kernel_code
stringlengths
49
1.22M
kernel void test_fa1_3_1_kern(global float* a0, global float* a1, global float* a2, global float* b0) { unsigned int gtid = get_global_id(0); float tmp0 = a0[gtid] + 0.1f; float tmp1 = a1[gtid] + 1.1f; float tmp2 = a2[gtid] + 2.1f; barrier(0x01); b0[gtid] = (0.0f + 1.1f) * (+tmp0 + tmp1 + tmp2); }
kernel void evaluateCylinder(global const float* parameter, global const float* samplingValues, global float* outputBuffer, int numKnotsU, int numKnotsV) { int numElements = numKnotsU * numKnotsV; int iGID = get_global_id(0); if (iGID >= numElements) { return; } int gridWidth = (int)sqrt((float)numElem...
kernel void addBySaveVal(global float* grid, const float value, int const num_elements) { int global_id = get_global_id(0); if (global_id >= num_elements) return; if (isnan(grid[global_id]) || isinf(grid[global_id])) grid[global_id] = 0; grid[global_id] += value; }
kernel void serial_adjacent_find(const unsigned int size, global unsigned int* output, global int* _buf0) { unsigned int result = size; for (unsigned int i = 0; i < size - 1; i++) { if (_buf0[i] == _buf0[i + 1]) { result = i; break; } } *output = result; }
kernel void rdft(global float* x, global float* y, const int N) { int num_vectors = N / 4; float X_real = 0.0f; float X_imag = 0.0f; float4 input, arg, w_real, w_imag; float two_pi_k_over_N = 2 * 3.14159265358979323846264338327950288f * get_global_id(0) / N; for (int i = 0; i < num_vectors; i++) { ar...
inline void order(unsigned int* x, unsigned int a, unsigned int b, bool asc) { if (asc ^ (x[a] < x[b])) { unsigned int auxa = x[a]; unsigned int auxb = x[b]; x[a] = auxb; x[b] = auxa; } } inline void merge2(unsigned int* x, bool asc) { for (int j = 0; j < 1; j++) order(x, j, j + 1, asc); ; ...
kernel void vector_mult_opencl(unsigned int nx, global float* val, float factor) { const int i = get_global_id(0); if (i < nx) { val[i] *= factor; } }
kernel void half_log_float2(global float2* src_0, global float2* dst) { int gid = get_global_id(0); dst[gid] = half_log(src_0[gid]); }
kernel void same_location(global float* buffer, global float* dummyBuffer) { *dummyBuffer = buffer[0]; }
kernel void multiply_floatfloat(global float* src_0, global float* src_1, global float* dst) { int gid = get_global_id(0); dst[gid] = src_0[gid] * src_1[gid]; }
kernel void clkernel_sigmoid(const int num, global const float* in, global float* out) { const int i = get_global_id(0); if (i >= num) return; out[i] = 1 / (1 + exp(-(in[i]))); }
kernel void foo(global int* a) { *a = *a + 1; }
kernel void bitselect_long16long16long16(global long16* src_0, global long16* src_1, global long16* src_2, global long16* dst) { int gid = get_global_id(0); dst[gid] = bitselect(src_0[gid], src_1[gid], src_2[gid]); }
kernel void prefix_sum_init(global int* counts, global unsigned long* prefix_sum) { int thread_idx = get_global_id(0); int n = get_global_size(0); prefix_sum[thread_idx] = (thread_idx > 0) ? counts[thread_idx - 1] : 0; }
kernel void buffSizeTest(global int* v) { unsigned int i = get_global_id(0); v[i] = 0; }
kernel void copy(global int* _buf0, global int* _buf1, const unsigned int count) { unsigned int index = get_local_id(0) + (32 * get_group_id(0)); for (unsigned int i = 0; i < 4; i++) { if (index < count) { _buf0[index] = _buf1[index]; index += 8; } } }
kernel void cl_copy_weigthed_blend(global const float4* in, global float4* out) { int gid = get_global_id(0); float4 in_v = in[gid]; out[gid] = in_v; }
kernel void test_block_4_5_kern(global int* a0, global int* a1, global int* a2, global int* a3, global int* b0, global int* b1, global int* b2, global int* b3, global int* b4, local int* c0, local int* c1, local int* c2, local int* c3) { int i; unsigned int gtid = get_global_id(0); unsigned int ltid = get_local_i...
