quda-ref/v0.7.0/cov_dev__mu1__core_8h_source.html

 // *** CUDA DSLASH ***


 //#define DSLASH_SHARED_FLOATS_PER_THREAD 0


 #if ((CUDA_VERSION >= 4010) && (__COMPUTE_CAPABILITY__ >= 200)) // NVVM compiler

 #define VOLATILE

 #else // Open64 compiler

 #define VOLATILE volatile

 #endif

 // input spinor

 #ifdef SPINOR_DOUBLE

 #define spinorFloat double

 #define i00_re I0.x

 #define i00_im I0.y

 #define i01_re I1.x

 #define i01_im I1.y

 #define i02_re I2.x

 #define i02_im I2.y

 #define i10_re I3.x

 #define i10_im I3.y

 #define i11_re I4.x

 #define i11_im I4.y

 #define i12_re I5.x

 #define i12_im I5.y

 #define i20_re I6.x

 #define i20_im I6.y

 #define i21_re I7.x

 #define i21_im I7.y

 #define i22_re I8.x

 #define i22_im I8.y

 #define i30_re I9.x

 #define i30_im I9.y

 #define i31_re I10.x

 #define i31_im I10.y

 #define i32_re I11.x

 #define i32_im I11.y

 #else

 #define spinorFloat float

 #define i00_re I0.x

 #define i00_im I0.y

 #define i01_re I0.z

 #define i01_im I0.w

 #define i02_re I1.x

 #define i02_im I1.y

 #define i10_re I1.z

 #define i10_im I1.w

 #define i11_re I2.x

 #define i11_im I2.y

 #define i12_re I2.z

 #define i12_im I2.w

 #define i20_re I3.x

 #define i20_im I3.y

 #define i21_re I3.z

 #define i21_im I3.w

 #define i22_re I4.x

 #define i22_im I4.y

 #define i30_re I4.z

 #define i30_im I4.w

 #define i31_re I5.x

 #define i31_im I5.y

 #define i32_re I5.z

 #define i32_im I5.w

 #endif // SPINOR_DOUBLE


 // gauge link

 #ifdef GAUGE_FLOAT2

 #define g00_re G0.x

 #define g00_im G0.y

 #define g01_re G1.x

 #define g01_im G1.y

 #define g02_re G2.x

 #define g02_im G2.y

 #define g10_re G3.x

 #define g10_im G3.y

 #define g11_re G4.x

 #define g11_im G4.y

 #define g12_re G5.x

 #define g12_im G5.y

 #define g20_re G6.x

 #define g20_im G6.y

 #define g21_re G7.x

 #define g21_im G7.y

 #define g22_re G8.x

 #define g22_im G8.y


 #else

 #define g00_re G0.x

 #define g00_im G0.y

 #define g01_re G0.z

 #define g01_im G0.w

 #define g02_re G1.x

 #define g02_im G1.y

 #define g10_re G1.z

 #define g10_im G1.w

 #define g11_re G2.x

 #define g11_im G2.y

 #define g12_re G2.z

 #define g12_im G2.w

 #define g20_re G3.x

 #define g20_im G3.y

 #define g21_re G3.z

 #define g21_im G3.w

 #define g22_re G4.x

 #define g22_im G4.y


 #endif // GAUGE_DOUBLE


 // conjugated gauge link

 #define gT00_re (+g00_re)

 #define gT00_im (-g00_im)

 #define gT01_re (+g10_re)

 #define gT01_im (-g10_im)

 #define gT02_re (+g20_re)

 #define gT02_im (-g20_im)

 #define gT10_re (+g01_re)

 #define gT10_im (-g01_im)

 #define gT11_re (+g11_re)

 #define gT11_im (-g11_im)

 #define gT12_re (+g21_re)

 #define gT12_im (-g21_im)

 #define gT20_re (+g02_re)

 #define gT20_im (-g02_im)

 #define gT21_re (+g12_re)

 #define gT21_im (-g12_im)

 #define gT22_re (+g22_re)

 #define gT22_im (-g22_im)


 // output spinor

 VOLATILE spinorFloat o00_re;

