Skip to content

Reduce redundant CUDA Jacobian uploads during a linear solve #2806

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

Open
LwhJesse wants to merge 3 commits into
base: develop
Choose a base branch
from
+5 −1
Open

Reduce redundant CUDA Jacobian uploads during a linear solve #2806

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
02730c0
Reduce redundant CUDA Jacobian uploads
LwhJesse Apr 30, 2026
55385dd
Move CUDA Jacobian upload into CSysMatrixVectorProduct
LwhJesse May 6, 2026
e38dfc0
Defer CUDA matrix upload to first matvec use
LwhJesse May 8, 2026
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
5 changes: 5 additions & 0 deletions Common/include/linear_algebra/CMatrixVectorProduct.hpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -72,6 +72,7 @@ class CSysMatrixVectorProduct final : public CMatrixVectorProduct<ScalarType> {
const CSysMatrix<ScalarType>& matrix; /*!< \brief pointer to matrix that defines the product. */
CGeometry* geometry; /*!< \brief geometry associated with the matrix. */
const CConfig* config; /*!< \brief config of the problem. */
mutable bool matrix_uploaded = false; /*!< \brief Upload the matrix lazily on the first actual GPU matvec. */

public:
/*!
Expand All @@ -97,6 +98,10 @@ class CSysMatrixVectorProduct final : public CMatrixVectorProduct<ScalarType> {
inline void operator()(const CSysVector<ScalarType>& u, CSysVector<ScalarType>& v) const override {
if (config->GetCUDA()) {
#ifdef HAVE_CUDA
if (!matrix_uploaded) {
matrix.HtDTransfer();
matrix_uploaded = true;
}
matrix.GPUMatrixVectorProduct(u, v, geometry, config);
#else
SU2_MPI::Error(
Expand Down
1 change: 0 additions & 1 deletion Common/src/linear_algebra/CSysMatrixGPU.cu
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -70,7 +70,6 @@ void CSysMatrix<ScalarType>::GPUMatrixVectorProduct(const CSysVector<ScalarType>
ScalarType* d_vec = vec.GetDevicePointer();
ScalarType* d_prod = prod.GetDevicePointer();

HtDTransfer();
vec.HtDTransfer();
prod.GPUSetVal(0.0);

Expand Down