From 54816087cfa8795d9e36c82e4ae65324b14f3e95 Mon Sep 17 00:00:00 2001
From: Florent Pruvost <florent.pruvost@inria.fr>
Date: Thu, 17 Sep 2015 15:58:10 +0000
Subject: [PATCH] add eztrace chameleon_cuda module - not checked yet
---
cudablas/eztrace_module/CMakeLists.txt | 78 ++
.../eztrace_module/cudablas_eztrace_module | 1234 +++++++++++++++++
2 files changed, 1312 insertions(+)
create mode 100644 cudablas/eztrace_module/CMakeLists.txt
create mode 100644 cudablas/eztrace_module/cudablas_eztrace_module
diff --git a/cudablas/eztrace_module/CMakeLists.txt b/cudablas/eztrace_module/CMakeLists.txt
new file mode 100644
index 000000000..339accb92
--- /dev/null
+++ b/cudablas/eztrace_module/CMakeLists.txt
@@ -0,0 +1,78 @@
+###
+#
+# @copyright (c) 2009-2014 The University of Tennessee and The University
+# of Tennessee Research Foundation.
+# All rights reserved.
+# @copyright (c) 2012-2015 Inria. All rights reserved.
+# @copyright (c) 2012-2014 Bordeaux INP, CNRS (LaBRI UMR 5800), Inria, Univ. Bordeaux. All rights reserved.
+#
+###
+#
+# @file CMakeLists.txt
+#
+# @project MORSE
+# MORSE is a software package provided by:
+# Inria Bordeaux - Sud-Ouest,
+# Univ. of Tennessee,
+# King Abdullah Univesity of Science and Technology
+# Univ. of California Berkeley,
+# Univ. of Colorado Denver.
+#
+# @version 0.9.0
+# @author Florent Pruvost
+# @date 16-09-2015
+#
+###
+
+if (NOT EZTRACE_FOUND)
+ find_package(EZTRACE)
+endif()
+
+if (EZTRACE_FOUND AND EZTRACE_DIR_FOUND)
+
+ set(EZTRACE_eztrace_create_plugin_DIR "EZTRACE_eztrace_create_plugin_DIR-NOTFOUND")
+ find_path(EZTRACE_eztrace_create_plugin_DIR
+ NAMES eztrace_create_plugin
+ HINTS ${EZTRACE_DIR_FOUND}/bin)
+
+ if (EZTRACE_eztrace_create_plugin_DIR)
+
+ set(EZTRACE_CREATE_PLUGIN "${EZTRACE_eztrace_create_plugin_DIR}/eztrace_create_plugin")
+
+ add_custom_command(
+ OUTPUT ${CMAKE_CURRENT_BINARY_DIR}/output
+ COMMAND ${EZTRACE_CREATE_PLUGIN}
+ ARGS ${CMAKE_CURRENT_SOURCE_DIR}/cudablas_eztrace_module
+ DEPENDS ${CMAKE_CURRENT_SOURCE_DIR}/cudablas_eztrace_module
+ )
+ add_custom_target(
+ eztrace-module-chameleon_cuda-dir ALL
+ DEPENDS ${CMAKE_CURRENT_BINARY_DIR}/output
+ )
+ add_custom_command(
+ OUTPUT libeztrace-convert-chameleon_cuda.so
+ COMMAND make
+ WORKING_DIRECTORY ${CMAKE_CURRENT_BINARY_DIR}/output
+ DEPENDS ${CMAKE_CURRENT_BINARY_DIR}/output
+ )
+ add_custom_target(
+ eztrace-module-chameleon_cuda-libs ALL
+ DEPENDS libeztrace-convert-chameleon_cuda.so
+ )
+ # installation
+ # ------------
+ install(
+ FILES
+ ${CMAKE_CURRENT_BINARY_DIR}/output/libeztrace-autostart-chameleon_cuda.so
+ ${CMAKE_CURRENT_BINARY_DIR}/output/libeztrace-chameleon_cuda.so
+ ${CMAKE_CURRENT_BINARY_DIR}/output/libeztrace-convert-chameleon_cuda.so
+ DESTINATION ${EZTRACE_LIBRARY_DIRS}
+ )
+
+ endif (EZTRACE_eztrace_create_plugin_DIR)
+
+endif (EZTRACE_FOUND AND EZTRACE_DIR_FOUND)
+
+###
+### END CMakeLists.txt
+###
diff --git a/cudablas/eztrace_module/cudablas_eztrace_module b/cudablas/eztrace_module/cudablas_eztrace_module
new file mode 100644
index 000000000..4834a2653
--- /dev/null
+++ b/cudablas/eztrace_module/cudablas_eztrace_module
@@ -0,0 +1,1234 @@
+BEGIN_MODULE
+NAME chameleon_cuda
+DESC "Module for Chameleon CUDA functions"
+ID 7771
+
+int CUDA_cgelqt(
+ void* m, void* n, void* nb,
+ void *da, void* ldda,
+ void *v, void* ldv,
+ void *dt, void* lddt,
+ void *t, void* ldt,
+ void *dd,
+ void *d, void* ldd,
+ void *tau,
+ void *hwork,
+ void *dwork,
+ void* stream);
+int CUDA_cgemerge(
+ void* side, void* diag,
+ void* M, void* N,
+ void *A, void* LDA,
+ void *B, void* LDB,
+ void* stream);
+int CUDA_cgemm_V2(
+ void* transa, void* transb,
+ int m, int n, int k,
+ void *alpha,
+ const void *A, int lda,
+ const void *B, int ldb,
+ void *beta,
+ void *C, int ldc,
+ void* stream);
+int CUDA_cgemm(
+ void* transa, void* transb,
+ int m, int n, int k,