 VOLATILE spinorFloat o00_im;

 VOLATILE spinorFloat o01_re;

 VOLATILE spinorFloat o01_im;

 VOLATILE spinorFloat o02_re;

 VOLATILE spinorFloat o02_im;

 VOLATILE spinorFloat o10_re;

 VOLATILE spinorFloat o10_im;

 VOLATILE spinorFloat o11_re;

 VOLATILE spinorFloat o11_im;

 VOLATILE spinorFloat o12_re;

 VOLATILE spinorFloat o12_im;

 VOLATILE spinorFloat o20_re;

 VOLATILE spinorFloat o20_im;

 VOLATILE spinorFloat o21_re;

 VOLATILE spinorFloat o21_im;

 VOLATILE spinorFloat o22_re;

 VOLATILE spinorFloat o22_im;

 VOLATILE spinorFloat o30_re;

 VOLATILE spinorFloat o30_im;

 VOLATILE spinorFloat o31_re;

 VOLATILE spinorFloat o31_im;

 VOLATILE spinorFloat o32_re;

 VOLATILE spinorFloat o32_im;


 #include "read_gauge.h"

 #include "io_spinor.h"


 int x1, x2, x3, x4;

 int X;


 #if (defined MULTI_GPU) && (DD_PREC==2) // half precision

 int sp_norm_idx;

 #endif // MULTI_GPU half precision


 int sid;


 sid = blockIdx.x*blockDim.x + threadIdx.x;

 if (sid >= param.threads) return;


 #ifdef MULTI_GPU

 int face_idx;

 if (kernel_type == INTERIOR_KERNEL) {

 #endif


   // Inline by hand for the moment and assume even dimensions

   //coordsFromIndex(X, x1, x2, x3, x4, sid, param.parity);


   X = 2*sid;

   int aux1 = X / X1;

   x1 = X - aux1 * X1;

   int aux2 = aux1 / X2;

   x2 = aux1 - aux2 * X2;

   x4 = aux2 / X3;

   x3 = aux2 - x4 * X3;

   aux1 = (param.parity + x4 + x3 + x2) & 1;

   x1 += aux1;

   X += aux1;


   o00_re = 0;  o00_im = 0;

   o01_re = 0;  o01_im = 0;

   o02_re = 0;  o02_im = 0;

   o10_re = 0;  o10_im = 0;

   o11_re = 0;  o11_im = 0;

   o12_re = 0;  o12_im = 0;

   o20_re = 0;  o20_im = 0;

   o21_re = 0;  o21_im = 0;

   o22_re = 0;  o22_im = 0;

   o30_re = 0;  o30_im = 0;

   o31_re = 0;  o31_im = 0;

   o32_re = 0;  o32_im = 0;


 #ifdef MULTI_GPU

 } else { // exterior kernel


   const int dim = static_cast<int>(kernel_type);

   const int face_volume = param.threads;           // volume of one face

   const int face_num = 1;


   face_idx = sid;               // index into the respective face


   // ghostOffset is scaled to include body (includes stride) and number of FloatN arrays (SPINOR_HOP)

   // face_idx not sid since faces are spin projected and share the same volume index (modulo UP/DOWN reading)

   //sp_idx = face_idx + param.ghostOffset[dim];


 #if (DD_PREC==2) // half precision

   sp_norm_idx = sid + param.ghostNormOffset[static_cast<int>(kernel_type)];

 #endif


   const int dims[] = {X1, X2, X3, X4};

   coordsFromFaceIndex<1>(X, sid, x1, x2, x3, x4, face_idx, face_volume, dim, face_num, param.parity, dims);


   o00_re = 0.;  o00_im = 0.;

   o01_re = 0.;  o01_im = 0.;

   o02_re = 0.;  o02_im = 0.;

   o10_re = 0.;  o10_im = 0.;

   o11_re = 0.;  o11_im = 0.;

   o12_re = 0.;  o12_im = 0.;

   o20_re = 0.;  o20_im = 0.;

   o21_re = 0.;  o21_im = 0.;

   o22_re = 0.;  o22_im = 0.;

   o30_re = 0.;  o30_im = 0.;

   o31_re = 0.;  o31_im = 0.;

   o32_re = 0.;  o32_im = 0.;

