cubeluffa: Fix indent and add some static prefixes

use git "show -w <commithash>" to see changes

Duplicated functions in merged Cube+Luffa could be cross linked without

x11/cuda_x11_echo.cu

@@ -61,7 +61,7 @@ void cuda_echo_round(
 
 	k0 = 512 + 8;
 
-	#pragma unroll
+	#pragma unroll 4
 	for (int idx = 0; idx < 16; idx += 4)
 	{
 		AES_2ROUND(sharedMemory,
@@ -221,7 +221,7 @@ void cuda_echo_round(
 		#pragma unroll 4
 		for (int i = 0; i < 4; i++) // Schleife über je 2*uint32_t
 		{
-			#pragma unroll 64
+			#pragma unroll 4
 			for (int idx = 0; idx < 64; idx += 16) // Schleife über die elemnte
 			{
 				uint32_t a = W[idx + i];

x11/cuda_x11_luffa512_Cubehash.cu

@@ -93,8 +93,8 @@ typedef struct {
 	b0 ^= c1;
 
 /* initial values of chaining variables */
-__device__ __constant__ uint32_t c_IV[40];
+__device__ static __constant__ uint32_t c_IV[40];
-const uint32_t h_IV[40] = {
+static const uint32_t h_IV[40] = {
 	0x6d251e69,0x44b051e0,0x4eaa6fb4,0xdbf78465,
 	0x6e292011,0x90152df4,0xee058139,0xdef610bb,
 	0xc3b44b95,0xd9d2f256,0x70eee9a0,0xde099fa3,
@@ -106,8 +106,8 @@ const uint32_t h_IV[40] = {
 	0x6c68e9be,0x5ec41e22,0xc825b7c7,0xaffb4363,
 	0xf5df3999,0x0fc688f1,0xb07224cc,0x03e86cea};
 
-__device__ __constant__ uint32_t c_CNS[80];
+__device__ static __constant__ uint32_t c_CNS[80];
-const uint32_t h_CNS[80] = {
+static const uint32_t h_CNS[80] = {
 	0x303994a6,0xe0337818,0xc0e65299,0x441ba90d,
 	0x6cc33a12,0x7f34d442,0xdc56983e,0x9389217f,
 	0x1e00108f,0xe5a8bce6,0x7800423d,0x5274baf4,
@@ -132,7 +132,7 @@ const uint32_t h_CNS[80] = {
 
 /***************************************************/
 __device__ __forceinline__
-void rnd512(hashState *state)
+static void rnd512(hashState *state)
 {
 	int i,j;
 	uint32_t t[40];
@@ -279,7 +279,7 @@ void rnd512(hashState *state)
 
 
 __device__ __forceinline__
-void Update512(hashState *state, const BitSequence *data)
+static void Update512(hashState *state, const BitSequence *data)
 {
 #pragma unroll 8
 	for(int i=0;i<8;i++) state->buffer[i] = cuda_swab32(((uint32_t*)data)[i]);
@@ -293,7 +293,7 @@ void Update512(hashState *state, const BitSequence *data)
 
 /***************************************************/
 __device__ __forceinline__
-void finalization512(hashState *state, uint32_t *b)
+static void finalization512(hashState *state, uint32_t *b)
 {
 	int i,j;
 
@@ -332,7 +332,7 @@ void finalization512(hashState *state, uint32_t *b)
 	}
 }
 
-typedef unsigned char BitSequence;
+//typedef unsigned char BitSequence;
 
 #define CUBEHASH_ROUNDS 16 /* this is r for CubeHashr/b */
 #define CUBEHASH_BLOCKBYTES 32 /* this is b for CubeHashr/b */
@@ -480,7 +480,8 @@ static __device__ __forceinline__ void rrounds(uint32_t x[2][2][2][2][2])
 }
 
 
-static __device__ __forceinline__ void block_tox(uint32_t *in, uint32_t x[2][2][2][2][2])
+__device__ __forceinline__
+static void block_tox(uint32_t *in, uint32_t x[2][2][2][2][2])
 {
 	int k;
 	int l;
@@ -496,7 +497,8 @@ static __device__ __forceinline__ void block_tox(uint32_t *in, uint32_t x[2][2][
 				x[0][0][k][l][m] ^= *in++;
 }
 
-static __device__ __forceinline__ void hash_fromx(uint32_t *out, uint32_t x[2][2][2][2][2])
+__device__ __forceinline__
+static void hash_fromx(uint32_t *out, uint32_t x[2][2][2][2][2])
 {
 	int j;
 	int k;
@@ -556,7 +558,8 @@ void __device__ __forceinline__ Init(uint32_t x[2][2][2][2][2])
 #endif
 }
 
-void __device__ __forceinline__ Update32(uint32_t x[2][2][2][2][2], const BitSequence *data)
+__device__ __forceinline__
+static void Update32(uint32_t x[2][2][2][2][2], const BitSequence *data)
 {
 	/* "xor the block into the first b bytes of the state" */
 	/* "and then transform the state invertibly through r identical rounds" */
@@ -564,7 +567,8 @@ void __device__ __forceinline__ Update32(uint32_t x[2][2][2][2][2], const BitSeq
 	rrounds(x);
 }
 
-void __device__ __forceinline__ Final(uint32_t x[2][2][2][2][2], BitSequence *hashval)
+__device__ __forceinline__
+static void Final(uint32_t x[2][2][2][2][2], BitSequence *hashval)
 {
 	int i;
 
@@ -581,7 +585,7 @@ void __device__ __forceinline__ Final(uint32_t x[2][2][2][2][2], BitSequence *ha
 
 
 /***************************************************/
-// Die Hash-Funktion
+// Hash Function
 __global__
 void x11_luffaCubehash512_gpu_hash_64(int threads, uint32_t startNounce, uint64_t *g_hash, uint32_t *g_nonceVector)
 {
@@ -619,25 +623,23 @@ void x11_luffaCubehash512_gpu_hash_64(int threads, uint32_t startNounce, uint64_
 }
 
 
-// Setup-Funktionen
+// Setup
-__host__ void x11_luffaCubehash512_cpu_init(int thr_id, int threads)
+__host__
+void x11_luffaCubehash512_cpu_init(int thr_id, int threads)
 {
 	cudaMemcpyToSymbol(c_IV, h_IV, sizeof(h_IV), 0, cudaMemcpyHostToDevice);
 	cudaMemcpyToSymbol(c_CNS, h_CNS, sizeof(h_CNS), 0, cudaMemcpyHostToDevice);
 }
 
-__host__ void x11_luffaCubehash512_cpu_hash_64(int thr_id, int threads, uint32_t startNounce, uint32_t *d_nonceVector, uint32_t *d_hash, int order)
+__host__
+void x11_luffaCubehash512_cpu_hash_64(int thr_id, int threads, uint32_t startNounce, uint32_t *d_nonceVector, uint32_t *d_hash, int order)
 {
 	const int threadsperblock = 256;
 
-    // berechne wie viele Thread Blocks wir brauchen
 	dim3 grid((threads + threadsperblock-1)/threadsperblock);
 	dim3 block(threadsperblock);
 
-    // Größe des dynamischen Shared Memory Bereichs
+	x11_luffaCubehash512_gpu_hash_64 <<< grid, block >>> (threads, startNounce, (uint64_t*)d_hash, d_nonceVector);
-    size_t shared_size = 0;
-
-	x11_luffaCubehash512_gpu_hash_64 << <grid, block, shared_size >> >(threads, startNounce, (uint64_t*)d_hash, d_nonceVector);
 	MyStreamSynchronize(NULL, order, thr_id);
 }