diff --git a/.github/workflows/pr.yml b/.github/workflows/pr.yml
index 99a106cce..98c1b6614 100644
--- a/.github/workflows/pr.yml
+++ b/.github/workflows/pr.yml
@@ -31,6 +31,7 @@ jobs:
     runs-on: ubuntu-latest
     # One runner for each domain
     strategy:
+      fail-fast: false
       matrix:
         include:
         - config: portBLAS
@@ -77,13 +78,16 @@ jobs:
         # 64 bit int
         cmake ${SHARED_OPT} -DBUILD_INDEX64=on -B lapack/build64
         cmake --build lapack/build64 ${PARALLEL} --target install
-    - name: Install oneapi
+    - name: Install compiler
       if: steps.domain_check.outputs.result == 'true'
-      uses: rscohn2/setup-oneapi@2ad0cf6b74bc2426bdcee825cf88f9db719dd727 # v0.1.0
-      with:
-        components: |
-          icx@2024.1.0
-          mkl@2024.1.0
+      run: |
+        wget --progress=dot:giga https://registrationcenter-download.intel.com/akdlm/IRC_NAS/ac92f2bb-4818-4e53-a432-f8b34d502f23/intel-dpcpp-cpp-compiler-2025.0.0.740_offline.sh
+        sudo bash intel-dpcpp-cpp-compiler-2025.0.0.740_offline.sh -s -a -s --action install --eula accept
+    - name: Install Intel oneMKL
+      if: steps.domain_check.outputs.result == 'true'
+      run: |
+        wget --progress=dot:giga https://registrationcenter-download.intel.com/akdlm/IRC_NAS/79153e0f-74d7-45af-b8c2-258941adf58a/intel-onemkl-2025.0.0.940_offline.sh
+        sudo bash intel-onemkl-2025.0.0.940_offline.sh -s -a -s --action install --eula accept
     - name: Configure/Build for a domain
       if: steps.domain_check.outputs.result == 'true'
       run: |
diff --git a/src/lapack/backends/mkl_common/mkl_lapack_backend.hpp b/src/lapack/backends/mkl_common/mkl_lapack_backend.hpp
index d0ba37e7a..992659a67 100644
--- a/src/lapack/backends/mkl_common/mkl_lapack_backend.hpp
+++ b/src/lapack/backends/mkl_common/mkl_lapack_backend.hpp
@@ -523,33 +523,35 @@ sycl::event getrf(sycl::queue& queue, std::int64_t m, std::int64_t n, std::compl
                   std::int64_t lda, std::int64_t* ipiv, std::complex<double>* scratchpad,
                   std::int64_t scratchpad_size, const std::vector<sycl::event>& dependencies = {});
 sycl::event getri(sycl::queue& queue, std::int64_t n, std::complex<float>* a, std::int64_t lda,
-                  std::int64_t* ipiv, std::complex<float>* scratchpad, std::int64_t scratchpad_size,
-                  const std::vector<sycl::event>& dependencies = {});
+                  const std::int64_t* ipiv, std::complex<float>* scratchpad,
+                  std::int64_t scratchpad_size, const std::vector<sycl::event>& dependencies = {});
 sycl::event getri(sycl::queue& queue, std::int64_t n, double* a, std::int64_t lda,
-                  std::int64_t* ipiv, double* scratchpad, std::int64_t scratchpad_size,
+                  const std::int64_t* ipiv, double* scratchpad, std::int64_t scratchpad_size,
                   const std::vector<sycl::event>& dependencies = {});
 sycl::event getri(sycl::queue& queue, std::int64_t n, float* a, std::int64_t lda,
-                  std::int64_t* ipiv, float* scratchpad, std::int64_t scratchpad_size,
+                  const std::int64_t* ipiv, float* scratchpad, std::int64_t scratchpad_size,
                   const std::vector<sycl::event>& dependencies = {});
 sycl::event getri(sycl::queue& queue, std::int64_t n, std::complex<double>* a, std::int64_t lda,
-                  std::int64_t* ipiv, std::complex<double>* scratchpad,
+                  const std::int64_t* ipiv, std::complex<double>* scratchpad,
                   std::int64_t scratchpad_size, const std::vector<sycl::event>& dependencies = {});
 sycl::event getrs(sycl::queue& queue, oneapi::mkl::transpose trans, std::int64_t n,
-                  std::int64_t nrhs, std::complex<float>* a, std::int64_t lda, std::int64_t* ipiv,
-                  std::complex<float>* b, std::int64_t ldb, std::complex<float>* scratchpad,
-                  std::int64_t scratchpad_size, const std::vector<sycl::event>& dependencies = {});
+                  std::int64_t nrhs, const std::complex<float>* a, std::int64_t lda,
+                  const std::int64_t* ipiv, std::complex<float>* b, std::int64_t ldb,
+                  std::complex<float>* scratchpad, std::int64_t scratchpad_size,
+                  const std::vector<sycl::event>& dependencies = {});
 sycl::event getrs(sycl::queue& queue, oneapi::mkl::transpose trans, std::int64_t n,
-                  std::int64_t nrhs, double* a, std::int64_t lda, std::int64_t* ipiv, double* b,
-                  std::int64_t ldb, double* scratchpad, std::int64_t scratchpad_size,
+                  std::int64_t nrhs, const double* a, std::int64_t lda, const std::int64_t* ipiv,
+                  double* b, std::int64_t ldb, double* scratchpad, std::int64_t scratchpad_size,
                   const std::vector<sycl::event>& dependencies = {});
 sycl::event getrs(sycl::queue& queue, oneapi::mkl::transpose trans, std::int64_t n,
-                  std::int64_t nrhs, float* a, std::int64_t lda, std::int64_t* ipiv, float* b,
-                  std::int64_t ldb, float* scratchpad, std::int64_t scratchpad_size,
+                  std::int64_t nrhs, const float* a, std::int64_t lda, const std::int64_t* ipiv,
+                  float* b, std::int64_t ldb, float* scratchpad, std::int64_t scratchpad_size,
