Skip to content

Use oneMKL LAPACK gesv for dpnp.linalg.solve() #2558

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
wants to merge 3 commits into
base: master
Choose a base branch
from
Open

Use oneMKL LAPACK gesv for dpnp.linalg.solve() #2558

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
6797cf3
Use gesv instead of getrf/getrs in dpnp_solve
vlad-perevezentsev Aug 14, 2025
3d737dc
Update changelog
vlad-perevezentsev Aug 14, 2025
a9fa229
Merge master into use_gesv_solve
vlad-perevezentsev Aug 19, 2025
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
1 change: 1 addition & 0 deletions CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -34,6 +34,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
* Replaced `ci` section in `.pre-commit-config.yaml` with a new GitHub workflow with scheduled run to autoupdate the `pre-commit` configuration [#2542](https://github.com/IntelPython/dpnp/pull/2542)
* FFT module is updated to perform in-place FFT in intermediate steps of ND FFT [#2543](https://github.com/IntelPython/dpnp/pull/2543)
* Reused dpctl tensor include to enable experimental SYCL namespace for complex types [#2546](https://github.com/IntelPython/dpnp/pull/2546)
* Refactored backend implementation of `dpnp.linalg.solve` to use oneMKL LAPACK `gesv` directly [#2558](https://github.com/IntelPython/dpnp/pull/2558)

### Deprecated

Expand Down
49 changes: 9 additions & 40 deletions dpnp/linalg/dpnp_utils_linalg.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2619,15 +2619,9 @@ def dpnp_solve(a, b):
a_usm_arr = dpnp.get_usm_ndarray(a)
b_usm_arr = dpnp.get_usm_ndarray(b)

# Due to MKLD-17226 (bug with incorrect checking ldb parameter
# in oneapi::mkl::lapack::gesv_scratchad_size that raises an error
# `invalid argument` when nrhs > n) we can not use _gesv directly.
# This w/a uses _getrf and _getrs instead
# to handle cases where nrhs > n for a.shape = (n x n)
# and b.shape = (n x nrhs).

# oneMKL LAPACK getrf overwrites `a`.
a_h = dpnp.empty_like(a, order="C", dtype=res_type, usm_type=res_usm_type)
# oneMKL LAPACK getrs overwrites `a` and assumes fortran-like array as
# input
a_h = dpnp.empty_like(a, order="F", dtype=res_type, usm_type=res_usm_type)

_manager = dpu.SequentialOrderManager[exec_q]
dev_evs = _manager.submitted_events
Expand Down Expand Up @@ -2658,39 +2652,14 @@ def dpnp_solve(a, b):
)
_manager.add_event_pair(ht_ev, b_copy_ev)

n = a.shape[0]

ipiv_h = dpnp.empty_like(
a,
shape=(n,),
dtype=dpnp.int64,
)
dev_info_h = [0]

# Call the LAPACK extension function _getrf
# to perform LU decomposition of the input matrix
ht_ev, getrf_ev = li._getrf(
exec_q,
a_h.get_array(),
ipiv_h.get_array(),
dev_info_h,
depends=[a_copy_ev],
# Call the LAPACK extension function _gesv to solve the system of linear
# equations with the coefficient square matrix and
# the dependent variables array.
ht_lapack_ev, gesv_ev = li._gesv(
exec_q, a_h.get_array(), b_h.get_array(), [a_copy_ev, b_copy_ev]
)
_manager.add_event_pair(ht_ev, getrf_ev)

_check_lapack_dev_info(dev_info_h)
Copy link
Collaborator

@vtavana vtavana Aug 14, 2025 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

With these changes, lines 733-735 are not covered with any test.

dpnp/dpnp/linalg/dpnp_utils_linalg.py

Line 733 in 3d737dc

error_msg = error_msg or "Singular matrix"

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Good point!
That is because gesv raises an error internally.
I will update tests for dpnp.linalg.inv() to cover this.
Thank you

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@vtavana this #2559 will help cover this line


# Call the LAPACK extension function _getrs
# to solve the system of linear equations with an LU-factored
# coefficient square matrix, with multiple right-hand sides.
ht_ev, getrs_ev = li._getrs(
exec_q,
a_h.get_array(),
ipiv_h.get_array(),
b_h.get_array(),
depends=[b_copy_ev, getrf_ev],
)
_manager.add_event_pair(ht_ev, getrs_ev)
_manager.add_event_pair(ht_lapack_ev, gesv_ev)
return b_h


Expand Down
Loading