17 files changed, 234 insertions, 7 deletions

diff --git a/volk/include/volk/volk_16sc_magnitude_16s_aligned16.h b/volk/include/volk/volk_16sc_magnitude_16s_aligned16.h
index 1482ab82e..9f3222aa6 100644
--- a/volk/include/volk/volk_16sc_magnitude_16s_aligned16.h
+++ b/volk/include/volk/volk_16sc_magnitude_16s_aligned16.h
@@ -164,7 +164,7 @@ static inline void volk_16sc_magnitude_16s_aligned16_generic(int16_t* magnitudeV
   const int16_t* complexVectorPtr = (const int16_t*)complexVector;
   int16_t* magnitudeVectorPtr = magnitudeVector;
   unsigned int number = 0;
-  const float scalar = 32786.0;
+  const float scalar = 32768.0;
   for(number = 0; number < num_points; number++){
     float real = ((float)(*complexVectorPtr++)) / scalar;
     float imag = ((float)(*complexVectorPtr++)) / scalar;
@@ -173,7 +173,18 @@ static inline void volk_16sc_magnitude_16s_aligned16_generic(int16_t* magnitudeV
 }
 #endif /* LV_HAVE_GENERIC */
 
-
+#if LV_HAVE_ORC
+/*!
+  \brief Calculates the magnitude of the complexVector and stores the results in the magnitudeVector
+  \param complexVector The vector containing the complex input values
+  \param magnitudeVector The vector containing the real output values
+  \param num_points The number of complex values in complexVector to be calculated and stored into cVector
+*/
+extern void volk_16sc_magnitude_16s_aligned16_orc_impl(int16_t* magnitudeVector, const lv_16sc_t* complexVector, unsigned int num_points);
+static inline void volk_16sc_magnitude_16s_aligned16_orc(int16_t* magnitudeVector, const lv_16sc_t* complexVector, unsigned int num_points){
+    volk_16sc_magnitude_16s_aligned16_orc_impl(magnitudeVector, complexVector, num_points);
+}
+#endif /* LV_HAVE_ORC */
 
 
 #endif /* INCLUDED_VOLK_16sc_MAGNITUDE_16s_ALIGNED16_H */
diff --git a/volk/include/volk/volk_16sc_magnitude_32f_aligned16.h b/volk/include/volk/volk_16sc_magnitude_32f_aligned16.h
index 9c2a48835..e063ae432 100644
--- a/volk/include/volk/volk_16sc_magnitude_32f_aligned16.h
+++ b/volk/include/volk/volk_16sc_magnitude_32f_aligned16.h
@@ -161,7 +161,19 @@ static inline void volk_16sc_magnitude_32f_aligned16_generic(float* magnitudeVec
 }
 #endif /* LV_HAVE_GENERIC */
 
-
+#if LV_HAVE_ORC
+/*!
+  \brief Calculates the magnitude of the complexVector and stores the results in the magnitudeVector
+  \param complexVector The vector containing the complex input values
+  \param magnitudeVector The vector containing the real output values
+  \param scalar The data value to be divided against each input data value of the input complex vector
+  \param num_points The number of complex values in complexVector to be calculated and stored into cVector
+*/
+extern void volk_16sc_magnitude_32f_aligned16_orc_impl(float* magnitudeVector, const lv_16sc_t* complexVector, const float scalar, unsigned int num_points);
+static inline void volk_16sc_magnitude_32f_aligned16_orc(float* magnitudeVector, const lv_16sc_t* complexVector, const float scalar, unsigned int num_points){
+    volk_16sc_magnitude_32f_aligned16_orc_impl(magnitudeVector, complexVector, scalar, num_points);
+}
+#endif /* LV_HAVE_ORC */
 
 
 #endif /* INCLUDED_VOLK_16sc_MAGNITUDE_32f_ALIGNED16_H */
diff --git a/volk/include/volk/volk_32fc_magnitude_16s_aligned16.h b/volk/include/volk/volk_32fc_magnitude_16s_aligned16.h
index 4e64d8c22..4e590e120 100644
--- a/volk/include/volk/volk_32fc_magnitude_16s_aligned16.h
+++ b/volk/include/volk/volk_32fc_magnitude_16s_aligned16.h
@@ -140,7 +140,19 @@ static inline void volk_32fc_magnitude_16s_aligned16_generic(int16_t* magnitudeV
 }
 #endif /* LV_HAVE_GENERIC */
 
