; ; jquanti.asm - sample data conversion and quantization (SSE2) ; ; Copyright 2009 Pierre Ossman for Cendio AB ; Copyright (C) 2016, D. R. Commander. ; ; Based on the x86 SIMD extension for IJG JPEG library ; Copyright (C) 1999-2006, MIYASAKA Masaru. ; For conditions of distribution and use, see copyright notice in jsimdext.inc ; ; This file should be assembled with NASM (Netwide Assembler), ; can *not* be assembled with Microsoft's MASM or any compatible ; assembler (including Borland's Turbo Assembler). ; NASM is available from http://nasm.sourceforge.net/ or ; http://sourceforge.net/project/showfiles.php?group_id=6208 %include "jsimdext.inc" %include "jdct.inc" ; -------------------------------------------------------------------------- SECTION SEG_TEXT BITS 32 ; ; Load data into workspace, applying unsigned->signed conversion ; ; GLOBAL(void) ; jsimd_convsamp_sse2(JSAMPARRAY sample_data, JDIMENSION start_col, ; DCTELEM *workspace); ; %define sample_data ebp + 8 ; JSAMPARRAY sample_data %define start_col ebp + 12 ; JDIMENSION start_col %define workspace ebp + 16 ; DCTELEM *workspace align 32 GLOBAL_FUNCTION(jsimd_convsamp_sse2) EXTN(jsimd_convsamp_sse2): push ebp mov ebp, esp push ebx ; push ecx ; need not be preserved ; push edx ; need not be preserved push esi push edi pxor xmm6, xmm6 ; xmm6=(all 0's) pcmpeqw xmm7, xmm7 psllw xmm7, 7 ; xmm7={0xFF80 0xFF80 0xFF80 0xFF80 ..} mov esi, JSAMPARRAY [sample_data] ; (JSAMPROW *) mov eax, JDIMENSION [start_col] mov edi, POINTER [workspace] ; (DCTELEM *) mov ecx, DCTSIZE/4 alignx 16, 7 .convloop: mov ebx, JSAMPROW [esi+0*SIZEOF_JSAMPROW] ; (JSAMPLE *) mov edx, JSAMPROW [esi+1*SIZEOF_JSAMPROW] ; (JSAMPLE *) movq xmm0, XMM_MMWORD [ebx+eax*SIZEOF_JSAMPLE] ; xmm0=(01234567) movq xmm1, XMM_MMWORD [edx+eax*SIZEOF_JSAMPLE] ; xmm1=(89ABCDEF) mov ebx, JSAMPROW [esi+2*SIZEOF_JSAMPROW] ; (JSAMPLE *) mov edx, JSAMPROW [esi+3*SIZEOF_JSAMPROW] ; (JSAMPLE *) movq xmm2, XMM_MMWORD [ebx+eax*SIZEOF_JSAMPLE] ; xmm2=(GHIJKLMN) movq xmm3, XMM_MMWORD [edx+eax*SIZEOF_JSAMPLE] ; xmm3=(OPQRSTUV) punpcklbw xmm0, xmm6 ; xmm0=(01234567) punpcklbw xmm1, xmm6 ; xmm1=(89ABCDEF) paddw xmm0, xmm7 paddw xmm1, xmm7 punpcklbw xmm2, xmm6 ; xmm2=(GHIJKLMN) punpcklbw xmm3, xmm6 ; xmm3=(OPQRSTUV) paddw xmm2, xmm7 paddw xmm3, xmm7 movdqa XMMWORD [XMMBLOCK(0,0,edi,SIZEOF_DCTELEM)], xmm0 movdqa XMMWORD [XMMBLOCK(1,0,edi,SIZEOF_DCTELEM)], xmm1 movdqa XMMWORD [XMMBLOCK(2,0,edi,SIZEOF_DCTELEM)], xmm2 movdqa XMMWORD [XMMBLOCK(3,0,edi,SIZEOF_DCTELEM)], xmm3 add esi, byte 4*SIZEOF_JSAMPROW add edi, byte 4*DCTSIZE*SIZEOF_DCTELEM dec ecx jnz short .convloop pop edi pop esi ; pop edx ; need not be preserved ; pop ecx ; need not be preserved pop ebx pop ebp ret ; -------------------------------------------------------------------------- ; ; Quantize/descale the coefficients, and store into coef_block ; ; This