Merge pull request #47 from falseywinchnet/d4ptfr-main/optimize-bfft-performance-in-benchmark

Build release binaries without debug asserts

Author: falseywinchnet

Makefile

@@ -16,7 +16,7 @@ CXXWARNFLAGS ?= -Wall -Wextra -Wpedantic
 COPTFLAGS ?= -O2
 CWARNFLAGS ?= -Wall -Wextra -Wpedantic
 CPPFLAGS ?=
-CXXFLAGS ?= $(CXXOPTFLAGS) -std=$(CXX_STD) -fPIC $(CXXWARNFLAGS)
+CXXFLAGS ?= $(CXXOPTFLAGS) -DNDEBUG -std=$(CXX_STD) -fPIC $(CXXWARNFLAGS)
 CFLAGS ?= $(COPTFLAGS) -std=c11 $(CWARNFLAGS)
 LDFLAGS ?=
 LDLIBS ?= -lm

setup.py

@@ -57,7 +57,7 @@ def _optimization_flags(cxx: str) -> list:
     BFFT_NO_NATIVE=1 to skip CPU-native codegen and BFFT_NO_FAST_MATH=1 to keep
     strict IEEE math.
     """
-    flags = ["-O3"]
+    flags = ["-O3", "-DNDEBUG"]
 
     if not _env_off("BFFT_NO_NATIVE"):
         # x86 / older clang / gcc use -march=native; Apple-silicon clang wants