From f9ee6a55cb397f9302769a25a8c959fa162354f0 Mon Sep 17 00:00:00 2001
From: Nick Foster <nick@nerdnetworks.org>
Date: Tue, 14 Dec 2010 22:58:33 -0800
Subject: Volk: Some new basic Orc implementations with QA code

---
 volk/include/volk/volk_16u_byteswap_aligned16.h |  4 ++--
 volk/include/volk/volk_32f_divide_aligned16.h   | 13 +++++++++++++
 volk/include/volk/volk_32f_multiply_aligned16.h | 14 +++++++++++++-
 volk/include/volk/volk_32f_subtract_aligned16.h | 14 ++++++++++++++
 4 files changed, 42 insertions(+), 3 deletions(-)

(limited to 'volk/include')

diff --git a/volk/include/volk/volk_16u_byteswap_aligned16.h b/volk/include/volk/volk_16u_byteswap_aligned16.h
index 8e628ab17a..9d19d1a456 100644
--- a/volk/include/volk/volk_16u_byteswap_aligned16.h
+++ b/volk/include/volk/volk_16u_byteswap_aligned16.h
@@ -67,9 +67,9 @@ static inline void volk_16u_byteswap_aligned16_generic(uint16_t* intsToSwap, uns
   \param intsToSwap The vector of data to byte swap
   \param numDataPoints The number of data points
 */
-extern void volk_16u_byteswap_aligned16_orc_impl(uint16_t* intsToSwap, uint16_t* intsToSwapAgain, unsigned int num_points);
+extern void volk_16u_byteswap_aligned16_orc_impl(uint16_t* intsToSwap, unsigned int num_points);
 static inline void volk_16u_byteswap_aligned16_orc(uint16_t* intsToSwap, unsigned int num_points){
-    volk_16u_byteswap_aligned16_orc_impl(intsToSwap, intsToSwap, num_points);
+    volk_16u_byteswap_aligned16_orc_impl(intsToSwap, num_points);
 }
 #endif /* LV_HAVE_ORC */
 
diff --git a/volk/include/volk/volk_32f_divide_aligned16.h b/volk/include/volk/volk_32f_divide_aligned16.h
index c00700cd8a..c595b5e92c 100644
--- a/volk/include/volk/volk_32f_divide_aligned16.h
+++ b/volk/include/volk/volk_32f_divide_aligned16.h
@@ -63,6 +63,19 @@ static inline void volk_32f_divide_aligned16_generic(float* cVector, const float
 }
 #endif /* LV_HAVE_GENERIC */
 
+#if LV_HAVE_ORC
+/*!
+  \brief Divides the two input vectors and store their results in the third vector
+  \param cVector The vector where the results will be stored
+  \param aVector The vector to be divideed
+  \param bVector The divisor vector
+  \param num_points The number of values in aVector and bVector to be divideed together and stored into cVector
+*/
+extern void volk_32f_divide_aligned16_orc_impl(float* cVector, const float* aVector, const float* bVector, unsigned int num_points);
+static inline void volk_32f_divide_aligned16_orc(float* cVector, const float* aVector, const float* bVector, unsigned int num_points){
+    volk_32f_divide_aligned16_orc_impl(cVector, aVector, bVector, num_points);
+}
+#endif /* LV_HAVE_ORC */
 
 
 
diff --git a/volk/include/volk/volk_32f_multiply_aligned16.h b/volk/include/volk/volk_32f_multiply_aligned16.h
index b557580aba..87ae7bcf8c 100644
--- a/volk/include/volk/volk_32f_multiply_aligned16.h
+++ b/volk/include/volk/volk_32f_multiply_aligned16.h
@@ -63,7 +63,19 @@ static inline void volk_32f_multiply_aligned16_generic(float* cVector, const flo
 }
 #endif /* LV_HAVE_GENERIC */
 
-
+#if LV_HAVE_ORC
+/*!
+  \brief Multiplys 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 multiplied
+  \param bVector One of the vectors to be multiplied
+  \param num_points The number of values in aVector and bVector to be multiplied together and stored into cVector
+*/
+extern void volk_32f_multiply_aligned16_orc_impl(float* cVector, const float* aVector, const float* bVector, unsigned int num_points);
+static inline void volk_32f_multiply_aligned16_orc(float* cVector, const float* aVector, const float* bVector, unsigned int num_points){
+    volk_32f_multiply_aligned16_orc_impl(cVector, aVector, bVector, num_points);
+}
+#endif /* LV_HAVE_ORC */
 
 
 #endif /* INCLUDED_VOLK_32f_MULTIPLY_ALIGNED16_H */
diff --git a/volk/include/volk/volk_32f_subtract_aligned16.h b/volk/include/volk/volk_32f_subtract_aligned16.h
index ac3f5e5d17..e152429016 100644
--- a/volk/include/volk/volk_32f_subtract_aligned16.h
+++ b/volk/include/volk/volk_32f_subtract_aligned16.h
@@ -63,5 +63,19 @@ static inline void volk_32f_subtract_aligned16_generic(float* cVector, const flo
 }
 #endif /* LV_HAVE_GENERIC */
 
+#if LV_HAVE_ORC
+/*!
+  \brief Subtracts bVector form aVector and store their results in the cVector
+  \param cVector The vector where the results will be stored
+  \param aVector The initial vector
+  \param bVector The vector to be subtracted
+  \param num_points The number of values in aVector and bVector to be subtracted together and stored into cVector
+*/
+extern void volk_32f_subtract_aligned16_orc_impl(float* cVector, const float* aVector, const float* bVector, unsigned int num_points);
+static inline void volk_32f_subtract_aligned16_orc(float* cVector, const float* aVector, const float* bVector, unsigned int num_points){
+    volk_32f_subtract_aligned16_orc_impl(cVector, aVector, bVector, num_points);
+}
+#endif /* LV_HAVE_ORC */
+
 
 #endif /* INCLUDED_VOLK_32f_SUBTRACT_ALIGNED16_H */
-- 
cgit v1.2.3