+ void *alpha,
+ const void *A, int lda,
+ const void *B, int ldb,
+ void *beta,
+ void *C, int ldc,
+ void* stream);
+int CUDA_cgeqrt(
+ void* m, void* n, void* nb,
+ void *da, void* ldda,
+ void *v, void* ldv,
+ void *dt, void* lddt,
+ void *t, void* ldt,
+ void *dd,
+ void *d, void* ldd,
+ void *tau,
+ void *hwork,
+ void *dwork,
+ void* stream);
+int CUDA_cgessm(
+ char storev, void* m, void* n, void* k, void* ib,
+ void* *ipiv,
+ void *dL1, void* lddl1,
+ void *dL, void* lddl,
+ void *dA, void* ldda,
+ void* info);
+int CUDA_cgetrf_incpiv(
+ char storev, void* m, void* n, void* ib,
+ void *hA, void* ldha, void *dA, void* ldda,
+ void *hL, void* ldhl, void *dL, void* lddl,
+ void* *ipiv,
+ void *dwork, void* lddwork,
+ void* info);
+int CUDA_cgetrf_nopiv(
+ void* m, void* n,
+ void *dA, void* ldda,
+ void* info);
+int CUDA_chemm_V2(
+ void* side, void* uplo,
+ int m, int n,
+ void *alpha,
+ const void *A, int lda,
+ const void *B, int ldb,
+ void *beta,
+ void *C, int ldc,
+ void* stream);
+int CUDA_chemm(
+ void* side, void* uplo,
+ int m, int n,
+ void *alpha,
+ const void *A, int lda,
+ const void *B, int ldb,
+ void *beta,
+ void *C, int ldc,
+ void* stream);
+int CUDA_cher2k_V2(
+ void* uplo, void* trans,
+ int n, int k,
+ void *alpha,
+ const void *A, int lda,
+ const void *B, int ldb,
+ float *beta,
+ void *C, int ldc,
+ void* stream);
+int CUDA_cher2k(
+ void* uplo, void* trans,
+ int n, int k,
+ void *alpha,
+ const void *A, int lda,
+ const void *B, int ldb,
+ float *beta,
+ void *C, int ldc,
+ void* stream);
+int CUDA_cherk_V2(
+ void* uplo, void* trans,
+ int n, int k,
+ float *alpha,
+ const void *A, int lda,
+ float *beta,
+ void *B, int ldb,
+ void* stream);
+int CUDA_cherk(
+ void* uplo, void* trans,
+ int n, int k,
+ float *alpha,
+ const void *A, int lda,
+ float *beta,
+ void *B, int ldb,
+ void* stream);
+int CUDA_clauum(
+ char uplo, void* n,
+ void *dA, void* ldda, void* info);
+int CUDA_cparfb(
+ void* side, void* trans,
+ void* direct, void* storev,
+ void* M1, void* N1,
+ void* M2, void* N2,
+ void* K, void* L,
+ void *A1, void* LDA1,
+ void *A2, void* LDA2,
+ void *V, void* LDV,
+ void *T, void* LDT,
+ void *WORK, void* LDWORK,
+ void *WORKC, void* LDWORKC,
+ void* stream);
+int CUDA_cpotrf(
+ void* uplo, void* n,
+ void *dA, void* ldda, void* info);
+int CUDA_cssssm(
+ void* storev, void* m1, void* n1,
+ void* m2, void* n2, void* k, void* ib,
+ void *dA1, void* ldda1,
+ void *dA2, void* ldda2,
+ void *dL1, void* lddl1,
+ void *dL2, void* lddl2,
+ void* *IPIV, void* info);
+int CUDA_csymm_V2(
+ void* side, void* uplo,
+ int m, int n,
+ void *alpha,
+ const void *A, int lda,
+ const void *B, int ldb,
+ void *beta,
+ void *C, int ldc,
+ void* stream);
+int CUDA_csymm(
+ void* side, void* uplo,
+ int m, int n,
+ void *alpha,
+ const void *A, int lda,
+ const void *B, int ldb,
+ void *beta,
+ void *C, int ldc,
+ void* stream);
+int CUDA_csyr2k_V2(
+ void* uplo, void* trans,
+ int n, int k,
+ void *alpha,
+ const void *A, int lda,
+ const void *B, int ldb,
+ void *beta,
+ void *C, int ldc,
+ void* stream);
+int CUDA_csyr2k(
+ void* uplo, void* trans,
+ int n, int k,
+ void *alpha,
+ const void *A, int lda,
+ const void *B, int ldb,
+ void *beta,
+ void *C, int ldc,
+ void* stream);
+int CUDA_csyrk_V2(
+ void* uplo, void* trans,
+ int n, int k,
+ void *alpha,
+ const void *A, int lda,
+ void *beta,
+ void *C, int ldc,
+ void* stream);
+int CUDA_csyrk(
+ void* uplo, void* trans,
+ int n, int k,
+ void *alpha,
+ const void *A, int lda,
+ void *beta,
+ void *C, int ldc,
+ void* stream);
+int CUDA_ctrmm_V2(
+ void* side, void* uplo,
+ void* transa, void* diag,
+ int m, int n,
+ void *alpha,
+ const void *A, int lda,
+ const void *B, int ldb,
+ void *C, int ldc,
+ void* stream);
+int CUDA_ctrmm(
+ void* side, void* uplo,
+ void* transa, void* diag,
