X-Git-Url: https://git.sesse.net/?p=stockfish;a=blobdiff_plain;f=src%2FMakefile;h=c00b60b5a56abfc6737689a1e2c2604719853f5a;hp=4741e722bfdd125d5fbdd260d31558767703577f;hb=ea6220f3813e5b76b444a02905eaf2c556bdb368;hpb=84f3e867903f62480c33243dd0ecbffd342796fc diff --git a/src/Makefile b/src/Makefile index 4741e722..c00b60b5 100644 --- a/src/Makefile +++ b/src/Makefile @@ -68,10 +68,8 @@ endif # prefetch = yes/no --- -DUSE_PREFETCH --- Use prefetch asm-instruction # popcnt = yes/no --- -DUSE_POPCNT --- Use popcnt asm-instruction # sse = yes/no --- -msse --- Use Intel Streaming SIMD Extensions -# sse3 = yes/no --- -msse3 --- Use Intel Streaming SIMD Extensions 3 # ssse3 = yes/no --- -mssse3 --- Use Intel Supplemental Streaming SIMD Extensions 3 # sse41 = yes/no --- -msse4.1 --- Use Intel Streaming SIMD Extensions 4.1 -# sse42 = yes/no --- -msse4.2 --- Use Intel Streaming SIMD Extensions 4.2 # avx2 = yes/no --- -mavx2 --- Use Intel Advanced Vector Extensions 2 # pext = yes/no --- -DUSE_PEXT --- Use pext x86_64 asm-instruction # avx512 = yes/no --- -mavx512bw --- Use Intel Advanced Vector Extensions 512 @@ -88,11 +86,10 @@ sanitize = no bits = 64 prefetch = no popcnt = no +mmx = no sse = no -sse3 = no ssse3 = no sse41 = no -sse42 = no avx2 = no pext = no avx512 = no @@ -114,6 +111,7 @@ ifeq ($(ARCH),x86-32) arch = i386 bits = 32 prefetch = yes + mmx = yes sse = yes endif @@ -127,18 +125,10 @@ ifeq ($(ARCH),x86-64) sse = yes endif -ifeq ($(ARCH),x86-64-sse3) - arch = x86_64 - prefetch = yes - sse = yes - sse3 = yes -endif - ifeq ($(ARCH),x86-64-sse3-popcnt) arch = x86_64 prefetch = yes sse = yes - sse3 = yes popcnt = yes endif @@ -146,39 +136,25 @@ ifeq ($(ARCH),x86-64-ssse3) arch = x86_64 prefetch = yes sse = yes - sse3 = yes ssse3 = yes endif -ifeq ($(ARCH),x86-64-sse41) - arch = x86_64 - prefetch = yes - popcnt = yes - sse = yes - sse3 = yes - ssse3 = yes - sse41 = yes -endif - ifeq ($(ARCH),x86-64-modern) arch = x86_64 prefetch = yes popcnt = yes sse = yes - sse3 = yes ssse3 = yes sse41 = yes endif -ifeq ($(ARCH),x86-64-sse42) +ifeq ($(ARCH),x86-64-sse41-popcnt) arch = x86_64 prefetch = yes popcnt = yes sse = yes - sse3 = yes ssse3 = yes sse41 = yes - sse42 = yes endif ifeq ($(ARCH),x86-64-avx2) @@ -186,10 +162,8 @@ ifeq ($(ARCH),x86-64-avx2) prefetch = yes popcnt = yes sse = yes - sse3 = yes ssse3 = yes sse41 = yes - sse42 = yes avx2 = yes endif @@ -198,10 +172,8 @@ ifeq ($(ARCH),x86-64-bmi2) prefetch = yes popcnt = yes sse = yes - sse3 = yes ssse3 = yes sse41 = yes - sse42 = yes avx2 = yes pext = yes endif @@ -211,10 +183,8 @@ ifeq ($(ARCH),x86-64-avx512) prefetch = yes popcnt = yes sse = yes - sse3 = yes ssse3 = yes sse41 = yes - sse42 = yes avx2 = yes pext = yes avx512 = yes @@ -282,6 +252,9 @@ ifeq ($(COMP),gcc) ifneq ($(KERNEL),Darwin) LDFLAGS += -Wl,--no-as-needed endif + + gccversion = $(shell $(CXX) --version) + gccisclang = $(findstring clang,$(gccversion)) endif ifeq ($(COMP),mingw) @@ -351,8 +324,8 @@ endif endif ifeq ($(KERNEL),Darwin) - CXXFLAGS += -arch $(arch) -mmacosx-version-min=10.15 - LDFLAGS += -arch $(arch) -mmacosx-version-min=10.15 + CXXFLAGS += -arch $(arch) -mmacosx-version-min=10.14 + LDFLAGS += -arch $(arch) -mmacosx-version-min=10.14 endif ### Travis CI script uses COMPILER to overwrite CXX @@ -443,14 +416,7 @@ endif ifeq ($(avx512),yes) CXXFLAGS += -DUSE_AVX512 ifeq ($(comp),$(filter $(comp),gcc clang mingw)) - CXXFLAGS += -mavx512bw - endif -endif - -ifeq ($(sse42),yes) - CXXFLAGS += -DUSE_SSE42 - ifeq ($(comp),$(filter $(comp),gcc clang mingw)) - CXXFLAGS += -msse4.2 + CXXFLAGS += -mavx512f -mavx512bw endif endif @@ -468,10 +434,10 @@ ifeq ($(ssse3),yes) endif endif -ifeq ($(sse3),yes) - CXXFLAGS += -DUSE_SSE3 +ifeq ($(mmx),yes) + CXXFLAGS += -DUSE_MMX ifeq ($(comp),$(filter $(comp),gcc clang mingw)) - CXXFLAGS += -msse3 + CXXFLAGS += -mmmx endif endif @@ -496,18 +462,33 @@ endif ### needs access to the optimization flags. ifeq ($(optimize),yes) ifeq ($(debug), no) - ifeq ($(comp),$(filter $(comp),gcc clang)) + ifeq ($(comp),clang) + CXXFLAGS += -flto=thin + LDFLAGS += $(CXXFLAGS) + +# GCC and CLANG use different methods for parallelizing LTO and CLANG pretends to be +# GCC on some systems. + else ifeq ($(comp),gcc) + ifeq ($(gccisclang),) CXXFLAGS += -flto + LDFLAGS += $(CXXFLAGS) -flto=jobserver + ifneq ($(findstring MINGW,$(KERNEL)),) + LDFLAGS += -save-temps + else ifneq ($(findstring MSYS,$(KERNEL)),) + LDFLAGS += -save-temps + endif + else + CXXFLAGS += -flto=thin LDFLAGS += $(CXXFLAGS) endif # To use LTO and static linking on windows, the tool chain requires a recent gcc: # gcc version 10.1 in msys2 or TDM-GCC version 9.2 are know to work, older might not. # So, only enable it for a cross from Linux by default. - ifeq ($(comp),mingw) + else ifeq ($(comp),mingw) ifeq ($(KERNEL),Linux) CXXFLAGS += -flto - LDFLAGS += $(CXXFLAGS) + LDFLAGS += $(CXXFLAGS) -flto=jobserver endif endif endif @@ -544,14 +525,12 @@ help: @echo "x86-64-avx512 > x86 64-bit with avx512 support" @echo "x86-64-bmi2 > x86 64-bit with bmi2 support" @echo "x86-64-avx2 > x86 64-bit with avx2 support" - @echo "x86-64-sse42 > x86 64-bit with sse42 support" - @echo "x86-64-modern > x86 64-bit with sse41 support (x86-64-sse41)" - @echo "x86-64-sse41 > x86 64-bit with sse41 support" + @echo "x86-64-sse41-popcnt > x86 64-bit with sse41 and popcnt support" + @echo "x86-64-modern > the same as previous (x86-64-sse41-popcnt)" @echo "x86-64-ssse3 > x86 64-bit with ssse3 support" @echo "x86-64-sse3-popcnt > x86 64-bit with sse3 and popcnt support" - @echo "x86-64-sse3 > x86 64-bit with sse3 support" @echo "x86-64 > x86 64-bit generic" - @echo "x86-32 > x86 32-bit (also enables SSE)" + @echo "x86-32 > x86 32-bit (also enables MMX and SSE)" @echo "x86-32-old > x86 32-bit fall back for old hardware" @echo "ppc-64 > PPC 64-bit" @echo "ppc-32 > PPC 32-bit" @@ -590,7 +569,7 @@ help: build: config-sanity $(MAKE) ARCH=$(ARCH) COMP=$(COMP) all -profile-build: config-sanity objclean profileclean +profile-build: config-sanity objclean profileclean net @echo "" @echo "Step 1/4. Building instrumented executable ..." $(MAKE) ARCH=$(ARCH) COMP=$(COMP) $(profile_make) @@ -631,7 +610,7 @@ objclean: # clean auxiliary profiling files profileclean: @rm -rf profdir - @rm -f bench.txt *.gcda *.gcno ./syzygy/*.gcda ./nnue/*.gcda ./nnue/features/*.gcda + @rm -f bench.txt *.gcda *.gcno ./syzygy/*.gcda ./nnue/*.gcda ./nnue/features/*.gcda *.s @rm -f stockfish.profdata *.profraw default: @@ -656,10 +635,8 @@ config-sanity: @echo "prefetch: '$(prefetch)'" @echo "popcnt: '$(popcnt)'" @echo "sse: '$(sse)'" - @echo "sse3: '$(sse3)'" @echo "ssse3: '$(ssse3)'" @echo "sse41: '$(sse41)'" - @echo "sse42: '$(sse42)'" @echo "avx2: '$(avx2)'" @echo "pext: '$(pext)'" @echo "avx512: '$(avx512)'" @@ -682,10 +659,8 @@ config-sanity: @test "$(prefetch)" = "yes" || test "$(prefetch)" = "no" @test "$(popcnt)" = "yes" || test "$(popcnt)" = "no" @test "$(sse)" = "yes" || test "$(sse)" = "no" - @test "$(sse3)" = "yes" || test "$(sse3)" = "no" @test "$(ssse3)" = "yes" || test "$(ssse3)" = "no" @test "$(sse41)" = "yes" || test "$(sse41)" = "no" - @test "$(sse42)" = "yes" || test "$(sse42)" = "no" @test "$(avx2)" = "yes" || test "$(avx2)" = "no" @test "$(pext)" = "yes" || test "$(pext)" = "no" @test "$(avx512)" = "yes" || test "$(avx512)" = "no" @@ -693,7 +668,7 @@ config-sanity: @test "$(comp)" = "gcc" || test "$(comp)" = "icc" || test "$(comp)" = "mingw" || test "$(comp)" = "clang" $(EXE): $(OBJS) - $(CXX) -o $@ $(OBJS) $(LDFLAGS) + +$(CXX) -o $@ $(OBJS) $(LDFLAGS) clang-profile-make: $(MAKE) ARCH=$(ARCH) COMP=$(COMP) \