cwpearson/astaroth
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

The previous attempt (dsl_feature_completeness_2019-08-23) to enable arbitrary kernel functions was a failure: we get significant performance loss (25-100%) if step_number is not passed as a template parameter to the integration kernel. Apparently the CUDA compiler cannot perform some optimizations if there is a if/else construct in a performance-critical part which cannot be evaluated at compile time. This branch keeps step_number as a template parameter but takes rest of the user parameters as uniforms (dt is no longer passed as a function parameter but as an uniform with the DSL instead).

Browse Source
This commit is contained in:
jpekkila
2019-08-27 17:36:33 +03:00
parent 9fa39b66fc
commit 20138263f4
4 changed files with 21 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
acc/mhd_solver/stencil_definition.sdh
View File

@@ -15,6 +15,7 @@ uniform int AC_bin_steps;
uniform int AC_bc_type;
// Real params
uniform Scalar AC_dt;
// Spacing
uniform Scalar AC_dsx;
uniform Scalar AC_dsy;

3
acc/mhd_solver/stencil_process.sps
View File

@@ -294,8 +294,9 @@ out ScalarField out_tt(VTXBUF_TEMPERATURE);
#endif
Kernel void
solve(Scalar dt)
solve()
{
Scalar dt = AC_dt;
out_lnrho = rk3(out_lnrho, lnrho, continuity(uu, lnrho), dt);
#if LMAGNETIC

18
acc/src/code_generator.c
View File

@@ -62,9 +62,9 @@ static const char* translation_table[TRANSLATION_TABLE_SIZE] = {
[MATRIX] = "AcMatrix",
[SCALARFIELD] = "AcReal",
// Type qualifiers
[KERNEL] = "template <int step_number> static "
"__global__", //__launch_bounds__(RK_THREADBLOCK_SIZE,
// RK_LAUNCH_BOUND_MIN_BLOCKS),
[KERNEL] = "template <int step_number> static __global__",
//__launch_bounds__(RK_THREADBLOCK_SIZE,
// RK_LAUNCH_BOUND_MIN_BLOCKS),
[PREPROCESSED] = "static __device__ "
"__forceinline__",
[CONSTANT] = "const",
@@ -318,9 +318,15 @@ traverse(const ASTNode* node)
inside_kernel = true;
// Kernel parameter boilerplate
const char* kernel_parameter_boilerplate = "GEN_KERNEL_PARAM_BOILERPLATE, ";
if (inside_kernel && node->type == NODE_FUNCTION_PARAMETER_DECLARATION)
printf("%s ", kernel_parameter_boilerplate);
const char* kernel_parameter_boilerplate = "GEN_KERNEL_PARAM_BOILERPLATE";
if (inside_kernel && node->type == NODE_FUNCTION_PARAMETER_DECLARATION) {
printf("%s", kernel_parameter_boilerplate);
if (node->lhs != NULL) {
printf("Compilation error: function parameters for Kernel functions not allowed!\n");
exit(EXIT_FAILURE);
}
}
// Kernel builtin variables boilerplate (read input/output arrays and setup
// indices)