 }

 #endif // MULTI_GPU


 #ifdef MULTI_GPU

 if ( (kernel_type == INTERIOR_KERNEL && (!param.ghostDim[1] || x2<X2m1)) ||

      (kernel_type == EXTERIOR_KERNEL_Y && x2==X2m1) )

 #endif

 {

   // Projector P1-

   // 1 0 0 -1

   // 0 1 1 0

   // 0 1 1 0

   // -1 0 0 1


 #ifdef MULTI_GPU

   const int sp_idx = (kernel_type == INTERIOR_KERNEL) ? (x2==X2m1 ? X-X2X1mX1 : X+X1) >> 1 :

     face_idx + param.ghostOffset[static_cast<int>(kernel_type)];

 #else

   const int sp_idx = (x2==X2m1 ? X-X2X1mX1 : X+X1) >> 1;

 #endif


   const int ga_idx = sid;


   // read gauge matrix from device memory

   READ_GAUGE_MATRIX(G, GAUGE0TEX, 2, ga_idx, ga_stride);


 #ifdef MULTI_GPU

   if (kernel_type == INTERIOR_KERNEL) {

 #endif


     // read spinor from device memory

     READ_SPINOR(SPINORTEX, param.sp_stride, sp_idx, sp_idx);


     // Do nothing useful with the spinors

   // reconstruct gauge matrix

   RECONSTRUCT_GAUGE_MATRIX(2);


   // multiply row 0

   o00_re += g00_re * i00_re;

   o00_re -= g00_im * i00_im;

   o00_re += g01_re * i01_re;

   o00_re -= g01_im * i01_im;

   o00_re += g02_re * i02_re;

   o00_re -= g02_im * i02_im;

   o00_im += g00_re * i00_im;

   o00_im += g00_im * i00_re;

   o00_im += g01_re * i01_im;

   o00_im += g01_im * i01_re;

   o00_im += g02_re * i02_im;

   o00_im += g02_im * i02_re;

   o10_re += g00_re * i10_re;

   o10_re -= g00_im * i10_im;

   o10_re += g01_re * i11_re;

   o10_re -= g01_im * i11_im;

   o10_re += g02_re * i12_re;

   o10_re -= g02_im * i12_im;

   o10_im += g00_re * i10_im;

   o10_im += g00_im * i10_re;

   o10_im += g01_re * i11_im;

   o10_im += g01_im * i11_re;

   o10_im += g02_re * i12_im;

   o10_im += g02_im * i12_re;

   o20_re += g00_re * i20_re;

   o20_re -= g00_im * i20_im;

   o20_re += g01_re * i21_re;

   o20_re -= g01_im * i21_im;

   o20_re += g02_re * i22_re;

   o20_re -= g02_im * i22_im;

   o20_im += g00_re * i20_im;

   o20_im += g00_im * i20_re;

   o20_im += g01_re * i21_im;

   o20_im += g01_im * i21_re;

   o20_im += g02_re * i22_im;

   o20_im += g02_im * i22_re;

   o30_re += g00_re * i30_re;

   o30_re -= g00_im * i30_im;

   o30_re += g01_re * i31_re;

   o30_re -= g01_im * i31_im;

   o30_re += g02_re * i32_re;

   o30_re -= g02_im * i32_im;

   o30_im += g00_re * i30_im;

   o30_im += g00_im * i30_re;

   o30_im += g01_re * i31_im;

   o30_im += g01_im * i31_re;

   o30_im += g02_re * i32_im;

   o30_im += g02_im * i32_re;