-
+#if LV_HAVE_ORC
+/*!
+  \brief Calculates the magnitude of the complexVector, scales the resulting value and stores the results in the magnitudeVector
+  \param complexVector The vector containing the complex input values
+  \param scalar The scale value multiplied to the magnitude of each complex vector
+  \param magnitudeVector The vector containing the real output values
+  \param num_points The number of complex values in complexVector to be calculated and stored into cVector
+*/
+extern void volk_32fc_magnitude_16s_aligned16_orc_impl(int16_t* magnitudeVector, const lv_32fc_t* complexVector, const float scalar, unsigned int num_points);
+static inline void volk_32fc_magnitude_16s_aligned16_orc(int16_t* magnitudeVector, const lv_32fc_t* complexVector, const float scalar, unsigned int num_points){
+    volk_32fc_magnitude_16s_aligned16_orc_impl(magnitudeVector, complexVector, scalar, num_points);
+}
+#endif /* LV_HAVE_ORC */
 
 
 #endif /* INCLUDED_VOLK_32fc_MAGNITUDE_16s_ALIGNED16_H */
diff --git a/volk/include/volk/volk_32fc_magnitude_32f_aligned16.h b/volk/include/volk/volk_32fc_magnitude_32f_aligned16.h
index 7a8fd1ef9..3ea62da6a 100644
--- a/volk/include/volk/volk_32fc_magnitude_32f_aligned16.h
+++ b/volk/include/volk/volk_32fc_magnitude_32f_aligned16.h
@@ -115,7 +115,18 @@ static inline void volk_32fc_magnitude_32f_aligned16_generic(float* magnitudeVec
 }
 #endif /* LV_HAVE_GENERIC */
 
-
+#if LV_HAVE_ORC
+  /*!
+    \brief Calculates the magnitude of the complexVector and stores the results in the magnitudeVector
+    \param complexVector The vector containing the complex input values
+    \param magnitudeVector The vector containing the real output values
+    \param num_points The number of complex values in complexVector to be calculated and stored into cVector
+  */
+extern void volk_32fc_magnitude_32f_aligned16_orc_impl(float* magnitudeVector, const lv_32fc_t* complexVector, unsigned int num_points);
+static inline void volk_32fc_magnitude_32f_aligned16_orc(float* magnitudeVector, const lv_32fc_t* complexVector, unsigned int num_points){
+    volk_32fc_magnitude_32f_aligned16_orc_impl(magnitudeVector, complexVector, num_points);
+}
+#endif /* LV_HAVE_ORC */
 
 
 #endif /* INCLUDED_VOLK_32fc_MAGNITUDE_32f_ALIGNED16_H */
diff --git a/volk/include/volk/volk_32s_or_aligned16.h b/volk/include/volk/volk_32s_or_aligned16.h
index f4c427c4d..64748d535 100644
--- a/volk/include/volk/volk_32s_or_aligned16.h
+++ b/volk/include/volk/volk_32s_or_aligned16.h
@@ -63,7 +63,19 @@ static inline void volk_32s_or_aligned16_generic(int32_t* cVector, const int32_t
 }
 #endif /* LV_HAVE_GENERIC */
 
-
+#if LV_HAVE_ORC
+/*!
+  \brief Ors the two input vectors and store their results in the third vector
+  \param cVector The vector where the results will be stored
+  \param aVector One of the vectors to be ored
+  \param bVector One of the vectors to be ored
+  \param num_points The number of values in aVector and bVector to be ored together and stored into cVector
+*/
+extern void volk_32s_or_aligned16_orc_impl(int32_t* cVector, const int32_t* aVector, const int32_t* bVector, unsigned int num_points);
+static inline void volk_32s_or_aligned16_orc(int32_t* cVector, const int32_t* aVector, const int32_t* bVector, unsigned int num_points){
+    volk_32s_or_aligned16_orc_impl(cVector, aVector, bVector, num_points);
+}
+#endif /* LV_HAVE_ORC */
 