+ int m, int n,
+ void *alpha,
+ const void *A, int lda,
+ void *B, int ldb,
+ void* stream);
+int CUDA_ctrsm_V2(
+ void* side, void* uplo,
+ void* transa, void* diag,
+ int m, int n,
+ const void *alpha,
+ const void *A, int lda,
+ void *B, int ldb,
+ void* stream);
+int CUDA_ctrsm(
+ void* side, void* uplo,
+ void* transa, void* diag,
+ int m, int n,
+ void *alpha,
+ const void *A, int lda,
+ void *B, int ldb,
+ void* stream);
+int CUDA_ctrtri(
+ void* uplo, void* diag, void* n,
+ void *dA, void* ldda, void* info);
+int CUDA_ctslqt(
+ void* m, void* n, void* nb,
+ void *da1, void* ldda1,
+ void *da2, void* ldda2,
+ void *a2, void* lda2,
+ void *dt, void* lddt,
+ void *t, void* ldt,
+ void *dd,
+ void *d, void* ldd,
+ void *tau,
+ void *hwork,
+ void *dwork,
+ void* stream);
+int CUDA_ctsmlq(
+ void* side, void* trans,
+ void* M1, void* N1,
+ void* M2, void* N2,
+ void* K, void* IB,
+ void *A1, void* LDA1,
+ void *A2, void* LDA2,
+ void *V, void* LDV,
+ void *T, void* LDT,
+ void *WORK, void* LDWORK,
+ void *WORKC, void* LDWORKC,
+ void* stream);
+int CUDA_ctsmqr(
+ void* side, void* trans,
+ void* M1, void* N1,
+ void* M2, void* N2,
+ void* K, void* IB,
+ void *A1, void* LDA1,
+ void *A2, void* LDA2,
+ void *V, void* LDV,
+ void *T, void* LDT,
+ void *WORK, void* LDWORK,
+ void *WORKC, void* LDWORKC,
+ void* stream);
+int CUDA_ctsqrt(
+ void* m, void* n, void* nb,
+ void *da1, void* ldda1,
+ void *da2, void* ldda2,
+ void *a2, void* lda2,
+ void *dt, void* lddt,
+ void *t, void* ldt,
+ void *dd,
+ void *d, void* ldd,
+ void *tau,
+ void *hwork,
+ void *dwork,
+ void* stream);
+int CUDA_ctstrf(
+ char storev, void* m, void* n, void* ib, void* nb,
+ void *hU, void* ldhu, void *dU, void* lddu,
+ void *hA, void* ldha, void *dA, void* ldda,
+ void *hL, void* ldhl, void *dL, void* lddl,
+ void* *ipiv,
+ void *hwork, void* ldhwork,
+ void *dwork, void* lddwork,
+ void* info);
+int CUDA_cunmlqt(
+ void* side, void* trans,
+ void* M, void* N, void* K, void* IB,
+ void *A, void* LDA,
+ void *T, void* LDT,
+ void *C, void* LDC,
+ void *WORK, void* LDWORK );
+int CUDA_cunmqrt(
+ void* side, void* trans,
+ void* M, void* N, void* K, void* IB,
+ void *A, void* LDA,
+ void *T, void* LDT,
+ void *C, void* LDC,
+ void *WORK, void* LDWORK );
+
+int CUDA_dgelqt(
+ void* m, void* n, void* nb,
+ double *da, void* ldda,
+ double *v, void* ldv,
+ double *dt, void* lddt,
+ double *t, void* ldt,
+ double *dd,
+ double *d, void* ldd,
+ double *tau,
+ double *hwork,
+ double *dwork,
+ void* stream);
+int CUDA_dgemerge(
+ void* side, void* diag,
+ void* M, void* N,
+ double *A, void* LDA,
+ double *B, void* LDB,
+ void* stream);
+int CUDA_dgemm_V2(
+ void* transa, void* transb,
+ int m, int n, int k,
+ double *alpha,
+ const double *A, int lda,
+ const double *B, int ldb,
+ double *beta,
+ double *C, int ldc,
+ void* stream);
+int CUDA_dgemm(
+ void* transa, void* transb,
+ int m, int n, int k,
+ double *alpha,
+ const double *A, int lda,
+ const double *B, int ldb,
+ double *beta,
+ double *C, int ldc,
+ void* stream);
+int CUDA_dgeqrt(
+ void* m, void* n, void* nb,
+ double *da, void* ldda,
+ double *v, void* ldv,
+ double *dt, void* lddt,
+ double *t, void* ldt,
+ double *dd,
+ double *d, void* ldd,
+ double *tau,
+ double *hwork,
+ double *dwork,
+ void* stream);
+int CUDA_dgessm(
+ char storev, void* m, void* n, void* k, void* ib,
+ void* *ipiv,
+ double *dL1, void* lddl1,
+ double *dL, void* lddl,
+ double *dA, void* ldda,
+ void* info);
+int CUDA_dgetrf_incpiv(
+ char storev, void* m, void* n, void* ib,
+ double *hA, void* ldha, double *dA, void* ldda,
+ double *hL, void* ldhl, double *dL, void* lddl,
+ void* *ipiv,
+ double *dwork, void* lddwork,
+ void* info);
+int CUDA_dgetrf_nopiv(
+ void* m, void* n,
+ double *dA, void* ldda,
+ void* info);
+int CUDA_dsymm_V2(
+ void* side, void* uplo,
+ int m, int n,
