Feat/arm64 support

.github/workflows/ci.yaml

@@ -0,0 +1,68 @@
+name: CI
+
+on:
+  push:
+    branches:
+    - master
+  pull_request:
+
+jobs:
+  build-linux-x8664:
+    name: Linux x86_64
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        compiler: [gcc, clang]
+
+    steps:
+    - name: Checkout ksw2
+      uses: actions/checkout@v4
+
+    - name: Compile with ${{ matrix.compiler }}
+      run: |
+        make CC=${{ matrix.compiler }}
+        file ksw2-test | grep x86-64
+
+  build-linux-aarch64:
+    name: Linux aarch64
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        compiler: [gcc]
+
+    steps:
+    - name: Checkout ksw2
+      uses: actions/checkout@v4
+
+    - name: Compile with ${{ matrix.compiler }}
+      uses: uraimo/run-on-arch-action@v2
+      with:
+        arch: aarch64
+        distro: ubuntu22.04
+        githubToken: ${{ github.token }}
+        dockerRunArgs: |
+          --volume "${PWD}:/ksw2"
+        install: |
+          apt-get update -q -y
+          apt-get install -q -y make ${{ matrix.compiler }} zlib1g-dev file
+        run: |
+          cd /ksw2
+          make CC=${{ matrix.compiler }} arm_neon=1 aarch64=1 -j
+          file ksw2-test | grep aarch64
+
+  build-mac-arm64:
+    name: Mac ARM64
+    runs-on: macos-14
+    strategy:
+      matrix:
+        compiler: [clang]
+
+    steps:
+    - name: Checkout ksw2
+      uses: actions/checkout@v4
+
+    - name: Compile with ${{ matrix.compiler }}
+      run: |
+        make CC=${{ matrix.compiler }} arm_neon=1 aarch64=1 -j
+        file ksw2-test | grep arm64
+

Makefile

@@ -1,12 +1,36 @@
 CC=			gcc
 CFLAGS=		-g -Wall -Wextra -Wc++-compat -O2
-CPPFLAGS=	-DHAVE_KALLOC
+CPPFLAGS=	-DHAVE_KALLOC -DKSW_SSE2_ONLY -D__SSE2__
 INCLUDES=	-I.
 OBJS=		ksw2_gg.o ksw2_gg2.o ksw2_gg2_sse.o ksw2_extz.o ksw2_extz2_sse.o \
 			ksw2_extd.o ksw2_extd2_sse.o ksw2_extf2_sse.o ksw2_exts2_sse.o
 PROG=		ksw2-test
 LIBS=		-lz
 
+ifneq ($(aarch64),)
+    arm_neon=1
+endif
+
+ifneq ($(arm_neon),) # if arm_neon is not defined
+    INCLUDES+=-Isse2neon
+ifeq ($(aarch64),)  #if aarch64 is not defined
+    CFLAGS+=-D_FILE_OFFSET_BITS=64 -mfpu=neon -fsigned-char
+else                #if aarch64 is defined
+    CFLAGS+=-D_FILE_OFFSET_BITS=64 -fsigned-char
+endif
+endif
+
+ifneq ($(asan),)
+    CFLAGS+=-fsanitize=address
+    LIBS+=-fsanitize=address -ldl
+endif
+
+ifneq ($(tsan),)
+    CFLAGS+=-fsanitize=thread
+    LIBS+=-fsanitize=thread -ldl
+endif
+
+
 ifneq ($(gaba),) # gaba source code directory
 	CPPFLAGS += -DHAVE_GABA
 	INCLUDES += -I$(gaba)

README.md

@@ -46,7 +46,14 @@ documentations. TeX file [ksw2.tex](tex/ksw2.tex) gives brief derivation.
 
 To compile the test program `ksw-test`, just type `make`. It takes the
 advantage of SSE4.1 when available. To compile with SSE2 only, use `make
-sse2=1` instead. If you have installed [parasail][para], use `make
+sse2=1` instead, which will make minimap2 slightly slower.
+
+`ksw2` also works with ARM CPUs supporting the NEON instruction sets. To
+compile for 32 bit ARM architectures (such as ARMv7), use `make arm_neon=1`. To
+compile for for 64 bit ARM architectures (such as ARMv8), use `make arm_neon=1
+aarch64=1`.
+
+If you have installed [parasail][para], use `make
 parasail=prefix`, where `prefix` points to the parasail install directory (e.g.
 `/usr/local`).