                   const std::vector<sycl::event>& dependencies = {});
 sycl::event getrs(sycl::queue& queue, oneapi::mkl::transpose trans, std::int64_t n,
-                  std::int64_t nrhs, std::complex<double>* a, std::int64_t lda, std::int64_t* ipiv,
-                  std::complex<double>* b, std::int64_t ldb, std::complex<double>* scratchpad,
-                  std::int64_t scratchpad_size, const std::vector<sycl::event>& dependencies = {});
+                  std::int64_t nrhs, const std::complex<double>* a, std::int64_t lda,
+                  const std::int64_t* ipiv, std::complex<double>* b, std::int64_t ldb,
+                  std::complex<double>* scratchpad, std::int64_t scratchpad_size,
+                  const std::vector<sycl::event>& dependencies = {});
 sycl::event gesvd(sycl::queue& queue, oneapi::mkl::jobsvd jobu, oneapi::mkl::jobsvd jobvt,
                   std::int64_t m, std::int64_t n, double* a, std::int64_t lda, double* s, double* u,
                   std::int64_t ldu, double* vt, std::int64_t ldvt, double* scratchpad,
@@ -603,46 +605,48 @@ sycl::event hetrf(sycl::queue& queue, oneapi::mkl::uplo uplo, std::int64_t n,
                   std::complex<double>* scratchpad, std::int64_t scratchpad_size,
                   const std::vector<sycl::event>& dependencies = {});
 sycl::event orgbr(sycl::queue& queue, oneapi::mkl::generate vec, std::int64_t m, std::int64_t n,
-                  std::int64_t k, float* a, std::int64_t lda, float* tau, float* scratchpad,
+                  std::int64_t k, float* a, std::int64_t lda, const float* tau, float* scratchpad,
                   std::int64_t scratchpad_size, const std::vector<sycl::event>& dependencies = {});
 sycl::event orgbr(sycl::queue& queue, oneapi::mkl::generate vec, std::int64_t m, std::int64_t n,
-                  std::int64_t k, double* a, std::int64_t lda, double* tau, double* scratchpad,
-                  std::int64_t scratchpad_size, const std::vector<sycl::event>& dependencies = {});
-sycl::event orgqr(sycl::queue& queue, std::int64_t m, std::int64_t n, std::int64_t k, double* a,
-                  std::int64_t lda, double* tau, double* scratchpad, std::int64_t scratchpad_size,
+                  std::int64_t k, double* a, std::int64_t lda, const double* tau,
+                  double* scratchpad, std::int64_t scratchpad_size,
                   const std::vector<sycl::event>& dependencies = {});
+sycl::event orgqr(sycl::queue& queue, std::int64_t m, std::int64_t n, std::int64_t k, double* a,
+                  std::int64_t lda, const double* tau, double* scratchpad,
+                  std::int64_t scratchpad_size, const std::vector<sycl::event>& dependencies = {});
 sycl::event orgqr(sycl::queue& queue, std::int64_t m, std::int64_t n, std::int64_t k, float* a,
-                  std::int64_t lda, float* tau, float* scratchpad, std::int64_t scratchpad_size,
-                  const std::vector<sycl::event>& dependencies = {});
+                  std::int64_t lda, const float* tau, float* scratchpad,
+                  std::int64_t scratchpad_size, const std::vector<sycl::event>& dependencies = {});
 sycl::event orgtr(sycl::queue& queue, oneapi::mkl::uplo uplo, std::int64_t n, float* a,
-                  std::int64_t lda, float* tau, float* scratchpad, std::int64_t scratchpad_size,
-                  const std::vector<sycl::event>& dependencies = {});
+                  std::int64_t lda, const float* tau, float* scratchpad,
+                  std::int64_t scratchpad_size, const std::vector<sycl::event>& dependencies = {});
 sycl::event orgtr(sycl::queue& queue, oneapi::mkl::uplo uplo, std::int64_t n, double* a,
-                  std::int64_t lda, double* tau, double* scratchpad, std::int64_t scratchpad_size,
-                  const std::vector<sycl::event>& dependencies = {});
-sycl::event ormtr(sycl::queue& queue, oneapi::mkl::side side, oneapi::mkl::uplo uplo,
-                  oneapi::mkl::transpose trans, std::int64_t m, std::int64_t n, float* a,
-                  std::int64_t lda, float* tau, float* c, std::int64_t ldc, float* scratchpad,
+                  std::int64_t lda, const double* tau, double* scratchpad,
                   std::int64_t scratchpad_size, const std::vector<sycl::event>& dependencies = {});
 sycl::event ormtr(sycl::queue& queue, oneapi::mkl::side side, oneapi::mkl::uplo uplo,
-                  oneapi::mkl::transpose trans, std::int64_t m, std::int64_t n, double* a,
-                  std::int64_t lda, double* tau, double* c, std::int64_t ldc, double* scratchpad,
+                  oneapi::mkl::transpose trans, std::int64_t m, std::int64_t n, const float* a,
+                  std::int64_t lda, const float* tau, float* c, std::int64_t ldc, float* scratchpad,
                   std::int64_t scratchpad_size, const std::vector<sycl::event>& dependencies = {});
+sycl::event ormtr(sycl::queue& queue, oneapi::mkl::side side, oneapi::mkl::uplo uplo,
+                  oneapi::mkl::transpose trans, std::int64_t m, std::int64_t n, const double* a,
+                  std::int64_t lda, const double* tau, double* c, std::int64_t ldc,
+                  double* scratchpad, std::int64_t scratchpad_size,
+                  const std::vector<sycl::event>& dependencies = {});
 sycl::event ormrq(sycl::queue& queue, oneapi::mkl::side side, oneapi::mkl::transpose trans,