 
 #endif /* INCLUDED_VOLK_32s_OR_ALIGNED16_H */
diff --git a/volk/lib/qa_16sc_magnitude_16s_aligned16.cc b/volk/lib/qa_16sc_magnitude_16s_aligned16.cc
index b14610757..c8f13ff84 100644
--- a/volk/lib/qa_16sc_magnitude_16s_aligned16.cc
+++ b/volk/lib/qa_16sc_magnitude_16s_aligned16.cc
@@ -23,6 +23,7 @@ void qa_16sc_magnitude_16s_aligned16::t1() {
   std::complex<int16_t> input0[vlen] __attribute__ ((aligned (16)));
   
   int16_t output_generic[vlen] __attribute__ ((aligned (16)));
+  int16_t output_orc[vlen] __attribute__ ((aligned (16)));
   int16_t output_sse[vlen] __attribute__ ((aligned (16)));
   int16_t output_sse3[vlen] __attribute__ ((aligned (16)));
 
@@ -41,6 +42,13 @@ void qa_16sc_magnitude_16s_aligned16::t1() {
   printf("generic_time: %f\n", total);
   start = clock();
   for(int count = 0; count < ITERS; ++count) {
+    volk_16sc_magnitude_16s_aligned16_manual(output_orc, input0, vlen, "orc");
+  }
+  end = clock();
+  total = (double)(end-start)/(double)CLOCKS_PER_SEC;
+  printf("orc_time: %f\n", total);
+  start = clock();
+  for(int count = 0; count < ITERS; ++count) {
     volk_16sc_magnitude_16s_aligned16_manual(output_sse, input0, vlen, "sse");
   }
   end = clock();
@@ -64,6 +72,7 @@ void qa_16sc_magnitude_16s_aligned16::t1() {
     //printf("%d...%d\n", output0[i], output01[i]);
     CPPUNIT_ASSERT_DOUBLES_EQUAL(output_generic[i], output_sse[i], 1.1);
     CPPUNIT_ASSERT_DOUBLES_EQUAL(output_generic[i], output_sse3[i], 1.1);
+    CPPUNIT_ASSERT_DOUBLES_EQUAL(output_generic[i], output_orc[i], 1.1);
   }
 }
 
diff --git a/volk/lib/qa_16sc_magnitude_32f_aligned16.cc b/volk/lib/qa_16sc_magnitude_32f_aligned16.cc
index 2c9e48f6e..e7178863c 100644
--- a/volk/lib/qa_16sc_magnitude_32f_aligned16.cc
+++ b/volk/lib/qa_16sc_magnitude_32f_aligned16.cc
@@ -15,6 +15,7 @@ void qa_16sc_magnitude_32f_aligned16::t1() {
   std::complex<int16_t> input0[vlen] __attribute__ ((aligned (16)));
   
   float output_generic[vlen] __attribute__ ((aligned (16)));
+  float output_orc[vlen] __attribute__ ((aligned (16)));
   float output_known[vlen] __attribute__ ((aligned (16)));
 
   int16_t* inputLoad = (int16_t*)input0;
@@ -37,6 +38,14 @@ void qa_16sc_magnitude_32f_aligned16::t1() {
   end = clock();
   total = (double)(end-start)/(double)CLOCKS_PER_SEC;
   printf("generic_time: %f\n", total);
+  
+  start = clock();
+  for(int count = 0; count < ITERS; ++count) {
+    volk_16sc_magnitude_32f_aligned16_manual(output_orc, input0, scale, vlen, "orc");
+  }
+  end = clock();
+  total = (double)(end-start)/(double)CLOCKS_PER_SEC;
+  printf("orc_time: %f\n", total);
 
