BLD: `utest` port

interface/meson.build

@@ -1,7 +1,7 @@
 interface_args = ['-include', f'@bldroot@/simd_conf.h']
-if not build_without_cblas
+# if not build_without_cblas
-  interface_args += '-DCBLAS'
+#   interface_args += '-DCBLAS'
-endif
+# endif
 _blas_roots = [
   # NOTE: q, qx, x, xq do not have cblas_ rules in the Makefile
   # NOTE: https://developer.arm.com/documentation/101004/2310/BLAS-Basic-Linear-Algebra-Subprograms/CBLAS-functions?lang=en

kernel/meson.build

@@ -56,12 +56,18 @@ base_kops = [
     'modes': {
         's': {'exts': {'_k': {'dir': 'x86_64', 'kernel': 'saxpy.c'}}},
         'd': {'exts': {'_k': {'dir': 'x86_64', 'kernel': 'daxpy.c'}}},
-        'c': {'exts': {'_k': {'dir': 'x86_64', 'kernel': 'caxpy.c'}}},
+        'c': {'exts': {'_k': {'dir': 'x86_64', 'kernel': 'caxpy.c', 'addl': ['-UCONJ']}}},
-        'z': {'exts': {'_k': {'dir': 'x86_64', 'kernel': 'zaxpy.c'}}},
+        'z': {'exts': {'_k': {'dir': 'x86_64', 'kernel': 'zaxpy.c', 'addl': ['-UCONJ']}}},
         # 'q': {'exts': {'_k': {'dir': 'x86_64', 'kernel': 'axpy.S'}}},
         # 'x': {'exts': {'_k': {'dir': 'x86_64', 'kernel': 'zaxpy.S'}}},
     },
   },
+  { 'base': '?axpyc',
+    'modes': {
+        'c': {'exts': {'_k': {'dir': 'x86_64', 'kernel': 'caxpy.c', 'addl': ['-DCONJ']}}},
+        'z': {'exts': {'_k': {'dir': 'x86_64', 'kernel': 'zaxpy.c', 'addl': ['-DCONJ']}}},
+    },
+  },
   { 'base': '?dot',
     'modes': {
         's': {'exts': {'_k': {'dir': 'x86_64', 'kernel': 'sdot.c'}}},
@@ -84,7 +90,7 @@ base_kops = [
   { 'base': '?dsdot',
     'modes': {
         's': {'exts': {'_k': {'dir': 'x86_64', 'kernel': 'sdot.c'}}},
-        'd': {'exts': {'_k': {'dir': 'x86_64', 'kernel': 'sdot.c'}}},
+        '': {'exts': {'_k': {'dir': 'x86_64', 'kernel': 'sdot.c', 'addl': ['-DDSDOT']}}},
     },
   },
   # TODO(rg): Add dsdotkernel conditionals
@@ -863,15 +869,82 @@ base_kops = [
   { 'base': '?omatcopy_k',
     'modes': {
         's': {'exts': {
-            '_cn': {'dir': 'arm', 'kernel': 'omatcopy_cn.c',
+            '_cn': {'dir': 'arm', 'kernel': 'omatcopy_cn.c', 'addl': ['-UROWM']},
-                          'addl': ['-UROWM']},
+            '_rn': {'dir': 'arm', 'kernel': 'omatcopy_rn.c', 'addl': ['-DROWM']},
-            '_rn': {'dir': 'arm', 'kernel': 'omatcopy_rn.c',
+            '_ct': {'dir': 'arm', 'kernel': 'omatcopy_ct.c', 'addl': ['-UROWM']},
-                          'addl': ['-DROWM']},
+            '_rt': {'dir': 'arm', 'kernel': 'omatcopy_rt.c', 'addl': ['-DROWM']},
-            '_ct': {'dir': 'arm', 'kernel': 'omatcopy_ct.c',