-                  std::int64_t m, std::int64_t n, std::int64_t k, float* a, std::int64_t lda,
-                  float* tau, float* c, std::int64_t ldc, float* scratchpad,
+                  std::int64_t m, std::int64_t n, std::int64_t k, const float* a, std::int64_t lda,
+                  const float* tau, float* c, std::int64_t ldc, float* scratchpad,
                   std::int64_t scratchpad_size, const std::vector<sycl::event>& dependencies = {});
 sycl::event ormrq(sycl::queue& queue, oneapi::mkl::side side, oneapi::mkl::transpose trans,
-                  std::int64_t m, std::int64_t n, std::int64_t k, double* a, std::int64_t lda,
-                  double* tau, double* c, std::int64_t ldc, double* scratchpad,
+                  std::int64_t m, std::int64_t n, std::int64_t k, const double* a, std::int64_t lda,
+                  const double* tau, double* c, std::int64_t ldc, double* scratchpad,
                   std::int64_t scratchpad_size, const std::vector<sycl::event>& dependencies = {});
 sycl::event ormqr(sycl::queue& queue, oneapi::mkl::side side, oneapi::mkl::transpose trans,
-                  std::int64_t m, std::int64_t n, std::int64_t k, double* a, std::int64_t lda,
-                  double* tau, double* c, std::int64_t ldc, double* scratchpad,
+                  std::int64_t m, std::int64_t n, std::int64_t k, const double* a, std::int64_t lda,
+                  const double* tau, double* c, std::int64_t ldc, double* scratchpad,
                   std::int64_t scratchpad_size, const std::vector<sycl::event>& dependencies = {});
 sycl::event ormqr(sycl::queue& queue, oneapi::mkl::side side, oneapi::mkl::transpose trans,
-                  std::int64_t m, std::int64_t n, std::int64_t k, float* a, std::int64_t lda,
-                  float* tau, float* c, std::int64_t ldc, float* scratchpad,
+                  std::int64_t m, std::int64_t n, std::int64_t k, const float* a, std::int64_t lda,
+                  const float* tau, float* c, std::int64_t ldc, float* scratchpad,
                   std::int64_t scratchpad_size, const std::vector<sycl::event>& dependencies = {});
 sycl::event potrf(sycl::queue& queue, oneapi::mkl::uplo uplo, std::int64_t n, float* a,
                   std::int64_t lda, float* scratchpad, std::int64_t scratchpad_size,
@@ -669,17 +673,18 @@ sycl::event potri(sycl::queue& queue, oneapi::mkl::uplo uplo, std::int64_t n,
                   std::complex<double>* a, std::int64_t lda, std::complex<double>* scratchpad,
                   std::int64_t scratchpad_size, const std::vector<sycl::event>& dependencies = {});
 sycl::event potrs(sycl::queue& queue, oneapi::mkl::uplo uplo, std::int64_t n, std::int64_t nrhs,
-                  float* a, std::int64_t lda, float* b, std::int64_t ldb, float* scratchpad,
+                  const float* a, std::int64_t lda, float* b, std::int64_t ldb, float* scratchpad,
                   std::int64_t scratchpad_size, const std::vector<sycl::event>& dependencies = {});
 sycl::event potrs(sycl::queue& queue, oneapi::mkl::uplo uplo, std::int64_t n, std::int64_t nrhs,
-                  double* a, std::int64_t lda, double* b, std::int64_t ldb, double* scratchpad,
-                  std::int64_t scratchpad_size, const std::vector<sycl::event>& dependencies = {});
+                  const double* a, std::int64_t lda, double* b, std::int64_t ldb,
+                  double* scratchpad, std::int64_t scratchpad_size,
+                  const std::vector<sycl::event>& dependencies = {});
 sycl::event potrs(sycl::queue& queue, oneapi::mkl::uplo uplo, std::int64_t n, std::int64_t nrhs,
-                  std::complex<float>* a, std::int64_t lda, std::complex<float>* b,
+                  const std::complex<float>* a, std::int64_t lda, std::complex<float>* b,
                   std::int64_t ldb, std::complex<float>* scratchpad, std::int64_t scratchpad_size,
                   const std::vector<sycl::event>& dependencies = {});
 sycl::event potrs(sycl::queue& queue, oneapi::mkl::uplo uplo, std::int64_t n, std::int64_t nrhs,
-                  std::complex<double>* a, std::int64_t lda, std::complex<double>* b,
+                  const std::complex<double>* a, std::int64_t lda, std::complex<double>* b,
                   std::int64_t ldb, std::complex<double>* scratchpad, std::int64_t scratchpad_size,
                   const std::vector<sycl::event>& dependencies = {});
 sycl::event syevd(sycl::queue& queue, oneapi::mkl::job jobz, oneapi::mkl::uplo uplo, std::int64_t n,
@@ -717,75 +722,75 @@ sycl::event sytrf(sycl::queue& queue, oneapi::mkl::uplo uplo, std::int64_t n,
                   std::complex<double>* scratchpad, std::int64_t scratchpad_size,
                   const std::vector<sycl::event>& dependencies = {});
 sycl::event trtrs(sycl::queue& queue, oneapi::mkl::uplo uplo, oneapi::mkl::transpose trans,
-                  oneapi::mkl::diag diag, std::int64_t n, std::int64_t nrhs, std::complex<float>* a,
-                  std::int64_t lda, std::complex<float>* b, std::int64_t ldb,
-                  std::complex<float>* scratchpad, std::int64_t scratchpad_size,
+                  oneapi::mkl::diag diag, std::int64_t n, std::int64_t nrhs,
+                  const std::complex<float>* a, std::int64_t lda, std::complex<float>* b,
+                  std::int64_t ldb, std::complex<float>* scratchpad, std::int64_t scratchpad_size,
