libavutil: Add fixed_dsp

Signed-off-by: Nedeljko Babic <nbabic@mips.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

libavutil/Makefile

@@ -89,6 +89,7 @@ OBJS = adler32.o                                                        \
        file.o                                                           \
        file_open.o                                                      \
        float_dsp.o                                                      \
+       fixed_dsp.o                                                      \
        frame.o                                                          \
        hash.o                                                           \
        hmac.o                                                           \

libavutil/fixed_dsp.c

@@ -0,0 +1,98 @@
+/*
+ * Copyright (c) 2012
+ *      MIPS Technologies, Inc., California.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ * 3. Neither the name of the MIPS Technologies, Inc., nor the names of its
+ *    contributors may be used to endorse or promote products derived from
+ *    this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE MIPS TECHNOLOGIES, INC. ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE MIPS TECHNOLOGIES, INC. BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ *
+ * Author:  Nedeljko Babic (nbabic@mips.com)
+ *
+ * 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
+ */
+
+#include "fixed_dsp.h"
+
+static void vector_fmul_window_fixed_scaled_c(int16_t *dst, const int32_t *src0,
+                                       const int32_t *src1, const int32_t *win,
+                                       int len, uint8_t bits)
+{
+    int32_t s0, s1, wi, wj, i,j, round;
+
+    dst += len;
+    win += len;
+    src0+= len;
+    round = bits? 1 << (bits-1) : 0;
+
+    for (i=-len, j=len-1; i<0; i++, j--) {
+        s0 = src0[i];
+        s1 = src1[j];
+        wi = win[i];
+        wj = win[j];
+        dst[i] = av_clip_int16_c(((((int64_t)s0*wj - (int64_t)s1*wi + 0x40000000) >> 31) + round) >> bits);
+        dst[j] = av_clip_int16_c(((((int64_t)s0*wi + (int64_t)s1*wj + 0x40000000) >> 31) + round) >> bits);
+    }
+}
+
+static void vector_fmul_window_fixed_c(int32_t *dst, const int32_t *src0,
+                                       const int32_t *src1, const int32_t *win,
+                                       int len)
+{
+    int32_t s0, s1, wi, wj, i, j;
+
+    dst += len;
+    win += len;
+    src0+= len;
+
+    for (i=-len, j=len-1; i<0; i++, j--) {
+        s0 = src0[i];
+        s1 = src1[j];
+        wi = win[i];
+        wj = win[j];
+        dst[i] = ((int64_t)s0*wj - (int64_t)s1*wi + 0x40000000) >> 31;
+        dst[j] = ((int64_t)s0*wi + (int64_t)s1*wj + 0x40000000) >> 31;
+    }
+}
+
+AVFixedDSPContext * avpriv_alloc_fixed_dsp(int bit_exact)
+{
+    AVFixedDSPContext * fdsp = (AVFixedDSPContext*)av_malloc(sizeof(AVFixedDSPContext));
+    fdsp->vector_fmul_window_scaled = vector_fmul_window_fixed_scaled_c;
+    fdsp->vector_fmul_window = vector_fmul_window_fixed_c;
+
+    return fdsp;
+}

libavutil/fixed_dsp.h

@@ -0,0 +1,144 @@
+/*
+ * Copyright (c) 2012
+ *      MIPS Technologies, Inc., California.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ * 3. Neither the name of the MIPS Technologies, Inc., nor the names of its
+ *    contributors may be used to endorse or promote products derived from
+ *    this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE MIPS TECHNOLOGIES, INC. ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE MIPS TECHNOLOGIES, INC. BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ *
+ * Author:  Nedeljko Babic (nbabic@mips.com)
+ *
+ * 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
+ */
+
+#ifndef AVUTIL_FIXED_DSP_H
+#define AVUTIL_FIXED_DSP_H
+
+#include <stdint.h>
+#include "attributes.h"
+#include "common.h"
+#include "libavcodec/mathops.h"
+
+typedef struct AVFixedDSPContext {
+    /**
+     * Overlap/add with window function.
+     * Used primarily by MDCT-based audio codecs.
+     * Source and destination vectors must overlap exactly or not at all.
+     *
+     * @param dst  result vector
+     *             constraints: 16-byte aligned
+     * @param src0 first source vector
+     *             constraints: 16-byte aligned
+     * @param src1 second source vector
+     *             constraints: 16-byte aligned
+     * @param win  half-window vector
+     *             constraints: 16-byte aligned
+     * @param len  length of vector
+     *             constraints: multiple of 4
+     * @param bits scaling parameter
+     *
+     */
+    void (*vector_fmul_window_scaled)(int16_t *dst, const int32_t *src0, const int32_t *src1, const int32_t *win, int len, uint8_t bits);
+
+    /**
+     * Overlap/add with window function.
+     * Used primarily by MDCT-based audio codecs.
+     * Source and destination vectors must overlap exactly or not at all.
+     *
+     * @param dst  result vector
+     *             constraints: 32-byte aligned
+     * @param src0 first source vector
+     *             constraints: 16-byte aligned
+     * @param src1 second source vector
+     *             constraints: 16-byte aligned
+     * @param win  half-window vector
+     *             constraints: 16-byte aligned
+     * @param len  length of vector
+     *             constraints: multiple of 4
+     */
+    void (*vector_fmul_window)(int32_t *dst, const int32_t *src0, const int32_t *src1, const int32_t *win, int len);
+
+} AVFixedDSPContext;
+
+/**
+ * Allocate and initialize a fixed DSP context.
+ * note: should be freed with a av_free call when no longer needed.
+ *
+ * @param strict  setting to non-zero avoids using functions which may not be IEEE-754 compliant
+ */
+AVFixedDSPContext * avpriv_alloc_fixed_dsp(int strict);
+
+/**
+ * Calculate the square root
+ *
+ * @param x    input fixed point number
+ *
+ * @param bits format of fixed point number (32 - bits).bits
+ *
+ * note: input is normalized to (0, 1) fixed point value
+ */
+
+static av_always_inline int fixed_sqrt(int x, int bits)
+{
+    int retval, bit_mask, guess, square, i;
+    int64_t accu;
+    int shift1 = 30 - bits;
+    int shift2 = bits - 15;
+
+    if (shift1 > 0) retval = ff_sqrt(x << shift1);
+    else retval = ff_sqrt(x >> -shift1);
+
+    if (shift2 > 0) {
+        retval = retval << shift2;
+        bit_mask = (1 << (shift2 - 1));
+
+        for (i=0; i<shift2; i++){
+            guess = retval + bit_mask;
+            accu = (int64_t)guess * guess;
+            square = (int)((accu + bit_mask) >> bits);
+            if (x >= square)
+                retval += bit_mask;
+            bit_mask >>= 1;
+        }
+
+    }
+    else retval >>= (-shift2);
+
+    return retval;
+}
+
+#endif /* AVUTIL_FIXED_DSP_H */

libavutil/version.h

@@ -56,7 +56,7 @@
  */
 
 #define LIBAVUTIL_VERSION_MAJOR  52
-#define LIBAVUTIL_VERSION_MINOR  70
+#define LIBAVUTIL_VERSION_MINOR  71
 #define LIBAVUTIL_VERSION_MICRO 100
 
 #define LIBAVUTIL_VERSION_INT   AV_VERSION_INT(LIBAVUTIL_VERSION_MAJOR, \