-                          'addl': ['-UROWM']},
-            '_rt': {'dir': 'x86_64', 'kernel': 'omatcopy_rt.c',
-                          'addl': ['-DROWM']},
         }},
+        'd': {'exts': {
+            '_cn': {'dir': 'arm', 'kernel': 'omatcopy_cn.c', 'addl': ['-UROWM']},
+            '_rn': {'dir': 'arm', 'kernel': 'omatcopy_rn.c', 'addl': ['-DROWM']},
+            '_ct': {'dir': 'arm', 'kernel': 'omatcopy_ct.c', 'addl': ['-UROWM']},
+            '_rt': {'dir': 'arm', 'kernel': 'omatcopy_rt.c', 'addl': ['-DROWM']},
+        }},
+        'c': {'exts': {
+            '_cn': {'dir': 'arm', 'kernel': 'zomatcopy_cn.c', 'addl': ['-UROWM', '-UCONJ']},
+            '_rn': {'dir': 'arm', 'kernel': 'zomatcopy_rn.c', 'addl': ['-DROWM', '-UCONJ']},
+            '_ct': {'dir': 'arm', 'kernel': 'zomatcopy_ct.c', 'addl': ['-UROWM', '-UCONJ']},
+            '_rt': {'dir': 'arm', 'kernel': 'zomatcopy_rt.c', 'addl': ['-DROWM', '-UCONJ']},
+            '_cnc': {'dir': 'arm', 'kernel': 'zomatcopy_cnc.c', 'addl': ['-UROWM', '-DCONJ']},
+            '_rnc': {'dir': 'arm', 'kernel': 'zomatcopy_rnc.c', 'addl': ['-DROWM', '-DCONJ']},
+            '_ctc': {'dir': 'arm', 'kernel': 'zomatcopy_ctc.c', 'addl': ['-UROWM', '-DCONJ']},
+            '_rtc': {'dir': 'arm', 'kernel': 'zomatcopy_rtc.c', 'addl': ['-DROWM', '-DCONJ']},
+        }},
+        'z': {'exts': {
+            '_cn': {'dir': 'arm', 'kernel': 'zomatcopy_cn.c', 'addl': ['-UROWM', '-UCONJ']},
+            '_rn': {'dir': 'arm', 'kernel': 'zomatcopy_rn.c', 'addl': ['-DROWM', '-UCONJ']},
+            '_ct': {'dir': 'arm', 'kernel': 'zomatcopy_ct.c', 'addl': ['-UROWM', '-UCONJ']},
+            '_rt': {'dir': 'arm', 'kernel': 'zomatcopy_rt.c', 'addl': ['-DROWM', '-UCONJ']},
+            '_cnc': {'dir': 'arm', 'kernel': 'zomatcopy_cnc.c', 'addl': ['-UROWM', '-DCONJ']},
+            '_rnc': {'dir': 'arm', 'kernel': 'zomatcopy_rnc.c', 'addl': ['-DROWM', '-DCONJ']},
+            '_ctc': {'dir': 'arm', 'kernel': 'zomatcopy_ctc.c', 'addl': ['-UROWM', '-DCONJ']},
+            '_rtc': {'dir': 'arm', 'kernel': 'zomatcopy_rtc.c', 'addl': ['-DROWM', '-DCONJ']},
+        }},
+    },
+  },
+  { 'base': '?imatcopy_k',
+    'modes': {
+        's': {'exts': {
+            '_cn': {'dir': 'generic', 'kernel': 'imatcopy_cn.c', 'addl': ['-UROWM']},
+            '_rn': {'dir': 'generic', 'kernel': 'imatcopy_rn.c', 'addl': ['-DROWM']},
+            '_ct': {'dir': 'generic', 'kernel': 'imatcopy_ct.c', 'addl': ['-UROWM']},
+            '_rt': {'dir': 'generic', 'kernel': 'imatcopy_rt.c', 'addl': ['-DROWM']},
