Skip to content
This repository was archived by the owner on Mar 1, 2024. It is now read-only.
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
39 changes: 35 additions & 4 deletions Makefile.in
Original file line number Diff line number Diff line change
Expand Up @@ -45,6 +45,13 @@ TAR=tar
SHELL=/bin/sh
EXE=

FILL_WINDOW_SSE_o=
FILL_WINDOW_SSE_lo=
CRC_FOLDING_o=
CRC_FOLDING_lo=
DEFLATE_QUICK_o=
DEFLATE_QUICK_lo=

prefix = /usr/local
exec_prefix = ${prefix}
libdir = ${exec_prefix}/lib
Expand All @@ -54,11 +61,11 @@ mandir = ${prefix}/share/man
man3dir = ${mandir}/man3
pkgconfigdir = ${libdir}/pkgconfig

OBJZ = adler32.o crc32.o deflate.o infback.o inffast.o inflate.o inftrees.o trees.o zutil.o
OBJZ = adler32.o ${CRC_FOLDING_o} ${DEFLATE_QUICK_o} crc32.o ${FILL_WINDOW_SSE_o} deflate.o infback.o inffast.o inflate.o inftrees.o trees.o zutil.o
OBJG = compress.o uncompr.o gzclose.o gzlib.o gzread.o gzwrite.o
OBJC = $(OBJZ) $(OBJG)

PIC_OBJZ = adler32.lo crc32.lo deflate.lo infback.lo inffast.lo inflate.lo inftrees.lo trees.lo zutil.lo
PIC_OBJZ = adler32.lo ${CRC_FOLDING_lo} ${DEFLATE_QUICK_lo} crc32.lo ${FILL_WINDOW_SSE_lo} deflate.lo infback.lo inffast.lo inflate.lo inftrees.lo trees.lo zutil.lo
PIC_OBJG = compress.lo uncompr.lo gzclose.lo gzlib.lo gzread.lo gzwrite.lo
PIC_OBJC = $(PIC_OBJZ) $(PIC_OBJG)

Expand Down Expand Up @@ -113,6 +120,30 @@ test64: all64
fi; \
rm -f $$TMP64

fill_window_sse.lo: fill_window_sse.c
-@mkdir objs 2>/dev/null || test -d objs
$(CC) $(SFLAGS) -msse2 -DPIC -c -o objs/$*.o $<
-@mv objs/$*.o $@

fill_window_sse.o: fill_window_sse.c
${CC} ${CFLAGS} -msse2 -I. -c -o $@ fill_window_sse.c

crc_folding.lo: crc_folding.c
-@mkdir objs 2>/dev/null || test -d objs
$(CC) $(SFLAGS) -mpclmul -msse4 -DPIC -c -o objs/$*.o $<
-@mv objs/$*.o $@

crc_folding.o: crc_folding.c
${CC} ${CFLAGS} -mpclmul -msse4 -I. -c -o $@ crc_folding.c

deflate_quick.lo: deflate_quick.c
-@mkdir objs 2>/dev/null || test -d objs
$(CC) $(SFLAGS) -msse4 -DPIC -c -o objs/$*.o $<
-@mv objs/$*.o $@

deflate_quick.o: deflate_quick.c
${CC} ${CFLAGS} -msse4 -I. -c -o $@ $<

infcover.o: test/infcover.c zlib.h zconf.h
$(CC) $(CFLAGS) -I. -c -o $@ test/infcover.c

