From 6a7c4d60a1498929c2a366f2ef4ccc35621a4358 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Denis-Courmont?= <remi@remlab.net>
Date: Wed, 29 May 2024 17:43:21 +0300
Subject: [PATCH] lavu/float_dsp: add double-precision scalar product

The function pointer is appended to the structure for backward binary
compatibility. Fortunately, this is allocated by libavutil, not by the
user, so increasing the structure size is safe.
---
 libavutil/float_dsp.c | 12 ++++++++++++
 libavutil/float_dsp.h | 34 ++++++++++++++++++++++++++++++----
 2 files changed, 42 insertions(+), 4 deletions(-)

diff --git a/libavutil/float_dsp.c b/libavutil/float_dsp.c
index e9fb023466..08bbc85e3e 100644
--- a/libavutil/float_dsp.c
+++ b/libavutil/float_dsp.c
@@ -132,6 +132,17 @@ float avpriv_scalarproduct_float_c(const float *v1, const float *v2, int len)
     return p;
 }
 
+double ff_scalarproduct_double_c(const double *v1, const double *v2,
+                                 size_t len)
+{
+    double p = 0.0;
+
+    for (size_t i = 0; i < len; i++)
+        p += v1[i] * v2[i];
+
+    return p;
+}
+
 av_cold AVFloatDSPContext *avpriv_float_dsp_alloc(int bit_exact)
 {
     AVFloatDSPContext *fdsp = av_mallocz(sizeof(AVFloatDSPContext));
@@ -149,6 +160,7 @@ av_cold AVFloatDSPContext *avpriv_float_dsp_alloc(int bit_exact)
     fdsp->vector_fmul_reverse = vector_fmul_reverse_c;
     fdsp->butterflies_float = butterflies_float_c;
     fdsp->scalarproduct_float = avpriv_scalarproduct_float_c;
+    fdsp->scalarproduct_double = ff_scalarproduct_double_c;
 
 #if ARCH_AARCH64
     ff_float_dsp_init_aarch64(fdsp);
diff --git a/libavutil/float_dsp.h b/libavutil/float_dsp.h
index 342a8715c5..ca05fd9382 100644
--- a/libavutil/float_dsp.h
+++ b/libavutil/float_dsp.h
@@ -19,6 +19,8 @@
 #ifndef AVUTIL_FLOAT_DSP_H
 #define AVUTIL_FLOAT_DSP_H
 
+#include <stddef.h>
+
 typedef struct AVFloatDSPContext {
     /**
      * Calculate the entry wise product of two vectors of floats and store the result in
@@ -177,20 +179,32 @@ typedef struct AVFloatDSPContext {
      * a vector of doubles.
      *
      * @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_dmul)(double *dst, const double *src0, const double *src1,
                         int len);
+
+    /**
+     * Calculate the scalar product of two vectors of doubles.
+     *
+     * @param v1  first vector
+     *             constraints: 32-byte aligned
+     * @param v2  second vector
+     *             constraints: 32-byte aligned
+     * @param len length of vectors
+     *             constraints: 32-byte aligned
+     *
+     * @return inner product of the vectors
+     */
+    double (*scalarproduct_double)(const double *v1, const double *v2,
+                                   size_t len);
 } AVFloatDSPContext;
 
 /**
- * Return the scalar product of two vectors.
+ * Return the scalar product of two vectors of floats.
  *
  * @param v1  first input vector
  * @param v2  first input vector
@@ -200,6 +214,18 @@ typedef struct AVFloatDSPContext {
  */
 float avpriv_scalarproduct_float_c(const float *v1, const float *v2, int len);
 
+/**
+ * Return the scalar product of two vectors of doubles.
+ *
+ * @param v1  first input vector
+ * @param v2  first input vector
+ * @param len number of elements
+ *
+ * @return inner product of the vectors
+ */
+double ff_scalarproduct_double_c(const double *v1, const double *v2,
+                                 size_t len);
+
 void ff_float_dsp_init_aarch64(AVFloatDSPContext *fdsp);
 void ff_float_dsp_init_arm(AVFloatDSPContext *fdsp);
 void ff_float_dsp_init_ppc(AVFloatDSPContext *fdsp, int strict);