SIZEOF_JCOEF)] michael@0: por xmm1, XMMWORD [XMMBLOCK(6,0,esi,SIZEOF_JCOEF)] michael@0: por xmm0, XMMWORD [XMMBLOCK(7,0,esi,SIZEOF_JCOEF)] michael@0: por xmm1,xmm0 michael@0: packsswb xmm1,xmm1 michael@0: packsswb xmm1,xmm1 michael@0: movd eax,xmm1 michael@0: test eax,eax michael@0: jnz short .columnDCT michael@0: michael@0: ; -- AC terms all zero michael@0: michael@0: movdqa xmm0, XMMWORD [XMMBLOCK(0,0,esi,SIZEOF_JCOEF)] michael@0: pmullw xmm0, XMMWORD [XMMBLOCK(0,0,edx,SIZEOF_ISLOW_MULT_TYPE)] michael@0: michael@0: movdqa xmm7,xmm0 ; xmm0=in0=(00 01 02 03 04 05 06 07) michael@0: punpcklwd xmm0,xmm0 ; xmm0=(00 00 01 01 02 02 03 03) michael@0: punpckhwd xmm7,xmm7 ; xmm7=(04 04 05 05 06 06 07 07) michael@0: michael@0: pshufd xmm6,xmm0,0x00 ; xmm6=col0=(00 00 00 00 00 00 00 00) michael@0: pshufd xmm2,xmm0,0x55 ; xmm2=col1=(01 01 01 01 01 01 01 01) michael@0: pshufd xmm5,xmm0,0xAA ; xmm5=col2=(02 02 02 02 02 02 02 02) michael@0: pshufd xmm0,xmm0,0xFF ; xmm0=col3=(03 03 03 03 03 03 03 03) michael@0: pshufd xmm1,xmm7,0x00 ; xmm1=col4=(04 04 04 04 04 04 04 04) michael@0: pshufd xmm4,xmm7,0x55 ; xmm4=col5=(05 05 05 05 05 05 05 05) michael@0: pshufd xmm3,xmm7,0xAA ; xmm3=col6=(06 06 06 06 06 06 06 06) michael@0: pshufd xmm7,xmm7,0xFF ; xmm7=col7=(07 07 07 07 07 07 07 07) michael@0: michael@0: movdqa XMMWORD [wk(0)], xmm2 ; wk(0)=col1 michael@0: movdqa XMMWORD [wk(1)], xmm0 ; wk(1)=col3 michael@0: jmp near .column_end michael@0: alignx 16,7 michael@0: %endif michael@0: .columnDCT: michael@0: michael@0: ; -- Even part michael@0: michael@0: movdqa xmm0, XMMWORD [XMMBLOCK(0,0,esi,SIZEOF_JCOEF)] michael@0: movdqa xmm1, XMMWORD [XMMBLOCK(2,0,esi,SIZEOF_JCOEF)] michael@0: pmullw xmm0, XMMWORD [XMMBLOCK(0,0,edx,SIZEOF_IFAST_MULT_TYPE)] michael@0: pmullw xmm1, XMMWORD [XMMBLOCK(2,0,edx,SIZEOF_IFAST_MULT_TYPE)] michael@0: movdqa xmm2, XMMWORD [XMMBLOCK(4,0,esi,SIZEOF_JCOEF)] michael@0: movdqa xmm3, XMMWORD [XMMBLOCK(6,0,esi,SIZEOF_JCOEF)] michael@0: pmullw xmm2, XMMWORD [XMMBLOCK(4,0,edx,SIZEOF_IFAST_MULT_TYPE)] michael@0: pmullw xmm3, XMMWORD [XMMBLOCK(6,0,edx,SIZEOF_IFAST_MULT_TYPE)] michael@0: michael@0: movdqa xmm4,xmm0 michael@0: movdqa xmm5,xmm1 michael@0: psubw xmm0,xmm2 ; xmm0=tmp11 michael@0: psubw xmm1,xmm3 michael@0: paddw xmm4,xmm2 ; xmm4=tmp10 michael@0: paddw xmm5,xmm3 ; xmm5=tmp13 michael@0: michael@0: psllw xmm1,PRE_MULTIPLY_SCALE_BITS