implementation is based on an algorithm described in ; "How to optimize for the Pentium family of microprocessors" ; (http://www.agner.org/assem/). ; ; GLOBAL(void) ; jsimd_quantize_sse2(JCOEFPTR coef_block, DCTELEM *divisors, ; DCTELEM *workspace); ; %define RECIPROCAL(m, n, b) \ XMMBLOCK(DCTSIZE * 0 + (m), (n), (b), SIZEOF_DCTELEM) %define CORRECTION(m, n, b) \ XMMBLOCK(DCTSIZE * 1 + (m), (n), (b), SIZEOF_DCTELEM) %define SCALE(m, n, b) \ XMMBLOCK(DCTSIZE * 2 + (m), (n), (b), SIZEOF_DCTELEM) %define coef_block ebp + 8 ; JCOEFPTR coef_block %define divisors ebp + 12 ; DCTELEM *divisors %define workspace ebp + 16 ; DCTELEM *workspace align 32 GLOBAL_FUNCTION(jsimd_quantize_sse2) EXTN(jsimd_quantize_sse2): push ebp mov ebp, esp ; push ebx ; unused ; push ecx ; unused ; push edx ; need not be preserved push esi push edi mov esi, POINTER [workspace] mov edx, POINTER [divisors] mov edi, JCOEFPTR [coef_block] mov eax, DCTSIZE2/32 alignx 16, 7 .quantloop: movdqa xmm4, XMMWORD [XMMBLOCK(0,0,esi,SIZEOF_DCTELEM)] movdqa xmm5, XMMWORD [XMMBLOCK(1,0,esi,SIZEOF_DCTELEM)] movdqa xmm6, XMMWORD [XMMBLOCK(2,0,esi,SIZEOF_DCTELEM)] movdqa xmm7, XMMWORD [XMMBLOCK(3,0,esi,SIZEOF_DCTELEM)] movdqa xmm0, xmm4 movdqa xmm1, xmm5 movdqa xmm2, xmm6 movdqa xmm3, xmm7 psraw xmm4, (WORD_BIT-1) psraw xmm5, (WORD_BIT-1) psraw xmm6, (WORD_BIT-1) psraw xmm7, (WORD_BIT-1) pxor xmm0, xmm4 pxor xmm1, xmm5 pxor xmm2, xmm6 pxor xmm3, xmm7 psubw xmm0, xmm4 ; if (xmm0 < 0) xmm0 = -xmm0; psubw xmm1, xmm5 ; if (xmm1 < 0) xmm1 = -xmm1; psubw xmm2, xmm6 ; if (xmm2 < 0) xmm2 = -xmm2; psubw xmm3, xmm7 ; if (xmm3 < 0) xmm3 = -xmm3; paddw xmm0, XMMWORD [CORRECTION(0,0,edx)] ; correction + roundfactor paddw xmm1, XMMWORD [CORRECTION(1,0,edx)] paddw xmm2, XMMWORD [CORRECTION(2,0,edx)] paddw xmm3, XMMWORD [CORRECTION(3,0,edx)] pmulhuw xmm0, XMMWORD [RECIPROCAL(0,0,edx)] ; reciprocal pmulhuw xmm1, XMMWORD [RECIPROCAL(1,0,edx)] pmulhuw xmm2, XMMWORD [RECIPROCAL(2,0,edx)] pmulhuw xmm3, XMMWORD [RECIPROCAL(3,0,edx)] pmulhuw xmm0, XMMWORD [SCALE(0,0,edx)] ; scale pmulhuw xmm1, XMMWORD [SCALE(1,0,edx)] pmulhuw xmm2, XMMWORD [SCALE(2,0,edx)] pmulhuw xmm3, XMMWORD [SCALE(3,0,edx)] pxor xmm0, xmm4 pxor xmm1, xmm5 pxor xmm2, xmm6 pxor xmm3, xmm7 psubw xmm0, xmm4 psubw xmm1, xmm5 psubw xmm2, xmm6 psubw xmm3, xmm7 movdqa XMMWORD [XMMBLOCK(0,0,edi,SIZEOF_DCTELEM)], xmm0 movdqa XMMWORD [XMMBLOCK(1,0,edi,SIZEOF_DCTELEM)], xmm1 movdqa XMMWORD [XMMBLOCK(2,0,edi,SIZEOF_DCTELEM)], xmm2 movdqa XMMWORD [XMMBLOCK(3,0,edi,SIZEOF_DCTELEM)], xmm3 add esi, byte 32*SIZEOF_DCTELEM add edx, byte 32*SIZEOF_DCTELEM add edi, byte 32*SIZEOF_JCOEF dec eax jnz near .quantloop pop edi pop esi ; pop edx ; need not be preserved ; pop ecx ; unused ; pop ebx ; unused pop ebp ret ; For some reason, the OS X linker does not honor the request to align the ; segment unless we do this. align 32