                   const std::vector<sycl::event>& dependencies = {});
 sycl::event trtrs(sycl::queue& queue, oneapi::mkl::uplo uplo, oneapi::mkl::transpose trans,
-                  oneapi::mkl::diag diag, std::int64_t n, std::int64_t nrhs, double* a,
+                  oneapi::mkl::diag diag, std::int64_t n, std::int64_t nrhs, const double* a,
                   std::int64_t lda, double* b, std::int64_t ldb, double* scratchpad,
                   std::int64_t scratchpad_size, const std::vector<sycl::event>& dependencies = {});
 sycl::event trtrs(sycl::queue& queue, oneapi::mkl::uplo uplo, oneapi::mkl::transpose trans,
-                  oneapi::mkl::diag diag, std::int64_t n, std::int64_t nrhs, float* a,
+                  oneapi::mkl::diag diag, std::int64_t n, std::int64_t nrhs, const float* a,
                   std::int64_t lda, float* b, std::int64_t ldb, float* scratchpad,
                   std::int64_t scratchpad_size, const std::vector<sycl::event>& dependencies = {});
 sycl::event trtrs(sycl::queue& queue, oneapi::mkl::uplo uplo, oneapi::mkl::transpose trans,
                   oneapi::mkl::diag diag, std::int64_t n, std::int64_t nrhs,
-                  std::complex<double>* a, std::int64_t lda, std::complex<double>* b,
+                  const std::complex<double>* a, std::int64_t lda, std::complex<double>* b,
                   std::int64_t ldb, std::complex<double>* scratchpad, std::int64_t scratchpad_size,
                   const std::vector<sycl::event>& dependencies = {});
 sycl::event ungbr(sycl::queue& queue, oneapi::mkl::generate vec, std::int64_t m, std::int64_t n,
                   std::int64_t k, std::complex<float>* a, std::int64_t lda,
-                  std::complex<float>* tau, std::complex<float>* scratchpad,
+                  const std::complex<float>* tau, std::complex<float>* scratchpad,
                   std::int64_t scratchpad_size, const std::vector<sycl::event>& dependencies = {});
 sycl::event ungbr(sycl::queue& queue, oneapi::mkl::generate vec, std::int64_t m, std::int64_t n,
                   std::int64_t k, std::complex<double>* a, std::int64_t lda,
-                  std::complex<double>* tau, std::complex<double>* scratchpad,
+                  const std::complex<double>* tau, std::complex<double>* scratchpad,
                   std::int64_t scratchpad_size, const std::vector<sycl::event>& dependencies = {});
 sycl::event ungqr(sycl::queue& queue, std::int64_t m, std::int64_t n, std::int64_t k,
-                  std::complex<float>* a, std::int64_t lda, std::complex<float>* tau,
+                  std::complex<float>* a, std::int64_t lda, const std::complex<float>* tau,
                   std::complex<float>* scratchpad, std::int64_t scratchpad_size,
                   const std::vector<sycl::event>& dependencies = {});
 sycl::event ungqr(sycl::queue& queue, std::int64_t m, std::int64_t n, std::int64_t k,
-                  std::complex<double>* a, std::int64_t lda, std::complex<double>* tau,
+                  std::complex<double>* a, std::int64_t lda, const std::complex<double>* tau,
                   std::complex<double>* scratchpad, std::int64_t scratchpad_size,
                   const std::vector<sycl::event>& dependencies = {});
 sycl::event ungtr(sycl::queue& queue, oneapi::mkl::uplo uplo, std::int64_t n,
-                  std::complex<float>* a, std::int64_t lda, std::complex<float>* tau,
+                  std::complex<float>* a, std::int64_t lda, const std::complex<float>* tau,
                   std::complex<float>* scratchpad, std::int64_t scratchpad_size,
                   const std::vector<sycl::event>& dependencies = {});
 sycl::event ungtr(sycl::queue& queue, oneapi::mkl::uplo uplo, std::int64_t n,
-                  std::complex<double>* a, std::int64_t lda, std::complex<double>* tau,
+                  std::complex<double>* a, std::int64_t lda, const std::complex<double>* tau,
                   std::complex<double>* scratchpad, std::int64_t scratchpad_size,
                   const std::vector<sycl::event>& dependencies = {});
 sycl::event unmrq(sycl::queue& queue, oneapi::mkl::side side, oneapi::mkl::transpose trans,
-                  std::int64_t m, std::int64_t n, std::int64_t k, std::complex<float>* a,
-                  std::int64_t lda, std::complex<float>* tau, std::complex<float>* c,
+                  std::int64_t m, std::int64_t n, std::int64_t k, const std::complex<float>* a,
+                  std::int64_t lda, const std::complex<float>* tau, std::complex<float>* c,
                   std::int64_t ldc, std::complex<float>* scratchpad, std::int64_t scratchpad_size,
                   const std::vector<sycl::event>& dependencies = {});
 sycl::event unmrq(sycl::queue& queue, oneapi::mkl::side side, oneapi::mkl::transpose trans,
-                  std::int64_t m, std::int64_t n, std::int64_t k, std::complex<double>* a,
-                  std::int64_t lda, std::complex<double>* tau, std::complex<double>* c,
+                  std::int64_t m, std::int64_t n, std::int64_t k, const std::complex<double>* a,
+                  std::int64_t lda, const std::complex<double>* tau, std::complex<double>* c,
                   std::int64_t ldc, std::complex<double>* scratchpad, std::int64_t scratchpad_size,
                   const std::vector<sycl::event>& dependencies = {});