   // multiply row 1

   o01_re += g10_re * i00_re;

   o01_re -= g10_im * i00_im;

   o01_re += g11_re * i01_re;

   o01_re -= g11_im * i01_im;

   o01_re += g12_re * i02_re;

   o01_re -= g12_im * i02_im;

   o01_im += g10_re * i00_im;

   o01_im += g10_im * i00_re;

   o01_im += g11_re * i01_im;

   o01_im += g11_im * i01_re;

   o01_im += g12_re * i02_im;

   o01_im += g12_im * i02_re;

   o11_re += g10_re * i10_re;

   o11_re -= g10_im * i10_im;

   o11_re += g11_re * i11_re;

   o11_re -= g11_im * i11_im;

   o11_re += g12_re * i12_re;

   o11_re -= g12_im * i12_im;

   o11_im += g10_re * i10_im;

   o11_im += g10_im * i10_re;

   o11_im += g11_re * i11_im;

   o11_im += g11_im * i11_re;

   o11_im += g12_re * i12_im;

   o11_im += g12_im * i12_re;

   o21_re += g10_re * i20_re;

   o21_re -= g10_im * i20_im;

   o21_re += g11_re * i21_re;

   o21_re -= g11_im * i21_im;

   o21_re += g12_re * i22_re;

   o21_re -= g12_im * i22_im;

   o21_im += g10_re * i20_im;

   o21_im += g10_im * i20_re;

   o21_im += g11_re * i21_im;

   o21_im += g11_im * i21_re;

   o21_im += g12_re * i22_im;

   o21_im += g12_im * i22_re;

   o31_re += g10_re * i30_re;

   o31_re -= g10_im * i30_im;

   o31_re += g11_re * i31_re;

   o31_re -= g11_im * i31_im;

   o31_re += g12_re * i32_re;

   o31_re -= g12_im * i32_im;

   o31_im += g10_re * i30_im;

   o31_im += g10_im * i30_re;

   o31_im += g11_re * i31_im;

   o31_im += g11_im * i31_re;

   o31_im += g12_re * i32_im;

   o31_im += g12_im * i32_re;


   // multiply row 2

   o02_re += g20_re * i00_re;

   o02_re -= g20_im * i00_im;

   o02_re += g21_re * i01_re;

   o02_re -= g21_im * i01_im;

   o02_re += g22_re * i02_re;

   o02_re -= g22_im * i02_im;

   o02_im += g20_re * i00_im;

   o02_im += g20_im * i00_re;

   o02_im += g21_re * i01_im;

   o02_im += g21_im * i01_re;

   o02_im += g22_re * i02_im;

   o02_im += g22_im * i02_re;

   o12_re += g20_re * i10_re;

   o12_re -= g20_im * i10_im;

   o12_re += g21_re * i11_re;

   o12_re -= g21_im * i11_im;

   o12_re += g22_re * i12_re;

   o12_re -= g22_im * i12_im;

   o12_im += g20_re * i10_im;

   o12_im += g20_im * i10_re;

   o12_im += g21_re * i11_im;

   o12_im += g21_im * i11_re;

   o12_im += g22_re * i12_im;

   o12_im += g22_im * i12_re;

   o22_re += g20_re * i20_re;

   o22_re -= g20_im * i20_im;

   o22_re += g21_re * i21_re;

   o22_re -= g21_im * i21_im;

   o22_re += g22_re * i22_re;

   o22_re -= g22_im * i22_im;

   o22_im += g20_re * i20_im;

   o22_im += g20_im * i20_re;

   o22_im += g21_re * i21_im;

   o22_im += g21_im * i21_re;

   o22_im += g22_re * i22_im;

   o22_im += g22_im * i22_re;

   o32_re += g20_re * i30_re;

   o32_re -= g20_im * i30_im;

   o32_re += g21_re * i31_re;

   o32_re -= g21_im * i31_im;

   o32_re += g22_re * i32_re;

   o32_re -= g22_im * i32_im;

   o32_im += g20_re * i30_im;

   o32_im += g20_im * i30_re;

   o32_im += g21_re * i31_im;

   o32_im += g21_im * i31_re;

   o32_im += g22_re * i32_im;

   o32_im += g22_im * i32_re;


 #ifdef MULTI_GPU

   //JARLLLL

   } else {


     const int sp_stride_pad = ghostFace[static_cast<int>(kernel_type)];


     // read full spinor from device memory

     READ_SPINOR(SPINORTEX, sp_stride_pad, sp_idx /*+ (SPINOR_HOP)*sp_stride_pad*/, sp_norm_idx);