michael@0: pmulhw xmm1,[GOTOFF(ebx,PW_F1414)] michael@0: psubw xmm1,xmm5 ; xmm1=tmp12 michael@0: michael@0: movdqa xmm6,xmm4 michael@0: movdqa xmm7,xmm0 michael@0: psubw xmm4,xmm5 ; xmm4=tmp3 michael@0: psubw xmm0,xmm1 ; xmm0=tmp2 michael@0: paddw xmm6,xmm5 ; xmm6=tmp0 michael@0: paddw xmm7,xmm1 ; xmm7=tmp1 michael@0: michael@0: movdqa XMMWORD [wk(1)], xmm4 ; wk(1)=tmp3 michael@0: movdqa XMMWORD [wk(0)], xmm0 ; wk(0)=tmp2 michael@0: michael@0: ; -- Odd part michael@0: michael@0: movdqa xmm2, XMMWORD [XMMBLOCK(1,0,esi,SIZEOF_JCOEF)] michael@0: movdqa xmm3, XMMWORD [XMMBLOCK(3,0,esi,SIZEOF_JCOEF)] michael@0: pmullw xmm2, XMMWORD [XMMBLOCK(1,0,edx,SIZEOF_IFAST_MULT_TYPE)] michael@0: pmullw xmm3, XMMWORD [XMMBLOCK(3,0,edx,SIZEOF_IFAST_MULT_TYPE)] michael@0: movdqa xmm5, XMMWORD [XMMBLOCK(5,0,esi,SIZEOF_JCOEF)] michael@0: movdqa xmm1, XMMWORD [XMMBLOCK(7,0,esi,SIZEOF_JCOEF)] michael@0: pmullw xmm5, XMMWORD [XMMBLOCK(5,0,edx,SIZEOF_IFAST_MULT_TYPE)] michael@0: pmullw xmm1, XMMWORD [XMMBLOCK(7,0,edx,SIZEOF_IFAST_MULT_TYPE)] michael@0: michael@0: movdqa xmm4,xmm2 michael@0: movdqa xmm0,xmm5 michael@0: psubw xmm2,xmm1 ; xmm2=z12 michael@0: psubw xmm5,xmm3 ; xmm5=z10 michael@0: paddw xmm4,xmm1 ; xmm4=z11 michael@0: paddw xmm0,xmm3 ; xmm0=z13 michael@0: michael@0: movdqa xmm1,xmm5 ; xmm1=z10(unscaled) michael@0: psllw xmm2,PRE_MULTIPLY_SCALE_BITS michael@0: psllw xmm5,PRE_MULTIPLY_SCALE_BITS michael@0: michael@0: movdqa xmm3,xmm4 michael@0: psubw xmm4,xmm0 michael@0: paddw xmm3,xmm0 ; xmm3=tmp7 michael@0: michael@0: psllw xmm4,PRE_MULTIPLY_SCALE_BITS michael@0: pmulhw xmm4,[GOTOFF(ebx,PW_F1414)] ; xmm4=tmp11 michael@0: michael@0: ; To avoid overflow... michael@0: ; michael@0: ; (Original) michael@0: ; tmp12 = -2.613125930 * z10 + z5; michael@0: ; michael@0: ; (This implementation) michael@0: ; tmp12 = (-1.613125930 - 1) * z10 + z5; michael@0: ; = -1.613125930 * z10 - z10 + z5; michael@0: michael@0: movdqa xmm0,xmm5 michael@0: paddw xmm5,xmm2 michael@0: pmulhw xmm5,[GOTOFF(ebx,PW_F1847)] ; xmm5=z5 michael@0: pmulhw xmm0,[GOTOFF(ebx,PW_MF1613)] michael@0: pmulhw xmm2,[GOTOFF(ebx,PW_F1082)] michael@0: psubw xmm0,xmm1 michael@0: psubw xmm2,xmm5 ; xmm2=tmp10 michael@0: paddw xmm0,xmm5 ; xmm0=tmp12 michael@0: michael@0: ; -- Final output stage michael@0: michael@0: psubw xmm0,xmm3 ; xmm0=tmp6 michael@0: movdqa xmm1,xmm6 