Extract inline_threshold value from targetdescr before use.

    - Addresses review comments on how to properly set the
      inline_threshold target option

Author: Diptorup Deb

numba_dpex/core/descriptor.py

@@ -8,6 +8,8 @@
 from numba.core.cpu import CPUTargetOptions
 from numba.core.descriptors import TargetDescriptor
 
+from numba_dpex import config
+
 from .targets.dpjit_target import DPEX_TARGET_NAME, DpexTargetContext
 from .targets.kernel_target import (
     DPEX_KERNEL_TARGET_NAME,
@@ -50,7 +52,12 @@ def finalize(self, flags, options):
         _inherit_if_not_set(flags, options, "release_gil", False)
         _inherit_if_not_set(flags, options, "no_compile", True)
         _inherit_if_not_set(flags, options, "use_mlir", False)
-        _inherit_if_not_set(flags, options, "inline_threshold", 0)
+        if config.INLINE_THRESHOLD is not None:
+            _inherit_if_not_set(
+                flags, options, "inline_threshold", config.INLINE_THRESHOLD
+            )
+        else:
+            _inherit_if_not_set(flags, options, "inline_threshold", 0)
         _inherit_if_not_set(
             flags, options, "_compilation_mode", CompilationMode.KERNEL
         )

numba_dpex/experimental/kernel_dispatcher.py

@@ -11,7 +11,7 @@
 
 import numba.core.event as ev
 from numba.core import errors, sigutils, types
-from numba.core.compiler import CompileResult
+from numba.core.compiler import CompileResult, Flags
 from numba.core.compiler_lock import global_compiler_lock
 from numba.core.dispatcher import Dispatcher, _FunctionCompiler
 from numba.core.target_extension import dispatcher_registry, target_registry
@@ -84,14 +84,15 @@ def _compile_to_spirv(
         kernel_fn = kernel_targetctx.prepare_spir_kernel(
             kernel_func, kernel_fndesc.argtypes
         )
+        # Get the compiler flags that were passed through the target descriptor
+        flags = Flags()
+        self.targetdescr.options.parse_as_flags(flags, self.targetoptions)
+
         # If the inline_threshold option was set then set the property in the
         # kernel_library to force inlining ``overload`` calls into a kernel.
-        try:
-            kernel_library.inline_threshold = self.targetoptions[
-                "inline_threshold"
-            ]
-        except KeyError:
-            pass
+        inline_threshold = flags.inline_threshold  # pylint: disable=E1101
+        kernel_library.inline_threshold = inline_threshold
+
         # Call finalize on the LLVM module. Finalization will result in
         # all linking libraries getting linked together and final optimization
         # including inlining of functions if an inlining level is specified.
@@ -202,9 +203,6 @@ class KernelDispatcher(Dispatcher):
     an executable binary, the dispatcher compiles it to SPIR-V and then caches
     that SPIR-V bitcode.
 
-    FIXME: Fix issues identified by pylint with this class.
-    https://github.com/IntelPython/numba-dpex/issues/1196
-
     """
 
     targetdescr = dpex_exp_kernel_target

numba_dpex/tests/experimental/test_inline_threshold_config.py

@@ -0,0 +1,3 @@
+# SPDX-FileCopyrightText: 2023 Intel Corporation
+#
+# SPDX-License-Identifier: Apache-2.0