 sycl::event unmqr(sycl::queue& queue, oneapi::mkl::side side, oneapi::mkl::transpose trans,
-                  std::int64_t m, std::int64_t n, std::int64_t k, std::complex<float>* a,
-                  std::int64_t lda, std::complex<float>* tau, std::complex<float>* c,
+                  std::int64_t m, std::int64_t n, std::int64_t k, const std::complex<float>* a,
+                  std::int64_t lda, const std::complex<float>* tau, std::complex<float>* c,
                   std::int64_t ldc, std::complex<float>* scratchpad, std::int64_t scratchpad_size,
                   const std::vector<sycl::event>& dependencies = {});
 sycl::event unmqr(sycl::queue& queue, oneapi::mkl::side side, oneapi::mkl::transpose trans,
-                  std::int64_t m, std::int64_t n, std::int64_t k, std::complex<double>* a,
-                  std::int64_t lda, std::complex<double>* tau, std::complex<double>* c,
+                  std::int64_t m, std::int64_t n, std::int64_t k, const std::complex<double>* a,
+                  std::int64_t lda, const std::complex<double>* tau, std::complex<double>* c,
                   std::int64_t ldc, std::complex<double>* scratchpad, std::int64_t scratchpad_size,
                   const std::vector<sycl::event>& dependencies = {});
 sycl::event unmtr(sycl::queue& queue, oneapi::mkl::side side, oneapi::mkl::uplo uplo,
                   oneapi::mkl::transpose trans, std::int64_t m, std::int64_t n,
-                  std::complex<float>* a, std::int64_t lda, std::complex<float>* tau,
+                  const std::complex<float>* a, std::int64_t lda, const std::complex<float>* tau,
                   std::complex<float>* c, std::int64_t ldc, std::complex<float>* scratchpad,
                   std::int64_t scratchpad_size, const std::vector<sycl::event>& dependencies = {});
 sycl::event unmtr(sycl::queue& queue, oneapi::mkl::side side, oneapi::mkl::uplo uplo,
                   oneapi::mkl::transpose trans, std::int64_t m, std::int64_t n,
-                  std::complex<double>* a, std::int64_t lda, std::complex<double>* tau,
+                  const std::complex<double>* a, std::int64_t lda, const std::complex<double>* tau,
                   std::complex<double>* c, std::int64_t ldc, std::complex<double>* scratchpad,
                   std::int64_t scratchpad_size, const std::vector<sycl::event>& dependencies = {});
 sycl::event geqrf_batch(sycl::queue& queue, std::int64_t m, std::int64_t n, float* a,
@@ -865,106 +870,110 @@ sycl::event getrf_batch(sycl::queue& queue, std::int64_t* m, std::int64_t* n,
                         std::complex<double>* scratchpad, std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event getri_batch(sycl::queue& queue, std::int64_t n, float* a, std::int64_t lda,
-                        std::int64_t stride_a, std::int64_t* ipiv, std::int64_t stride_ipiv,
+                        std::int64_t stride_a, const std::int64_t* ipiv, std::int64_t stride_ipiv,
                         std::int64_t batch_size, float* scratchpad, std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event getri_batch(sycl::queue& queue, std::int64_t n, double* a, std::int64_t lda,
-                        std::int64_t stride_a, std::int64_t* ipiv, std::int64_t stride_ipiv,
+                        std::int64_t stride_a, const std::int64_t* ipiv, std::int64_t stride_ipiv,
                         std::int64_t batch_size, double* scratchpad, std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event getri_batch(sycl::queue& queue, std::int64_t n, std::complex<float>* a,
-                        std::int64_t lda, std::int64_t stride_a, std::int64_t* ipiv,
+                        std::int64_t lda, std::int64_t stride_a, const std::int64_t* ipiv,
                         std::int64_t stride_ipiv, std::int64_t batch_size,
                         std::complex<float>* scratchpad, std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event getri_batch(sycl::queue& queue, std::int64_t n, std::complex<double>* a,
-                        std::int64_t lda, std::int64_t stride_a, std::int64_t* ipiv,
+                        std::int64_t lda, std::int64_t stride_a, const std::int64_t* ipiv,
                         std::int64_t stride_ipiv, std::int64_t batch_size,
                         std::complex<double>* scratchpad, std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event getri_batch(sycl::queue& queue, std::int64_t* n, float** a, std::int64_t* lda,
-                        std::int64_t** ipiv, std::int64_t group_count, std::int64_t* group_sizes,
-                        float* scratchpad, std::int64_t scratchpad_size,
+                        const std::int64_t* const* ipiv, std::int64_t group_count,
+                        std::int64_t* group_sizes, float* scratchpad, std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event getri_batch(sycl::queue& queue, std::int64_t* n, double** a, std::int64_t* lda,
-                        std::int64_t** ipiv, std::int64_t group_count, std::int64_t* group_sizes,
-                        double* scratchpad, std::int64_t scratchpad_size,
+                        const std::int64_t* const* ipiv, std::int64_t group_count,
+                        std::int64_t* group_sizes, double* scratchpad, std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event getri_batch(sycl::queue& queue, std::int64_t* n, std::complex<float>** a,