kernel void average(global const int* a, global const int* b, global int* out, int n) { int i = get_global_id(0); if (i >= n) return; out[i] = (95 * b[i]) / 100 + (5 * a[i]) / 100; }
kernel void test_switch_more(global int* out, global const int* in, global const int* cond) { unsigned int gid = get_global_id(0); int val = in[gid]; switch (cond[gid]) { case 0: out[gid] = -val; break; case -1: out[gid] = val * 5; break; case 1: out[gid] = val * 17; ...
inline float sigmoid(float z) { return 1.0f / (1.0f + exp(-z)); } inline float sigmoid_gradient(float y) { return y * (1.0f - y); } inline float relu(float z) { return z > 0 ? z : 0; } inline float relu_gradient(float y) { return y > 0 ? 1 : 0; } inline float tanh_gradient(float y) { return 1.0f - y * y; } ...
kernel void add_sat_ulong8ulong8(global ulong8* src_0, global ulong8* src_1, global ulong8* dst) { int gid = get_global_id(0); dst[gid] = add_sat(src_0[gid], src_1[gid]); }
kernel void isordered_float2float2(global float2* src_0, global float2* src_1, global int2* dst) { int gid = get_global_id(0); dst[gid] = (int2)(isordered(src_0[gid], src_1[gid])); }
kernel void pre_increment_uchar(global uchar* src_0, global uchar* dst) { int gid = get_global_id(0); dst[gid] = ++src_0[gid]; }
kernel void test_arg_7_1_kern(global float* a0, global float* a1, global float* a2, global float* a3, global float* a4, global float* a5, global float* a6, global float* b0) { unsigned int gtid = get_global_id(0); float tmp0 = a0[gtid] + 0.1f; float tmp1 = a1[gtid] + 1.1f; float tmp2 = a2[gtid] + 2.1f; float ...
kernel void atomTest3(global int* out) { int lid = get_local_id(0); int t1 = atomic_add(&out[0], lid); }
float calculation(global float* dataEndListArg, global float* dataTurnoverListArg, global float* successListArg, int dataSize, int samplingSize, global float* parameterListArgs, int parameterCellSize, global float* resultListArgs, int resultCellSize, int i, int x) { float result = 0; for (int j = samplingSize - 1;...
kernel void kernel0_local_read(global char* ptr, unsigned long memsize, volatile global unsigned int* err_count, global unsigned long* err_addr, global unsigned long* err_expect, global unsigned long* err_current, global unsigned long* err_second_read) { int i; global unsigned long* buf = (global unsigned long*)ptr...
kernel void default_kernel(global uchar* dst, global const uchar* src, const unsigned int size) { int gid = get_global_id(0); int lid = get_local_id(0); dst[gid] = src[gid]; }
float Phi(float X) { float y, absX, t; const float c1 = 0.319381530f; const float c2 = -0.356563782f; const float c3 = 1.781477937f; const float c4 = -1.821255978f; const float c5 = 1.330274429f; const float oneBySqrt2pi = 0.398942280f; absX = fabs(X); t = 1.0f / (1.0f + 0.2316419f * absX); y = ...
kernel void AtomicSum(global int* s) { const int i = 1; atomic_add(s, i); }
kernel void pre_decrement_char16(global char16* src_0, global char16* dst) { int gid = get_global_id(0); dst[gid] = --src_0[gid]; }
kernel void foo(global char* out, int a, int b) { char x; if (a == 10 || b == 2) x = 2; else if (b == 12 && a == 4) x = 3; else if (b != 3 || a != 5) x = 1; else x = 0; *out = x; }
kernel void hypot_float4float4(global float4* src_0, global float4* src_1, global float4* dst) { int gid = get_global_id(0); dst[gid] = hypot(src_0[gid], src_1[gid]); }
kernel void sub_sat_uintuint(global unsigned int* src_0, global unsigned int* src_1, global unsigned int* dst) { int gid = get_global_id(0); dst[gid] = sub_sat(src_0[gid], src_1[gid]); }
kernel void isless_float4float4(global float4* src_0, global float4* src_1, global int4* dst) { int gid = get_global_id(0); dst[gid] = (int4)(isless(src_0[gid], src_1[gid])); }
kernel void mc_kernel_gridGrad(int volnx, int volny, int volnz, float edgex, float edgey, float edgez, global float4* mcgrid, global float4* mcgridGrad) { size_t gid = get_global_id(0); if (mcgridGrad[gid].w != -1.0) { mcgridGrad[gid].x = (mcgrid[gid - 1].w - mcgrid[gid + 1].w) / edgex; mcgridGrad[gid].y =...