Expand Down Expand Up @@ -271,7 +302,7 @@ adler32.o zutil.o: zutil.h zlib.h zconf.h
gzclose.o gzlib.o gzread.o gzwrite.o: zlib.h zconf.h gzguts.h
compress.o example.o minigzip.o uncompr.o: zlib.h zconf.h
crc32.o: zutil.h zlib.h zconf.h crc32.h
deflate.o: deflate.h zutil.h zlib.h zconf.h
deflate.o: deflate.h zutil.h zlib.h zconf.h ${DEFLATE_QUICK_o}
infback.o inflate.o: zutil.h zlib.h zconf.h inftrees.h inflate.h inffast.h inffixed.h
inffast.o: zutil.h zlib.h zconf.h inftrees.h inflate.h inffast.h
inftrees.o: zutil.h zlib.h zconf.h inftrees.h
Expand All @@ -281,7 +312,7 @@ adler32.lo zutil.lo: zutil.h zlib.h zconf.h
gzclose.lo gzlib.lo gzread.lo gzwrite.lo: zlib.h zconf.h gzguts.h
compress.lo example.lo minigzip.lo uncompr.lo: zlib.h zconf.h
crc32.lo: zutil.h zlib.h zconf.h crc32.h
deflate.lo: deflate.h zutil.h zlib.h zconf.h
deflate.lo: deflate.h zutil.h zlib.h zconf.h ${DEFLATE_QUICK_lo}
infback.lo inflate.lo: zutil.h zlib.h zconf.h inftrees.h inflate.h inffast.h inffixed.h
inffast.lo: zutil.h zlib.h zconf.h inftrees.h inflate.h inffast.h
inftrees.lo: zutil.h zlib.h zconf.h inftrees.h
Expand Down
16 changes: 16 additions & 0 deletions adler32.c
Original file line number Diff line number Diff line change
Expand Up @@ -104,21 +104,37 @@ uLong ZEXPORT adler32(adler, buf, len)
/* do length NMAX blocks -- requires just one modulo operation */
while (len >= NMAX) {
len -= NMAX;
#ifndef ADLER32_UNROLL_LESS
n = NMAX / 16; /* NMAX is divisible by 16 */
#else
n = NMAX / 8; /* NMAX is divisible by 8 */
#endif
do {
#ifndef ADLER32_UNROLL_LESS
DO16(buf); /* 16 sums unrolled */
buf += 16;
#else
DO8(buf,0); /* 8 sums unrolled */
buf += 8;
#endif
} while (--n);
MOD(adler);
MOD(sum2);
}

/* do remaining bytes (less than NMAX, still just one modulo) */
if (len) { /* avoid modulos if none remaining */
#ifndef ADLER32_UNROLL_LESS
while (len >= 16) {
len -= 16;
DO16(buf);
buf += 16;
#else
while (len >= 8) {
len -= 8;
DO8(buf, 0);
buf += 8;
#endif
}
while (len--) {
adler += *buf++;
Expand Down
168 changes: 168 additions & 0 deletions configure
Original file line number Diff line number Diff line change
Expand Up @@ -22,8 +22,28 @@ date >> configure.log
if [ -n "${CHOST}" ]; then
uname="`echo "${CHOST}" | sed -e 's/^[^-]*-\([^-]*\)$/\1/' -e 's/^[^-]*-[^-]*-\([^-]*\)$/\1/' -e 's/^[^-]*-[^-]*-\([^-]*\)-.*$/\1/'`"
CROSS_PREFIX="${CHOST}-"
ARCH="`echo "${CHOST}" | sed -e 's/-.*//'`"
else
ARCH="`uname -m`"
fi

case "${ARCH}" in
x86_64)
case "${CFLAGS}" in
*-m32*)
ARCH=i686
;;
esac
;;
i386 | i486 | i586 | i686)
case "${CFLAGS}" in
*-m64*)
ARCH=x86_64
;;
esac
;;
esac

# destination name for static library
STATICLIB=libz.a

Expand Down Expand Up @@ -740,6 +760,142 @@ EOF
fi
fi

# Check for SSE2 intrinsics
cat > $test.c << EOF
#include <immintrin.h>
int main(void)
{
__m128i zero = _mm_setzero_si128();
return 0;
}
EOF
if try ${CC} ${CFLAGS} -msse2 $test.c; then
echo "Checking for SSE2 intrinsics ... Yes." | tee -a configure.log
HAVE_SSE2_INTRIN=1
else
echo "Checking for SSE2 intrinsics ... No." | tee -a configure.log
HAVE_SSE2_INTRIN=0
fi

