All pull requests

Pull requests

• [blas][generic] Enable the generic SYCL BLAS backend on Windows

  #787·Foadsf opened Oct 13, 2026·2 tasks done·

  ·

• [blas] Add bfloat16 gemm_batch

  #785·zjin-lcf opened Oct 6, 2026Contributor·2 of 3 tasks·

  ·
  7

• [DFT][rocFFT] Add regression tests for stride validation and rank handling

  #783·zjin-lcf opened Sep 30, 2026Contributor·5 tasks done·

  ·
  1

  2

• [lapack][rocSOLVER] Implement gerqf and getri

  #782·zjin-lcf opened Sep 18, 2026Contributor·4 of 5 tasks·

  ·

• [lapack][rocSOLVER] Implement QR batch routines

  #781·zjin-lcf opened Sep 18, 2026Contributor·5 tasks done·

  ·

• [lapack][rocSOLVER] Implement LU batch routines

  #780·zjin-lcf opened Sep 18, 2026Contributor·4 of 5 tasks·

  ·

• [lapack][rocSOLVER] Implement Cholesky batch routines

  #779·zjin-lcf opened Sep 18, 2026Contributor·4 tasks done·

  ·

• [lapack][rocSOLVER] Report info for every matrix in a batch

  #778·zjin-lcf opened Sep 18, 2026Contributor·4 tasks done·

  ·

• [lapack][cuSOLVER] Implement trtrs with cuBLAS trsm

  #774·zjin-lcf opened Sep 18, 2026Contributor·2 of 3 tasks·

  ·

• [lapack][cuSOLVER] Implement grouped getri_batch

  #775·zjin-lcf opened Sep 18, 2026Contributor·2 of 3 tasks·

  ·

• [lapack][cuSOLVER] Support multiple RHS in potrs_batch

  #777·zjin-lcf opened Sep 18, 2026Contributor·2 of 3 tasks·

  ·

• [lapack][cuSOLVER] Support wide matrices in gesvd with gesvdj

  #776·zjin-lcf opened Sep 18, 2026Contributor·2 of 3 tasks·

  ·

• [lapack][cuSOLVER] Report info for every matrix in a batch

  #773·zjin-lcf opened Sep 18, 2026Contributor·2 of 3 tasks·

  ·

• [blas] Add bfloat16-output GEMM support

  #770·zjin-lcf opened Aug 24, 2026Contributor·5 tasks done·

  ·
  6

• [blas] Add half and bfloat16 dot support

  #769·zjin-lcf opened Aug 24, 2026Contributor·5 tasks done·

  ·
  7

• [blas] Implement omatcopy2 for cuBLAS and rocBLAS

  #766·zjin-lcf opened Aug 15, 2026Contributor·6 tasks done·

  ·
  2

• [blas][rocblas] Support int8 inputs with float output in gemm_batch

  #763·zjin-lcf opened Aug 15, 2026Contributor·

  ·
  1

• [lapack][rocsolver] Use the 64-bit pivot API for getrf and getrs

  #762·zjin-lcf opened Aug 14, 2026Contributor·6 tasks done·

  ·

• [lapack][cusolver] Use the 64-bit pivot API for getrf and getrs

  #759·zjin-lcf opened Aug 14, 2026Contributor·5 tasks done·

  ·
  1

• [LAPACK][rocSOLVER] Fix incorrect QR results on repeated runs (#626)

  #758·zjin-lcf opened Aug 12, 2026Contributor·2 of 3 tasks·

  ·

• [cmake][docs] Handle AMD code object ABI mismatch for ROCm < 6.3

  #757·zjin-lcf opened Aug 11, 2026Contributor·

  ·
  1

  4

• [cublas] Implement dgmm_batch backend

  #756·zjin-lcf opened Aug 11, 2026Contributor·3 tasks done·

  ·
  1

  2

• [cmake] Combine -fsycl-targets across all enabled GPU backends (fixes #708)

  #753·zjin-lcf opened Aug 11, 2026Contributor·3 tasks·

  ·
  1

  3

• Add compile-time dispatching examples for generic BLAS and portFFT backends

  #752·zjin-lcf opened Aug 11, 2026Contributor·

  ·
  1

  9

• [LAPACK][cuSOLVER] Fix incorrect QR results on repeated runs (#626)

  #748·zjin-lcf opened Jul 28, 2026Contributor·2 of 3 tasks·

  ·
  1

  4