float boundRange(float x, float lowerLimit, float upperLimit) { return (x < lowerLimit ? lowerLimit : (x > upperLimit ? upperLimit : x)); } float Logistic_fn(float x) { if (x < 88.722839f) { if (x > -88.722839f) return (float)1.0 / ((float)1.0 + exp(-x)); else return 0; } return 1; } floa...
kernel void serial_insertion_sort_by_key_float(local float* keys, local float4* data, unsigned int n, global float* _buf0, global float4* _buf1) { for (unsigned int i = 0; i < n; i++) { keys[i] = _buf0[i]; data[i] = _buf1[i]; } for (unsigned int i = 1; i < n; i++) { const float key = keys[i]; cons...
kernel void clkernel_divide_scalar_matx(const int num, global const float* in1, const float x, global float* out) { const int i = get_global_id(0); if (i >= num) return; out[i] = in1[i] / x; }
kernel void relational_less_than_ucharuchar(global uchar* src_0, global uchar* src_1, global int* dst) { int gid = get_global_id(0); dst[gid] = (int)(src_0[gid] < src_1[gid]); }
kernel void SignedDistanceTransformUnsignedByte(global const unsigned char* vIn, global unsigned char* vOut, int iDx, int iDy, int max_dist) { int iGID = get_global_id(0); if (iGID >= (iDx * iDy)) { return; } int idX = iGID % iDy; int idY = iGID / iDy; float2 pos = (float2)(idX, idY); int minX = cla...
kernel void multiply_kernel2D(global float2* projFFT, int4 fftDimension, global float2* kernelFFT) { const unsigned int idx = get_global_id(0); const unsigned int w = idx % (fftDimension.x * fftDimension.y); int j = w / fftDimension.x; int i = w % fftDimension.x; long int kernel_idx = i + j * fftDimension.x...
kernel void search(unsigned int p_count, global char* _buf0, global char* _buf1, global unsigned int* _buf2) { unsigned int i = get_global_id(0); const unsigned int i1 = i; unsigned int j; for (j = 0; j < p_count; j++, i++) { if (_buf0[j] != _buf1[i]) j = p_count + 1; } if (j == p_count) _buf2...
kernel void test_min(const global float* x, const global float* y, global float* out) { unsigned int gid = get_global_id(0); out[gid] = min(x[gid], y[gid]); }
kernel void div(global int* out, int a, int b) { out[0] = a / b; }
double dev_round_double(double value) { int newValue = (int)(value); if (value - newValue < .5f) return newValue; else return newValue++; } double calcLikelihoodSum(global unsigned char* I, global int* ind, int numOnes, int index) { double likelihoodSum = 0.0; int x; for (x = 0; x < numOnes; x++) ...
float boundRange(float x, float lowerLimit, float upperLimit) { return (x < lowerLimit ? lowerLimit : (x > upperLimit ? upperLimit : x)); } float Logistic_fn(float x) { if (x < 88.722839f) { if (x > -88.722839f) return (float)1.0 / ((float)1.0 + exp(-x)); else return 0; } return 1; } floa...
kernel void square(global int* data, unsigned int size) { unsigned int gid = get_global_id(0); if (gid < size) { data[gid] = data[gid] * data[gid]; } }
kernel void matvec_mul4_float(global const float* A, int offA, unsigned int A_col_size, unsigned int trail_item, global const float* v, int offv, float alpha, float beta, global float4* result, int offr, local float4* work) { unsigned int row_gid = get_group_id(0); unsigned int lid = get_local_id(0); const global...
inline void addData1024(volatile local unsigned int* s_WarpHist, unsigned int data, unsigned int tag) { unsigned int count; do { count = s_WarpHist[data] & 0x07FFFFFFU; count = tag | (count + 1); s_WarpHist[data] = count; } while (s_WarpHist[data] != count); } kernel void histogram1024Kernel(global u...