michael@0: movdqa xmm5,xmm7 michael@0: paddw xmm6,xmm3 ; xmm6=data0=(00 01 02 03 04 05 06 07) michael@0: paddw xmm7,xmm0 ; xmm7=data1=(10 11 12 13 14 15 16 17) michael@0: psubw xmm1,xmm3 ; xmm1=data7=(70 71 72 73 74 75 76 77) michael@0: psubw xmm5,xmm0 ; xmm5=data6=(60 61 62 63 64 65 66 67) michael@0: psubw xmm4,xmm0 ; xmm4=tmp5 michael@0: michael@0: movdqa xmm3,xmm6 ; transpose coefficients(phase 1) michael@0: punpcklwd xmm6,xmm7 ; xmm6=(00 10 01 11 02 12 03 13) michael@0: punpckhwd xmm3,xmm7 ; xmm3=(04 14 05 15 06 16 07 17) michael@0: movdqa xmm0,xmm5 ; transpose coefficients(phase 1) michael@0: punpcklwd xmm5,xmm1 ; xmm5=(60 70 61 71 62 72 63 73) michael@0: punpckhwd xmm0,xmm1 ; xmm0=(64 74 65 75 66 76 67 77) michael@0: michael@0: movdqa xmm7, XMMWORD [wk(0)] ; xmm7=tmp2 michael@0: movdqa xmm1, XMMWORD [wk(1)] ; xmm1=tmp3 michael@0: michael@0: movdqa XMMWORD [wk(0)], xmm5 ; wk(0)=(60 70 61 71 62 72 63 73) michael@0: movdqa XMMWORD [wk(1)], xmm0 ; wk(1)=(64 74 65 75 66 76 67 77) michael@0: michael@0: paddw xmm2,xmm4 ; xmm2=tmp4 michael@0: movdqa xmm5,xmm7 michael@0: movdqa xmm0,xmm1 michael@0: paddw xmm7,xmm4 ; xmm7=data2=(20 21 22 23 24 25 26 27) michael@0: paddw xmm1,xmm2 ; xmm1=data4=(40 41 42 43 44 45 46 47) michael@0: psubw xmm5,xmm4 ; xmm5=data5=(50 51 52 53 54 55 56 57) michael@0: psubw xmm0,xmm2 ; xmm0=data3=(30 31 32 33 34 35 36 37) michael@0: michael@0: movdqa xmm4,xmm7 ; transpose coefficients(phase 1) michael@0: punpcklwd xmm7,xmm0 ; xmm7=(20 30 21 31 22 32 23 33) michael@0: punpckhwd xmm4,xmm0 ; xmm4=(24 34 25 35 26 36 27 37) michael@0: movdqa xmm2,xmm1 ; transpose coefficients(phase 1) michael@0: punpcklwd xmm1,xmm5 ; xmm1=(40 50 41 51 42 52 43 53) michael@0: punpckhwd xmm2,xmm5 ; xmm2=(44 54 45 55 46 56 47 57) michael@0: michael@0: movdqa xmm0,xmm3 ; transpose coefficients(phase 2) michael@0: punpckldq xmm3,xmm4 ; xmm3=(04 14 24 34 05 15 25 35) michael@0: punpckhdq xmm0,xmm4 ; xmm0=(06 16 26 36 07 17 27 37) michael@0: movdqa xmm5,xmm6 ; transpose coefficients(phase 2) michael@0: punpckldq xmm6,xmm7 ; xmm6=(00 10 20 30 01 11 21 31) michael@0: punpckhdq xmm5,xmm7 ; xmm5=(02 12 22 32 03 13 23 33) michael@0: michael@0: movdqa xmm4, XMMWORD [wk(0)] ; xmm4=(60 70 61 71 62 72 63 73) michael@0: movdqa xmm7, XMMWORD [wk(1)] ; xmm7=(64 74 65 75 66 76 67 77) michael@0: michael@0: movdqa XMMWORD [wk(0)], xmm3 ; wk(0)=(04 