+ double *alpha,
+ const double *A, int lda,
+ const double *B, int ldb,
+ double *beta,
+ double *C, int ldc,
+ void* stream);
+int CUDA_dsymm(
+ void* side, void* uplo,
+ int m, int n,
+ double *alpha,
+ const double *A, int lda,
+ const double *B, int ldb,
+ double *beta,
+ double *C, int ldc,
+ void* stream);
+int CUDA_dsyr2k_V2(
+ void* uplo, void* trans,
+ int n, int k,
+ double *alpha,
+ const double *A, int lda,
+ const double *B, int ldb,
+ double *beta,
+ double *C, int ldc,
+ void* stream);
+int CUDA_dsyr2k(
+ void* uplo, void* trans,
+ int n, int k,
+ double *alpha,
+ const double *A, int lda,
+ const double *B, int ldb,
+ double *beta,
+ double *C, int ldc,
+ void* stream);
+int CUDA_dsyrk_V2(
+ void* uplo, void* trans,
+ int n, int k,
+ double *alpha,
+ const double *A, int lda,
+ double *beta,
+ double *B, int ldb,
+ void* stream);
+int CUDA_dsyrk(
+ void* uplo, void* trans,
+ int n, int k,
+ double *alpha,
+ const double *A, int lda,
+ double *beta,
+ double *B, int ldb,
+ void* stream);
+int CUDA_dlauum(
+ char uplo, void* n,
+ double *dA, void* ldda, void* info);
+int CUDA_dparfb(
+ void* side, void* trans,
+ void* direct, void* storev,
+ void* M1, void* N1,
+ void* M2, void* N2,
+ void* K, void* L,
+ double *A1, void* LDA1,
+ double *A2, void* LDA2,
+ const double *V, void* LDV,
+ const double *T, void* LDT,
+ double *WORK, void* LDWORK,
+ double *WORKC, void* LDWORKC,
+ void* stream);
+int CUDA_dpotrf(
+ void* uplo, void* n,
+ double *dA, void* ldda, void* info);
+int CUDA_dssssm(
+ void* storev, void* m1, void* n1,
+ void* m2, void* n2, void* k, void* ib,
+ double *dA1, void* ldda1,
+ double *dA2, void* ldda2,
+ double *dL1, void* lddl1,
+ double *dL2, void* lddl2,
+ void* *IPIV, void* info);
+int CUDA_dsymm_V2(
+ void* side, void* uplo,
+ int m, int n,
+ double *alpha,
+ const double *A, int lda,
+ const double *B, int ldb,
+ double *beta,
+ double *C, int ldc,
+ void* stream);
+int CUDA_dsymm(
+ void* side, void* uplo,
+ int m, int n,
+ double *alpha,
+ const double *A, int lda,
+ const double *B, int ldb,
+ double *beta,
+ double *C, int ldc,
+ void* stream);
+int CUDA_dsyr2k_V2(
+ void* uplo, void* trans,
+ int n, int k,
+ double *alpha,
+ const double *A, int lda,
+ const double *B, int ldb,
+ double *beta,
+ double *C, int ldc,
+ void* stream);
+int CUDA_dsyr2k(
+ void* uplo, void* trans,
+ int n, int k,
+ double *alpha,
+ const double *A, int lda,
+ const double *B, int ldb,
+ double *beta,
+ double *C, int ldc,
+ void* stream);
+int CUDA_dsyrk_V2(
+ void* uplo, void* trans,
+ int n, int k,
+ double *alpha,
+ const double *A, int lda,
+ double *beta,
+ double *C, int ldc,
+ void* stream);
+int CUDA_dsyrk(
+ void* uplo, void* trans,
+ int n, int k,
+ double *alpha,
+ const double *A, int lda,
+ double *beta,
+ double *C, int ldc,
+ void* stream);
+int CUDA_dtrmm_V2(
+ void* side, void* uplo,
+ void* transa, void* diag,
+ int m, int n,
+ double *alpha,
+ const double *A, int lda,
+ const double *B, int ldb,
+ double *C, int ldc,
+ void* stream);
+int CUDA_dtrmm(
+ void* side, void* uplo,
+ void* transa, void* diag,
+ int m, int n,
+ double *alpha,
+ const double *A, int lda,
+ double *B, int ldb,
+ void* stream);
+int CUDA_dtrsm_V2(
+ void* side, void* uplo,
+ void* transa, void* diag,
+ int m, int n,
+ const double *alpha,
+ const double *A, int lda,
+ double *B, int ldb,
+ void* stream);
+int CUDA_dtrsm(
+ void* side, void* uplo,
+ void* transa, void* diag,
+ int m, int n,
+ double *alpha,
+ const double *A, int lda,
+ double *B, int ldb,
+ void* stream);
+int CUDA_dtrtri(
+ void* uplo, void* diag, void* n,
+ double *dA, void* ldda, void* info);
+int CUDA_dtslqt(
+ void* m, void* n, void* nb,
+ double *da1, void* ldda1,
+ double *da2, void* ldda2,
+ double *a2, void* lda2,
+ double *dt, void* lddt,
+ double *t, void* ldt,
+ double *dd,
