libavutil: Add new fixed dsp functions.

Add functions needed for implementation of fixed point aac dec.
Signed-off-by: Nedeljko Babic <nedeljko.babic@imgtec.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

libavutil/fixed_dsp.c

@@ -47,6 +47,28 @@
 
 #include "fixed_dsp.h"
 
+static void vector_fmul_add_c(int *dst, const int *src0, const int *src1, const int *src2, int len){
+    int i;
+    int64_t accu;
+
+    for (i=0; i<len; i++) {
+        accu = (int64_t)src0[i] * src1[i];
+        dst[i] = src2[i] + (int)((accu + 0x40000000) >> 31);
+    }
+}
+
+static void vector_fmul_reverse_c(int *dst, const int *src0, const int *src1, int len)
+{
+    int i;
+    int64_t accu;
+
+    src1 += len-1;
+    for (i=0; i<len; i++) {
+        accu = (int64_t)src0[i] * src1[-i];
+        dst[i] = (int)((accu+0x40000000) >> 31);
+    }
+}
+
 static void vector_fmul_window_scaled_c(int16_t *dst, const int32_t *src0,
                                        const int32_t *src1, const int32_t *win,
                                        int len, uint8_t bits)
@@ -88,6 +110,41 @@ static void vector_fmul_window_c(int32_t *dst, const int32_t *src0,
     }
 }
 
+static void vector_fmul_c(int *dst, const int *src0, const int *src1, int len)
+{
+    int i;
+    int64_t accu;
+
+    for (i = 0; i < len; i++){
+        accu = (int64_t)src0[i] * src1[i];
+        dst[i] = (int)((accu+0x40000000) >> 31);
+    }
+}
+
+static int ff_scalarproduct_fixed_c(const int *v1, const int *v2, int len)
+{
+    /** p is initialized with 0x40000000 so that the proper rounding will occur
+      * at the end */
+    int64_t p = 0x40000000;
+    int i;
+
+    for (i = 0; i < len; i++)
+        p += (int64_t)v1[i] * v2[i];
+
+    return (int)(p >> 31);
+}
+
+static void butterflies_fixed_c(int *v1, int *v2, int len)
+{
+    int i;
+
+    for (i = 0; i < len; i++){
+        int t = v1[i] - v2[i];
+        v1[i] += v2[i];
+        v2[i] = t;
+    }
+}
+
 AVFixedDSPContext * avpriv_alloc_fixed_dsp(int bit_exact)
 {
     AVFixedDSPContext * fdsp = av_malloc(sizeof(AVFixedDSPContext));
@@ -97,6 +154,11 @@ AVFixedDSPContext * avpriv_alloc_fixed_dsp(int bit_exact)
 
     fdsp->vector_fmul_window_scaled = vector_fmul_window_scaled_c;
     fdsp->vector_fmul_window = vector_fmul_window_c;
+    fdsp->vector_fmul = vector_fmul_c;
+    fdsp->vector_fmul_add = vector_fmul_add_c;
+    fdsp->vector_fmul_reverse = vector_fmul_reverse_c;
+    fdsp->butterflies_fixed = butterflies_fixed_c;
+    fdsp->scalarproduct_fixed = ff_scalarproduct_fixed_c;
 
     return fdsp;
 }

libavutil/fixed_dsp.h

@@ -54,8 +54,13 @@
 #include "libavcodec/mathops.h"
 
 typedef struct AVFixedDSPContext {
+    /* Assume len is a multiple of 16, and arrays are 32-byte aligned */
+    /* Results of multiplications are scaled down by 31 bit (and rounded) if not
+     * stated otherwise */
+
     /**
      * Overlap/add with window function.
+     * Result is scaled down by "bits" bits.
      * Used primarily by MDCT-based audio codecs.
      * Source and destination vectors must overlap exactly or not at all.
      *
@@ -92,6 +97,60 @@ typedef struct AVFixedDSPContext {
      */
     void (*vector_fmul_window)(int32_t *dst, const int32_t *src0, const int32_t *src1, const int32_t *win, int len);
 
+    /**
+     * Fixed-point multiplication that calculates the entry wise product of two
+     * vectors of integers and stores the result in a vector of integers.
+     *
+     * @param dst  output vector
+     *             constraints: 32-byte aligned
+     * @param src0 first input vector
+     *             constraints: 32-byte aligned
+     * @param src1 second input vector
+     *             constraints: 32-byte aligned
+     * @param len  number of elements in the input
+     *             constraints: multiple of 16
+     */
+    void (*vector_fmul)(int *dst, const int *src0, const int *src1,
+                        int len);
+
+    void (*vector_fmul_reverse)(int *dst, const int *src0, const int *src1, int len);
+    /**
+     * Calculate the entry wise product of two vectors of integers, add a third vector of
+     * integers and store the result in a vector of integers.
+     *
+     * @param dst  output vector
+     *             constraints: 32-byte aligned
+     * @param src0 first input vector
+     *             constraints: 32-byte aligned
+     * @param src1 second input vector
+     *             constraints: 32-byte aligned
+     * @param src2 third input vector
+     *             constraints: 32-byte aligned
+     * @param len  number of elements in the input
+     *             constraints: multiple of 16
+     */
+    void (*vector_fmul_add)(int *dst, const int *src0, const int *src1,
+                            const int *src2, int len);
+
+    /**
+     * Calculate the scalar product of two vectors of integers.
+     *
+     * @param v1  first vector, 16-byte aligned
+     * @param v2  second vector, 16-byte aligned
+     * @param len length of vectors, multiple of 4
+     *
+     * @return sum of elementwise products
+     */
+    int (*scalarproduct_fixed)(const int *v1, const int *v2, int len);
+
+    /**
+     * Calculate the sum and difference of two vectors of integers.
+     *
+     * @param v1  first input vector, sum output, 16-byte aligned
+     * @param v2  second input vector, difference output, 16-byte aligned
+     * @param len length of vectors, multiple of 4
+     */
+    void (*butterflies_fixed)(int *av_restrict v1, int *av_restrict v2, int len);
 } AVFixedDSPContext;
 
 /**