Commit 52d45816429956fc8f1b71bfdd1a5321f34a9b03
2012-02-21T13:08:56
small changes to speedup no vec for AMD 898.1 OCL runtime
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38
diff --git a/diakgcn120216.cl b/diakgcn120216.cl
index da3b2a2..88ddb74 100644
--- a/diakgcn120216.cl
+++ b/diakgcn120216.cl
@@ -1,4 +1,4 @@
-// DiaKGCN 18-02-2012 - OpenCL kernel by Diapolo
+// DiaKGCN 20-02-2012 - OpenCL kernel by Diapolo
//
// Parts and / or ideas for this kernel are based upon the public-domain poclbm project, the phatk kernel by Phateus and the DiabloMiner kernel by DiabloD3.
// The kernel was rewritten by me (Diapolo) and is still public-domain!
@@ -20,15 +20,9 @@
#define ma(x, y, z) amd_bytealign(z ^ x, y, x)
#else
#define ch(x, y, z) bitselect(z, y, x)
- #if defined(VECTORS2) || defined(VECTORS4) || defined(VECTORS8)
- // GCN - VEC2 or VEC4 or VEC8
- #define ma(z, x, y) bitselect(z, y, z ^ x)
- #else
- // GCN - no VEC
- #define ma(z, x, y) ch(z ^ x, y, x)
- #endif
+ #define ma(z, x, y) bitselect(z, y, z ^ x)
#endif
-#else //BITALIGN
+#else
#define ch(x, y, z) (z ^ (x & (y ^ z)))
#define ma(x, y, z) ((x & z) | (y & (x | z)))
#endif
@@ -58,8 +52,8 @@ __kernel
const uint state5A, const uint state6A, const uint state7A,
__global uint * output)
{
- u W[16];
u V[8];
+ u W[16];
#ifdef VECTORS8
#ifdef GOFFSET