+ double *d, void* ldd,
+ double *tau,
+ double *hwork,
+ double *dwork,
+ void* stream);
+int CUDA_dtsmlq(
+ void* side, void* trans,
+ void* M1, void* N1,
+ void* M2, void* N2,
+ void* K, void* IB,
+ double *A1, void* LDA1,
+ double *A2, void* LDA2,
+ const double *V, void* LDV,
+ const double *T, void* LDT,
+ double *WORK, void* LDWORK,
+ double *WORKC, void* LDWORKC,
+ void* stream);
+int CUDA_dtsmqr(
+ void* side, void* trans,
+ void* M1, void* N1,
+ void* M2, void* N2,
+ void* K, void* IB,
+ double *A1, void* LDA1,
+ double *A2, void* LDA2,
+ const double *V, void* LDV,
+ const double *T, void* LDT,
+ double *WORK, void* LDWORK,
+ double *WORKC, void* LDWORKC,
+ void* stream);
+int CUDA_dtsqrt(
+ void* m, void* n, void* nb,
+ double *da1, void* ldda1,
+ double *da2, void* ldda2,
+ double *a2, void* lda2,
+ double *dt, void* lddt,
+ double *t, void* ldt,
+ double *dd,
+ double *d, void* ldd,
+ double *tau,
+ double *hwork,
+ double *dwork,
+ void* stream);
+int CUDA_dtstrf(
+ char storev, void* m, void* n, void* ib, void* nb,
+ double *hU, void* ldhu, double *dU, void* lddu,
+ double *hA, void* ldha, double *dA, void* ldda,
+ double *hL, void* ldhl, double *dL, void* lddl,
+ void* *ipiv,
+ double *hwork, void* ldhwork,
+ double *dwork, void* lddwork,
+ void* info);
+int CUDA_dormlqt(
+ void* side, void* trans,
+ void* M, void* N, void* K, void* IB,
+ const double *A, void* LDA,
+ const double *T, void* LDT,
+ double *C, void* LDC,
+ double *WORK, void* LDWORK );
+int CUDA_dormqrt(
+ void* side, void* trans,
+ void* M, void* N, void* K, void* IB,
+ const double *A, void* LDA,
+ const double *T, void* LDT,
+ double *C, void* LDC,
+ double *WORK, void* LDWORK );
+
+int CUDA_sgelqt(
+ void* m, void* n, void* nb,
+ float *da, void* ldda,
+ float *v, void* ldv,
+ float *dt, void* lddt,
+ float *t, void* ldt,
+ float *dd,
+ float *d, void* ldd,
+ float *tau,
+ float *hwork,
+ float *dwork,
+ void* stream);
+int CUDA_sgemerge(
+ void* side, void* diag,
+ void* M, void* N,
+ float *A, void* LDA,
+ float *B, void* LDB,
+ void* stream);
+int CUDA_sgemm_V2(
+ void* transa, void* transb,
+ int m, int n, int k,
+ float *alpha,
+ const float *A, int lda,
+ const float *B, int ldb,
+ float *beta,
+ float *C, int ldc,
+ void* stream);
+int CUDA_sgemm(
+ void* transa, void* transb,
+ int m, int n, int k,
+ float *alpha,
+ const float *A, int lda,
+ const float *B, int ldb,
+ float *beta,
+ float *C, int ldc,
+ void* stream);
+int CUDA_sgeqrt(
+ void* m, void* n, void* nb,
+ float *da, void* ldda,
+ float *v, void* ldv,
+ float *dt, void* lddt,
+ float *t, void* ldt,
+ float *dd,
+ float *d, void* ldd,
+ float *tau,
+ float *hwork,
+ float *dwork,
+ void* stream);
+int CUDA_sgessm(
+ char storev, void* m, void* n, void* k, void* ib,
+ void* *ipiv,
+ float *dL1, void* lddl1,
+ float *dL, void* lddl,
+ float *dA, void* ldda,
+ void* info);
+int CUDA_sgetrf_incpiv(
+ char storev, void* m, void* n, void* ib,
+ float *hA, void* ldha, float *dA, void* ldda,
+ float *hL, void* ldhl, float *dL, void* lddl,
+ void* *ipiv,
+ float *dwork, void* lddwork,
+ void* info);
+int CUDA_sgetrf_nopiv(
+ void* m, void* n,
+ float *dA, void* ldda,
+ void* info);
+int CUDA_ssymm_V2(
+ void* side, void* uplo,
+ int m, int n,
+ float *alpha,
+ const float *A, int lda,
+ const float *B, int ldb,
+ float *beta,
+ float *C, int ldc,
+ void* stream);
+int CUDA_ssymm(
+ void* side, void* uplo,
+ int m, int n,
+ float *alpha,
+ const float *A, int lda,
+ const float *B, int ldb,
+ float *beta,
+ float *C, int ldc,
+ void* stream);
+int CUDA_ssyr2k_V2(
+ void* uplo, void* trans,
+ int n, int k,
+ float *alpha,
+ const float *A, int lda,
+ const float *B, int ldb,
+ float *beta,
+ float *C, int ldc,
+ void* stream);
+int CUDA_ssyr2k(
+ void* uplo, void* trans,
+ int n, int k,
+ float *alpha,
+ const float *A, int lda,
+ const float *B, int ldb,
