Accepting request 138143 from home:mvyskocil:branches:devel:libraries:c_c++

- add longest_match performance patch (fate#314093)
  * suggested by IBM, sent upstream
- rename the main library package to libz1 according Shared
  Library Policy
- profiling build can be enabled via build --with profiling
- use the human-readable package description from zlib.net
- add rpmlintrc

OBS-URL: https://build.opensuse.org/request/show/138143
OBS-URL: https://build.opensuse.org/package/show/devel:libraries:c_c++/zlib?expand=0&rev=4

baselibs.conf

@@ -1,3 +1,3 @@
-zlib
+libz1
   targettype x86 provides "baselibs-x86:<prefix>/lib/libz.so.1"
 zlib-devel

zlib-1.2.7-improve-longest_match-performance.patch

@@ -0,0 +1,105 @@
+# http://mail.madler.net/pipermail/zlib-devel_madler.net/2012-June/002907.html
+# http://mail.madler.net/pipermail/zlib-devel_madler.net/2012-August/002977.html
+From: Andreas Krebbel krebbel at linux.vnet.ibm.com 
+Subject: RFC: Improve longest_match performance
+
+The code currently generated for longest_match looks far from optimal
+due to a bunch of pointless zero/sign extend instructions.
+
+By just promoting a few data types in the function I was able to get
+rid of all but two. The new hotloop is almost half the size of the
+original version providing quite a performance win for S/390:
+
+Measured on a zEnterprise z196
+zlib compiled with upstream GCC 4.8 branch
+
+32 bit	      	   	old 	new
+			
+256MB randomdata:	11.65s	10.92s	6.68%	
+100MB manpages:		4.23s	4.14s	2.17%
+217MB PDF:		10.54s	9.44s	11.65%
+
+unaligned ok
+
+256MB randomdata:	10.90s	10.54s	3.41%
+100MB manpages:		3.94s	3.87s	1.81%
+217MB PDF:		8.77s	8.64s   1.50%
+
+64 bit	      	   	old 	new
+			
+256MB randomdata:	11.90s	11.43s	4.11%	
+100MB manpages:		4.51s	4.44s	1.58%
+217MB PDF:		10.11s	9.89s	2.22%
+
+unaligned ok
+
+256MB randomdata:	11.51s	11.15s	3.23%
+100MB manpages:		4.33s	3.99s	8.52%
+217MB PDF:		9.81s	9.02s   8.76%
+
+I also did some measurements on x86 and Power:
+
+For Power (64 bit, unaligned_ok) an additional zero extend
+appears. However, the impact is not measurable. There are minor wins
+and minor regressions. The overall result is flat.
+
+For Core2 32 bit the patch is a clear winner with up to 9% for the pdf
+test.  Also on 64 bit the code optimized for Core2 gets a bit smaller
+but unfortunately causes some regressions which I cannot explain.
+
+For mainframe customers the performance of zlib is very important so I
+would be very happy to see the patch integrated into upstream
+zlib. Given that the patch might cause minor regressions on other
+targets, would it be possible to enable it arch-dependent?
+
+See below for the patch and some code snippets from my tests.
+
+Bye,
+
+-Andreas-
+
+---
+ deflate.c |   15 ++++++++-------
+ 1 file changed, 8 insertions(+), 7 deletions(-)
+
+Index: zlib-1.2.7/deflate.c
+===================================================================
+--- zlib-1.2.7.orig/deflate.c	2012-02-13 01:15:47.000000000 +0100
++++ zlib-1.2.7/deflate.c	2012-09-27 13:39:57.942762946 +0200
+@@ -1143,15 +1143,16 @@
+ /* For 80x86 and 680x0, an optimized version will be provided in match.asm or
+  * match.S. The code will be functionally equivalent.
+  */
+-local uInt longest_match(s, cur_match)
++local uInt longest_match(s, pcur_match)
+     deflate_state *s;
+-    IPos cur_match;                             /* current match */
++    IPos pcur_match;                             /* current match */
+ {
++    ptrdiff_t cur_match = pcur_match; /* extend to pointer width */
+     unsigned chain_length = s->max_chain_length;/* max hash chain length */
+     register Bytef *scan = s->window + s->strstart; /* current string */
+     register Bytef *match;                       /* matched string */
+     register int len;                           /* length of current match */
+-    int best_len = s->prev_length;              /* best match length so far */
++    ptrdiff_t best_len = s->prev_length;              /* best match length so far */
+     int nice_match = s->nice_match;             /* stop if match long enough */
+     IPos limit = s->strstart > (IPos)MAX_DIST(s) ?
+         s->strstart - (IPos)MAX_DIST(s) : NIL;
+@@ -1166,12 +1167,12 @@
+      * Try with and without -DUNALIGNED_OK to check.
+      */
+     register Bytef *strend = s->window + s->strstart + MAX_MATCH - 1;
+-    register ush scan_start = *(ushf*)scan;
+-    register ush scan_end   = *(ushf*)(scan+best_len-1);
++    register uInt scan_start = *(ushf*)scan;
++    register uInt scan_end   = *(ushf*)(scan+best_len-1);
+ #else
+     register Bytef *strend = s->window + s->strstart + MAX_MATCH;
+-    register Byte scan_end1  = scan[best_len-1];
+-    register Byte scan_end   = scan[best_len];
++    register uInt scan_end1  = scan[best_len-1];
++    register uInt scan_end   = scan[best_len];
+ #endif
+ 
+     /* The code is optimized for HASH_BITS >= 8 and MAX_MATCH-2 multiple of 16.