-                        std::int64_t* lda, std::int64_t** ipiv, std::int64_t group_count,
-                        std::int64_t* group_sizes, std::complex<float>* scratchpad,
-                        std::int64_t scratchpad_size,
+                        std::int64_t* lda, const std::int64_t* const* ipiv,
+                        std::int64_t group_count, std::int64_t* group_sizes,
+                        std::complex<float>* scratchpad, std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event getri_batch(sycl::queue& queue, std::int64_t* n, std::complex<double>** a,
-                        std::int64_t* lda, std::int64_t** ipiv, std::int64_t group_count,
-                        std::int64_t* group_sizes, std::complex<double>* scratchpad,
-                        std::int64_t scratchpad_size,
+                        std::int64_t* lda, const std::int64_t* const* ipiv,
+                        std::int64_t group_count, std::int64_t* group_sizes,
+                        std::complex<double>* scratchpad, std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event getrs_batch(sycl::queue& queue, oneapi::mkl::transpose trans, std::int64_t n,
-                        std::int64_t nrhs, float* a, std::int64_t lda, std::int64_t stride_a,
-                        std::int64_t* ipiv, std::int64_t stride_ipiv, float* b, std::int64_t ldb,
-                        std::int64_t stride_b, std::int64_t batch_size, float* scratchpad,
-                        std::int64_t scratchpad_size,
+                        std::int64_t nrhs, const float* a, std::int64_t lda, std::int64_t stride_a,
+                        const std::int64_t* ipiv, std::int64_t stride_ipiv, float* b,
+                        std::int64_t ldb, std::int64_t stride_b, std::int64_t batch_size,
+                        float* scratchpad, std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event getrs_batch(sycl::queue& queue, oneapi::mkl::transpose trans, std::int64_t n,
-                        std::int64_t nrhs, double* a, std::int64_t lda, std::int64_t stride_a,
-                        std::int64_t* ipiv, std::int64_t stride_ipiv, double* b, std::int64_t ldb,
-                        std::int64_t stride_b, std::int64_t batch_size, double* scratchpad,
-                        std::int64_t scratchpad_size,
+                        std::int64_t nrhs, const double* a, std::int64_t lda, std::int64_t stride_a,
+                        const std::int64_t* ipiv, std::int64_t stride_ipiv, double* b,
+                        std::int64_t ldb, std::int64_t stride_b, std::int64_t batch_size,
+                        double* scratchpad, std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event getrs_batch(sycl::queue& queue, oneapi::mkl::transpose trans, std::int64_t n,
-                        std::int64_t nrhs, std::complex<float>* a, std::int64_t lda,
-                        std::int64_t stride_a, std::int64_t* ipiv, std::int64_t stride_ipiv,
+                        std::int64_t nrhs, const std::complex<float>* a, std::int64_t lda,
+                        std::int64_t stride_a, const std::int64_t* ipiv, std::int64_t stride_ipiv,
                         std::complex<float>* b, std::int64_t ldb, std::int64_t stride_b,
                         std::int64_t batch_size, std::complex<float>* scratchpad,
                         std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event getrs_batch(sycl::queue& queue, oneapi::mkl::transpose trans, std::int64_t n,
-                        std::int64_t nrhs, std::complex<double>* a, std::int64_t lda,
-                        std::int64_t stride_a, std::int64_t* ipiv, std::int64_t stride_ipiv,
+                        std::int64_t nrhs, const std::complex<double>* a, std::int64_t lda,
+                        std::int64_t stride_a, const std::int64_t* ipiv, std::int64_t stride_ipiv,
                         std::complex<double>* b, std::int64_t ldb, std::int64_t stride_b,
                         std::int64_t batch_size, std::complex<double>* scratchpad,
                         std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event getrs_batch(sycl::queue& queue, oneapi::mkl::transpose* trans, std::int64_t* n,
-                        std::int64_t* nrhs, float** a, std::int64_t* lda, std::int64_t** ipiv,
-                        float** b, std::int64_t* ldb, std::int64_t group_count,
-                        std::int64_t* group_sizes, float* scratchpad, std::int64_t scratchpad_size,
+                        std::int64_t* nrhs, const float* const* a, std::int64_t* lda,
+                        const std::int64_t* const* ipiv, float** b, std::int64_t* ldb,
+                        std::int64_t group_count, std::int64_t* group_sizes, float* scratchpad,
+                        std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event getrs_batch(sycl::queue& queue, oneapi::mkl::transpose* trans, std::int64_t* n,
-                        std::int64_t* nrhs, double** a, std::int64_t* lda, std::int64_t** ipiv,
-                        double** b, std::int64_t* ldb, std::int64_t group_count,
-                        std::int64_t* group_sizes, double* scratchpad, std::int64_t scratchpad_size,
+                        std::int64_t* nrhs, const double* const* a, std::int64_t* lda,