+ float *beta,
+ float *C, int ldc,
+ void* stream);
+int CUDA_ssyrk_V2(
+ void* uplo, void* trans,
+ int n, int k,
+ float *alpha,
+ const float *A, int lda,
+ float *beta,
+ float *B, int ldb,
+ void* stream);
+int CUDA_ssyrk(
+ void* uplo, void* trans,
+ int n, int k,
+ float *alpha,
+ const float *A, int lda,
+ float *beta,
+ float *B, int ldb,
+ void* stream);
+int CUDA_slauum(
+ char uplo, void* n,
+ float *dA, void* ldda, void* info);
+int CUDA_sparfb(
+ void* side, void* trans,
+ void* direct, void* storev,
+ void* M1, void* N1,
+ void* M2, void* N2,
+ void* K, void* L,
+ float *A1, void* LDA1,
+ float *A2, void* LDA2,
+ const float *V, void* LDV,
+ const float *T, void* LDT,
+ float *WORK, void* LDWORK,
+ float *WORKC, void* LDWORKC,
+ void* stream);
+int CUDA_spotrf(
+ void* uplo, void* n,
+ float *dA, void* ldda, void* info);
+int CUDA_sssssm(
+ void* storev, void* m1, void* n1,
+ void* m2, void* n2, void* k, void* ib,
+ float *dA1, void* ldda1,
+ float *dA2, void* ldda2,
+ float *dL1, void* lddl1,
+ float *dL2, void* lddl2,
+ void* *IPIV, void* info);
+int CUDA_ssymm_V2(
+ void* side, void* uplo,
+ int m, int n,
+ float *alpha,
+ const float *A, int lda,
+ const float *B, int ldb,
+ float *beta,
+ float *C, int ldc,
+ void* stream);
+int CUDA_ssymm(
+ void* side, void* uplo,
+ int m, int n,
+ float *alpha,
+ const float *A, int lda,
+ const float *B, int ldb,
+ float *beta,
+ float *C, int ldc,
+ void* stream);
+int CUDA_ssyr2k_V2(
+ void* uplo, void* trans,
+ int n, int k,
+ float *alpha,
+ const float *A, int lda,
+ const float *B, int ldb,
+ float *beta,
+ float *C, int ldc,
+ void* stream);
+int CUDA_ssyr2k(
+ void* uplo, void* trans,
+ int n, int k,
+ float *alpha,
+ const float *A, int lda,
+ const float *B, int ldb,
+ float *beta,
+ float *C, int ldc,
+ void* stream);
+int CUDA_ssyrk_V2(
+ void* uplo, void* trans,
+ int n, int k,
+ float *alpha,
+ const float *A, int lda,
+ float *beta,
+ float *C, int ldc,
+ void* stream);
+int CUDA_ssyrk(
+ void* uplo, void* trans,
+ int n, int k,
+ float *alpha,
+ const float *A, int lda,
+ float *beta,
+ float *C, int ldc,
+ void* stream);
+int CUDA_strmm_V2(
+ void* side, void* uplo,
+ void* transa, void* diag,
+ int m, int n,
+ float *alpha,
+ const float *A, int lda,
+ const float *B, int ldb,
+ float *C, int ldc,
+ void* stream);
+int CUDA_strmm(
+ void* side, void* uplo,
+ void* transa, void* diag,
+ int m, int n,
+ float *alpha,
+ const float *A, int lda,
+ float *B, int ldb,
+ void* stream);
+int CUDA_strsm_V2(
+ void* side, void* uplo,
+ void* transa, void* diag,
+ int m, int n,
+ const float *alpha,
+ const float *A, int lda,
+ float *B, int ldb,
+ void* stream);
+int CUDA_strsm(
+ void* side, void* uplo,
+ void* transa, void* diag,
+ int m, int n,
+ float *alpha,
+ const float *A, int lda,
+ float *B, int ldb,
+ void* stream);
+int CUDA_strtri(
+ void* uplo, void* diag, void* n,
+ float *dA, void* ldda, void* info);
+int CUDA_stslqt(
+ void* m, void* n, void* nb,
+ float *da1, void* ldda1,
+ float *da2, void* ldda2,
+ float *a2, void* lda2,
+ float *dt, void* lddt,
+ float *t, void* ldt,
+ float *dd,
+ float *d, void* ldd,
+ float *tau,
+ float *hwork,
+ float *dwork,
+ void* stream);
+int CUDA_stsmlq(
+ void* side, void* trans,
+ void* M1, void* N1,
+ void* M2, void* N2,
+ void* K, void* IB,
+ float *A1, void* LDA1,
+ float *A2, void* LDA2,
+ const float *V, void* LDV,
+ const float *T, void* LDT,
+ float *WORK, void* LDWORK,
+ float *WORKC, void* LDWORKC,
+ void* stream);
+int CUDA_stsmqr(
+ void* side, void* trans,
+ void* M1, void* N1,
+ void* M2, void* N2,
+ void* K, void* IB,
+ float *A1, void* LDA1,
+ float *A2, void* LDA2,
+ const float *V, void* LDV,
+ const float *T, void* LDT,
+ float *WORK, void* LDWORK,
+ float *WORKC, void* LDWORKC,
+ void* stream);
+int CUDA_stsqrt(
+ void* m, void* n, void* nb,