14 24 34 05 15 25 35) michael@0: movdqa XMMWORD [wk(1)], xmm0 ; wk(1)=(06 16 26 36 07 17 27 37) michael@0: michael@0: movdqa xmm3,xmm1 ; transpose coefficients(phase 2) michael@0: punpckldq xmm1,xmm4 ; xmm1=(40 50 60 70 41 51 61 71) michael@0: punpckhdq xmm3,xmm4 ; xmm3=(42 52 62 72 43 53 63 73) michael@0: movdqa xmm0,xmm2 ; transpose coefficients(phase 2) michael@0: punpckldq xmm2,xmm7 ; xmm2=(44 54 64 74 45 55 65 75) michael@0: punpckhdq xmm0,xmm7 ; xmm0=(46 56 66 76 47 57 67 77) michael@0: michael@0: movdqa xmm4,xmm6 ; transpose coefficients(phase 3) michael@0: punpcklqdq xmm6,xmm1 ; xmm6=col0=(00 10 20 30 40 50 60 70) michael@0: punpckhqdq xmm4,xmm1 ; xmm4=col1=(01 11 21 31 41 51 61 71) michael@0: movdqa xmm7,xmm5 ; transpose coefficients(phase 3) michael@0: punpcklqdq xmm5,xmm3 ; xmm5=col2=(02 12 22 32 42 52 62 72) michael@0: punpckhqdq xmm7,xmm3 ; xmm7=col3=(03 13 23 33 43 53 63 73) michael@0: michael@0: movdqa xmm1, XMMWORD [wk(0)] ; xmm1=(04 14 24 34 05 15 25 35) michael@0: movdqa xmm3, XMMWORD [wk(1)] ; xmm3=(06 16 26 36 07 17 27 37) michael@0: michael@0: movdqa XMMWORD [wk(0)], xmm4 ; wk(0)=col1 michael@0: movdqa XMMWORD [wk(1)], xmm7 ; wk(1)=col3 michael@0: michael@0: movdqa xmm4,xmm1 ; transpose coefficients(phase 3) michael@0: punpcklqdq xmm1,xmm2 ; xmm1=col4=(04 14 24 34 44 54 64 74) michael@0: punpckhqdq xmm4,xmm2 ; xmm4=col5=(05 15 25 35 45 55 65 75) michael@0: movdqa xmm7,xmm3 ; transpose coefficients(phase 3) michael@0: punpcklqdq xmm3,xmm0 ; xmm3=col6=(06 16 26 36 46 56 66 76) michael@0: punpckhqdq xmm7,xmm0 ; xmm7=col7=(07 17 27 37 47 57 67 77) michael@0: .column_end: michael@0: michael@0: ; -- Prefetch the next coefficient block michael@0: michael@0: prefetchnta [esi + DCTSIZE2*SIZEOF_JCOEF + 0*32] michael@0: prefetchnta [esi + DCTSIZE2*SIZEOF_JCOEF + 1*32] michael@0: prefetchnta [esi + DCTSIZE2*SIZEOF_JCOEF + 2*32] michael@0: prefetchnta [esi + DCTSIZE2*SIZEOF_JCOEF + 3*32] michael@0: michael@0: ; ---- Pass 2: process rows from work array, store into output array. michael@0: michael@0: mov eax, [original_ebp] michael@0: mov edi, JSAMPARRAY [output_buf(eax)] ; (JSAMPROW *) michael@0: mov eax, JDIMENSION [output_col(eax)] michael@0: michael@0: ; -- Even part michael@0: michael@0: ; xmm6=col0, xmm5=col2, xmm1=col4, xmm3=col6 michael@0: michael@0: movdqa xmm2,xmm6 michael@0: movdqa xmm0,xmm5 michael@0: psubw xmm6,xmm1 ; xmm6=tmp11 michael@0: psubw xmm5,xmm3 michael@0: paddw xmm2,xmm1 ; xmm2=tmp10 michael@0: paddw xmm0,xmm3 ; xmm0=tmp13 michael@0: michael@0: psllw xmm5,PRE_MULTIPLY_SCALE_BITS michael@0: pmulhw xmm5,[GOTOFF(ebx,PW_F1414)] michael@0: psubw xmm5,xmm0 ; xmm5=tmp12 michael@0: michael@0: movdqa xmm1,xmm2 michael@0: movdqa xmm3,xmm6 michael@0: psubw xmm2,xmm0 ; xmm2=tmp3 michael@0: psubw xmm6,xmm5 ; xmm6=tmp2 michael@0: paddw xmm1,xmm0 ; xmm1=tmp0 michael@0: paddw xmm3,xmm5 ; xmm3=tmp1 michael@0: michael@0: movdqa xmm0, XMMWORD [wk(0)] ; xmm0=col1 michael@0: movdqa xmm5, XMMWORD [wk(1)] ; xmm5=col3 michael@0: michael@0: movdqa XMMWORD [wk(0)], xmm2 ; wk(0)=tmp3 michael@0: movdqa XMMWORD [wk(1)], xmm6 ; wk(1)=tmp2 michael@0: michael@0: ; -- Odd part michael@0: michael@0: ; xmm0=col1, xmm5=col3, xmm4=col5, xmm7=col7 michael@0: michael@0: movdqa xmm2,xmm0 michael@0: movdqa xmm6,xmm4 michael@0: psubw xmm0,xmm7 ; xmm0=z12 michael@0: psubw xmm4,xmm5 ; xmm4=z10 michael@0: paddw xmm2,xmm7 ; xmm2=z11 michael@0: paddw xmm6,xmm5 ; xmm6=z13 michael@0: michael@0: movdqa xmm7,xmm4 ; xmm7=z10(unscaled) michael@0: psllw xmm0,PRE_MULTIPLY_SCALE_BITS michael@0: psllw xmm4,PRE_MULTIPLY_SCALE_BITS michael@0: michael@0: movdqa xmm5,xmm2 michael@0: psubw xmm2,xmm6 michael@0: paddw xmm5,xmm6 ; xmm5=tmp7 michael@0: michael@0: psllw xmm2,PRE_MULTIPLY_SCALE_BITS michael@0: pmulhw xmm2,[GOTOFF(ebx,PW_F1414)] ; xmm2=tmp11 michael@0: michael@0: ; To avoid overflow... michael@0: ; michael@0: ; (Original) michael@0: ; tmp12 = -2.613125930 * z10 + z5; michael@0: ; michael@0: ; (This implementation) michael@0: ; tmp12 = (-1.613125930 - 1) * z10 + z5; michael@0: ; = -1.613125930 * z10 - z10 + z5; michael@0: michael@0: movdqa xmm6,xmm4 michael@0: paddw xmm4,xmm0 michael@0: pmulhw xmm4,[GOTOFF(ebx,PW_F1847)] ; xmm4=z5 michael@0: pmulhw xmm6,[GOTOFF(ebx,PW_MF1613)] michael@0: pmulhw xmm0,[GOTOFF(ebx,PW_F1082)] michael@0: psubw xmm6,xmm7 michael@0: psubw xmm0,xmm4 ; xmm0=tmp10 michael@0: paddw xmm6,xmm4 ; xmm6=tmp12 michael@0: michael@0: ; -- Final output stage michael@0: michael@0: psubw xmm6,xmm5 ; xmm6=tmp6 michael@0: movdqa xmm7,xmm1 michael@0: movdqa xmm4,xmm3 michael@0: paddw xmm1,xmm5 ; xmm1=data0=(00 10 20 30 40 50 60 70) michael@0: paddw xmm3,xmm6 ; xmm3=data1=(01 11 21 31 41 51 61 71) michael@0: psraw xmm1,(PASS1_BITS+3) ; descale