   /*
   for(int i = 0; i < 100; ++i) {
@@ -48,6 +57,7 @@ void qa_16sc_magnitude_32f_aligned16::t1() {
   for(int i = 0; i < vlen; ++i) {
     //printf("%d...%d\n", output0[i], output01[i]);
     CPPUNIT_ASSERT_DOUBLES_EQUAL(output_generic[i], output_known[i], fabs(output_generic[i])*1e-4);
+    CPPUNIT_ASSERT_DOUBLES_EQUAL(output_orc[i], output_known[i], fabs(output_generic[i])*1e-4);
   }
 }
 
@@ -63,6 +73,7 @@ void qa_16sc_magnitude_32f_aligned16::t1() {
   std::complex<int16_t> input0[vlen] __attribute__ ((aligned (16)));
   
   float output_generic[vlen] __attribute__ ((aligned (16)));
+  float output_orc[vlen] __attribute__ ((aligned (16)));
   float output_sse[vlen] __attribute__ ((aligned (16)));
   float output_sse3[vlen] __attribute__ ((aligned (16)));
 
@@ -81,6 +92,14 @@ void qa_16sc_magnitude_32f_aligned16::t1() {
   printf("generic_time: %f\n", total);
   start = clock();
   for(int count = 0; count < ITERS; ++count) {
+    volk_16sc_magnitude_32f_aligned16_manual(output_orc, input0, 32768.0, vlen, "orc");
+  }
+  end = clock();
+  total = (double)(end-start)/(double)CLOCKS_PER_SEC;
+  printf("orc_time: %f\n", total);
+
+  start = clock();
+  for(int count = 0; count < ITERS; ++count) {
     volk_16sc_magnitude_32f_aligned16_manual(output_sse, input0, 32768.0, vlen, "sse");
   }
   end = clock();
@@ -104,6 +123,7 @@ void qa_16sc_magnitude_32f_aligned16::t1() {
     //printf("%d...%d\n", output0[i], output01[i]);
     CPPUNIT_ASSERT_DOUBLES_EQUAL(output_generic[i], output_sse[i], fabs(output_generic[i])*1e-4);
     CPPUNIT_ASSERT_DOUBLES_EQUAL(output_generic[i], output_sse3[i], fabs(output_generic[i])*1e-4);
+    CPPUNIT_ASSERT_DOUBLES_EQUAL(output_generic[i], output_orc[i], fabs(output_generic[i])*1e-4);
   }
 }
 
diff --git a/volk/lib/qa_32f_divide_aligned16.cc b/volk/lib/qa_32f_divide_aligned16.cc
index f104e0443..b2c2ecf9a 100644
--- a/volk/lib/qa_32f_divide_aligned16.cc
+++ b/volk/lib/qa_32f_divide_aligned16.cc
@@ -88,6 +88,7 @@ void qa_32f_divide_aligned16::t1() {
   
   float output0[vlen] __attribute__ ((aligned (16)));
   float output01[vlen] __attribute__ ((aligned (16)));
+  float output02[vlen] __attribute__ ((aligned (16)));
 
   for(int i = 0; i < vlen; ++i) {   
     input0[i] = ((float) (rand() - (RAND_MAX/2))) / static_cast<float>((RAND_MAX/2));
@@ -104,6 +105,13 @@ void qa_32f_divide_aligned16::t1() {
   printf("generic_time: %f\n", total);
   start = clock();
   for(int count = 0; count < ITERS; ++count) {
+    volk_32f_divide_aligned16_manual(output02, input0, input1, vlen, "orc");
+  }
+  end = clock();
+  total = (double)(end-start)/(double)CLOCKS_PER_SEC;
+  printf("orc_time: %f\n", total);
+  start = clock();
+  for(int count = 0; count < ITERS; ++count) {
     volk_32f_divide_aligned16_manual(output01, input0, input1, vlen, "sse");
   }
   end = clock();
@@ -117,6 +125,7 @@ void qa_32f_divide_aligned16::t1() {
   for(int i = 0; i < vlen; ++i) {
     //printf("%d...%d\n", output0[i], output01[i]);
     CPPUNIT_ASSERT_EQUAL(output0[i], output01[i]);
+    CPPUNIT_ASSERT_EQUAL(output0[i], output02[i]);
   }
 }
 