# Check for PCLMULQDQ intrinsics
cat > $test.c << EOF
#include <immintrin.h>
#include <wmmintrin.h>
int main(void)
{
__m128i a = _mm_setzero_si128();
__m128i b = _mm_setzero_si128();
__m128i c = _mm_clmulepi64_si128(a, b, 0x10);
return 0;
}
EOF
if try ${CC} ${CFLAGS} -mpclmul $test.c; then
echo "Checking for PCLMULQDQ intrinsics ... Yes." | tee -a configure.log
HAVE_PCLMULQDQ_INTRIN=1
else
echo "Checking for PCLMULQDQ intrinsics ... No." | tee -a configure.log
HAVE_PCLMULQDQ_INTRIN=0
fi

DEFLATE_QUICK_o=""
DEFLATE_QUICK_lo=""

# Set ARCH specific FLAGS
case "${ARCH}" in
x86_64)
OBJC="${OBJC} x86.o"
PIC_OBJC="${PIC_OBJC} x86.lo"

CFLAGS="${CFLAGS} -DX86_64"
SFLAGS="${SFLAGS} -DX86_64"

CFLAGS="${CFLAGS} -DUNALIGNED_OK"
SFLAGS="${SFLAGS} -DUNALIGNED_OK"

CFLAGS="${CFLAGS} -DADLER32_UNROLL_LESS -DCRC32_UNROLL_LESS"
SFLAGS="${SFLAGS} -DADLER32_UNROLL_LESS -DCRC32_UNROLL_LESS"

if test ${HAVE_SSE2_INTRIN} -eq 1; then
CFLAGS="${CFLAGS} -UCHECK_SSE2 -DHAVE_SSE2"
SFLAGS="${SFLAGS} -UCHECK_SSE2 -DHAVE_SSE2"
FILL_WINDOW_SSE_o="fill_window_sse.o"
FILL_WINDOW_SSE_lo="fill_window_sse.lo"
OBJS="${OBJS} ${FILL_WINDOW_SSE_o}"
PIC_OBJS="${PIC_OBJS} ${FILL_WINDOW_SSE_lo}"
else
FILL_WINDOW_SSE_o=""
FILL_WINDOW_SSE_lo=""
fi

CFLAGS="${CFLAGS} -DUSE_SSE4_2_CRC_HASH"
SFLAGS="${SFLAGS} -DUSE_SSE4_2_CRC_HASH"

if test ${HAVE_PCLMULQDQ_INTRIN} -eq 1; then
CFLAGS="${CFLAGS} -DHAVE_PCLMULQDQ"
SFLAGS="${SFLAGS} -DHAVE_PCLMULQDQ"
CRC_FOLDING_o="crc_folding.o"
CRC_FOLDING_lo="crc_folding.lo"
OBJS="${OBJS} ${CRC_FOLDING_o}"
PIC_OBJS="${PIC_OBJS} ${CRC_FOLDING_lo}"
else
CRC_FOLDING_o=""
CRC_FOLDING_lo=""
fi

CFLAGS="${CFLAGS} -DUSE_QUICK -DUSE_MEDIUM"
SFLAGS="${SFLAGS} -DUSE_QUICK -DUSE_MEDIUM"

DEFLATE_QUICK_o="deflate_quick.o"
DEFLATE_QUICK_lo="deflate_quick.lo"
;;
i386 | i486 | i586 | i686)
OBJC="${OBJC} x86.o"
PIC_OBJC="${PIC_OBJC} x86.lo"

CFLAGS="${CFLAGS} -DX86"
SFLAGS="${SFLAGS} -DX86"

CFLAGS="${CFLAGS} -DUNALIGNED_OK"
SFLAGS="${SFLAGS} -DUNALIGNED_OK"