+        }},
+        'd': {'exts': {
+            '_cn': {'dir': 'generic', 'kernel': 'imatcopy_cn.c', 'addl': ['-UROWM']},
+            '_rn': {'dir': 'generic', 'kernel': 'imatcopy_rn.c', 'addl': ['-DROWM']},
+            '_ct': {'dir': 'generic', 'kernel': 'imatcopy_ct.c', 'addl': ['-UROWM']},
+            '_rt': {'dir': 'generic', 'kernel': 'imatcopy_rt.c', 'addl': ['-DROWM']},
+        }},
+        'c': {'exts': {
+            '_cn': {'dir': 'generic', 'kernel': 'zimatcopy_cn.c', 'addl': ['-UROWM', '-UCONJ']},
+            '_rn': {'dir': 'generic', 'kernel': 'zimatcopy_rn.c', 'addl': ['-DROWM', '-UCONJ']},
+            '_ct': {'dir': 'generic', 'kernel': 'zimatcopy_ct.c', 'addl': ['-UROWM', '-UCONJ']},
+            '_rt': {'dir': 'generic', 'kernel': 'zimatcopy_rt.c', 'addl': ['-DROWM', '-UCONJ']},
+            '_cnc': {'dir': 'generic', 'kernel': 'zimatcopy_cnc.c', 'addl': ['-UROWM', '-DCONJ']},
+            '_rnc': {'dir': 'generic', 'kernel': 'zimatcopy_rnc.c', 'addl': ['-DROWM', '-DCONJ']},
+            '_ctc': {'dir': 'generic', 'kernel': 'zimatcopy_ctc.c', 'addl': ['-UROWM', '-DCONJ']},
+            '_rtc': {'dir': 'generic', 'kernel': 'zimatcopy_rtc.c', 'addl': ['-DROWM', '-DCONJ']},
+        }},
+        'z': {'exts': {
+            '_cn': {'dir': 'generic', 'kernel': 'zimatcopy_cn.c', 'addl': ['-UROWM', '-UCONJ']},
+            '_rn': {'dir': 'generic', 'kernel': 'zimatcopy_rn.c', 'addl': ['-DROWM', '-UCONJ']},
+            '_ct': {'dir': 'generic', 'kernel': 'zimatcopy_ct.c', 'addl': ['-UROWM', '-UCONJ']},
+            '_rt': {'dir': 'generic', 'kernel': 'zimatcopy_rt.c', 'addl': ['-DROWM', '-UCONJ']},
+            '_cnc': {'dir': 'generic', 'kernel': 'zimatcopy_cnc.c', 'addl': ['-UROWM', '-DCONJ']},
+            '_rnc': {'dir': 'generic', 'kernel': 'zimatcopy_rnc.c', 'addl': ['-DROWM', '-DCONJ']},
+            '_ctc': {'dir': 'generic', 'kernel': 'zimatcopy_ctc.c', 'addl': ['-UROWM', '-DCONJ']},
+            '_rtc': {'dir': 'generic', 'kernel': 'zimatcopy_rtc.c', 'addl': ['-DROWM', '-DCONJ']},
+        }},
+    },
+  },
+  {
+    'base': '?geadd',
+    'modes': {
+        's': {'exts': {'_k': {'dir': 'generic', 'kernel': 'geadd.c', 'addl': ['-UROWM']}}},
+        'd': {'exts': {'_k': {'dir': 'generic', 'kernel': 'geadd.c', 'addl': ['-UROWM']}}},
+        'c': {'exts': {'_k': {'dir': 'generic', 'kernel': 'zgeadd.c', 'addl': ['-UROWM']}}},
+        'z': {'exts': {'_k': {'dir': 'generic', 'kernel': 'zgeadd.c', 'addl': ['-UROWM']}}},
     },
   },
   { 'base': '?gemm_small_kernel',