diff --git a/volk/lib/qa_32fc_magnitude_16s_aligned16.cc b/volk/lib/qa_32fc_magnitude_16s_aligned16.cc
index a4be1616b..c3e65866b 100644
--- a/volk/lib/qa_32fc_magnitude_16s_aligned16.cc
+++ b/volk/lib/qa_32fc_magnitude_16s_aligned16.cc
@@ -23,6 +23,7 @@ void qa_32fc_magnitude_16s_aligned16::t1() {
   std::complex<float> input0[vlen] __attribute__ ((aligned (16)));
   
   int16_t output_generic[vlen] __attribute__ ((aligned (16)));
+  int16_t output_orc[vlen] __attribute__ ((aligned (16)));
   int16_t output_sse[vlen] __attribute__ ((aligned (16)));
   int16_t output_sse3[vlen] __attribute__ ((aligned (16)));
 
@@ -41,6 +42,13 @@ void qa_32fc_magnitude_16s_aligned16::t1() {
   printf("generic_time: %f\n", total);
   start = clock();
   for(int count = 0; count < ITERS; ++count) {
+    volk_32fc_magnitude_16s_aligned16_manual(output_orc, input0, 32768.0, vlen, "orc");
+  }
+  end = clock();
+  total = (double)(end-start)/(double)CLOCKS_PER_SEC;
+  printf("orc_time: %f\n", total);
+  start = clock();
+  for(int count = 0; count < ITERS; ++count) {
     volk_32fc_magnitude_16s_aligned16_manual(output_sse, input0, 32768.0, vlen, "sse");
   }
   end = clock();
@@ -64,6 +72,7 @@ void qa_32fc_magnitude_16s_aligned16::t1() {
     //printf("%d...%d\n", output0[i], output01[i]);
     CPPUNIT_ASSERT_DOUBLES_EQUAL(output_generic[i], output_sse[i], 1.1);
     CPPUNIT_ASSERT_DOUBLES_EQUAL(output_generic[i], output_sse3[i], 1.1);
+    CPPUNIT_ASSERT_DOUBLES_EQUAL(output_generic[i], output_orc[i], 1.1);
   }
 }
 
diff --git a/volk/lib/qa_32fc_magnitude_32f_aligned16.cc b/volk/lib/qa_32fc_magnitude_32f_aligned16.cc
index d69ada408..6a1d46c7a 100644
--- a/volk/lib/qa_32fc_magnitude_32f_aligned16.cc
+++ b/volk/lib/qa_32fc_magnitude_32f_aligned16.cc
@@ -23,6 +23,7 @@ void qa_32fc_magnitude_32f_aligned16::t1() {
   std::complex<float> input0[vlen] __attribute__ ((aligned (16)));
   
   float output_generic[vlen] __attribute__ ((aligned (16)));
+  float output_orc[vlen] __attribute__ ((aligned (16)));
   float output_sse[vlen] __attribute__ ((aligned (16)));
   float output_sse3[vlen] __attribute__ ((aligned (16)));
 
@@ -41,6 +42,13 @@ void qa_32fc_magnitude_32f_aligned16::t1() {
   printf("generic_time: %f\n", total);
   start = clock();
   for(int count = 0; count < ITERS; ++count) {
+    volk_32fc_magnitude_32f_aligned16_manual(output_orc, input0, vlen, "orc");
+  }
+  end = clock();
+  total = (double)(end-start)/(double)CLOCKS_PER_SEC;
+  printf("orc_time: %f\n", total);
+  start = clock();
+  for(int count = 0; count < ITERS; ++count) {
     volk_32fc_magnitude_32f_aligned16_manual(output_sse, input0, vlen, "sse");
   }
   end = clock();
@@ -64,6 +72,7 @@ void qa_32fc_magnitude_32f_aligned16::t1() {
     //printf("%d...%d\n", output0[i], output01[i]);
     CPPUNIT_ASSERT_DOUBLES_EQUAL(output_generic[i], output_sse[i], fabs(output_generic[i])*1e-4);
     CPPUNIT_ASSERT_DOUBLES_EQUAL(output_generic[i], output_sse3[i], fabs(output_generic[i])*1e-4);
+    CPPUNIT_ASSERT_DOUBLES_EQUAL(output_generic[i], output_orc[i], fabs(output_generic[i])*1e-4);
   }
 }
 