     // reconstruct gauge matrix

     RECONSTRUCT_GAUGE_MATRIX(2);


     // multiply row 0

     o00_re += g00_re * i00_re;

     o00_re -= g00_im * i00_im;

     o00_re += g01_re * i01_re;

     o00_re -= g01_im * i01_im;

     o00_re += g02_re * i02_re;

     o00_re -= g02_im * i02_im;

     o00_im += g00_re * i00_im;

     o00_im += g00_im * i00_re;

     o00_im += g01_re * i01_im;

     o00_im += g01_im * i01_re;

     o00_im += g02_re * i02_im;

     o00_im += g02_im * i02_re;

     o10_re += g00_re * i10_re;

     o10_re -= g00_im * i10_im;

     o10_re += g01_re * i11_re;

     o10_re -= g01_im * i11_im;

     o10_re += g02_re * i12_re;

     o10_re -= g02_im * i12_im;

     o10_im += g00_re * i10_im;

     o10_im += g00_im * i10_re;

     o10_im += g01_re * i11_im;

     o10_im += g01_im * i11_re;

     o10_im += g02_re * i12_im;

     o10_im += g02_im * i12_re;

     o20_re += g00_re * i20_re;

     o20_re -= g00_im * i20_im;

     o20_re += g01_re * i21_re;

     o20_re -= g01_im * i21_im;

     o20_re += g02_re * i22_re;

     o20_re -= g02_im * i22_im;

     o20_im += g00_re * i20_im;

     o20_im += g00_im * i20_re;

     o20_im += g01_re * i21_im;

     o20_im += g01_im * i21_re;

     o20_im += g02_re * i22_im;

     o20_im += g02_im * i22_re;

     o30_re += g00_re * i30_re;

     o30_re -= g00_im * i30_im;

     o30_re += g01_re * i31_re;

     o30_re -= g01_im * i31_im;

     o30_re += g02_re * i32_re;

     o30_re -= g02_im * i32_im;

     o30_im += g00_re * i30_im;

     o30_im += g00_im * i30_re;

     o30_im += g01_re * i31_im;

     o30_im += g01_im * i31_re;

     o30_im += g02_re * i32_im;

     o30_im += g02_im * i32_re;


     // multiply row 1

     o01_re += g10_re * i00_re;

     o01_re -= g10_im * i00_im;

     o01_re += g11_re * i01_re;

     o01_re -= g11_im * i01_im;

     o01_re += g12_re * i02_re;

     o01_re -= g12_im * i02_im;

     o01_im += g10_re * i00_im;

     o01_im += g10_im * i00_re;

     o01_im += g11_re * i01_im;

     o01_im += g11_im * i01_re;

     o01_im += g12_re * i02_im;

     o01_im += g12_im * i02_re;

     o11_re += g10_re * i10_re;

     o11_re -= g10_im * i10_im;

     o11_re += g11_re * i11_re;

     o11_re -= g11_im * i11_im;

     o11_re += g12_re * i12_re;

     o11_re -= g12_im * i12_im;

     o11_im += g10_re * i10_im;

     o11_im += g10_im * i10_re;

     o11_im += g11_re * i11_im;

     o11_im += g11_im * i11_re;

     o11_im += g12_re * i12_im;

     o11_im += g12_im * i12_re;

     o21_re += g10_re * i20_re;

     o21_re -= g10_im * i20_im;

     o21_re += g11_re * i21_re;

     o21_re -= g11_im * i21_im;

     o21_re += g12_re * i22_re;

     o21_re -= g12_im * i22_im;

     o21_im += g10_re * i20_im;

     o21_im += g10_im * i20_re;

     o21_im += g11_re * i21_im;

     o21_im += g11_im * i21_re;

     o21_im += g12_re * i22_im;

     o21_im += g12_im * i22_re;

     o31_re += g10_re * i30_re;

     o31_re -= g10_im * i30_im;

     o31_re += g11_re * i31_re;

     o31_re -= g11_im * i31_im;

     o31_re += g12_re * i32_re;

     o31_re -= g12_im * i32_im;

     o31_im += g10_re * i30_im;

     o31_im += g10_im * i30_re;

     o31_im += g11_re * i31_im;

     o31_im += g11_im * i31_re;

     o31_im += g12_re * i32_im;

     o31_im += g12_im * i32_re;


