Fixing puppets

In kernels which output is shorter than num_points, memory is firtly
filled bu zeros and then the kernel is executed.

src/algorithms/libs/volk_gnsssdr_module/volk_gnsssdr/kernels/volk_gnsssdr/volk_gnsssdr_16ic_s32fc_x2_rotator_16ic.h

@@ -40,8 +40,6 @@
 #include <math.h>
 #include <stdio.h>
 
-#define ROTATOR_RELOAD 512
-
 
 #ifdef LV_HAVE_GENERIC
 
@@ -56,20 +54,9 @@
 static inline void volk_gnsssdr_16ic_s32fc_x2_rotator_16ic_generic(lv_16sc_t* outVector, const lv_16sc_t* inVector, const lv_32fc_t phase_inc, lv_32fc_t* phase, unsigned int num_points)
 {
     unsigned int i = 0;
-    int j = 0;
     lv_16sc_t tmp16;
     lv_32fc_t tmp32;
-    for(i = 0; i < (unsigned int)(num_points / ROTATOR_RELOAD); ++i)
+    for(i = 0; i < (unsigned int)(num_points); ++i)
-        {
-            for(j = 0; j < ROTATOR_RELOAD; ++j)
-                {
-                    tmp16 = *inVector++;
-                    tmp32 = lv_cmake((float)lv_creal(tmp16), (float)lv_cimag(tmp16)) * (*phase);
-                    *outVector++ = lv_cmake((int16_t)rintf(lv_creal(tmp32)), (int16_t)rintf(lv_cimag(tmp32)));
-                    (*phase) *= phase_inc;
-                }
-        }
-    for(i = 0; i < num_points % ROTATOR_RELOAD; ++i)
         {
             tmp16 = *inVector++;
             tmp32 = lv_cmake((float)lv_creal(tmp16), (float)lv_cimag(tmp16)) * (*phase);

src/algorithms/libs/volk_gnsssdr_module/volk_gnsssdr/kernels/volk_gnsssdr/volk_gnsssdr_16ic_x2_dotprodxnpuppet_16ic.h

@@ -51,6 +51,7 @@ static inline void volk_gnsssdr_16ic_x2_dotprodxnpuppet_16ic_generic(lv_16sc_t*
        in_a[n] = (lv_16sc_t*)volk_gnsssdr_malloc(sizeof(lv_16sc_t) * num_points, volk_gnsssdr_get_alignment());
        memcpy(in_a[n], in, sizeof(lv_16sc_t) * num_points);
     }
+    result = (lv_16sc_t*)calloc(num_points, sizeof(lv_16sc_t));
     volk_gnsssdr_16ic_x2_dot_prod_16ic_xn_generic(result, local_code, (const lv_16sc_t**) in_a, num_a_vectors, num_points);
 
     for(unsigned int n = 0; n < num_a_vectors; n++)
@@ -72,6 +73,7 @@ static inline void volk_gnsssdr_16ic_x2_dotprodxnpuppet_16ic_a_sse2(lv_16sc_t* r
        in_a[n] = (lv_16sc_t*)volk_gnsssdr_malloc(sizeof(lv_16sc_t) * num_points, volk_gnsssdr_get_alignment());
        memcpy((lv_16sc_t*)in_a[n], (lv_16sc_t*)in, sizeof(lv_16sc_t) * num_points);
     }
+    result = (lv_16sc_t*)calloc(num_points, sizeof(lv_16sc_t));
     volk_gnsssdr_16ic_x2_dot_prod_16ic_xn_a_sse2(result, local_code, (const lv_16sc_t**) in_a, num_a_vectors, num_points);
 