diff --git a/volk/lib/qa_32s_or_aligned16.cc b/volk/lib/qa_32s_or_aligned16.cc
index e09dfb91c..9ea5283a6 100644
--- a/volk/lib/qa_32s_or_aligned16.cc
+++ b/volk/lib/qa_32s_or_aligned16.cc
@@ -25,6 +25,7 @@ void qa_32s_or_aligned16::t1() {
   
   int32_t output0[vlen] __attribute__ ((aligned (16)));
   int32_t output01[vlen] __attribute__ ((aligned (16)));
+  int32_t output02[vlen] __attribute__ ((aligned (16)));
 
   for(int i = 0; i < vlen; ++i) {   
     input0[i] = ((int32_t) (rand() - (RAND_MAX/2)));
@@ -41,6 +42,13 @@ void qa_32s_or_aligned16::t1() {
   printf("generic_time: %f\n", total);
   start = clock();
   for(int count = 0; count < ITERS; ++count) {
+    volk_32s_or_aligned16_manual(output02, input0, input1, vlen, "orc");
+  }
+  end = clock();
+  total = (double)(end-start)/(double)CLOCKS_PER_SEC;
+  printf("orc_time: %f\n", total);
+  start = clock();
+  for(int count = 0; count < ITERS; ++count) {
     volk_32s_or_aligned16_manual(output01, input0, input1, vlen, "sse");
   }
   end = clock();
@@ -54,6 +62,7 @@ void qa_32s_or_aligned16::t1() {
   for(int i = 0; i < vlen; ++i) {
     //printf("%d...%d\n", output0[i], output01[i]);
     CPPUNIT_ASSERT_EQUAL(output0[i], output01[i]);
+    CPPUNIT_ASSERT_EQUAL(output0[i], output02[i]);
   }
 }
 
diff --git a/volk/orc/Makefile.am b/volk/orc/Makefile.am
index c71625d87..3f105fd80 100644
--- a/volk/orc/Makefile.am
+++ b/volk/orc/Makefile.am
@@ -29,11 +29,15 @@ volk_8s_convert_16s_aligned16_orc_impl.orc \
 volk_8s_convert_32f_aligned16_orc_impl.orc \
 volk_16u_byteswap_aligned16_orc_impl.orc \
 volk_32s_and_aligned16_orc_impl.orc \
+volk_32s_or_aligned16_orc_impl.orc \
 volk_32f_add_aligned16_orc_impl.orc \
 volk_32f_subtract_aligned16_orc_impl.orc \
 volk_32f_divide_aligned16_orc_impl.orc \
 volk_32f_multiply_aligned16_orc_impl.orc \
-volk_32f_sqrt_aligned16_orc_impl.orc
+volk_32f_sqrt_aligned16_orc_impl.orc \
+volk_16sc_magnitude_32f_aligned16_orc_impl.orc \
+volk_32fc_magnitude_32f_aligned16_orc_impl.orc \
+volk_32fc_magnitude_16s_aligned16_orc_impl.orc
 
 
 