     // multiply row 2

     o02_re += g20_re * i00_re;

     o02_re -= g20_im * i00_im;

     o02_re += g21_re * i01_re;

     o02_re -= g21_im * i01_im;

     o02_re += g22_re * i02_re;

     o02_re -= g22_im * i02_im;

     o02_im += g20_re * i00_im;

     o02_im += g20_im * i00_re;

     o02_im += g21_re * i01_im;

     o02_im += g21_im * i01_re;

     o02_im += g22_re * i02_im;

     o02_im += g22_im * i02_re;

     o12_re += g20_re * i10_re;

     o12_re -= g20_im * i10_im;

     o12_re += g21_re * i11_re;

     o12_re -= g21_im * i11_im;

     o12_re += g22_re * i12_re;

     o12_re -= g22_im * i12_im;

     o12_im += g20_re * i10_im;

     o12_im += g20_im * i10_re;

     o12_im += g21_re * i11_im;

     o12_im += g21_im * i11_re;

     o12_im += g22_re * i12_im;

     o12_im += g22_im * i12_re;

     o22_re += g20_re * i20_re;

     o22_re -= g20_im * i20_im;

     o22_re += g21_re * i21_re;

     o22_re -= g21_im * i21_im;

     o22_re += g22_re * i22_re;

     o22_re -= g22_im * i22_im;

     o22_im += g20_re * i20_im;

     o22_im += g20_im * i20_re;

     o22_im += g21_re * i21_im;

     o22_im += g21_im * i21_re;

     o22_im += g22_re * i22_im;

     o22_im += g22_im * i22_re;

     o32_re += g20_re * i30_re;

     o32_re -= g20_im * i30_im;

     o32_re += g21_re * i31_re;

     o32_re -= g21_im * i31_im;

     o32_re += g22_re * i32_re;

     o32_re -= g22_im * i32_im;

     o32_im += g20_re * i30_im;

     o32_im += g20_im * i30_re;

     o32_im += g21_re * i31_im;

     o32_im += g21_im * i31_re;

     o32_im += g22_re * i32_im;

     o32_im += g22_im * i32_re;


   }

 #endif // MULTI_GPU


 }


 // write spinor field back to device memory

 WRITE_SPINOR(param.sp_stride);


