Skip to content

Use same matrix list in benchmark and warmup/repetitions in solver #366

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
yhmtsai merged 11 commits into from
Oct 18, 2019
Merged

Use same matrix list in benchmark and warmup/repetitions in solver #366

Changes from 1 commit
Commits
Show all changes
11 commits
Select commit Hold shift + click to select a range
f5bd868
move the spmv and add the solver name
yhmtsai Oct 14, 2019
acc1f32
keep the same format as spmv's
yhmtsai Oct 14, 2019
8d16aa7
fix compile error
yhmtsai Oct 14, 2019
05b1d3c
split residual computation and timing component
yhmtsai Oct 15, 2019
e8c53a4
keep same matrix format argument
yhmtsai Oct 15, 2019
a03e81b
add warmup and repetitions in solver
yhmtsai Oct 15, 2019
5ae78a2
delete gflags declare
yhmtsai Oct 16, 2019
a72aeae
fix warning
yhmtsai Oct 17, 2019
18df523
fix sonar issues
yhmtsai Oct 17, 2019
a28219c
create func to handle duplication and rm disable
yhmtsai Oct 17, 2019
76cb7c1
fix throw in destructor
yhmtsai Oct 17, 2019
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
fix sonar issues
  • Loading branch information
@yhmtsai
yhmtsai committed Oct 17, 2019
commit 18df52366e0c0013bfa59ccde8ad1ff4b6d8ab10
35 changes: 19 additions & 16 deletions benchmark/utils/cuda_linops.hpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -74,11 +74,17 @@ class CuspBase : public gko::LinOp {
this->initialize_descr();
}

~CuspBase() = default;

CuspBase(const CuspBase &other) = delete;

CuspBase &operator=(const CuspBase &other)
{
gko::LinOp::operator=(other);
this->gpu_exec_ = other.gpu_exec_;
this->initialize_descr();
if (this != &other) {
gko::LinOp::operator=(other);
this->gpu_exec_ = other.gpu_exec_;
this->initialize_descr();
}
return *this;
}

Expand Down Expand Up @@ -312,16 +318,14 @@ class CuspCsrEx
const auto id = this->get_gpu_exec()->get_device_id();
gko::device_guard g{id};
if (set_buffer_) {
try {
GKO_ASSERT_NO_CUDA_ERRORS(cudaFree(buffer_));
} catch (const std::exception &e) {
std::cerr
<< "Error when unallocating CuspCsrEx temporary buffer: "
<< e.what() << std::endl;
}
GKO_ASSERT_NO_CUDA_ERRORS(cudaFree(buffer_));
}
}

CuspCsrEx(const CuspCsrEx &other) = delete;

CuspCsrEx &operator=(const CuspCsrEx &other) = default;

protected:
void apply_impl(const gko::LinOp *b, gko::LinOp *x) const override
{
Expand Down Expand Up @@ -418,14 +422,13 @@ class CuspHybrid
{
const auto id = this->get_gpu_exec()->get_device_id();
gko::device_guard g{id};
try {
GKO_ASSERT_NO_CUSPARSE_ERRORS(cusparseDestroyHybMat(hyb_));
} catch (const std::exception &e) {
std::cerr << "Error when unallocating CuspHybrid hyb_ matrix: "
<< e.what() << std::endl;
}
GKO_ASSERT_NO_CUSPARSE_ERRORS(cusparseDestroyHybMat(hyb_));
}

CuspHybrid(const CuspHybrid &other) = delete;

CuspHybrid &operator=(const CuspHybrid &other) = default;

protected:
void apply_impl(const gko::LinOp *b, gko::LinOp *x) const override
{
Expand Down