diff --git a/volk/orc/volk_16sc_magnitude_16s_aligned16_orc_impl.orc b/volk/orc/volk_16sc_magnitude_16s_aligned16_orc_impl.orc
new file mode 100644
index 000000000..f6c959c00
--- /dev/null
+++ b/volk/orc/volk_16sc_magnitude_16s_aligned16_orc_impl.orc
@@ -0,0 +1,24 @@
+.function volk_16sc_magnitude_16s_aligned16_orc_impl
+.source 4 src
+.dest 2 dst
+.temp 2 reals
+.temp 2 imags
+.temp 4 reall
+.temp 4 imagl
+.temp 4 realf
+.temp 4 imagf
+.temp 4 sumf
+.temp 4 rootf
+.temp 4 rootl
+
+splitlw reals, imags, src
+convwl reall, reals
+convwl imagl, imags
+convlf realf, reall
+convlf imagf, imagl
+mulf realf, realf, (1.0 / 32768.0)
+mulf imagf, imagf, (1.0 / 32768.0)
+addf sumf, realf, imagf
+sqrtf rootf, sumf
+convfl rootl, rootf
+conflw dst, rootl
diff --git a/volk/orc/volk_16sc_magnitude_32f_aligned16_orc_impl.orc b/volk/orc/volk_16sc_magnitude_32f_aligned16_orc_impl.orc
new file mode 100644
index 000000000..44654ad8e
--- /dev/null
+++ b/volk/orc/volk_16sc_magnitude_32f_aligned16_orc_impl.orc
@@ -0,0 +1,25 @@
+.function volk_16sc_magnitude_32f_aligned16_orc_impl
+.source 4 src
+.dest 4 dst
+.floatparam 4 scalar
+.temp 4 invscalar
+.temp 4 reall
+.temp 4 imagl
+.temp 2 reals
+.temp 2 imags
+.temp 4 realf
+.temp 4 imagf
+.temp 4 sumf
+
+divf invscalar, 1.0, scalar
+splitlw reals, imags, src
+convswl reall, reals
+convswl imagl, imags
+convlf realf, reall
+convlf imagf, imagl
+mulf realf, realf, invscalar
+mulf imagf, imagf, invscalar
+mulf realf, realf, realf
+mulf imagf, imagf, imagf
+addf sumf, realf, imagf
+sqrtf dst, sumf
diff --git a/volk/orc/volk_32fc_magnitude_16s_aligned16_orc_impl.orc b/volk/orc/volk_32fc_magnitude_16s_aligned16_orc_impl.orc
new file mode 100644
index 000000000..db8405e59
--- /dev/null
+++ b/volk/orc/volk_32fc_magnitude_16s_aligned16_orc_impl.orc
@@ -0,0 +1,25 @@
+.function volk_32fc_magnitude_16s_aligned16_orc_impl
+.source 8 src
+.dest 2 dst
+.floatparam 4 scalar
+.temp 4 invscalar
+.temp 4 reall
+.temp 4 imagl
+.temp 4 realf
+.temp 4 imagf
+.temp 4 sumf
+.temp 4 rootf
+.temp 4 rootl
+
+divf invscalar, 1.0, scalar
+splitql reall, imagl, src
+convlf realf, reall
+convlf imagf, imagl
+mulf realf, realf, invscalar
+mulf imagf, imagf, invscalar
+mulf realf, realf, realf
+mulf imagf, imagf, imagf
+addf sumf, realf, imagf
+sqrtf rootf, sumf
+convfl rootl, rootf
+convlw dst, rootl
diff --git a/volk/orc/volk_32fc_magnitude_32f_aligned16_orc_impl.orc b/volk/orc/volk_32fc_magnitude_32f_aligned16_orc_impl.orc
new file mode 100644
index 000000000..cc5c85b45
--- /dev/null
+++ b/volk/orc/volk_32fc_magnitude_32f_aligned16_orc_impl.orc
@@ -0,0 +1,21 @@
+.function volk_32fc_magnitude_32f_aligned16_orc_impl
+.source 8 src
+.dest 4 dst
+.floatparam 4 scalar
+.temp 4 invscalar
+.temp 4 reall
+.temp 4 imagl
+.temp 4 realf
+.temp 4 imagf
+.temp 4 sumf
+
+divf invscalar, 1.0, scalar
+splitql reall, imagl, src
+convlf realf, reall
+convlf imagf, imagl
+mulf realf, realf, invscalar
+mulf imagf, imagf, invscalar
+mulf realf, realf, realf
+mulf imagf, imagf, imagf
+addf sumf, realf, imagf
+sqrtf dst, sumf
diff --git a/volk/orc/volk_32s_or_aligned16_orc_impl.orc b/volk/orc/volk_32s_or_aligned16_orc_impl.orc
new file mode 100644
index 000000000..6d2a3859a
--- /dev/null
+++ b/volk/orc/volk_32s_or_aligned16_orc_impl.orc
@@ -0,0 +1,5 @@
+.function volk_32s_or_aligned16_orc_impl
+.dest 4 dst
+.source 4 src1
+.source 4 src2
+orl dst, src1, src2