/* * Copyright (C) 2020 Linux Studio Plugins Project * (C) 2020 Vladimir Sadovnikov * * This file is part of lsp-dsp-lib * Created on: 31 мар. 2020 г. * * lsp-dsp-lib is free software: you can redistribute it and/or modify * it under the terms of the GNU Lesser General Public License as published by * the Free Software Foundation, either version 3 of the License, or * any later version. * * lsp-dsp-lib is distributed in the hope that it will be useful, * but WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * GNU Lesser General Public License for more details. * * You should have received a copy of the GNU Lesser General Public License * along with lsp-dsp-lib. If not, see . */ #ifndef PRIVATE_DSP_ARCH_X86_AVX_FASTCONV_BUTTERFLY_H_ #define PRIVATE_DSP_ARCH_X86_AVX_FASTCONV_BUTTERFLY_H_ #ifndef PRIVATE_DSP_ARCH_X86_AVX_IMPL #error "This header should not be included directly" #endif /* PRIVATE_DSP_ARCH_X86_AVX_IMPL */ namespace lsp { namespace avx { #define __SKIP(x) // c = a - b // a' = a + b // b' = c * w #define FASTCONV_DIRECT_BUTTERFLY_BODY8(add_re, add_im, FMA_SEL) \ ARCH_X86_ASM \ ( \ /* Prepare angle */ \ __ASM_EMIT("vmovaps 0x00(%[ak]), %%ymm6") /* ymm6 = x_re */ \ __ASM_EMIT("vmovaps 0x20(%[ak]), %%ymm7") /* ymm7 = x_im */ \ /* Start loop */ \ __ASM_EMIT("1:") \ __ASM_EMIT("vmovups 0x00(%[dst], %[off1]), %%ymm0") /* ymm0 = a_re */ \ __ASM_EMIT("vmovups 0x20(%[dst], %[off1]), %%ymm1") /* ymm1 = a_im */ \ __ASM_EMIT("vmovups 0x00(%[dst], %[off2]), %%ymm2") /* ymm2 = b_re */ \ __ASM_EMIT("vmovups 0x20(%[dst], %[off2]), %%ymm3") /* ymm3 = b_im */ \ /* Perform butterfly */ \ __ASM_EMIT("vsubps %%ymm2, %%ymm0, %%ymm4") /* ymm4 = c_re = a_re - b_re */ \ __ASM_EMIT("vsubps %%ymm3, %%ymm1, %%ymm5") /* ymm5 = c_im = a_im - b_im */ \ __ASM_EMIT("vaddps %%ymm2, %%ymm0, %%ymm0") /* ymm0 = a_re' = a_re + b_re */ \ __ASM_EMIT("vaddps %%ymm3, %%ymm1, %%ymm1") /* ymm1 = a_im' = a_im + b_im */ \ __ASM_EMIT("vmulps %%ymm7, %%ymm4, %%ymm2") /* ymm2 = x_im * c_re */ \ __ASM_EMIT("vmulps %%ymm7, %%ymm5, %%ymm3") /* ymm3 = x_im * c_im */ \ __ASM_EMIT(FMA_SEL("vmulps %%ymm6, %%ymm4, %%ymm4", "")) /* ymm4 = x_re * c_re */ \ __ASM_EMIT(FMA_SEL("vmulps %%ymm6, %%ymm5, %%ymm5", "")) /* ymm5 = x_re * c_im */ \ __ASM_EMIT(FMA_SEL(add_re " %%ymm3, %%ymm4, %%ymm4", add_re " %%ymm6, %%ymm3, %%ymm4")) /* ymm4 = b_re = x_re * c_re +- x_im * c_im */ \ __ASM_EMIT(FMA_SEL(add_im " %%ymm2, %%ymm5, %%ymm5", add_im " %%ymm6, %%ymm2, %%ymm5")) /* ymm5 = b_im = x_re * c_im -+ x_im * c_re */ \ /* Store values */ \ __ASM_EMIT("vmovups %%ymm0, 0x00(%[dst], %[off1])") \ __ASM_EMIT("vmovups %%ymm1, 0x20(%[dst], %[off1])") \ __ASM_EMIT("vmovups %%ymm4, 0x00(%[dst], %[off2])") \ __ASM_EMIT("vmovups %%ymm5, 0x20(%[dst], %[off2])") \ __ASM_EMIT("add $0x40, %[off1]") \ __ASM_EMIT("add $0x40, %[off2]") \ __ASM_EMIT32("subl $8, %[np]") \ __ASM_EMIT64("subq $8, %[np]") \ __ASM_EMIT("jz 2f") \ /* Rotate angle */ \ __ASM_EMIT("vmovaps 0x00(%[wk]), %%ymm4") /* xmm4 = w_re */ \ __ASM_EMIT("vmovaps 0x20(%[wk]), %%ymm5") /* xmm5 = w_im */ \ __ASM_EMIT("vmulps %%ymm5, %%ymm6, %%ymm2") /* ymm2 = w_im * x_re */ \ __ASM_EMIT("vmulps %%ymm5, %%ymm7, %%ymm3") /* ymm3 = w_im * x_im */ \ __ASM_EMIT(FMA_SEL("vmulps %%ymm4, %%ymm6, %%ymm6", "")) /* ymm6 = w_re * x_re */ \ __ASM_EMIT(FMA_SEL("vmulps %%ymm4, %%ymm7, %%ymm7", "")) /* ymm7 = w_re * x_im */ \ __ASM_EMIT(FMA_SEL("vsubps %%ymm3, %%ymm6, %%ymm6", "vfmsub132ps %%ymm4, %%ymm3, %%ymm6")) /* ymm6 = x_re' = w_re * x_re - w_im * x_im */ \ __ASM_EMIT(FMA_SEL("vaddps %%ymm2, %%ymm7, %%ymm7", "vfmadd132ps %%ymm4, %%ymm2, %%ymm7")) /* ymm7 = x_im' = w_re * x_im + w_im * x_re */ \ /* Repeat loop */ \ __ASM_EMIT("jmp 1b") \ __ASM_EMIT("2:") \ \ : [off1] "+r" (off1), [off2] "+r" (off2), [np] __ASM_ARG_RW(np) \ : [dst] "r" (dst), [ak] "r" (ak), [wk] "r" (wk) \ : "cc", "memory", \ "%xmm0", "%xmm1", "%xmm2", "%xmm3", \ "%xmm4", "%xmm5", "%xmm6", "%xmm7" \ ); #define FASTCONV_DIRECT_BUTTERFLY_LAST(add_re, add_im, FMA_SEL) \ ARCH_X86_ASM( \ /* Loop 2x 4-element butterflies */ \ __ASM_EMIT("vmovaps 0x00 + %[FFT_A], %%ymm6") /* ymm6 = x_re */ \ __ASM_EMIT("vmovaps 0x20 + %[FFT_A], %%ymm7") /* ymm7 = x_im */ \ __ASM_EMIT("sub $2, %[nb]") \ __ASM_EMIT("jb 2f") \ __ASM_EMIT("1:") \ /* Load data to registers */ \ __ASM_EMIT("vmovups 0x00(%[dst]), %%xmm0") /* xmm0 = r0 r1 r2 r3 */ \ __ASM_EMIT("vmovups 0x10(%[dst]), %%xmm2") /* xmm2 = r4 r5 r6 r7 */ \ __ASM_EMIT("vmovups 0x20(%[dst]), %%xmm1") /* xmm1 = i0 i1 i2 i3 */ \ __ASM_EMIT("vmovups 0x30(%[dst]), %%xmm3") /* xmm3 = i4 i5 i6 i7 */ \ __ASM_EMIT("vinsertf128 $1, 0x40(%[dst]), %%ymm0, %%ymm0") /* ymm0 = a_re = r0 r1 r2 r3 r8 r9 r10 r11 */ \ __ASM_EMIT("vinsertf128 $1, 0x50(%[dst]), %%ymm2, %%ymm2") /* ymm2 = b_re = r4 r5 r6 r7 r12 r13 r14 r15 */ \ __ASM_EMIT("vinsertf128 $1, 0x60(%[dst]), %%ymm1, %%ymm1") /* ymm1 = a_im = i0 i1 i2 i3 i8 i9 i10 i11 */ \ __ASM_EMIT("vinsertf128 $1, 0x70(%[dst]), %%ymm3, %%ymm3") /* ymm3 = b_im = i4 i5 i6 i7 i12 i13 i14 i15 */ \ /* Perform 3rd-order butterflies */ \ __ASM_EMIT("vsubps %%ymm2, %%ymm0, %%ymm4") /* ymm4 = c_re = a_re - b_re */ \ __ASM_EMIT("vsubps %%ymm3, %%ymm1, %%ymm5") /* ymm5 = c_im = a_im - b_im */ \ __ASM_EMIT("vaddps %%ymm2, %%ymm0, %%ymm0") /* ymm0 = a_re' = a_re + b_re */ \ __ASM_EMIT("vaddps %%ymm3, %%ymm1, %%ymm1") /* ymm1 = a_im' = a_im + b_im */ \ __ASM_EMIT("vmulps %%ymm7, %%ymm4, %%ymm2") /* ymm2 = x_im * c_re */ \ __ASM_EMIT("vmulps %%ymm7, %%ymm5, %%ymm3") /* ymm3 = x_im * c_im */ \ __ASM_EMIT(FMA_SEL("vmulps %%ymm6, %%ymm4, %%ymm4", "")) /* ymm4 = x_re * c_re */ \ __ASM_EMIT(FMA_SEL("vmulps %%ymm6, %%ymm5, %%ymm5", "")) /* ymm5 = x_re * c_im */ \ __ASM_EMIT(FMA_SEL(add_re " %%ymm3, %%ymm4, %%ymm4", add_re " %%ymm6, %%ymm3, %%ymm4")) /* ymm4 = b_re = x_re * c_re +- x_im * c_im */ \ __ASM_EMIT(FMA_SEL(add_im " %%ymm2, %%ymm5, %%ymm5", add_im " %%ymm6, %%ymm2, %%ymm5")) /* ymm5 = b_im = x_re * c_im -+ x_im * c_re */ \ /* 2nd-order butterflies */ \ /* s0" = (r0 + r2) + j*(i0 + i2) + (r1 + r3) + j*(i1 + i3) */ \ /* s1" = (r0 + r2) + j*(i0 + i2) - (r1 + r3) - j*(i1 + i3) */ \ /* s2" = (r0 - r2) + j*(i0 - i2) + (i1 - i3) - j*(r1 - r3) */ \ /* s3" = (r0 - r2) + j*(i0 - i2) - (i1 - i3) + j*(r1 - r3) */ \ /* ymm0 = r0 r1 r2 r3 ... */ \ /* ymm1 = i0 i1 i2 i3 ... */ \ /* ymm4 = r4 r5 r6 r7 ... */ \ /* ymm5 = i4 i5 i6 i7 ... */ \ __ASM_EMIT("vshufps $0xd8, %%ymm0, %%ymm0, %%ymm0") /* ymm0 = r0 r2 r1 r3 */ \ __ASM_EMIT("vshufps $0xd8, %%ymm1, %%ymm1, %%ymm1") /* ymm1 = i0 i2 i1 i3 */ \ __ASM_EMIT("vshufps $0xd8, %%ymm4, %%ymm4, %%ymm4") \ __ASM_EMIT("vshufps $0xd8, %%ymm5, %%ymm5, %%ymm5") \ __ASM_EMIT("vhsubps %%ymm1, %%ymm0, %%ymm2") /* ymm2 = r0-r2 r1-r3 i0-i2 i1-i3 = r1' r3' i1' i3' */ \ __ASM_EMIT("vhsubps %%ymm5, %%ymm4, %%ymm3") \ __ASM_EMIT("vhaddps %%ymm1, %%ymm0, %%ymm0") /* ymm0 = r0+r2 r1+r3 i0+i2 i1+i3 = r0' r2' i0' i2' */ \ __ASM_EMIT("vhaddps %%ymm5, %%ymm4, %%ymm4") \ /* 1st-order 8x butterfly */ \ __ASM_EMIT("vshufps $0x6e, %%ymm2, %%ymm0, %%ymm1") /* ymm0 = i0' i2' i1' r3' */ \ __ASM_EMIT("vshufps $0x6e, %%ymm3, %%ymm4, %%ymm5") \ __ASM_EMIT("vshufps $0xc4, %%ymm2, %%ymm0, %%ymm0") /* ymm1 = r0' r2' r1' i3' */ \ __ASM_EMIT("vshufps $0xc4, %%ymm3, %%ymm4, %%ymm4") \ __ASM_EMIT("vhsubps %%ymm1, %%ymm0, %%ymm2") /* ymm2 = r0'-r2' r1'-i3' i0'-i2' i1'-r3' = r1" r3" i1" i2" */ \ __ASM_EMIT("vhsubps %%ymm5, %%ymm4, %%ymm3") \ __ASM_EMIT("vhaddps %%ymm1, %%ymm0, %%ymm0") /* ymm0 = r0'+r2' r1'+i3' i0'+i2' i1'+r3' = r0" r2" i0" i3" */ \ __ASM_EMIT("vhaddps %%ymm5, %%ymm4, %%ymm4") \ __ASM_EMIT("vblendps $0x88, %%ymm0, %%ymm2, %%ymm1") /* ymm1 = r1" r3" i1" i3" */ \ __ASM_EMIT("vblendps $0x88, %%ymm4, %%ymm3, %%ymm5") \ __ASM_EMIT("vblendps $0x88, %%ymm2, %%ymm0, %%ymm0") /* ymm0 = r0" r2" i0" i2" */ \ __ASM_EMIT("vblendps $0x88, %%ymm3, %%ymm4, %%ymm4") \ __ASM_EMIT("vunpckhps %%ymm1, %%ymm0, %%ymm3") /* ymm3 = r0" r1" r2" r3" */ \ __ASM_EMIT("vunpcklps %%ymm1, %%ymm0, %%ymm2") /* ymm2 = i0" i1" i2" i3" */ \ __ASM_EMIT("vunpckhps %%ymm5, %%ymm4, %%ymm1") \ __ASM_EMIT("vunpcklps %%ymm5, %%ymm4, %%ymm0") \ /* Store */ \ __ASM_EMIT("vmovups %%xmm2, 0x00(%[dst])") \ __ASM_EMIT("vmovups %%xmm0, 0x10(%[dst])") \ __ASM_EMIT("vmovups %%xmm3, 0x20(%[dst])") \ __ASM_EMIT("vmovups %%xmm1, 0x30(%[dst])") \ __ASM_EMIT("vextractf128 $1, %%ymm2, 0x40(%[dst])") \ __ASM_EMIT("vextractf128 $1, %%ymm0, 0x50(%[dst])") \ __ASM_EMIT("vextractf128 $1, %%ymm3, 0x60(%[dst])") \ __ASM_EMIT("vextractf128 $1, %%ymm1, 0x70(%[dst])") \ /* Move pointers and repeat*/ \ __ASM_EMIT("add $0x80, %[dst]") \ __ASM_EMIT("sub $2, %[nb]") \ __ASM_EMIT("jae 1b") \ __ASM_EMIT("2:") \ /* 1x 4-element butterflies */ \ __ASM_EMIT("add $1, %[nb]") \ __ASM_EMIT("jl 4f") \ __ASM_EMIT("vmovups 0x00(%[dst]), %%xmm0") /* xmm0 = r0 r1 r2 r3 */ \ __ASM_EMIT("vmovups 0x10(%[dst]), %%xmm2") /* xmm2 = r4 r5 r6 r7 */ \ __ASM_EMIT("vmovups 0x20(%[dst]), %%xmm1") /* xmm1 = i0 i1 i2 i3 */ \ __ASM_EMIT("vmovups 0x30(%[dst]), %%xmm3") /* xmm3 = i4 i5 i6 i7 */ \ /* Perform 3rd-order 8x butterfly */ \ __ASM_EMIT("vsubps %%xmm2, %%xmm0, %%xmm4") /* xmm4 = c_re = a_re - b_re */ \ __ASM_EMIT("vsubps %%xmm3, %%xmm1, %%xmm5") /* xmm5 = c_im = a_im - b_im */ \ __ASM_EMIT("vaddps %%xmm2, %%xmm0, %%xmm0") /* xmm0 = a_re' = a_re + b_re */ \ __ASM_EMIT("vaddps %%xmm3, %%xmm1, %%xmm1") /* xmm1 = a_im' = a_im + b_im */ \ __ASM_EMIT("vmulps %%xmm7, %%xmm4, %%xmm2") /* xmm2 = x_im * c_re */ \ __ASM_EMIT("vmulps %%xmm7, %%xmm5, %%xmm3") /* xmm3 = x_im * c_im */ \ __ASM_EMIT(FMA_SEL("vmulps %%xmm6, %%xmm4, %%xmm4", "")) /* xmm4 = x_re * c_re */ \ __ASM_EMIT(FMA_SEL("vmulps %%xmm6, %%xmm5, %%xmm5", "")) /* xmm5 = x_re * c_im */ \ __ASM_EMIT(FMA_SEL(add_re " %%xmm3, %%xmm4, %%xmm4", add_re " %%xmm6, %%xmm3, %%xmm4")) /* xmm4 = b_re = x_re * c_re +- x_im * c_im */ \ __ASM_EMIT(FMA_SEL(add_im " %%xmm2, %%xmm5, %%xmm5", add_im " %%xmm6, %%xmm2, %%xmm5")) /* xmm5 = b_im = x_re * c_im -+ x_im * c_re */ \ /* 2nd-order butterflies */ \ /* s0" = (r0 + r2) + j*(i0 + i2) + (r1 + r3) + j*(i1 + i3) */ \ /* s1" = (r0 + r2) + j*(i0 + i2) - (r1 + r3) - j*(i1 + i3) */ \ /* s2" = (r0 - r2) + j*(i0 - i2) + (i1 - i3) - j*(r1 - r3) */ \ /* s3" = (r0 - r2) + j*(i0 - i2) - (i1 - i3) + j*(r1 - r3) */ \ /* xmm0 = r0 r1 r2 r3 ... */ \ /* xmm1 = i0 i1 i2 i3 ... */ \ /* xmm4 = r4 r5 r6 r7 ... */ \ /* xmm5 = i4 i5 i6 i7 ... */ \ __ASM_EMIT("vshufps $0xd8, %%xmm0, %%xmm0, %%xmm0") /* xmm0 = r0 r2 r1 r3 */ \ __ASM_EMIT("vshufps $0xd8, %%xmm1, %%xmm1, %%xmm1") /* xmm1 = i0 i2 i1 i3 */ \ __ASM_EMIT("vshufps $0xd8, %%xmm4, %%xmm4, %%xmm4") \ __ASM_EMIT("vshufps $0xd8, %%xmm5, %%xmm5, %%xmm5") \ __ASM_EMIT("vhsubps %%xmm1, %%xmm0, %%xmm2") /* xmm2 = r0-r2 r1-r3 i0-i2 i1-i3 = r1' r3' i1' i3' */ \ __ASM_EMIT("vhsubps %%xmm5, %%xmm4, %%xmm3") \ __ASM_EMIT("vhaddps %%xmm1, %%xmm0, %%xmm0") /* xmm0 = r0+r2 r1+r3 i0+i2 i1+i3 = r0' r2' i0' i2' */ \ __ASM_EMIT("vhaddps %%xmm5, %%xmm4, %%xmm4") \ /* 1st-order butterflies */ \ __ASM_EMIT("vshufps $0x6e, %%xmm2, %%xmm0, %%xmm1") /* xmm0 = i0' i2' i1' r3' */ \ __ASM_EMIT("vshufps $0x6e, %%xmm3, %%xmm4, %%xmm5") \ __ASM_EMIT("vshufps $0xc4, %%xmm2, %%xmm0, %%xmm0") /* xmm1 = r0' r2' r1' i3' */ \ __ASM_EMIT("vshufps $0xc4, %%xmm3, %%xmm4, %%xmm4") \ __ASM_EMIT("vmovups %%xmm0, 0x00(%[dst])") \ __ASM_EMIT("vmovups %%xmm1, 0x10(%[dst])") \ __ASM_EMIT("vmovups %%xmm4, 0x20(%[dst])") \ __ASM_EMIT("vmovups %%xmm5, 0x30(%[dst])") \ __ASM_EMIT("vhsubps %%xmm1, %%xmm0, %%xmm2") /* xmm2 = r0'-r2' r1'-i3' i0'-i2' i1'-r3' = r1" r3" i1" i2" */ \ __ASM_EMIT("vhsubps %%xmm5, %%xmm4, %%xmm3") \ __ASM_EMIT("vhaddps %%xmm1, %%xmm0, %%xmm0") /* xmm0 = r0'+r2' r1'+i3' i0'+i2' i1'+r3' = r0" r2" i0" i3" */ \ __ASM_EMIT("vhaddps %%xmm5, %%xmm4, %%xmm4") \ __ASM_EMIT("vblendps $0x88, %%xmm0, %%xmm2, %%xmm1") /* xmm1 = r1" r3" i1" i3" */ \ __ASM_EMIT("vblendps $0x88, %%xmm4, %%xmm3, %%xmm5") \ __ASM_EMIT("vblendps $0x88, %%xmm2, %%xmm0, %%xmm0") /* xmm0 = r0" r2" i0" i2" */ \ __ASM_EMIT("vblendps $0x88, %%xmm3, %%xmm4, %%xmm4") \ __ASM_EMIT("vunpckhps %%xmm1, %%xmm0, %%xmm3") /* xmm3 = i0" i1" i2" i3" */ \ __ASM_EMIT("vunpcklps %%xmm1, %%xmm0, %%xmm2") /* xmm2 = r0" r1" r2" r3" */ \ __ASM_EMIT("vunpckhps %%xmm5, %%xmm4, %%xmm1") \ __ASM_EMIT("vunpcklps %%xmm5, %%xmm4, %%xmm0") \ /* Store */ \ __ASM_EMIT("vmovups %%xmm2, 0x00(%[dst])") \ __ASM_EMIT("vmovups %%xmm0, 0x10(%[dst])") \ __ASM_EMIT("vmovups %%xmm3, 0x20(%[dst])") \ __ASM_EMIT("vmovups %%xmm1, 0x30(%[dst])") \ __ASM_EMIT("4:") \ \ : [dst] "+r" (dst), [nb] "+r" (nb) \ : [FFT_A] "o" (FFT_A) \ : "cc", "memory", \ "%xmm0", "%xmm1", "%xmm2", "%xmm3", \ "%xmm4", "%xmm5", "%xmm6", "%xmm7" \ ); #define FASTCONV_REVERSE_BUTTERFLY_BODY8(add_re, add_im, FMA_SEL) \ ARCH_X86_ASM \ ( \ /* Prepare angle */ \ __ASM_EMIT("vmovaps 0x00(%[ak]), %%ymm6") /* ymm6 = x_re */ \ __ASM_EMIT("vmovaps 0x20(%[ak]), %%ymm7") /* ymm7 = x_im */ \ /* Start loop */ \ __ASM_EMIT("1:") \ __ASM_EMIT("vmovups 0x00(%[dst], %[off1]), %%ymm0") /* ymm0 = a_re */ \ __ASM_EMIT("vmovups 0x20(%[dst], %[off1]), %%ymm1") /* ymm1 = a_im */ \ __ASM_EMIT("vmovups 0x00(%[dst], %[off2]), %%ymm2") /* ymm2 = b_re */ \ __ASM_EMIT("vmovups 0x20(%[dst], %[off2]), %%ymm3") /* ymm3 = b_im */ \ /* Calculate complex multiplication */ \ __ASM_EMIT("vmulps %%ymm7, %%ymm2, %%ymm4") /* ymm4 = x_im * b_re */ \ __ASM_EMIT("vmulps %%ymm7, %%ymm3, %%ymm5") /* ymm5 = x_im * b_im */ \ __ASM_EMIT(FMA_SEL("vmulps %%ymm6, %%ymm2, %%ymm2", "")) /* ymm2 = x_re * b_re */ \ __ASM_EMIT(FMA_SEL("vmulps %%ymm6, %%ymm3, %%ymm3", "")) /* ymm3 = x_re * b_im */ \ __ASM_EMIT(FMA_SEL(add_re " %%ymm5, %%ymm2, %%ymm5", add_re " %%ymm6, %%ymm2, %%ymm5")) /* ymm5 = c_re = x_re * b_re +- x_im * b_im */ \ __ASM_EMIT(FMA_SEL(add_im " %%ymm4, %%ymm3, %%ymm4", add_im " %%ymm6, %%ymm3, %%ymm4")) /* ymm4 = c_im = x_re * b_im -+ x_im * b_re */ \ /* Perform butterfly */ \ __ASM_EMIT("vsubps %%ymm5, %%ymm0, %%ymm2") /* ymm2 = a_re - c_re */ \ __ASM_EMIT("vsubps %%ymm4, %%ymm1, %%ymm3") /* ymm3 = a_im - c_im */ \ __ASM_EMIT("vaddps %%ymm5, %%ymm0, %%ymm0") /* ymm0 = a_re + c_re */ \ __ASM_EMIT("vaddps %%ymm4, %%ymm1, %%ymm1") /* ymm1 = a_im + c_im */ \ /* Store values */ \ __ASM_EMIT("vmovups %%ymm0, 0x00(%[dst], %[off1])") \ __ASM_EMIT("vmovups %%ymm1, 0x20(%[dst], %[off1])") \ __ASM_EMIT("vmovups %%ymm2, 0x00(%[dst], %[off2])") \ __ASM_EMIT("vmovups %%ymm3, 0x20(%[dst], %[off2])") \ __ASM_EMIT("add $0x40, %[off1]") \ __ASM_EMIT("add $0x40, %[off2]") \ __ASM_EMIT32("subl $8, %[np]") \ __ASM_EMIT64("subq $8, %[np]") \ __ASM_EMIT("jz 2f") \ /* Rotate angle */ \ __ASM_EMIT("vmovaps 0x00(%[wk]), %%ymm4") /* xmm4 = w_re */ \ __ASM_EMIT("vmovaps 0x20(%[wk]), %%ymm5") /* xmm5 = w_im */ \ __ASM_EMIT("vmulps %%ymm5, %%ymm6, %%ymm2") /* ymm2 = w_im * x_re */ \ __ASM_EMIT("vmulps %%ymm5, %%ymm7, %%ymm3") /* ymm3 = w_im * x_im */ \ __ASM_EMIT(FMA_SEL("vmulps %%ymm4, %%ymm6, %%ymm6", "")) /* ymm6 = w_re * x_re */ \ __ASM_EMIT(FMA_SEL("vmulps %%ymm4, %%ymm7, %%ymm7", "")) /* ymm7 = w_re * x_im */ \ __ASM_EMIT(FMA_SEL("vsubps %%ymm3, %%ymm6, %%ymm6", "vfmsub132ps %%ymm4, %%ymm3, %%ymm6")) /* ymm6 = x_re' = w_re * x_re - w_im * x_im */ \ __ASM_EMIT(FMA_SEL("vaddps %%ymm2, %%ymm7, %%ymm7", "vfmadd132ps %%ymm4, %%ymm2, %%ymm7")) /* ymm7 = x_im' = w_re * x_im + w_im * x_re */ \ /* Repeat loop */ \ __ASM_EMIT("jmp 1b") \ __ASM_EMIT("2:") \ \ : [off1] "+r" (off1), [off2] "+r" (off2), [np] __ASM_ARG_RW(np) \ : [dst] "r" (dst), [ak] "r" (ak), [wk] "r" (wk) \ : "cc", "memory", \ "%xmm0", "%xmm1", "%xmm2", "%xmm3", \ "%xmm4", "%xmm5", "%xmm6", "%xmm7" \ ); #define FASTCONV_REVERSE_BUTTERFLY_BODY_LAST(add_re, add_im, FMA_SEL, IF_ADD) \ size_t off; \ float norm = 0.5f / np; \ ARCH_X86_ASM \ ( \ /* Prepare angle */ \ __ASM_EMIT("vbroadcastss %[norm], %%ymm1") /* ymm1 = k */ \ __ASM_EMIT("lea (,%[np], 4), %[off]") /* off = np * 8 */ \ __ASM_EMIT("vmovaps 0x00(%[ak]), %%ymm6") /* ymm6 = x_re */ \ __ASM_EMIT("vmovaps 0x20(%[ak]), %%ymm7") /* ymm7 = x_im */ \ __ASM_EMIT("vmovaps 0x00(%[wk]), %%ymm4") /* xmm4 = w_re */ \ __ASM_EMIT("vmovaps 0x20(%[wk]), %%ymm5") /* xmm5 = w_im */ \ /* Start loop */ \ __ASM_EMIT("1:") \ __ASM_EMIT("vmovups 0x00(%[src]), %%ymm0") /* ymm0 = a_re */ \ __ASM_EMIT("vmovups 0x00(%[src], %[off], 