+                        const std::int64_t* const* ipiv, double** b, std::int64_t* ldb,
+                        std::int64_t group_count, std::int64_t* group_sizes, double* scratchpad,
+                        std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event getrs_batch(sycl::queue& queue, oneapi::mkl::transpose* trans, std::int64_t* n,
-                        std::int64_t* nrhs, std::complex<float>** a, std::int64_t* lda,
-                        std::int64_t** ipiv, std::complex<float>** b, std::int64_t* ldb,
+                        std::int64_t* nrhs, const std::complex<float>* const* a, std::int64_t* lda,
+                        const std::int64_t* const* ipiv, std::complex<float>** b, std::int64_t* ldb,
                         std::int64_t group_count, std::int64_t* group_sizes,
                         std::complex<float>* scratchpad, std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event getrs_batch(sycl::queue& queue, oneapi::mkl::transpose* trans, std::int64_t* n,
-                        std::int64_t* nrhs, std::complex<double>** a, std::int64_t* lda,
-                        std::int64_t** ipiv, std::complex<double>** b, std::int64_t* ldb,
-                        std::int64_t group_count, std::int64_t* group_sizes,
+                        std::int64_t* nrhs, const std::complex<double>* const* a, std::int64_t* lda,
+                        const std::int64_t* const* ipiv, std::complex<double>** b,
+                        std::int64_t* ldb, std::int64_t group_count, std::int64_t* group_sizes,
                         std::complex<double>* scratchpad, std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event orgqr_batch(sycl::queue& queue, std::int64_t m, std::int64_t n, std::int64_t k,
-                        float* a, std::int64_t lda, std::int64_t stride_a, float* tau,
+                        float* a, std::int64_t lda, std::int64_t stride_a, const float* tau,
                         std::int64_t stride_tau, std::int64_t batch_size, float* scratchpad,
                         std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event orgqr_batch(sycl::queue& queue, std::int64_t m, std::int64_t n, std::int64_t k,
-                        double* a, std::int64_t lda, std::int64_t stride_a, double* tau,
+                        double* a, std::int64_t lda, std::int64_t stride_a, const double* tau,
                         std::int64_t stride_tau, std::int64_t batch_size, double* scratchpad,
                         std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event orgqr_batch(sycl::queue& queue, std::int64_t* m, std::int64_t* n, std::int64_t* k,
-                        float** a, std::int64_t* lda, float** tau, std::int64_t group_count,
-                        std::int64_t* group_sizes, float* scratchpad, std::int64_t scratchpad_size,
+                        float** a, std::int64_t* lda, const float* const* tau,
+                        std::int64_t group_count, std::int64_t* group_sizes, float* scratchpad,
+                        std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event orgqr_batch(sycl::queue& queue, std::int64_t* m, std::int64_t* n, std::int64_t* k,
-                        double** a, std::int64_t* lda, double** tau, std::int64_t group_count,
-                        std::int64_t* group_sizes, double* scratchpad, std::int64_t scratchpad_size,
+                        double** a, std::int64_t* lda, const double* const* tau,
+                        std::int64_t group_count, std::int64_t* group_sizes, double* scratchpad,
+                        std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event potrf_batch(sycl::queue& queue, oneapi::mkl::uplo uplo, std::int64_t n, float* a,
                         std::int64_t lda, std::int64_t stride_a, std::int64_t batch_size,
@@ -1003,68 +1012,72 @@ sycl::event potrf_batch(sycl::queue& queue, oneapi::mkl::uplo* uplo, std::int64_
                         std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event potrs_batch(sycl::queue& queue, oneapi::mkl::uplo uplo, std::int64_t n,
-                        std::int64_t nrhs, float* a, std::int64_t lda, std::int64_t stride_a,
+                        std::int64_t nrhs, const float* a, std::int64_t lda, std::int64_t stride_a,
                         float* b, std::int64_t ldb, std::int64_t stride_b, std::int64_t batch_size,
                         float* scratchpad, std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event potrs_batch(sycl::queue& queue, oneapi::mkl::uplo uplo, std::int64_t n,
-                        std::int64_t nrhs, double* a, std::int64_t lda, std::int64_t stride_a,
+                        std::int64_t nrhs, const double* a, std::int64_t lda, std::int64_t stride_a,
                         double* b, std::int64_t ldb, std::int64_t stride_b, std::int64_t batch_size,
                         double* scratchpad, std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event potrs_batch(sycl::queue& queue, oneapi::mkl::uplo uplo, std::int64_t n,
-                        std::int64_t nrhs, std::complex<float>* a, std::int64_t lda,
+                        std::int64_t nrhs, const std::complex<float>* a, std::int64_t lda,
                         std::int64_t stride_a, std::complex<float>* b, std::int64_t ldb,
                         std::int64_t stride_b, std::int64_t batch_size,