kernel void FillBufferLeftLeftover(global uchar* pDst, unsigned int dstOffsetInBytes, const global uchar* pPattern, const unsigned int patternSizeInEls) { unsigned int gid = get_global_id(0); pDst[gid + dstOffsetInBytes] = pPattern[gid & (patternSizeInEls - 1)]; }
kernel void mergeData(global const int* input, unsigned int dataSize, unsigned int partsCount, global int* output, unsigned int outputSize) { int id = get_global_id(0); int itemsPerThread = outputSize / get_global_size(0); for (int i = 0; i < itemsPerThread; i++) { output[(id * itemsPerThread) + i] = input[...
kernel void fir(global int* in, global int* coeff, global int* out, int filter_len) { int index = get_global_id(0); int i = 0, acc = 0; do { acc += in[index + i] * coeff[i]; i++; } while (i != filter_len); out[index] = acc; }
kernel void unifiedBlockScan(global unsigned int* output, global unsigned int* input, local unsigned int* sharedMem, const unsigned int block_size) { int id = get_local_id(0); int gid = get_global_id(0); int bid = get_group_id(0); sharedMem[id] = input[gid]; unsigned int cache = sharedMem[0]; for (int st...
kernel void testWorkGroupSMax(int a, global int* res) { res[0] = work_group_reduce_max(a); }
kernel void subtractBy(global float* gridA, const global float* gridB, int const num_elements) { int global_id = get_global_id(0); if (global_id >= num_elements) return; gridA[global_id] -= gridB[global_id]; }
kernel void WipeBufferBoolKernel(global bool* buffer, const int nBufferElements) { const unsigned int i = get_global_id(0); if (i < nBufferElements) { buffer[i] = false; } }
kernel void builtin_tgamma(global float* src, global float* dst) { int i = get_global_id(0); dst[i] = tgamma(src[i]); }
kernel void prefixSum(global int* restrict output, global int* restrict input, const unsigned int in_size) { local int p_in[512]; local int p_out[512]; int base = 0; int cnt = in_size / 512; for (int i = 0; i < cnt; i++) { for (int b = 0; b < 512; b++) p_in[b] = input[i * 512 + b]; p_out[0] = ...
kernel void expontial(global float* grid, int const num_elements) { int global_id = get_global_id(0); if (global_id >= num_elements) return; grid[global_id] = exp(grid[global_id]); }
kernel void cosh_float16(global float16* src_0, global float16* dst) { int gid = get_global_id(0); dst[gid] = cosh(src_0[gid]); }
kernel void cube(global int* input, global int* output) { int i = get_global_id(0); output[i] = input[i] * input[i] * input[i]; }
kernel void test_clamp(global char3* out, global char3* in) { *out = clamp(*in, (char3)7, (char3)42); }
kernel void builtin_shuffle(global float* src1, global float* src2, global float* dst1, global float* dst2) { int i = get_global_id(0); float2 src = (float2)(src1[i], src2[i]); uint2 mask = (uint2)(1, 0); float2 dst = shuffle(src, mask); dst1[i] = dst.s0; dst2[i] = dst.s1; }
kernel void expm1_float2(global float2* src_0, global float2* dst) { int gid = get_global_id(0); dst[gid] = expm1(src_0[gid]); }
kernel void kernelFloat(float2 input, global float2* output) { output[0] = input; }
kernel void inverse(global float* a, global float* result) { int gid = get_global_id(0); result[gid] = 1.0f / a[gid]; }
int divRndUp(int n, int d) { return (n / d) + ((n % d) ? 1 : 0); } void storeComponents(global int* d_r, global int* d_g, global int* d_b, int r, int g, int b, int pos) { d_r[pos] = r - 128; d_g[pos] = g - 128; d_b[pos] = b - 128; } void storeComponent(global int* d_c, const int c, int pos) { d_c[pos] = c - 1...
