QUDA
1.0.0
|
Public Member Functions | |
GenericPackGhostLauncher (Arg &arg, const ColorSpinorField &meta, MemoryLocation *destination) | |
virtual | ~GenericPackGhostLauncher () |
void | apply (const cudaStream_t &stream) |
void | setColorSpinBlock (TuneParam ¶m) const |
bool | advanceBlockDim (TuneParam ¶m) const |
int | blockStep () const |
int | blockMin () const |
bool | advanceAux (TuneParam ¶m) const |
TuneKey | tuneKey () const |
virtual void | initTuneParam (TuneParam ¶m) const |
virtual void | defaultTuneParam (TuneParam ¶m) const |
long long | flops () const |
long long | bytes () const |
![]() | |
TunableVectorYZ (unsigned int vector_length_y, unsigned int vector_length_z) | |
void | resizeVector (int y, int z) const |
void | resizeStep (int y, int z) const |
![]() | |
TunableVectorY (unsigned int vector_length_y) | |
void | resizeVector (int y) const |
void | resizeStep (int y) const |
![]() | |
Tunable () | |
virtual | ~Tunable () |
virtual void | preTune () |
virtual void | postTune () |
virtual int | tuningIter () const |
virtual std::string | paramString (const TuneParam ¶m) const |
virtual std::string | perfString (float time) const |
virtual bool | advanceTuneParam (TuneParam ¶m) const |
void | checkLaunchParam (TuneParam ¶m) |
CUresult | jitifyError () const |
CUresult & | jitifyError () |
Private Member Functions | |
unsigned int | minThreads () const |
bool | tuneGridDim () const |
bool | tuneAuxDim () const |
Private Attributes | |
Arg & | arg |
const ColorSpinorField & | meta |
Additional Inherited Members | |
![]() | |
virtual unsigned int | sharedBytesPerThread () const |
virtual unsigned int | sharedBytesPerBlock (const TuneParam ¶m) const |
![]() | |
virtual bool | tuneSharedBytes () const |
virtual bool | advanceGridDim (TuneParam ¶m) const |
virtual unsigned int | maxBlockSize (const TuneParam ¶m) const |
virtual unsigned int | maxGridSize () const |
virtual unsigned int | minGridSize () const |
virtual int | gridStep () const |
gridStep sets the step size when iterating the grid size in advanceGridDim. More... | |
virtual void | resetBlockDim (TuneParam ¶m) const |
unsigned int | maxBlocksPerSM () const |
For some reason this can't be queried from the device properties, so here we set set this. Based on Table 14 of the CUDA Programming Guide 10.0 (Technical Specifications per Compute Capability) More... | |
template<typename F > | |
void | setMaxDynamicSharedBytesPerBlock (F *func) const |
Enable the maximum dynamic shared bytes for the kernel "func" (values given by maxDynamicSharedBytesPerBlock()). More... | |
unsigned int | maxDynamicSharedBytesPerBlock () const |
This can't be correctly queried in CUDA for all architectures so here we set set this. Based on Table 14 of the CUDA Programming Guide 10.0 (Technical Specifications per Compute Capability). More... | |
virtual unsigned int | maxSharedBytesPerBlock () const |
The maximum shared memory that a CUDA thread block can use in the autotuner. This isn't necessarily the same as maxDynamicSharedMemoryPerBlock since that may need explicit opt in to enable (by calling setMaxDynamicSharedBytes for the kernel in question). If the CUDA kernel in question does this opt in then this function can be overloaded to return maxDynamicSharedBytesPerBlock. More... | |
virtual bool | advanceSharedBytes (TuneParam ¶m) const |
int | writeAuxString (const char *format,...) |
![]() | |
unsigned int | vector_length_y |
unsigned int | step_y |
bool | tune_block_x |
![]() | |
char | aux [TuneKey::aux_n] |
CUresult | jitify_error |
Definition at line 46 of file color_spinor_pack.cu.
|
inline |
Definition at line 54 of file color_spinor_pack.cu.
References quda::Tunable::aux, quda::LatticeField::AuxString(), comm_dim_partitioned(), comm_dim_partitioned_string(), comm_dim_topology_string(), quda::compile_type_str(), quda::Device, quda::Host, quda::LatticeField::Location(), and QUDA_CUDA_FIELD_LOCATION.
|
inlinevirtual |
Definition at line 83 of file color_spinor_pack.cu.
|
inlinevirtual |
Reimplemented from quda::Tunable.
Definition at line 137 of file color_spinor_pack.cu.
