Use new faster auto lowpp class implementation

Author: mdboom

cuda_bindings/cuda/bindings/_internal/cufile_linux.pyx

@@ -111,12 +111,15 @@ cdef void* load_library() except* with gil:
     return <void*>handle
 
 
-cdef int __check_or_init_cufile() except -1 nogil:
+cdef int _init_cufile() except -1 nogil:
     global __py_cufile_init
 
     cdef void* handle = NULL
 
     with gil, __symbol_lock:
+        # Recheck the flag after obtaining the locks
+        if __py_cufile_init:
+            return 0
         # Load function
         global __cuFileHandleRegister
         __cuFileHandleRegister = dlsym(RTLD_DEFAULT, 'cuFileHandleRegister')
@@ -427,7 +430,7 @@ cdef inline int _check_or_init_cufile() except -1 nogil:
     if __py_cufile_init:
         return 0
 
-    return __check_or_init_cufile()
+    return _init_cufile()
 
 
 cdef dict func_ptrs = None

cuda_bindings/cuda/bindings/_internal/nvjitlink_linux.pyx

@@ -80,12 +80,16 @@ cdef void* load_library() except* with gil:
     return <void*>handle
 
 
-cdef int __check_or_init_nvjitlink() except -1 nogil:
+cdef int _init_nvjitlink() except -1 nogil:
     global __py_nvjitlink_init
 
     cdef void* handle = NULL
 
     with gil, __symbol_lock:
+        # Recheck the flag after obtaining the locks
+        if __py_nvjitlink_init:
+            return 0
+
         # Load function
         global __nvJitLinkCreate
         __nvJitLinkCreate = dlsym(RTLD_DEFAULT, 'nvJitLinkCreate')
@@ -193,7 +197,7 @@ cdef inline int _check_or_init_nvjitlink() except -1 nogil:
     if __py_nvjitlink_init:
         return 0
 
-    return __check_or_init_nvjitlink()
+    return _init_nvjitlink()
 
 cdef dict func_ptrs = None
 

cuda_bindings/cuda/bindings/_internal/nvjitlink_windows.pyx

@@ -93,10 +93,14 @@ cdef void* __nvJitLinkGetInfoLog = NULL
 cdef void* __nvJitLinkVersion = NULL
 
 
-cdef int __check_or_init_nvjitlink() except -1 nogil:
+cdef int _init_nvjitlink() except -1 nogil:
     global __py_nvjitlink_init
 
     with gil, __symbol_lock:
+        # Recheck the flag after obtaining the locks
+        if __py_nvjitlink_init:
+            return 0
+
         # Load library
         handle = load_nvidia_dynamic_lib("nvJitLink")._handle_uint
 
@@ -151,7 +155,7 @@ cdef inline int _check_or_init_nvjitlink() except -1 nogil:
     if __py_nvjitlink_init:
         return 0
 
-    return __check_or_init_nvjitlink()
+    return _init_nvjitlink()
 
 
 cdef dict func_ptrs = None

cuda_bindings/cuda/bindings/_internal/nvvm_linux.pyx

@@ -79,12 +79,16 @@ cdef void* load_library() except* with gil:
     return <void*>handle
 
 
-cdef int __check_or_init_nvvm() except -1 nogil:
+cdef int _init_nvvm() except -1 nogil:
     global __py_nvvm_init
 
     cdef void* handle = NULL
 
     with gil, __symbol_lock:
+        # Recheck the flag after obtaining the locks
+        if __py_nvvm_init:
+            return 0
+
         # Load function
         global __nvvmGetErrorString
         __nvvmGetErrorString = dlsym(RTLD_DEFAULT, 'nvvmGetErrorString')
@@ -185,7 +189,7 @@ cdef inline int _check_or_init_nvvm() except -1 nogil:
     if __py_nvvm_init:
         return 0
 
-    return __check_or_init_nvvm()
+    return _init_nvvm()
 
 
 cdef dict func_ptrs = None

cuda_bindings/cuda/bindings/_internal/nvvm_windows.pyx

@@ -92,10 +92,14 @@ cdef void* __nvvmGetProgramLogSize = NULL
 cdef void* __nvvmGetProgramLog = NULL
 
 
-cdef int __check_or_init_nvvm() except -1 nogil:
+cdef int _init_nvvm() except -1 nogil:
     global __py_nvvm_init
 
     with gil, __symbol_lock:
+        # Recheck the flag after obtaining the locks
+        if __py_nvvm_init:
+            return 0
+
         # Load library
         handle = load_nvidia_dynamic_lib("nvvm")._handle_uint
 
@@ -147,7 +151,7 @@ cdef inline int _check_or_init_nvvm() except -1 nogil:
     if __py_nvvm_init:
         return 0
 
-    return __check_or_init_nvvm()
+    return _init_nvvm()
 
 
 cdef dict func_ptrs = None