kernel void plus(global short* out, short in) { out[0] = +in; }
kernel void test_switch_2_default(global int* out, global const int* in, global const int* cond) { unsigned int gid = get_global_id(0); int val = in[gid]; switch (cond[gid]) { case 17: out[gid] = val; break; default: out[gid] = -val; } }
kernel void CopyBufferToBufferBytes(const global uchar* pSrc, global uchar* pDst, unsigned int srcOffsetInBytes, unsigned int dstOffsetInBytes, unsigned int bytesToRead) { pSrc += (srcOffsetInBytes + get_global_id(0)); pDst += (dstOffsetInBytes + get_global_id(0)); pDst[0] = pSrc[0]; }
kernel void kernel_write(global char* ptr, unsigned long memsize, unsigned long p1) { int i; global unsigned long* buf = (global unsigned long*)ptr; int idx = get_global_id(0); unsigned long n = memsize / sizeof(unsigned long); int total_num_threads = get_global_size(0); for (i = idx; i < n; i += total_num...
float addLogs(float A, float B) { float maxi = fmax(A, B); float corr = native_log(1.0 + native_exp(-fabs(A - B))); return maxi + corr; } kernel void dataOut(global float* alpha, global float* beta, global float* gamma, global float* dataOUT) { int _outputs[16] = {0, 0, 1, 1, 1, 1, 0, 0, 3, 3, 2, 2, 2, 2, 3, 3...
kernel void fill(global float* grid, const float value, int const num_elements) { int global_id = get_global_id(0); if (global_id >= num_elements) return; grid[global_id] = value; }
kernel void reduce4_sl(global float* restrict out, global const float4* restrict in, int n_hex) { int i = get_global_id(0); int gws = get_global_size(0); if (i >= n_hex) return; float4 v0 = in[i + 0 * gws]; float4 v1 = in[i + 1 * gws]; float4 v2 = in[i + 2 * gws]; float4 v3 = in[i + 3 * gws]; float...
float calcFirstDer(float x11, float x21, float x31, float x12, float x22, float x32, float x13, float x23, float x33, float inputNodataValue, float outputNodataValue, float zFactor, float mCellSize) { int weight = 0; float sum = 0; if (x31 != inputNodataValue && x11 != inputNodataValue) { sum += (x31 - x11);...
kernel void post_increment_uchar(global uchar* src_0, global uchar* dst) { int gid = get_global_id(0); dst[gid] = src_0[gid]++; }
kernel void prefix_sum(global int* restrict A, global int* restrict B, private int numOfIterations) { size_t tid = get_global_id(0); for (int d = 0; d < numOfIterations; ++d) { if (tid >= (1 << d)) { B[tid] = A[tid - (1 << d)] + A[tid]; } else { B[tid] = A[tid]; } barrier(0x02); gl...
kernel void test_arg_4_5_kern(global int4* a0, global int4* a1, global int4* a2, global int4* a3, global int4* b0, global int4* b1, global int4* b2, global int4* b3, global int4* b4) { int4 za = (int4)(0, 1, 2, 3); unsigned int gtid = get_global_id(0); int4 tmp0 = a0[gtid] + 0 + za; int4 tmp1 = a1[gtid] + 1 + z...
kernel void async_copy_loop(global int* data, local int* scratch) { int i = get_local_id(0); event_t event = 0; for (int j = 0; j < get_local_size(0); j++) { int offset = j; event = async_work_group_copy(scratch + offset, data + offset, 1, event); } wait_group_events(1, &event); data[get_local_siz...
kernel void ratxb_kernel(global const float* T, global const float* C, global float* RF, global float* RB, global const float* RKLOW, const float TCONV) { const float TEMP = T[get_global_id(0)] * TCONV; const float ALOGT = log((TEMP)); float CTOT = 0.0f; float PR, PCOR, PRLOG, FCENT, FCLOG, XN; float CPRLOG, ...
kernel void mainKernel(global const float4* a, const long num, global float4* c) { int iGID = get_global_id(0); c[iGID] = a[iGID]; }
int func_1(global int* in, int n) { return in[n]; } kernel void func_0(global int* in, global int* out, int n) { out[n] = func_1(in, n); }
kernel void Film_MergeBufferFinalize(const unsigned int filmWidth, const unsigned int filmHeight, global float* channel_IMAGEPIPELINE) { const size_t gid = get_global_id(0); if (gid > filmWidth * filmHeight) return; global float* channelBufferPixel = &channel_IMAGEPIPELINE[gid * 3]; const float pixelR = ch...
void max_pool_forward_impl_float(const int nthreads, global const float* bottom_data, const int num, const int channels, const int height, const int width, const int pooled_height, const int pooled_width, const int kernel_h, const int kernel_w, const int stride_h, const int stride_w, const int pad_h, const int pad_w, g...
