Mercurial > libavcodec.hg
view iirfilter.c @ 11032:01bd040f8607 libavcodec
Unroll main loop so the edge==0 case is seperate.
This allows many things to be simplified away.
h264 decoder is overall 1% faster with a mbaff sample and
0.1% slower with the cathedral sample, probably because the slow loop
filter code must be loaded into the code cache for each first MB of each
row but isnt used for the following MBs.
author | michael |
---|---|
date | Thu, 28 Jan 2010 01:24:25 +0000 |
parents | 0448219eb490 |
children | 7dd2a45249a9 |
line wrap: on
line source
/* * IIR filter * Copyright (c) 2008 Konstantin Shishkov * * This file is part of FFmpeg. * * FFmpeg 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 2.1 of the License, or (at your option) any later version. * * FFmpeg 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 FFmpeg; if not, write to the Free Software * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA */ /** * @file libavcodec/iirfilter.c * different IIR filters implementation */ #include "iirfilter.h" #include <math.h> /** * IIR filter global parameters */ typedef struct FFIIRFilterCoeffs{ int order; float gain; int *cx; float *cy; }FFIIRFilterCoeffs; /** * IIR filter state */ typedef struct FFIIRFilterState{ float x[1]; }FFIIRFilterState; /// maximum supported filter order #define MAXORDER 30 av_cold struct FFIIRFilterCoeffs* ff_iir_filter_init_coeffs(enum IIRFilterType filt_type, enum IIRFilterMode filt_mode, int order, float cutoff_ratio, float stopband, float ripple) { int i, j; FFIIRFilterCoeffs *c; double wa; double p[MAXORDER + 1][2]; if(filt_type != FF_FILTER_TYPE_BUTTERWORTH || filt_mode != FF_FILTER_MODE_LOWPASS) return NULL; if(order <= 1 || (order & 1) || order > MAXORDER || cutoff_ratio >= 1.0) return NULL; c = av_malloc(sizeof(FFIIRFilterCoeffs)); c->cx = av_malloc(sizeof(c->cx[0]) * ((order >> 1) + 1)); c->cy = av_malloc(sizeof(c->cy[0]) * order); c->order = order; wa = 2 * tan(M_PI * 0.5 * cutoff_ratio); c->cx[0] = 1; for(i = 1; i < (order >> 1) + 1; i++) c->cx[i] = c->cx[i - 1] * (order - i + 1LL) / i; p[0][0] = 1.0; p[0][1] = 0.0; for(i = 1; i <= order; i++) p[i][0] = p[i][1] = 0.0; for(i = 0; i < order; i++){ double zp[2]; double th = (i + (order >> 1) + 0.5) * M_PI / order; double a_re, a_im, c_re, c_im; zp[0] = cos(th) * wa; zp[1] = sin(th) * wa; a_re = zp[0] + 2.0; c_re = zp[0] - 2.0; a_im = c_im = zp[1]; zp[0] = (a_re * c_re + a_im * c_im) / (c_re * c_re + c_im * c_im); zp[1] = (a_im * c_re - a_re * c_im) / (c_re * c_re + c_im * c_im); for(j = order; j >= 1; j--) { a_re = p[j][0]; a_im = p[j][1]; p[j][0] = a_re*zp[0] - a_im*zp[1] + p[j-1][0]; p[j][1] = a_re*zp[1] + a_im*zp[0] + p[j-1][1]; } a_re = p[0][0]*zp[0] - p[0][1]*zp[1]; p[0][1] = p[0][0]*zp[1] + p[0][1]*zp[0]; p[0][0] = a_re; } c->gain = p[order][0]; for(i = 0; i < order; i++){ c->gain += p[i][0]; c->cy[i] = (-p[i][0] * p[order][0] + -p[i][1] * p[order][1]) / (p[order][0] * p[order][0] + p[order][1] * p[order][1]); } c->gain /= 1 << order; return c; } av_cold struct FFIIRFilterState* ff_iir_filter_init_state(int order) { FFIIRFilterState* s = av_mallocz(sizeof(FFIIRFilterState) + sizeof(s->x[0]) * (order - 1)); return s; } #define FILTER(i0, i1, i2, i3) \ in = *src * c->gain \ + c->cy[0]*s->x[i0] + c->cy[1]*s->x[i1] \ + c->cy[2]*s->x[i2] + c->cy[3]*s->x[i3]; \ res = (s->x[i0] + in )*1 \ + (s->x[i1] + s->x[i3])*4 \ + s->x[i2] *6; \ *dst = av_clip_int16(lrintf(res)); \ s->x[i0] = in; \ src += sstep; \ dst += dstep; \ void ff_iir_filter(const struct FFIIRFilterCoeffs *c, struct FFIIRFilterState *s, int size, const int16_t *src, int sstep, int16_t *dst, int dstep) { int i; if(c->order == 4){ for(i = 0; i < size; i += 4){ float in, res; FILTER(0, 1, 2, 3); FILTER(1, 2, 3, 0); FILTER(2, 3, 0, 1); FILTER(3, 0, 1, 2); } }else{ for(i = 0; i < size; i++){ int j; float in, res; in = *src * c->gain; for(j = 0; j < c->order; j++) in += c->cy[j] * s->x[j]; res = s->x[0] + in + s->x[c->order >> 1] * c->cx[c->order >> 1]; for(j = 1; j < c->order >> 1; j++) res += (s->x[j] + s->x[c->order - j]) * c->cx[j]; for(j = 0; j < c->order - 1; j++) s->x[j] = s->x[j + 1]; *dst = av_clip_int16(lrintf(res)); s->x[c->order - 1] = in; src += sstep; dst += sstep; } } } av_cold void ff_iir_filter_free_state(struct FFIIRFilterState *state) { av_free(state); } av_cold void ff_iir_filter_free_coeffs(struct FFIIRFilterCoeffs *coeffs) { if(coeffs){ av_free(coeffs->cx); av_free(coeffs->cy); } av_free(coeffs); } #ifdef TEST #define FILT_ORDER 4 #define SIZE 1024 int main(void) { struct FFIIRFilterCoeffs *fcoeffs = NULL; struct FFIIRFilterState *fstate = NULL; float cutoff_coeff = 0.4; int16_t x[SIZE], y[SIZE]; int i; FILE* fd; fcoeffs = ff_iir_filter_init_coeffs(FF_FILTER_TYPE_BUTTERWORTH, FF_FILTER_MODE_LOWPASS, FILT_ORDER, cutoff_coeff, 0.0, 0.0); fstate = ff_iir_filter_init_state(FILT_ORDER); for (i = 0; i < SIZE; i++) { x[i] = lrint(0.75 * INT16_MAX * sin(0.5*M_PI*i*i/SIZE)); } ff_iir_filter(fcoeffs, fstate, SIZE, x, 1, y, 1); fd = fopen("in.bin", "w"); fwrite(x, sizeof(x[0]), SIZE, fd); fclose(fd); fd = fopen("out.bin", "w"); fwrite(y, sizeof(y[0]), SIZE, fd); fclose(fd); ff_iir_filter_free_coeffs(fcoeffs); ff_iir_filter_free_state(fstate); return 0; } #endif /* TEST */