+ float *da1, void* ldda1,
+ float *da2, void* ldda2,
+ float *a2, void* lda2,
+ float *dt, void* lddt,
+ float *t, void* ldt,
+ float *dd,
+ float *d, void* ldd,
+ float *tau,
+ float *hwork,
+ float *dwork,
+ void* stream);
+int CUDA_ststrf(
+ char storev, void* m, void* n, void* ib, void* nb,
+ float *hU, void* ldhu, float *dU, void* lddu,
+ float *hA, void* ldha, float *dA, void* ldda,
+ float *hL, void* ldhl, float *dL, void* lddl,
+ void* *ipiv,
+ float *hwork, void* ldhwork,
+ float *dwork, void* lddwork,
+ void* info);
+int CUDA_sormlqt(
+ void* side, void* trans,
+ void* M, void* N, void* K, void* IB,
+ const float *A, void* LDA,
+ const float *T, void* LDT,
+ float *C, void* LDC,
+ float *WORK, void* LDWORK );
+int CUDA_sormqrt(
+ void* side, void* trans,
+ void* M, void* N, void* K, void* IB,
+ const float *A, void* LDA,
+ const float *T, void* LDT,
+ float *C, void* LDC,
+ float *WORK, void* LDWORK );
+
+int CUDA_zgelqt(
+ void* m, void* n, void* nb,
+ void *da, void* ldda,
+ void *v, void* ldv,
+ void *dt, void* lddt,
+ void *t, void* ldt,
+ void *dd,
+ void *d, void* ldd,
+ void *tau,
+ void *hwork,
+ void *dwork,
+ void* stream);
+int CUDA_zgemerge(
+ void* side, void* diag,
+ void* M, void* N,
+ void *A, void* LDA,
+ void *B, void* LDB,
+ void* stream);
+int CUDA_zgemm_V2(
+ void* transa, void* transb,
+ int m, int n, int k,
+ void *alpha,
+ const void *A, int lda,
+ const void *B, int ldb,
+ void *beta,
+ void *C, int ldc,
+ void* stream);
+int CUDA_zgemm(
+ void* transa, void* transb,
+ int m, int n, int k,
+ void *alpha,
+ const void *A, int lda,
+ const void *B, int ldb,
+ void *beta,
+ void *C, int ldc,
+ void* stream);
+int CUDA_zgeqrt(
+ void* m, void* n, void* nb,
+ void *da, void* ldda,
+ void *v, void* ldv,
+ void *dt, void* lddt,
+ void *t, void* ldt,
+ void *dd,
+ void *d, void* ldd,
+ void *tau,
+ void *hwork,
+ void *dwork,
+ void* stream);
+int CUDA_zgessm(
+ char storev, void* m, void* n, void* k, void* ib,
+ void* *ipiv,
+ void *dL1, void* lddl1,
+ void *dL, void* lddl,
+ void *dA, void* ldda,
+ void* info);
+int CUDA_zgetrf_incpiv(
+ char storev, void* m, void* n, void* ib,
+ void *hA, void* ldha, void *dA, void* ldda,
+ void *hL, void* ldhl, void *dL, void* lddl,
+ void* *ipiv,
+ void *dwork, void* lddwork,
+ void* info);
+int CUDA_zgetrf_nopiv(
+ void* m, void* n,
+ void *dA, void* ldda,
+ void* info);
+int CUDA_zhemm_V2(
+ void* side, void* uplo,
+ int m, int n,
+ void *alpha,
+ const void *A, int lda,
+ const void *B, int ldb,
+ void *beta,
+ void *C, int ldc,
+ void* stream);
+int CUDA_zhemm(
+ void* side, void* uplo,
+ int m, int n,
+ void *alpha,
+ const void *A, int lda,
+ const void *B, int ldb,
+ void *beta,
+ void *C, int ldc,
+ void* stream);
+int CUDA_zher2k_V2(
+ void* uplo, void* trans,
+ int n, int k,
+ void *alpha,
+ const void *A, int lda,
+ const void *B, int ldb,
+ double *beta,
+ void *C, int ldc,
+ void* stream);
+int CUDA_zher2k(
+ void* uplo, void* trans,
+ int n, int k,
+ void *alpha,
+ const void *A, int lda,
+ const void *B, int ldb,
+ double *beta,
+ void *C, int ldc,
+ void* stream);
+int CUDA_zherk_V2(
+ void* uplo, void* trans,
+ int n, int k,
+ double *alpha,
+ const void *A, int lda,
+ double *beta,
+ void *B, int ldb,
+ void* stream);
+int CUDA_zherk(
+ void* uplo, void* trans,
+ int n, int k,
+ double *alpha,
+ const void *A, int lda,
+ double *beta,
+ void *B, int ldb,
+ void* stream);
+int CUDA_zlauum(
+ char uplo, void* n,
+ void *dA, void* ldda, void* info);
+int CUDA_zparfb(
+ void* side, void* trans,
+ void* direct, void* storev,
+ void* M1, void* N1,
+ void* M2, void* N2,
+ void* K, void* L,
+ void *A1, void* LDA1,
+ void *A2, void* LDA2,
+ void *V, void* LDV,
+ void *T, void* LDT,
+ void *WORK, void* LDWORK,
+ void *WORKC, void* LDWORKC,
+ void* stream);
+int CUDA_zpotrf(
+ void* uplo, void* n,