kernel void Film_MergeRADIANCE_PER_SCREEN_NORMALIZED(const unsigned int filmWidth, const unsigned int filmHeight, global float* channel_IMAGEPIPELINE, global float* mergeBuffer, const float scaleR, const float scaleG, const float scaleB) { const size_t gid = get_global_id(0); if (gid > filmWidth * filmHeight) r...
kernel void serial_reduce_by_key(unsigned int count, global unsigned int* result_size, global int* _buf0, global int* _buf1, global int* _buf2, global int* _buf3) { int result = _buf0[0]; int previous_key = _buf1[0]; int value; int key; unsigned int size = 1; _buf2[0] = previous_key; _buf3[0] = result; ...
void SHA256(uint4* restrict state0, uint4* restrict state1, const uint4 block0, const uint4 block1, const uint4 block2, const uint4 block3) { uint4 S0 = *state0; uint4 S1 = *state1; uint4 W[4]; W[0].x = block0.x; S1.w += (rotate(S1.x, 26U) ^ rotate(S1.x, 21U) ^ rotate(S1.x, 7U)) + bitselect(S1.z, S1.y, S1.x)...
kernel void test_arg_4_3_kern(global int* a0, global int* a1, global int* a2, global int* a3, global int* b0, global int* b1, global int* b2) { unsigned int gtid = get_global_id(0); int tmp0 = a0[gtid] + 0; int tmp1 = a1[gtid] + 1; int tmp2 = a2[gtid] + 2; int tmp3 = a3[gtid] + 3; b0[gtid] = (0 + 1) * (+tmp...
kernel void kernelChar(char8 input, global char8* output) { output[0] = input; }
kernel void round_float4(global float4* src_0, global float4* dst) { int gid = get_global_id(0); dst[gid] = round(src_0[gid]); }
kernel void encrypt1(const global uint4* input, global uint4* output, const global uint2* constant_key, const global unsigned int* global_sbox1, const global unsigned int* global_sbox2, const global unsigned int* global_sbox3, const global unsigned int* global_sbox4, local unsigned int* sbox1, local unsigned int* sbox2...
kernel void kernel_round_withoutDD1(global float* result_round, int N) { float t1; float t2; float t3; float t4; float t5; float t6; float t7; float t8; float t9; float t10; float i = 1.0; float n = 0.3 * (float)(N); for (i = 1.0; i < n; i = i + 0.01) { t1 = round(i); t2 = round(i + 0...
kernel void MaxPoolingForward(global float* output, global float* input, global bool* switches, global int* poolingTable, const int inputVolume, const int inputArea, const int outputVolume, const int outputArea, const int miniBatchSize) { const int iOutput = get_global_id(0); if (iOutput < outputVolume * miniBatch...
; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; kernel void builtin_convert_float_to_short_sat(global float* src, global short* dst) { int i = get_global_id(0); dst[i] = convert_short_sat(src[i]); }
kernel void moveObjectsKernel(global float4* posOrnColors, int numObjects) { int iGID = get_global_id(0); if (iGID >= numObjects) return; global float4* positions = &posOrnColors[0]; if (iGID < 0.5 * numObjects) { positions[iGID].y += 0.01f; } global float4* colors = &posOrnColors[numObjects * 2]; ...
kernel void calculateBin(const unsigned int bin, global uchar4* sm_mapping) { unsigned char offset = bin % 4; unsigned char indexlo = (bin >> 2) % 256; unsigned char indexhi = (bin >> 10) % 2; unsigned char block = bin / 8; offset *= 8; uchar4 sm; sm.x = block; sm.y = indexhi; sm.z = indexlo; sm.w...
kernel void kernel_rsqrt_withDD8(global float* result_rsqrt, int N) { float t1 = 23456.234; float8 t2 = t1 + (float8)(1, 1, 1, 0, 1, 0, 1, 1.1); int i = 0; for (i = 0; i < N; i++) { t2 = rsqrt(t2); t2 = rsqrt(t2); t2 = rsqrt(t2); t2 = rsqrt(t2); t2 = rsqrt(t2); t2 = rsqrt(t2); t2 = r...
kernel void compiler_double_3(global float* src, global double* dst) { int i = get_global_id(0); float d = 1.234567890123456789f; dst[i] = i < 14 ? d : 14; }