 // undefine to prevent warning when precision is changed

 #undef spinorFloat

 #undef SHARED_STRIDE


 #undef g00_re

 #undef g00_im

 #undef gT00_re

 #undef gT00_im

 #undef g01_re

 #undef g01_im

 #undef gT01_re

 #undef gT01_im

 #undef g02_re

 #undef g02_im

 #undef gT02_re

 #undef gT02_im

 #undef g10_re

 #undef g10_im

 #undef gT10_re

 #undef gT10_im

 #undef g11_re

 #undef g11_im

 #undef gT11_re

 #undef gT11_im

 #undef g12_re

 #undef g12_im

 #undef gT12_re

 #undef gT12_im

 #undef g20_re

 #undef g20_im

 #undef gT20_re

 #undef gT20_im

 #undef g21_re

 #undef g21_im

 #undef gT21_re

 #undef gT21_im

 #undef g22_re

 #undef g22_im

 #undef gT22_re

 #undef gT22_im


 #undef i00_re

 #undef i00_im

 #undef i01_re

 #undef i01_im

 #undef i02_re

 #undef i02_im

 #undef i10_re

 #undef i10_im

 #undef i11_re

 #undef i11_im

 #undef i12_re

 #undef i12_im

 #undef i20_re

 #undef i20_im

 #undef i21_re

 #undef i21_im

 #undef i22_re

 #undef i22_im

 #undef i30_re

 #undef i30_im

 #undef i31_re

 #undef i31_im

 #undef i32_re

 #undef i32_im


i31_im
#define i31_im
Definition: covDev_mu1_core.h:61

X2
__constant__ int X2
Definition: dslash_constants.h:107

g10_im
#define g10_im
Definition: covDev_mu1_core.h:95

i22_re
#define i22_re
Definition: covDev_mu1_core.h:56

o21_re
VOLATILE spinorFloat o21_re
Definition: covDev_mu1_core.h:144

g11_re
#define g11_re
Definition: covDev_mu1_core.h:96

X2X1mX1
__constant__ int X2X1mX1
Definition: dslash_constants.h:126

i32_im
#define i32_im
Definition: covDev_mu1_core.h:63

g02_re
#define g02_re
Definition: covDev_mu1_core.h:92

i30_re
#define i30_re
Definition: covDev_mu1_core.h:58

o32_re
VOLATILE spinorFloat o32_re
Definition: covDev_mu1_core.h:152

o10_im
VOLATILE spinorFloat o10_im
Definition: covDev_mu1_core.h:137

spinorFloat
#define spinorFloat
Definition: covDev_mu1_core.h:39

o31_re
VOLATILE spinorFloat o31_re
Definition: covDev_mu1_core.h:150

X1
__constant__ int X1
Definition: dslash_constants.h:106

sid
int sid
Definition: covDev_mu1_core.h:165

sp_idx
int sp_idx
Definition: dw_dslash5inv_core.h:157

o21_im
VOLATILE spinorFloat o21_im
Definition: covDev_mu1_core.h:145

face_idx
int face_idx
Definition: tm_ndeg_fused_exterior_dslash_core.h:196

g12_re
#define g12_re
Definition: covDev_mu1_core.h:98

o00_im
VOLATILE spinorFloat o00_im
Definition: covDev_mu1_core.h:131

i22_im
#define i22_im
Definition: covDev_mu1_core.h:57

o11_re
VOLATILE spinorFloat o11_re
Definition: covDev_mu1_core.h:138

i12_im
#define i12_im
Definition: covDev_mu1_core.h:51

g01_re
#define g01_re
Definition: covDev_mu1_core.h:90

RECONSTRUCT_GAUGE_MATRIX
RECONSTRUCT_GAUGE_MATRIX(2)

i12_re
#define i12_re
Definition: covDev_mu1_core.h:50

X
int X
Definition: covDev_mu1_core.h:159

i31_re
#define i31_re
Definition: covDev_mu1_core.h:60

param
QudaGaugeParam param
Definition: pack_test.cpp:17

g22_re
#define g22_re
Definition: covDev_mu1_core.h:104

aux2
int aux2
Definition: covDev_mu1_core.h:181

i02_im
#define i02_im
Definition: covDev_mu1_core.h:45

ghostFace
__constant__ int ghostFace[QUDA_MAX_DIM+1]
Definition: dslash_constants.h:158

o22_im
VOLATILE spinorFloat o22_im
Definition: covDev_mu1_core.h:147

o01_im
VOLATILE spinorFloat o01_im
Definition: covDev_mu1_core.h:133

o01_re
VOLATILE spinorFloat o01_re
Definition: covDev_mu1_core.h:132

g02_im
#define g02_im
Definition: covDev_mu1_core.h:93

o30_re
VOLATILE spinorFloat o30_re
Definition: covDev_mu1_core.h:148

EXTERIOR_KERNEL_Y
Definition: dslash_constants.h:5

o02_re
VOLATILE spinorFloat o02_re
Definition: covDev_mu1_core.h:134

GAUGE0TEX
#define GAUGE0TEX
Definition: covDev.h:112

x4
int x4
Definition: covDev_mu1_core.h:158

x1
int x1
Definition: covDev_mu1_core.h:158

READ_SPINOR
READ_SPINOR(SPINORTEX, param.sp_stride, sp_idx, sp_idx)