CFLAGS="${CFLAGS} -DADLER32_UNROLL_LESS -DCRC32_UNROLL_LESS"
SFLAGS="${SFLAGS} -DADLER32_UNROLL_LESS -DCRC32_UNROLL_LESS"

if test ${HAVE_SSE2_INTRIN} -eq 1; then
CFLAGS="${CFLAGS} -DCHECK_SSE2 -DHAVE_SSE2"
SFLAGS="${SFLAGS} -DCHECK_SSE2 -DHAVE_SSE2"
FILL_WINDOW_SSE_o="fill_window_sse.o"
FILL_WINDOW_SSE_lo="fill_window_sse.lo"
OBJS="${OBJS} ${FILL_WINDOW_SSE_o}"
PIC_OBJS="${PIC_OBJS} ${FILL_WINDOW_SSE_lo}"
else
FILL_WINDOW_SSE_o=""
FILL_WINDOW_SSE_lo=""
fi

CFLAGS="${CFLAGS} -DUSE_SSE4_2_CRC_HASH"
SFLAGS="${SFLAGS} -DUSE_SSE4_2_CRC_HASH"

if test ${HAVE_PCLMULQDQ_INTRIN} -eq 1; then
CFLAGS="${CFLAGS} -DHAVE_PCLMULQDQ"
SFLAGS="${SFLAGS} -DHAVE_PCLMULQDQ"
CRC_FOLDING_o="crc_folding.o"
CRC_FOLDING_lo="crc_folding.lo"
OBJS="${OBJS} ${CRC_FOLDING_o}"
PIC_OBJS="${PIC_OBJS} ${CRC_FOLDING_lo}"
else
CRC_FOLDING_o=""
CRC_FOLDING_lo=""
fi

CFLAGS="${CFLAGS} -DUSE_QUICK -DUSE_MEDIUM"
SFLAGS="${SFLAGS} -DUSE_QUICK -DUSE_MEDIUM"

DEFLATE_QUICK_o="deflate_quick.o"
DEFLATE_QUICK_lo="deflate_quick.lo"
;;
esac

# show the results in the log
echo >> configure.log
echo ALL = $ALL >> configure.log
Expand Down Expand Up @@ -771,6 +927,12 @@ echo mandir = $mandir >> configure.log
echo prefix = $prefix >> configure.log
echo sharedlibdir = $sharedlibdir >> configure.log
echo uname = $uname >> configure.log
echo FILL_WINDOW_SSE_o = ${FILL_WINDOW_SSE_o} >> configure.log
echo FILL_WINDOW_SSE_lo= ${FILL_WINDOW_SSE_lo} >> configure.log
echo CRC_FOLDING_o = ${CRC_FOLDING_o} >> configure.log
echo CRC_FOLDING_lo= ${CRC_FOLDING_lo} >> configure.log
echo DEFLATE_QUICK_o=${DEFLATE_QUICK_o} >> configure.log
echo DEFLATE_QUICK_lo=${DEFLATE_QUICK_lo} >> configure.log

# udpate Makefile with the configure results
sed < Makefile.in "
Expand Down Expand Up @@ -800,6 +962,12 @@ sed < Makefile.in "
/^PIC_OBJC *=/s#=.*#= $PIC_OBJC#
/^all: */s#:.*#: $ALL#
/^test: */s#:.*#: $TEST#
/^FILL_WINDOW_SSE_o *=/s#=.*#=$FILL_WINDOW_SSE_o#
/^FILL_WINDOW_SSE_lo *=/s#=.*#=$FILL_WINDOW_SSE_lo#
/^CRC_FOLDING_o *=/s#=.*#=$CRC_FOLDING_o#
/^CRC_FOLDING_lo *=/s#=.*#=$CRC_FOLDING_lo#
/^DEFLATE_QUICK_o *=/s#=.*#=$DEFLATE_QUICK_o#
/^DEFLATE_QUICK_lo *=/s#=.*#=$DEFLATE_QUICK_lo#
" > Makefile

# create zlib.pc with the configure results
Expand Down
Loading