Several tweaks and cleanups: * Use .rept instead of repeated macros for repeating blocks. * Use MUL (variant) instead of MLA (variant) in the first step of the ARM scalarproduct() if there's no loop. * Unroll ARM assembler functions to 32 where not already done, plus the generic scalarproduct().

git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19144 a1c6a512-1295-4272-9138-f99709370657

1 files changed, 20 insertions, 2 deletions

diff --git a/apps/codecs/demac/libdemac/vector_math_generic.h b/apps/codecs/demac/libdemac/vector_math_generic.h
index 7b61db77be..00bf07a007 100644
--- a/apps/codecs/demac/libdemac/vector_math_generic.h
+++ b/apps/codecs/demac/libdemac/vector_math_generic.h
@@ -116,8 +116,8 @@ static inline int32_t scalarproduct(filter_int* v1, filter_int* v2)
 {
     int res = 0;
 
-#if ORDER > 16
-    int order = (ORDER >> 4);
+#if ORDER > 32
+    int order = (ORDER >> 5);
     while (order--)
 #endif
     {
@@ -137,6 +137,24 @@ static inline int32_t scalarproduct(filter_int* v1, filter_int* v2)
         res += *v1++ * *v2++;
         res += *v1++ * *v2++;
         res += *v1++ * *v2++;
+#if ORDER > 16
+        res += *v1++ * *v2++;
+        res += *v1++ * *v2++;
+        res += *v1++ * *v2++;
+        res += *v1++ * *v2++;
+        res += *v1++ * *v2++;
+        res += *v1++ * *v2++;
+        res += *v1++ * *v2++;
+        res += *v1++ * *v2++;
+        res += *v1++ * *v2++;
+        res += *v1++ * *v2++;
+        res += *v1++ * *v2++;
+        res += *v1++ * *v2++;
+        res += *v1++ * *v2++;
+        res += *v1++ * *v2++;
+        res += *v1++ * *v2++;
+        res += *v1++ * *v2++;
+#endif
     }
     return res;
 }