Squashed commit of the sb-vbs branch.

Includes the SD-VBS benchmarks modified to:
- Use libextra to loop as realtime jobs
- Preallocate memory before starting their main computation
- Accept input via stdin instead of via argc

Does not include the SD-VBS matlab code.

Fixes libextra execution in LITMUS^RT.

1 files changed, 76 insertions, 0 deletions

diff --git a/SD-VBS/benchmarks/stitch/src/c/dist2.c b/SD-VBS/benchmarks/stitch/src/c/dist2.c
new file mode 100644
index 0000000..d76fcb2
--- /dev/null
+++ b/SD-VBS/benchmarks/stitch/src/c/dist2.c
@@ -0,0 +1,76 @@
+/********************************
+Author: Sravanthi Kota Venkata
+********************************/
+
+#include "stitch.h"
+
+F2D* dist2(I2D* x, F2D* c)
+{
+    int ndata, dimx, ncentres, dimc, i, j, k;
+    F2D *n2, *t1, *t2;
+    float temp;
+    F2D *s1, *s2, *ctrans;
+    F2D *mult1, *mult2, *mult3;
+
+    ndata = x->height;
+    dimx = x->width;
+
+    ncentres = c->height;
+    dimc = c->width;
+
+    if(dimx != dimc)
+        return NULL;
+
+    s1 = fSetArray(ncentres, 1, 1);
+    s2 = fSetArray(ndata, 1, 1);
+    t1 = fMallocHandle(1, x->height);
+
+    for(j=0; j<t1->width; j++)
+    {
+        temp = 0;
+        for(i=0; i<t1->height; i++)
+        {
+            temp += subsref(x,j,i) * subsref(x,j,i);
+        }
+
+        asubsref(t1,j) = temp;
+    }
+
+    mult1 = fMtimes(s1, t1);
+    t2 = fMallocHandle(1, c->height);
+
+    for(j=0; j<t2->width; j++)
+    {
+        temp = 0;
+        for(i=0; i<t2->height; i++)
+        {
+            temp += subsref(c,j,i) * subsref(c,j,i);
+        }
+
+        asubsref(t2,j) = temp;
+    }
+
+    mult2 = fMtimes(s2, t2);
+    ctrans = fTranspose(c);
+    mult3 = ifMtimes(x, ctrans);
+
+    for(i=0; i<(mult3->height * mult3->width); i++)
+        asubsref(mult3,i) = asubsref(mult3,i) * 2;
+
+    free(t1);
+    free(t2);
+    free(s1);
+    free(s2);
+    free(ctrans);
+
+    n2 = fMallocHandle(ndata, ncentres);
+    for(i=0; i<(ndata*ncentres); i++)
+        asubsref(n2,i) = asubsref(mult1,i) + asubsref(mult2,i) - asubsref(mult3,i);
+
+    free(mult1);
+    free(mult2);
+    free(mult3);
+
+    return n2;
+
+}