X-Git-Url: https://git.sesse.net/?a=blobdiff_plain;f=src%2FMakefile;h=3d6432fd96b033c9c6964f6b11b0768968f625ad;hb=b4ad3a3c4b68f9c8736f444aeb3364f833247fdc;hp=775c72c36e9fcf5ce558d346ee359510f1e16f1d;hpb=037ef3e18dc7f5455cc671995ae38d5b4d1fce4a;p=stockfish diff --git a/src/Makefile b/src/Makefile index 775c72c3..3d6432fd 100644 --- a/src/Makefile +++ b/src/Makefile @@ -69,32 +69,33 @@ VPATH = syzygy:nnue:nnue/features ### Section 2. High-level Configuration ### ========================================================================== # -# flag --- Comp switch --- Description +# flag --- Comp switch --- Description # ---------------------------------------------------------------------------- # -# debug = yes/no --- -DNDEBUG --- Enable/Disable debug mode +# debug = yes/no --- -DNDEBUG --- Enable/Disable debug mode # sanitize = none/ ... (-fsanitize ) -# --- ( undefined ) --- enable undefined behavior checks -# --- ( thread ) --- enable threading error checks -# --- ( address ) --- enable memory access checks -# --- ...etc... --- see compiler documentation for supported sanitizers -# optimize = yes/no --- (-O3/-fast etc.) --- Enable/Disable optimizations -# arch = (name) --- (-arch) --- Target architecture -# bits = 64/32 --- -DIS_64BIT --- 64-/32-bit operating system -# prefetch = yes/no --- -DUSE_PREFETCH --- Use prefetch asm-instruction -# popcnt = yes/no --- -DUSE_POPCNT --- Use popcnt asm-instruction -# pext = yes/no --- -DUSE_PEXT --- Use pext x86_64 asm-instruction -# sse = yes/no --- -msse --- Use Intel Streaming SIMD Extensions -# mmx = yes/no --- -mmmx --- Use Intel MMX instructions -# sse2 = yes/no --- -msse2 --- Use Intel Streaming SIMD Extensions 2 -# ssse3 = yes/no --- -mssse3 --- Use Intel Supplemental Streaming SIMD Extensions 3 -# sse41 = yes/no --- -msse4.1 --- Use Intel Streaming SIMD Extensions 4.1 -# avx2 = yes/no --- -mavx2 --- Use Intel Advanced Vector Extensions 2 -# avxvnni = yes/no --- -mavxvnni --- Use Intel Vector Neural Network Instructions AVX -# avx512 = yes/no --- -mavx512bw --- Use Intel Advanced Vector Extensions 512 -# vnni256 = yes/no --- -mavx512vnni --- Use Intel Vector Neural Network Instructions 256 -# vnni512 = yes/no --- -mavx512vnni --- Use Intel Vector Neural Network Instructions 512 -# neon = yes/no --- -DUSE_NEON --- Use ARM SIMD architecture +# --- ( undefined ) --- enable undefined behavior checks +# --- ( thread ) --- enable threading error checks +# --- ( address ) --- enable memory access checks +# --- ...etc... --- see compiler documentation for supported sanitizers +# optimize = yes/no --- (-O3/-fast etc.) --- Enable/Disable optimizations +# arch = (name) --- (-arch) --- Target architecture +# bits = 64/32 --- -DIS_64BIT --- 64-/32-bit operating system +# prefetch = yes/no --- -DUSE_PREFETCH --- Use prefetch asm-instruction +# popcnt = yes/no --- -DUSE_POPCNT --- Use popcnt asm-instruction +# pext = yes/no --- -DUSE_PEXT --- Use pext x86_64 asm-instruction +# sse = yes/no --- -msse --- Use Intel Streaming SIMD Extensions +# mmx = yes/no --- -mmmx --- Use Intel MMX instructions +# sse2 = yes/no --- -msse2 --- Use Intel Streaming SIMD Extensions 2 +# ssse3 = yes/no --- -mssse3 --- Use Intel Supplemental Streaming SIMD Extensions 3 +# sse41 = yes/no --- -msse4.1 --- Use Intel Streaming SIMD Extensions 4.1 +# avx2 = yes/no --- -mavx2 --- Use Intel Advanced Vector Extensions 2 +# avxvnni = yes/no --- -mavxvnni --- Use Intel Vector Neural Network Instructions AVX +# avx512 = yes/no --- -mavx512bw --- Use Intel Advanced Vector Extensions 512 +# vnni256 = yes/no --- -mavx512vnni --- Use Intel Vector Neural Network Instructions 256 +# vnni512 = yes/no --- -mavx512vnni --- Use Intel Vector Neural Network Instructions 512 +# neon = yes/no --- -DUSE_NEON --- Use ARM SIMD architecture +# dotprod = yes/no --- -DUSE_NEON_DOTPROD --- Use ARM advanced SIMD Int8 dot product instructions # # Note that Makefile is space sensitive, so when adding new architectures # or modifying existing flags, you have to make sure there are no extra spaces @@ -116,7 +117,7 @@ ifeq ($(ARCH), $(filter $(ARCH), \ x86-64-vnni512 x86-64-vnni256 x86-64-avx512 x86-64-avxvnni x86-64-bmi2 \ x86-64-avx2 x86-64-sse41-popcnt x86-64-modern x86-64-ssse3 x86-64-sse3-popcnt \ x86-64 x86-32-sse41-popcnt x86-32-sse2 x86-32 ppc-64 ppc-32 e2k \ - armv7 armv7-neon armv8 apple-silicon general-64 general-32 riscv64)) + armv7 armv7-neon armv8 armv8-dotprod apple-silicon general-64 general-32 riscv64)) SUPPORTED_ARCH=true else SUPPORTED_ARCH=false @@ -140,6 +141,7 @@ avx512 = no vnni256 = no vnni512 = no neon = no +dotprod = no arm_version = 0 STRIP = strip @@ -308,11 +310,21 @@ ifeq ($(ARCH),armv8) arm_version = 8 endif +ifeq ($(ARCH),armv8-dotprod) + arch = armv8 + prefetch = yes + popcnt = yes + neon = yes + dotprod = yes + arm_version = 8 +endif + ifeq ($(ARCH),apple-silicon) arch = arm64 prefetch = yes popcnt = yes neon = yes + dotprod = yes arm_version = 8 endif @@ -675,6 +687,10 @@ ifeq ($(neon),yes) endif endif +ifeq ($(dotprod),yes) + CXXFLAGS += -march=armv8.2-a+dotprod -DUSE_NEON_DOTPROD +endif + ### 3.7 pext ifeq ($(pext),yes) CXXFLAGS += -DUSE_PEXT @@ -776,6 +792,7 @@ help: @echo "armv7 > ARMv7 32-bit" @echo "armv7-neon > ARMv7 32-bit with popcnt and neon" @echo "armv8 > ARMv8 64-bit with popcnt and neon" + @echo "armv8-dotprod > ARMv8 64-bit with popcnt, neon and dot product support" @echo "e2k > Elbrus 2000" @echo "apple-silicon > Apple silicon ARM64" @echo "general-64 > unspecified 64-bit"