                         std::complex<float>* scratchpad, std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event potrs_batch(sycl::queue& queue, oneapi::mkl::uplo uplo, std::int64_t n,
-                        std::int64_t nrhs, std::complex<double>* a, std::int64_t lda,
+                        std::int64_t nrhs, const std::complex<double>* a, std::int64_t lda,
                         std::int64_t stride_a, std::complex<double>* b, std::int64_t ldb,
                         std::int64_t stride_b, std::int64_t batch_size,
                         std::complex<double>* scratchpad, std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event potrs_batch(sycl::queue& queue, oneapi::mkl::uplo* uplo, std::int64_t* n,
-                        std::int64_t* nrhs, float** a, std::int64_t* lda, float** b,
+                        std::int64_t* nrhs, const float* const* a, std::int64_t* lda, float** b,
                         std::int64_t* ldb, std::int64_t group_count, std::int64_t* group_sizes,
                         float* scratchpad, std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event potrs_batch(sycl::queue& queue, oneapi::mkl::uplo* uplo, std::int64_t* n,
-                        std::int64_t* nrhs, double** a, std::int64_t* lda, double** b,
+                        std::int64_t* nrhs, const double* const* a, std::int64_t* lda, double** b,
                         std::int64_t* ldb, std::int64_t group_count, std::int64_t* group_sizes,
                         double* scratchpad, std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event potrs_batch(sycl::queue& queue, oneapi::mkl::uplo* uplo, std::int64_t* n,
-                        std::int64_t* nrhs, std::complex<float>** a, std::int64_t* lda,
+                        std::int64_t* nrhs, const std::complex<float>* const* a, std::int64_t* lda,
                         std::complex<float>** b, std::int64_t* ldb, std::int64_t group_count,
                         std::int64_t* group_sizes, std::complex<float>* scratchpad,
                         std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event potrs_batch(sycl::queue& queue, oneapi::mkl::uplo* uplo, std::int64_t* n,
-                        std::int64_t* nrhs, std::complex<double>** a, std::int64_t* lda,
+                        std::int64_t* nrhs, const std::complex<double>* const* a, std::int64_t* lda,
                         std::complex<double>** b, std::int64_t* ldb, std::int64_t group_count,
                         std::int64_t* group_sizes, std::complex<double>* scratchpad,
                         std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event ungqr_batch(sycl::queue& queue, std::int64_t m, std::int64_t n, std::int64_t k,
                         std::complex<float>* a, std::int64_t lda, std::int64_t stride_a,
-                        std::complex<float>* tau, std::int64_t stride_tau, std::int64_t batch_size,
-                        std::complex<float>* scratchpad, std::int64_t scratchpad_size,
+                        const std::complex<float>* tau, std::int64_t stride_tau,
+                        std::int64_t batch_size, std::complex<float>* scratchpad,
+                        std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event ungqr_batch(sycl::queue& queue, std::int64_t m, std::int64_t n, std::int64_t k,
                         std::complex<double>* a, std::int64_t lda, std::int64_t stride_a,
-                        std::complex<double>* tau, std::int64_t stride_tau, std::int64_t batch_size,
-                        std::complex<double>* scratchpad, std::int64_t scratchpad_size,
+                        const std::complex<double>* tau, std::int64_t stride_tau,
+                        std::int64_t batch_size, std::complex<double>* scratchpad,
+                        std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event ungqr_batch(sycl::queue& queue, std::int64_t* m, std::int64_t* n, std::int64_t* k,
-                        std::complex<float>** a, std::int64_t* lda, std::complex<float>** tau,
-                        std::int64_t group_count, std::int64_t* group_sizes,
-                        std::complex<float>* scratchpad, std::int64_t scratchpad_size,
+                        std::complex<float>** a, std::int64_t* lda,
+                        const std::complex<float>* const* tau, std::int64_t group_count,
+                        std::int64_t* group_sizes, std::complex<float>* scratchpad,
+                        std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 sycl::event ungqr_batch(sycl::queue& queue, std::int64_t* m, std::int64_t* n, std::int64_t* k,
-                        std::complex<double>** a, std::int64_t* lda, std::complex<double>** tau,
-                        std::int64_t group_count, std::int64_t* group_sizes,
-                        std::complex<double>* scratchpad, std::int64_t scratchpad_size,
+                        std::complex<double>** a, std::int64_t* lda,
+                        const std::complex<double>* const* tau, std::int64_t group_count,
+                        std::int64_t* group_sizes, std::complex<double>* scratchpad,
+                        std::int64_t scratchpad_size,
                         const std::vector<sycl::event>& dependencies = {});
 
 template <typename fp_type, internal::is_floating_point<fp_type> = nullptr>