michael@0: psraw xmm3,(PASS1_BITS+3) ; descale michael@0: psubw xmm7,xmm5 ; xmm7=data7=(07 17 27 37 47 57 67 77) michael@0: psubw xmm4,xmm6 ; xmm4=data6=(06 16 26 36 46 56 66 76) michael@0: psraw xmm7,(PASS1_BITS+3) ; descale michael@0: psraw xmm4,(PASS1_BITS+3) ; descale michael@0: psubw xmm2,xmm6 ; xmm2=tmp5 michael@0: michael@0: packsswb xmm1,xmm4 ; xmm1=(00 10 20 30 40 50 60 70 06 16 26 36 46 56 66 76) michael@0: packsswb xmm3,xmm7 ; xmm3=(01 11 21 31 41 51 61 71 07 17 27 37 47 57 67 77) michael@0: michael@0: movdqa xmm5, XMMWORD [wk(1)] ; xmm5=tmp2 michael@0: movdqa xmm6, XMMWORD [wk(0)] ; xmm6=tmp3 michael@0: michael@0: paddw xmm0,xmm2 ; xmm0=tmp4 michael@0: movdqa xmm4,xmm5 michael@0: movdqa xmm7,xmm6 michael@0: paddw xmm5,xmm2 ; xmm5=data2=(02 12 22 32 42 52 62 72) michael@0: paddw xmm6,xmm0 ; xmm6=data4=(04 14 24 34 44 54 64 74) michael@0: psraw xmm5,(PASS1_BITS+3) ; descale michael@0: psraw xmm6,(PASS1_BITS+3) ; descale michael@0: psubw xmm4,xmm2 ; xmm4=data5=(05 15 25 35 45 55 65 75) michael@0: psubw xmm7,xmm0 ; xmm7=data3=(03 13 23 33 43 53 63 73) michael@0: psraw xmm4,(PASS1_BITS+3) ; descale michael@0: psraw xmm7,(PASS1_BITS+3) ; descale michael@0: michael@0: movdqa xmm2,[GOTOFF(ebx,PB_CENTERJSAMP)] ; xmm2=[PB_CENTERJSAMP] michael@0: michael@0: packsswb xmm5,xmm6 ; xmm5=(02 12 22 32 42 52 62 72 04 14 24 34 44 54 64 74) michael@0: packsswb xmm7,xmm4 ; xmm7=(03 13 23 33 43 53 63 73 05 15 25 35 45 55 65 75) michael@0: michael@0: paddb xmm1,xmm2 michael@0: paddb xmm3,xmm2 michael@0: paddb xmm5,xmm2 michael@0: paddb xmm7,xmm2 michael@0: michael@0: movdqa xmm0,xmm1 ; transpose coefficients(phase 1) michael@0: punpcklbw xmm1,xmm3 ; xmm1=(00 01 10 11 20 21 30 31 40 41 50 51 60 61 70 71) michael@0: punpckhbw xmm0,xmm3 ; xmm0=(06 07 16 17 26 27 36 37 46 47 56 57 66 67 76 77) michael@0: movdqa xmm6,xmm5 ; transpose coefficients(phase 1) michael@0: punpcklbw xmm5,xmm7 ; xmm5=(02 03 12 13 22 23 32 33 42 43 52 53 62 63 72 73) michael@0: punpckhbw xmm6,xmm7 ; xmm6=(04 05 14 15 24 25 34 35 44 45 54 55 64 65 74 75) michael@0: michael@0: movdqa xmm4,xmm1 ; transpose coefficients(phase 2) michael@0: punpcklwd xmm1,xmm5 ; xmm1=(00 01 02 03 10 11 12 13 20 21 22 23 30 31 32 33) michael@0: punpckhwd xmm4,xmm5 ; xmm4=(40 41 42 43 50 51 52 53 60 61 62 63 70 71 72 73) michael@0: movdqa xmm2,xmm6 ; transpose coefficients(phase 