IntelPython / dpnp / 12200771689

81%

push

github

Update `dpnp.linalg.svd()` to run on CUDA  (#2212)

This PR suggests updating `dpnp.linagl.svd()` implementation to support
running on CUDA devices.
Since cuSolver gesvd only supports m>=n the previous implementation crashed with `Segmentation fault (core
dumped)`

This suggests adding checks for `m>=n` otherwise transpose the input
array.

Passing the transposed array to `oneapi::mkl::lapack::gesvd` increases
the performance of `dpnp.linalg.svd()` due to the reducing a matrix with
`m >= n` to bidiagonal form (inside `lapack::gesvd`) is more efficient

4425 of 10054 branches covered (44.01%)

Branch coverage included in aggregate %.

45 of 45 new or added lines in 1 file covered. (100.0%)

16534 of 20786 relevant lines covered (79.54%)

20529.92 hits per line