WRITE_SPINOR
WRITE_SPINOR(param.sp_stride)

dim
int dim
Definition: tm_ndeg_fused_exterior_dslash_core.h:195

i01_re
#define i01_re
Definition: covDev_mu1_core.h:42

o12_im
VOLATILE spinorFloat o12_im
Definition: covDev_mu1_core.h:141

g12_im
#define g12_im
Definition: covDev_mu1_core.h:99

io_spinor.h

X2m1
__constant__ int X2m1
Definition: dslash_constants.h:117

i20_im
#define i20_im
Definition: covDev_mu1_core.h:53

i11_im
#define i11_im
Definition: covDev_mu1_core.h:49

INTERIOR_KERNEL
Definition: dslash_constants.h:2

o31_im
VOLATILE spinorFloat o31_im
Definition: covDev_mu1_core.h:151

g22_im
#define g22_im
Definition: covDev_mu1_core.h:105

SPINORTEX
#define SPINORTEX
Definition: clover_def.h:40

g21_im
#define g21_im
Definition: covDev_mu1_core.h:103

g20_im
#define g20_im
Definition: covDev_mu1_core.h:101

g10_re
#define g10_re
Definition: covDev_mu1_core.h:94

i32_re
#define i32_re
Definition: covDev_mu1_core.h:62

i21_im
#define i21_im
Definition: covDev_mu1_core.h:55

aux1
int aux1
Definition: covDev_mu1_core.h:179

ga_idx
const int ga_idx
Definition: covDev_mu1_core.h:256

i21_re
#define i21_re
Definition: covDev_mu1_core.h:54

ga_stride
__constant__ int ga_stride
Definition: dslash_constants.h:156

x2
int x2
Definition: covDev_mu1_core.h:158

i10_re
#define i10_re
Definition: covDev_mu1_core.h:46

i10_im
#define i10_im
Definition: covDev_mu1_core.h:47

o22_re
VOLATILE spinorFloat o22_re
Definition: covDev_mu1_core.h:146

i01_im
#define i01_im
Definition: covDev_mu1_core.h:43

i00_re
#define i00_re
Definition: covDev_mu1_core.h:40

o20_re
VOLATILE spinorFloat o20_re
Definition: covDev_mu1_core.h:142

o12_re
VOLATILE spinorFloat o12_re
Definition: covDev_mu1_core.h:140

o20_im
VOLATILE spinorFloat o20_im
Definition: covDev_mu1_core.h:143

X3
__constant__ int X3
Definition: dslash_constants.h:108

i00_im
#define i00_im
Definition: covDev_mu1_core.h:41

x3
int x3
Definition: covDev_mu1_core.h:158

g01_im
#define g01_im
Definition: covDev_mu1_core.h:91

g00_re
#define g00_re
Definition: covDev_mu1_core.h:88

o10_re
VOLATILE spinorFloat o10_re
Definition: covDev_mu1_core.h:136

o30_im
VOLATILE spinorFloat o30_im
Definition: covDev_mu1_core.h:149

g20_re
#define g20_re
Definition: covDev_mu1_core.h:100

i30_im
#define i30_im
Definition: covDev_mu1_core.h:59

g21_re
#define g21_re
Definition: covDev_mu1_core.h:102

read_gauge.h

o11_im
VOLATILE spinorFloat o11_im
Definition: covDev_mu1_core.h:139

g00_im
#define g00_im
Definition: covDev_mu1_core.h:89

i11_re
#define i11_re
Definition: covDev_mu1_core.h:48

dims
const int dims[]
Definition: asym_wilson_clover_dslash_dagger_fermi_core.h:408

READ_GAUGE_MATRIX
READ_GAUGE_MATRIX(G, GAUGE0TEX, 2, ga_idx, ga_stride)

kernel_type
KernelType kernel_type
Definition: tm_ndeg_fused_exterior_dslash_core.h:13

o02_im
VOLATILE spinorFloat o02_im
Definition: covDev_mu1_core.h:135

g11_im
#define g11_im
Definition: covDev_mu1_core.h:97

VOLATILE
#define VOLATILE
Definition: covDev_mu1_core.h:9

o00_re
VOLATILE spinorFloat o00_re
Definition: covDev_mu1_core.h:130

X4
__constant__ int X4
Definition: dslash_constants.h:109

o32_im
VOLATILE spinorFloat o32_im
Definition: covDev_mu1_core.h:153

i20_re
#define i20_re
Definition: covDev_mu1_core.h:52

i02_re
#define i02_re
Definition: covDev_mu1_core.h:44