meson.build

@@ -550,3 +550,7 @@ pkg.generate(_openblas,
              description: 'OpenBLAS via meson build',
              version: f'@pkg_ver@_meson',
             )
+
+if build_testing
+  subdir('utest')
+endif

utest/meson.build

@@ -0,0 +1,106 @@
+sources_utest = [
+    config_h,
+    'utest_main.c',
+    'test_min.c',
+    'test_amax.c',
+    'test_ismin.c',
+    'test_rotmg.c',
+    'test_axpy.c',
+    'test_dotu.c',
+    'test_dsdot.c',
+    'test_swap.c',
+    'test_rot.c',
+    'test_dnrm2.c',
+    'test_zscal.c',
+    'test_amin.c',
+    'test_axpby.c',
+]
+
+dir_ext = 'test_extensions'
+sources_utest_ext = [
+    config_h,
+    'utest_main.c',
+    dir_ext / 'xerbla.c',
+    dir_ext / 'common.c',
+    dir_ext / 'test_isamin.c',
+    dir_ext / 'test_idamin.c',
+    dir_ext / 'test_icamin.c',
+    dir_ext / 'test_izamin.c',
+    dir_ext / 'test_ssum.c',
+    dir_ext / 'test_dsum.c',
+    dir_ext / 'test_scsum.c',
+    dir_ext / 'test_dzsum.c',
+    dir_ext / 'test_saxpby.c',
+    dir_ext / 'test_daxpby.c',
+    dir_ext / 'test_caxpby.c',
+    dir_ext / 'test_zaxpby.c',
+    dir_ext / 'test_zaxpyc.c',
+    dir_ext / 'test_caxpyc.c',
+    dir_ext / 'test_samin.c',
+    dir_ext / 'test_damin.c',
+    dir_ext / 'test_scamin.c',
+    dir_ext / 'test_dzamin.c',
+    dir_ext / 'test_scamax.c',
+    dir_ext / 'test_dzamax.c',
+    dir_ext / 'test_drotmg.c',
+    dir_ext / 'test_srotmg.c',
+    dir_ext / 'test_zrotg.c',
+    dir_ext / 'test_crotg.c',
+    dir_ext / 'test_crot.c',
+    dir_ext / 'test_zrot.c',
+    dir_ext / 'test_zscal.c',
+    dir_ext / 'test_cscal.c',
+    dir_ext / 'test_domatcopy.c',
+    dir_ext / 'test_somatcopy.c',
+    dir_ext / 'test_zomatcopy.c',
+    dir_ext / 'test_comatcopy.c',
+    dir_ext / 'test_simatcopy.c',
+    dir_ext / 'test_dimatcopy.c',
+    dir_ext / 'test_cimatcopy.c',
+    dir_ext / 'test_zimatcopy.c',
+    dir_ext / 'test_sgeadd.c',
+    dir_ext / 'test_dgeadd.c',
+    dir_ext / 'test_cgeadd.c',
+    dir_ext / 'test_zgeadd.c',
+    dir_ext / 'test_cgemv_t.c',
+    dir_ext / 'test_zgemv_t.c',
+    dir_ext / 'test_cgemv_n.c',
+    dir_ext / 'test_zgemv_n.c',
+    dir_ext / 'test_sgemmt.c',
+    dir_ext / 'test_dgemmt.c',
+    dir_ext / 'test_cgemmt.c',
+    dir_ext / 'test_zgemmt.c',
+    dir_ext / 'test_ztrmv.c',
+    dir_ext / 'test_ctrmv.c',
+    dir_ext / 'test_ztrsv.c',
+    dir_ext / 'test_ctrsv.c',
+    dir_ext / 'test_zgemm.c',
+    dir_ext / 'test_cgemm.c',
+    dir_ext / 'test_zgbmv.c',
+    dir_ext / 'test_cgbmv.c',
+    dir_ext / 'test_zspmv.c',
+    dir_ext / 'test_cspmv.c',
+    dir_ext / 'test_zsbmv.c',
+    dir_ext / 'test_csbmv.c',
+]
+
+openblas_utest_exec = executable(
+    'openblas_utest',
+    sources: sources_utest,
+    link_with: _openblas,
+    dependencies: [dependency('threads')],
+    include_directories: ['..', '.'],
+    c_args: _cargs,
+)
+
+openblas_utest_ext_exec = executable(
+    'openblas_utest_ext',
+    sources: sources_utest_ext,
+    link_with: _openblas,
+    dependencies: [dependency('threads')],
+    include_directories: ['..', '.'],
+    c_args: _cargs,
+)
+
+test('openblas_utest_tests', openblas_utest_exec)
+test('openblas_utest_tests', openblas_utest_ext_exec)