Add optimized crc32 for POWER8 and later processors

This commit adds an optimized version of the crc32 function based
on crc32-vpmsum from https://github.com/antonblanchard/crc32-vpmsum/ .
The code has been relicensed to the zlib license.

This is the C implementation created by Rogerio Alves <rogealve@br.ibm.com>

It makes use of vector instructions to speed up CRC32 algorithm. Decompression
times were improved by +30% on tests.

Based on Daniel Black's work for the original zlib (madler/zlib#478).

CMakeLists.txt

@@ -619,6 +619,10 @@ if(WITH_OPTIM)
                 list(APPEND ZLIB_ARCH_HDRS ${ARCHDIR}/power.h)
                 list(APPEND ZLIB_ARCH_SRCS ${ARCHDIR}/power.c)
                 set(POWER8_SRCS ${ARCHDIR}/adler32_power8.c ${ARCHDIR}/chunkset_power8.c ${ARCHDIR}/slide_hash_power8.c)
+                if("${ARCH}" MATCHES "powerpc64(le)?")
+                    add_definitions(-DPOWER8_VSX_CRC32)
+                    list(APPEND POWER8_SRCS ${ARCHDIR}/crc32_power8.c)
+                endif()
                 list(APPEND ZLIB_ARCH_SRCS ${POWER8_SRCS})
                 set_property(SOURCE ${POWER8_SRCS} PROPERTY COMPILE_FLAGS "${POWER8FLAG} ${NOLTOFLAG}")
             else()

arch/power/Makefile.in

@@ -21,6 +21,8 @@ all: power.o \
      adler32_power8.lo \
      chunkset_power8.o \
      chunkset_power8.lo \
+     crc32_power8.o \
+     crc32_power8.lo \
      slide_hash_power8.o \
      slide_hash_power8.lo
 
@@ -42,6 +44,12 @@ chunkset_power8.o:
 chunkset_power8.lo:
 	$(CC) $(SFLAGS) $(P8FLAGS) $(NOLTOFLAG) $(INCLUDES) -c -o $@ $(SRCDIR)/chunkset_power8.c
 
+crc32_power8.o:
+	$(CC) $(CFLAGS) $(P8FLAGS) $(NOLTOFLAG) $(INCLUDES) -c -o $@ $(SRCDIR)/crc32_power8.c
+
+crc32_power8.lo:
+	$(CC) $(SFLAGS) $(P8FLAGS) $(NOLTOFLAG) $(INCLUDES) -c -o $@ $(SRCDIR)/crc32_power8.c
+
 slide_hash_power8.o:
 	$(CC) $(CFLAGS) $(P8FLAGS) $(NOLTOFLAG) $(INCLUDES) -c -o $@ $(SRCDIR)/slide_hash_power8.c