+ void *dA, void* ldda, void* info);
+int CUDA_zssssm(
+ void* storev, void* m1, void* n1,
+ void* m2, void* n2, void* k, void* ib,
+ void *dA1, void* ldda1,
+ void *dA2, void* ldda2,
+ void *dL1, void* lddl1,
+ void *dL2, void* lddl2,
+ void* *IPIV, void* info);
+int CUDA_zsymm_V2(
+ void* side, void* uplo,
+ int m, int n,
+ void *alpha,
+ const void *A, int lda,
+ const void *B, int ldb,
+ void *beta,
+ void *C, int ldc,
+ void* stream);
+int CUDA_zsymm(
+ void* side, void* uplo,
+ int m, int n,
+ void *alpha,
+ const void *A, int lda,
+ const void *B, int ldb,
+ void *beta,
+ void *C, int ldc,
+ void* stream);
+int CUDA_zsyr2k_V2(
+ void* uplo, void* trans,
+ int n, int k,
+ void *alpha,
+ const void *A, int lda,
+ const void *B, int ldb,
+ void *beta,
+ void *C, int ldc,
+ void* stream);
+int CUDA_zsyr2k(
+ void* uplo, void* trans,
+ int n, int k,
+ void *alpha,
+ const void *A, int lda,
+ const void *B, int ldb,
+ void *beta,
+ void *C, int ldc,
+ void* stream);
+int CUDA_zsyrk_V2(
+ void* uplo, void* trans,
+ int n, int k,
+ void *alpha,
+ const void *A, int lda,
+ void *beta,
+ void *C, int ldc,
+ void* stream);
+int CUDA_zsyrk(
+ void* uplo, void* trans,
+ int n, int k,
+ void *alpha,
+ const void *A, int lda,
+ void *beta,
+ void *C, int ldc,
+ void* stream);
+int CUDA_ztrmm_V2(
+ void* side, void* uplo,
+ void* transa, void* diag,
+ int m, int n,
+ void *alpha,
+ const void *A, int lda,
+ const void *B, int ldb,
+ void *C, int ldc,
+ void* stream);
+int CUDA_ztrmm(
+ void* side, void* uplo,
+ void* transa, void* diag,
+ int m, int n,
+ void *alpha,
+ const void *A, int lda,
+ void *B, int ldb,
+ void* stream);
+int CUDA_ztrsm_V2(
+ void* side, void* uplo,
+ void* transa, void* diag,
+ int m, int n,
+ const void *alpha,
+ const void *A, int lda,
+ void *B, int ldb,
+ void* stream);
+int CUDA_ztrsm(
+ void* side, void* uplo,
+ void* transa, void* diag,
+ int m, int n,
+ void *alpha,
+ const void *A, int lda,
+ void *B, int ldb,
+ void* stream);
+int CUDA_ztrtri(
+ void* uplo, void* diag, void* n,
+ void *dA, void* ldda, void* info);
+int CUDA_ztslqt(
+ void* m, void* n, void* nb,
+ void *da1, void* ldda1,
+ void *da2, void* ldda2,
+ void *a2, void* lda2,
+ void *dt, void* lddt,
+ void *t, void* ldt,
+ void *dd,
+ void *d, void* ldd,
+ void *tau,
+ void *hwork,
+ void *dwork,
+ void* stream);
+int CUDA_ztsmlq(
+ void* side, void* trans,
+ void* M1, void* N1,
+ void* M2, void* N2,
+ void* K, void* IB,
+ void *A1, void* LDA1,
+ void *A2, void* LDA2,
+ void *V, void* LDV,
+ void *T, void* LDT,
+ void *WORK, void* LDWORK,
+ void *WORKC, void* LDWORKC,
+ void* stream);
+int CUDA_ztsmqr(
+ void* side, void* trans,
+ void* M1, void* N1,
+ void* M2, void* N2,
+ void* K, void* IB,
+ void *A1, void* LDA1,
+ void *A2, void* LDA2,
+ void *V, void* LDV,
+ void *T, void* LDT,
+ void *WORK, void* LDWORK,
+ void *WORKC, void* LDWORKC,
+ void* stream);
+int CUDA_ztsqrt(
+ void* m, void* n, void* nb,
+ void *da1, void* ldda1,
+ void *da2, void* ldda2,
+ void *a2, void* lda2,
+ void *dt, void* lddt,
+ void *t, void* ldt,
+ void *dd,
+ void *d, void* ldd,
+ void *tau,
+ void *hwork,
+ void *dwork,
+ void* stream);
+int CUDA_ztstrf(
+ char storev, void* m, void* n, void* ib, void* nb,
+ void *hU, void* ldhu, void *dU, void* lddu,
+ void *hA, void* ldha, void *dA, void* ldda,
+ void *hL, void* ldhl, void *dL, void* lddl,
+ void* *ipiv,
+ void *hwork, void* ldhwork,
+ void *dwork, void* lddwork,
+ void* info);
+int CUDA_zunmlqt(
+ void* side, void* trans,
+ void* M, void* N, void* K, void* IB,
+ void *A, void* LDA,
+ void *T, void* LDT,
+ void *C, void* LDC,
+ void *WORK, void* LDWORK );
+int CUDA_zunmqrt(
+ void* side, void* trans,
+ void* M, void* N, void* K, void* IB,
+ void *A, void* LDA,
+ void *T, void* LDT,
+ void *C, void* LDC,
+ void *WORK, void* LDWORK );
+
+END_MODULE
--
GitLab