References quda::TuneParam::aux, quda::TunableVectorYZ::initTuneParam(), quda::Arg< real, Ns, Nc, order >::nParity, quda::TunableVectorYZ::resizeVector(), and quda::GenericPackGhostLauncher< Float, block_float, Ns, Ms, Nc, Mc, Arg >::setColorSpinBlock().
|
inlinevirtual |
Reimplemented from quda::TunableVectorYZ.
Definition at line 124 of file color_spinor_pack.cu.
References quda::Tunable::advanceBlockDim(), quda::TunableVectorYZ::advanceBlockDim(), and quda::GenericPackGhostLauncher< Float, block_float, Ns, Ms, Nc, Mc, Arg >::setColorSpinBlock().
|
inlinevirtual |
Implements quda::Tunable.
Definition at line 85 of file color_spinor_pack.cu.
References quda::GenericPackGhostLauncher< Float, block_float, Ns, Ms, Nc, Mc, Arg >::arg, quda::TuneParam::aux, quda::TuneParam::block, getTuning(), getVerbosity(), quda::TuneParam::grid, quda::Tunable::jitify_error, quda::LatticeField::Location(), quda::Arg< real, Ns, Nc, order >::nParity, QUDA_CPU_FIELD_LOCATION, quda::TuneParam::shared_bytes, and quda::tuneLaunch().
Referenced by quda::genericPackGhost().
|
inlinevirtual |
Reimplemented from quda::Tunable.
Definition at line 135 of file color_spinor_pack.cu.
References quda::Tunable::blockMin().
|
inlinevirtual |
Reimplemented from quda::Tunable.
Definition at line 134 of file color_spinor_pack.cu.
References quda::Tunable::blockStep().
|
inlinevirtual |
Reimplemented from quda::Tunable.
Definition at line 169 of file color_spinor_pack.cu.
References comm_dim_partitioned(), quda::LatticeField::GhostPrecision(), quda::LatticeField::Precision(), and quda::LatticeField::SurfaceCB().
|
inlinevirtual |
sets default values for when tuning is disabled
Reimplemented from quda::TunableVectorYZ.
Definition at line 162 of file color_spinor_pack.cu.
References quda::TuneParam::aux, quda::TunableVectorYZ::defaultTuneParam(), and quda::GenericPackGhostLauncher< Float, block_float, Ns, Ms, Nc, Mc, Arg >::setColorSpinBlock().
|
inlinevirtual |
Implements quda::Tunable.
Definition at line 168 of file color_spinor_pack.cu.
|
inlinevirtual |
Reimplemented from quda::TunableVectorYZ.
Definition at line 156 of file color_spinor_pack.cu.
References quda::TuneParam::aux, quda::TunableVectorYZ::initTuneParam(), and quda::GenericPackGhostLauncher< Float, block_float, Ns, Ms, Nc, Mc, Arg >::setColorSpinBlock().
|
inlineprivatevirtual |
Reimplemented from quda::Tunable.
Definition at line 49 of file color_spinor_pack.cu.
References quda::Arg< real, Ns, Nc, order >::volumeCB.
|
inline |
Definition at line 117 of file color_spinor_pack.cu.
References quda::TuneParam::aux, quda::TuneParam::block, quda::TuneParam::grid, and quda::Arg< real, Ns, Nc, order >::nParity.
Referenced by quda::GenericPackGhostLauncher< Float, block_float, Ns, Ms, Nc, Mc, Arg >::advanceAux(), quda::GenericPackGhostLauncher< Float, block_float, Ns, Ms, Nc, Mc, Arg >::advanceBlockDim(), quda::GenericPackGhostLauncher< Float, block_float, Ns, Ms, Nc, Mc, Arg >::defaultTuneParam(), and quda::GenericPackGhostLauncher< Float, block_float, Ns, Ms, Nc, Mc, Arg >::initTuneParam().
|
inlineprivatevirtual |
Reimplemented from quda::Tunable.
Definition at line 51 of file color_spinor_pack.cu.
|
inlineprivatevirtual |
Reimplemented from quda::Tunable.
Definition at line 50 of file color_spinor_pack.cu.
|
inlinevirtual |
Implements quda::Tunable.
Definition at line 152 of file color_spinor_pack.cu.
References quda::Tunable::aux, and quda::LatticeField::VolString().
|
private |
Definition at line 47 of file color_spinor_pack.cu.
Referenced by quda::GenericPackGhostLauncher< Float, block_float, Ns, Ms, Nc, Mc, Arg >::apply(), and quda::genericPackGhost().
|
private |
Definition at line 48 of file color_spinor_pack.cu.