zlib-rpmlintrc

@@ -0,0 +1,4 @@
+# zlib-devel require libz1 - zlib produces libz, not libzlib
+addFilter("zlib-devel.*: W: no-dependency-on zlib*/zlib-libs/libzlib")
+# used only if build --with profiling
+addFilter("zlib.src.*: W: make-check-outside-check-section time make check")

zlib.changes

@@ -1,3 +1,14 @@
+-------------------------------------------------------------------
+Mon Oct 15 07:39:29 UTC 2012 - mvyskocil@suse.com
+
+- add longest_match performance patch (fate#314093)
+  * suggested by IBM, sent upstream
+- rename the main library package to libz1 according Shared
+  Library Policy
+- profiling build can be enabled via build --with profiling
+- use the human-readable package description from zlib.net
+- add rpmlintrc
+
 -------------------------------------------------------------------
 Mon May  7 12:34:45 UTC 2012 - joop.boonen@opensuse.org
 

zlib.spec

@@ -15,39 +15,59 @@
 # Please submit bugfixes or comments via http://bugs.opensuse.org/
 #
 
+%bcond_with profiling
 
 Name:           zlib
-Provides:       libz
-Obsoletes:      libz
-# bug437293
-%ifarch ppc64
-Obsoletes:      zlib-64bit
-%endif
-#
 Version:        1.2.7
 Release:        0
 Summary:        Data Compression Library
 License:        Zlib
 Group:          System/Libraries
 Url:            http://www.zlib.net/
-# git://github.com/kaffeemonster/zlib.git (branch adler32_vec)
+Source0:        http://zlib.net/zlib-%{version}.tar.bz2
-Source:         http://zlib.net/zlib-%{version}.tar.bz2
 Source1:        LICENSE
 Source2:        baselibs.conf
+Source3:        zlib-rpmlintrc
+#PATCH-FIX-SUSE: fate#314093, sent upstream by IBM
+Patch0:         zlib-1.2.7-improve-longest_match-performance.patch
 BuildRequires:  pkgconfig
 BuildRoot:      %{_tmppath}/%{name}-%{version}-build
+# bug437293
+%ifarch ppc64
+Obsoletes:      zlib-64bit
+%endif
 
 %description
-ftp://ds.internic.net/rfc/rfc1950.txt (zlib format), rfc1951.txt
+zlib is designed to be a free, general-purpose, legally unencumbered -- that
-(deflate format) and rfc1952.txt (gzip format). These documents are
+is, not covered by any patents -- lossless data-compression library for use on
-also available in other formats from
+virtually any computer hardware and operating system. the zlib data format is
-ftp://ftp.uu.net/graphics/png/documents/zlib/zdoc-index.html.
+itself portable across platforms. unlike the lzw compression method used in
+unix compress(1) and in the gif image format, the compression method currently
+used in zlib essentially never expands the data. (lzw can double or triple the
+file size in extreme cases.) zlib's memory footprint is also independent of the
+input data and can be reduced, if necessary, at some cost in compression.
+
+%package -n libz1
+Summary:        Include Files and Libraries mandatory for Development
+Group:          Development/Languages/C and C++
+Provides:       %{name} = %{version}-%{release}
+Obsoletes:      %{name} < %{version}-%{release}
+
+%description -n libz1
+zlib is designed to be a free, general-purpose, legally unencumbered -- that
+is, not covered by any patents -- lossless data-compression library for use on
+virtually any computer hardware and operating system. the zlib data format is
+itself portable across platforms. unlike the lzw compression method used in
+unix compress(1) and in the gif image format, the compression method currently
+used in zlib essentially never expands the data. (lzw can double or triple the
+file size in extreme cases.) zlib's memory footprint is also independent of the
+input data and can be reduced, if necessary, at some cost in compression.
 
 %package devel
 Summary:        Include Files and Libraries mandatory for Development
 Group:          Development/Languages/C and C++
 Requires:       glibc-devel
-Requires:       zlib = %{version}
+Requires:       libz1 = %{version}
 Provides:       libz:/usr/include/zlib.h
 # bug437293
 %ifarch ppc64
@@ -73,12 +93,12 @@ libraries.
 
 %prep
 %setup -q
+%patch0 -p1
 
 %build
 export LDFLAGS="-Wl,-z,relro,-z,now"
-# Marcus: breaks example64 in 32bit builds.
+# Marcus: breaks example64 in 32bit builds, so it's disabled by default
-%define do_profiling 0
+%if %{with profiling}
-%if %{do_profiling}
 profiledir=$(mktemp -d)
 trap "rm -rf $profiledir" EXIT
 CC="gcc" ./configure --shared --prefix=%{_prefix} --libdir=/%{_lib}
@@ -96,24 +116,23 @@ make %{?_smp_mflags}
 time make check
 
 %install
-#mkdir -p %{buildroot}%{_mandir}/man3
 mkdir -p %{buildroot}%{_libdir}
 %make_install
 ln -s -v /%{_lib}/$(readlink %{buildroot}/%{_lib}/libz.so) %{buildroot}%{_libdir}/libz.so
 rm -v %{buildroot}/%{_lib}/libz.so
 # static lib
 mv %{buildroot}/%{_lib}/libz.a %{buildroot}%{_libdir}
-# Move .pc file to %{_libdir}
+# Move .pc file to _libdir
 mv %{buildroot}/%{_lib}/pkgconfig %{buildroot}%{_libdir}
 # manpage
 install -m 644 zlib.3 %{buildroot}%{_mandir}/man3
 install -m 644 zutil.h %{buildroot}%{_includedir}
 
-%post -p /sbin/ldconfig
+%post -n libz1 -p /sbin/ldconfig
 
-%postun -p /sbin/ldconfig
+%postun -n libz1 -p /sbin/ldconfig
 
-%files
+%files -n libz1
 %defattr(-,root,root)
 /%{_lib}/libz.so.1.2.*
 /%{_lib}/libz.so.1