     for(unsigned int n = 0; n < num_a_vectors; n++)
@@ -94,6 +96,7 @@ static inline void volk_gnsssdr_16ic_x2_dotprodxnpuppet_16ic_u_sse2(lv_16sc_t* r
        in_a[n] = (lv_16sc_t*)volk_gnsssdr_malloc(sizeof(lv_16sc_t)*num_points, volk_gnsssdr_get_alignment());
        memcpy(in_a[n], in, sizeof(lv_16sc_t)*num_points);
     }
+    result = (lv_16sc_t*)calloc(num_points, sizeof(lv_16sc_t));
     volk_gnsssdr_16ic_x2_dot_prod_16ic_xn_u_sse2(result, local_code, (const lv_16sc_t**) in_a, num_a_vectors, num_points);
 
     for(unsigned int n = 0; n < num_a_vectors; n++)
@@ -116,6 +119,7 @@ static inline void volk_gnsssdr_16ic_x2_dotprodxnpuppet_16ic_neon(lv_16sc_t* res
        in_a[n] = (lv_16sc_t*)volk_gnsssdr_malloc(sizeof(lv_16sc_t)*num_points, volk_gnsssdr_get_alignment());
        memcpy(in_a[n], in, sizeof(lv_16sc_t)*num_points);
     }
+    result = (lv_16sc_t*)calloc(num_points, sizeof(lv_16sc_t));
     volk_gnsssdr_16ic_x2_dot_prod_16ic_xn_neon(result, local_code, (const lv_16sc_t**) in_a, num_a_vectors, num_points);
 
     for(unsigned int n = 0; n < num_a_vectors; n++)

src/algorithms/libs/volk_gnsssdr_module/volk_gnsssdr/kernels/volk_gnsssdr/volk_gnsssdr_16ic_x2_rotator_dotprodxnpuppet_16ic.h

@@ -59,6 +59,7 @@ static inline void volk_gnsssdr_16ic_x2_rotator_dotprodxnpuppet_16ic_generic(lv_
        in_a[n] = (lv_16sc_t*)volk_gnsssdr_malloc(sizeof(lv_16sc_t) * num_points, volk_gnsssdr_get_alignment());
        memcpy(in_a[n], in, sizeof(lv_16sc_t) * num_points);
     }
+    result = (lv_16sc_t*)calloc(num_points, sizeof(lv_16sc_t));
     volk_gnsssdr_16ic_x2_rotator_dot_prod_16ic_xn_generic(result, local_code, phase_inc[0], phase,(const lv_16sc_t**) in_a, num_a_vectors, num_points);
 
     for(unsigned int n = 0; n < num_a_vectors; n++)
@@ -88,6 +89,7 @@ static inline void volk_gnsssdr_16ic_x2_rotator_dotprodxnpuppet_16ic_a_sse3(lv_1
        in_a[n] = (lv_16sc_t*)volk_gnsssdr_malloc(sizeof(lv_16sc_t) * num_points, volk_gnsssdr_get_alignment());
        memcpy((lv_16sc_t*)in_a[n], (lv_16sc_t*)in, sizeof(lv_16sc_t) * num_points);
     }
+    result = (lv_16sc_t*)calloc(num_points, sizeof(lv_16sc_t));
     volk_gnsssdr_16ic_x2_rotator_dot_prod_16ic_xn_a_sse3(result, local_code, phase_inc[0], phase, (const lv_16sc_t**) in_a, num_a_vectors, num_points);
 
     for(unsigned int n = 0; n < num_a_vectors; n++)
@@ -118,6 +120,7 @@ static inline void volk_gnsssdr_16ic_x2_rotator_dotprodxnpuppet_16ic_u_sse3(lv_1
        in_a[n] = (lv_16sc_t*)volk_gnsssdr_malloc(sizeof(lv_16sc_t)*num_points, volk_gnsssdr_get_alignment());
        memcpy(in_a[n], in, sizeof(lv_16sc_t)*num_points);
     }
+    result = (lv_16sc_t*)calloc(num_points, sizeof(lv_16sc_t));
     volk_gnsssdr_16ic_x2_rotator_dot_prod_16ic_xn_u_sse3(result, local_code, phase_inc[0], phase, (const lv_16sc_t**) in_a, num_a_vectors, num_points);
 
     for(unsigned int n = 0; n < num_a_vectors; n++)