2), %%ymm2") /* ymm2 = b_re */ \ __ASM_EMIT("vmovups 0x20(%[src], %[off], 2), %%ymm3") /* ymm3 = b_im */ \ /* Calculate complex multiplication */ \ __ASM_EMIT("vmulps %%ymm7, %%ymm3, %%ymm3") /* ymm3 = x_im * b_im */ \ __ASM_EMIT(FMA_SEL("vmulps %%ymm6, %%ymm2, %%ymm2", "")) /* ymm2 = x_re * b_re */ \ __ASM_EMIT(FMA_SEL(add_re " %%ymm3, %%ymm2, %%ymm3", add_re " %%ymm6, %%ymm2, %%ymm3")) /* ymm3 = c_re = x_re * b_re +- x_im * b_im */ \ /* Perform butterfly */ \ __ASM_EMIT("vsubps %%ymm3, %%ymm0, %%ymm2") /* ymm2 = a_re - c_re */ \ __ASM_EMIT("vaddps %%ymm3, %%ymm0, %%ymm0") /* ymm0 = a_re + c_re */ \ __ASM_EMIT("vmulps %%ymm1, %%ymm2, %%ymm2") \ __ASM_EMIT("vmulps %%ymm1, %%ymm0, %%ymm0") \ /* Store values */ \ __ASM_EMIT(IF_ADD("vaddps 0x00(%[dst]), %%ymm0, %%ymm0")) \ __ASM_EMIT(IF_ADD("vaddps 0x00(%[dst], %[off]), %%ymm2, %%ymm2")) \ __ASM_EMIT("vmovups %%ymm0, 0x00(%[dst])") \ __ASM_EMIT("vmovups %%ymm2, 0x00(%[dst], %[off])") \ __ASM_EMIT("add $0x40, %[src]") \ __ASM_EMIT("add $0x20, %[dst]") \ __ASM_EMIT32("subl $8, %[np]") \ __ASM_EMIT64("subq $8, %[np]") \ __ASM_EMIT("jbe 2f") \ /* Rotate angle */ \ __ASM_EMIT("vmulps %%ymm5, %%ymm6, %%ymm2") /* ymm2 = w_im * x_re */ \ __ASM_EMIT("vmulps %%ymm5, %%ymm7, %%ymm3") /* ymm3 = w_im * x_im */ \ __ASM_EMIT(FMA_SEL("vmulps %%ymm4, %%ymm6, %%ymm6", "")) /* ymm6 = w_re * x_re */ \ __ASM_EMIT(FMA_SEL("vmulps %%ymm4, %%ymm7, %%ymm7", "")) /* ymm7 = w_re * x_im */ \ __ASM_EMIT(FMA_SEL("vsubps %%ymm3, %%ymm6, %%ymm6", "vfmsub132ps %%ymm4, %%ymm3, %%ymm6")) /* ymm6 = x_re' = w_re * x_re - w_im * x_im */ \ __ASM_EMIT(FMA_SEL("vaddps %%ymm2, %%ymm7, %%ymm7", "vfmadd132ps %%ymm4, %%ymm2, %%ymm7")) /* ymm7 = x_im' = w_re * x_im + w_im * x_re */ \ /* Repeat loop */ \ __ASM_EMIT("jmp 1b") \ __ASM_EMIT("2:") \ \ : [off] "=&r" (off), [np] __ASM_ARG_RW(np) \ : [dst] "r" (dst), [src] "r" (src), [ak] "r" (ak), [wk] "r" (wk), \ [norm] "o" (norm) \ : "cc", "memory", \ "%xmm0", "%xmm1", "%xmm2", "%xmm3", \ "%xmm4", "%xmm5", "%xmm6", "%xmm7" \ ); #define FMA_OFF(a, b) a #define FMA_ON(a, b) b #define FASTCONV_SET(x) #define FASTCONV_ADD(x) x static inline void fastconv_direct_butterfly(float *dst, const float *ak, const float *wk, size_t pairs, size_t nb) { size_t off1, off2, np; off1 = 0; size_t step = pairs << 3; for (size_t i=0; i