2) michael@0: punpcklwd xmm6,xmm0 ; xmm6=(04 05 06 07 14 15 16 17 24 25 26 27 34 35 36 37) michael@0: punpckhwd xmm2,xmm0 ; xmm2=(44 45 46 47 54 55 56 57 64 65 66 67 74 75 76 77) michael@0: michael@0: movdqa xmm3,xmm1 ; transpose coefficients(phase 3) michael@0: punpckldq xmm1,xmm6 ; xmm1=(00 01 02 03 04 05 06 07 10 11 12 13 14 15 16 17) michael@0: punpckhdq xmm3,xmm6 ; xmm3=(20 21 22 23 24 25 26 27 30 31 32 33 34 35 36 37) michael@0: movdqa xmm7,xmm4 ; transpose coefficients(phase 3) michael@0: punpckldq xmm4,xmm2 ; xmm4=(40 41 42 43 44 45 46 47 50 51 52 53 54 55 56 57) michael@0: punpckhdq xmm7,xmm2 ; xmm7=(60 61 62 63 64 65 66 67 70 71 72 73 74 75 76 77) michael@0: michael@0: pshufd xmm5,xmm1,0x4E ; xmm5=(10 11 12 13 14 15 16 17 00 01 02 03 04 05 06 07) michael@0: pshufd xmm0,xmm3,0x4E ; xmm0=(30 31 32 33 34 35 36 37 20 21 22 23 24 25 26 27) michael@0: pshufd xmm6,xmm4,0x4E ; xmm6=(50 51 52 53 54 55 56 57 40 41 42 43 44 45 46 47) michael@0: pshufd xmm2,xmm7,0x4E ; xmm2=(70 71 72 73 74 75 76 77 60 61 62 63 64 65 66 67) michael@0: michael@0: mov edx, JSAMPROW [edi+0*SIZEOF_JSAMPROW] michael@0: mov esi, JSAMPROW [edi+2*SIZEOF_JSAMPROW] michael@0: movq XMM_MMWORD [edx+eax*SIZEOF_JSAMPLE], xmm1 michael@0: movq XMM_MMWORD [esi+eax*SIZEOF_JSAMPLE], xmm3 michael@0: mov edx, JSAMPROW [edi+4*SIZEOF_JSAMPROW] michael@0: mov esi, JSAMPROW [edi+6*SIZEOF_JSAMPROW] michael@0: movq XMM_MMWORD [edx+eax*SIZEOF_JSAMPLE], xmm4 michael@0: movq XMM_MMWORD [esi+eax*SIZEOF_JSAMPLE], xmm7 michael@0: michael@0: mov edx, JSAMPROW [edi+1*SIZEOF_JSAMPROW] michael@0: mov esi, JSAMPROW [edi+3*SIZEOF_JSAMPROW] michael@0: movq XMM_MMWORD [edx+eax*SIZEOF_JSAMPLE], xmm5 michael@0: movq XMM_MMWORD [esi+eax*SIZEOF_JSAMPLE], xmm0 michael@0: mov edx, JSAMPROW [edi+5*SIZEOF_JSAMPROW] michael@0: mov esi, JSAMPROW [edi+7*SIZEOF_JSAMPROW] michael@0: movq XMM_MMWORD [edx+eax*SIZEOF_JSAMPLE], xmm6 michael@0: movq XMM_MMWORD [esi+eax*SIZEOF_JSAMPLE], xmm2 michael@0: michael@0: pop edi michael@0: pop esi michael@0: ; pop edx ; need not be preserved michael@0: ; pop ecx ; unused michael@0: poppic ebx michael@0: mov esp,ebp ; esp <- aligned ebp michael@0: pop esp ; esp <- original ebp michael@0: pop ebp michael@0: ret michael@0: michael@0: ; For some reason, the OS X linker does not honor the request to align the michael@0: ; segment unless we do this. michael@0: align 16