This library will be used by Audacious in the future. This library was created by nenolod after he noticed that GLists aren't really lists but queues. You can use the benchmarking program linked in the URL field to confirm that the library is operational on your architecture. I will forward any portability issues or concerns to upstream. If you have a patch, feel free to attach it here. Your cooperation is appreciated.
x86 has it keyworded, nothing to do :)
"GList" vs mowgli_list_t test. GList is in quotes because it's actually a queue, and thus is expected to be O(n). Scores are measured in seconds required to finish the job. 100 iterations: GList: 0.0000779629 Mowgli: 0.0000140667 1000 iterations: GList: 0.0016171932 Mowgli: 0.0001211166 10000 iterations: GList: 0.2303349972 Mowgli: 0.0021960735 25000 iterations: GList: 3.2247440815 Mowgli: 0.0219058990 50000 iterations: GList: 16.5825288296 Mowgli: 0.1605970860 100000 iterations: GList: 69.0515899658 Mowgli: 0.8810250759 Keyworded PPC & PPC64. Testing done with GCC 4.1.2; 64UL on the PPC64. For testers: gcc -o bench test.c `pkg-config libmowgli --cflags --libs` `pkg-config glib-2.0 --cflags --libs`
The emerge output contained the following warnings: mowgli_hash.c:44: warning: implicit declaration of function ‘rand’ mowgli_hash.c:51: warning: implicit declaration of function ‘tolower’ mowgli_hook.c:102: warning: implicit declaration of function ‘g_return_val_if_fa il’ mowgli_init.c:36: warning: implicit declaration of function ‘mowgli_node_init’ And then compilation of the test program failed thusly: gentoo/bugs/171580 # make gcc -O2 -pipe -lmowgli -I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include -lglib-2.0 -I/usr/include/libmowgli -ldl 171580_test.c -o 171580_test /usr/lib/gcc/hppa2.0-unknown-linux-gnu/4.1.1/../../../libmowgli.so: undefined reference to `g_return_val_if_fail' collect2: ld returned 1 exit status make: *** [171580_test] Error 1 # scanelf -a /usr/lib/libmowgli.so TYPE PAX STK/REL/PTL TEXTREL RPATH BIND FILE ET_DYN ---xe- --- --- RWX - - LAZY /usr/lib/libmowgli.so In short it needs work before it can be keyworded for HPPA. Alas, the build log is useless as it only displays the neat "kernel style" build messages. It become even more confusing or confused on closer inspection: /var/tmp/portage/dev-libs/libmowgli-0.1.1/work/libmowgli-0.1.1 # grep -R libmcs . ./configure.ac:PACKAGE="libmcs" ./mk/rules.mk:pkgdatadir ?= $(datadir)/libmcs ./mk/rules.mk:pkglibdir ?= $(libdir)/libmcs ./mk/rules.mk:pkgincludedir ?= $(includedir)/libmcs ./mk/rules.mk:PACKAGE ?= libmcs ./configure:PACKAGE="libmcs" ./config.log:PACKAGE='libmcs' ./config.status:s,@PACKAGE@,|#_!!_#|libmcs,g At which point I thought it best to wrap this up and wait for 0.1.2. :)
Created attachment 113941 [details] strace output
Using CFLAGS="-O2 -pipe -mschedule=8000 -march=2.0 -ggdb -Wall -Wl,-z,now" I got something half working: elmer /keeps/gentoo/bugs/171580 # make gcc -o 171580_test -O2 -pipe -I/usr/include/libmowgli -lmowgli -I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include -lglib-2.0 -I/usr/include/libmowgli -ldl 171580_test.c elmer /keeps/gentoo/bugs/171580 # ./171580_test "GList" vs mowgli_list_t test. GList is in quotes because it's actually a queue, and thus is expected to be O(n). Scores are measured in seconds required to finish the job. 100 iterations: Segmentation fault gdb says: (gdb) run Starting program: /keeps/gentoo/bugs/171580/171580_test "GList" vs mowgli_list_t test. GList is in quotes because it's actually a queue, and thus is expected to be O(n). Scores are measured in seconds required to finish the job. 100 iterations: Program received signal SIGSEGV, Segmentation fault. mowgli_node_add (data=0x0, n=0x210b8, l=0xfb150450) at mowgli_list.c:100 100 mowgli_list.c: No such file or directory. in mowgli_list.c (gdb) thread apply bt full (gdb) bt full #0 mowgli_node_add (data=0x0, n=0x210b8, l=0xfb150450) at mowgli_list.c:100 tn = (mowgli_node_t *) 0x1 __PRETTY_FUNCTION__ = "mowgli_node_add" #1 0x00010830 in bench (iterations=100) at 171580_test.c:25 tv = {tv_sec = 1174445053, tv_usec = 701249} i = 1 l = {head = 0x20f46, tail = 0x1, count = 4212457900} list = <value optimized out> Register (null) not available [strace output submitted separately]
0.1.2 fixes these defects. Sorry about that. (It also works on my HPPA hardware.)
Bumped in CVS. Do try again.
(In reply to comment #6) > 0.1.2 fixes these defects. Sorry about that. (It also works on my HPPA > hardware.) Same here: elmer /keeps/gentoo/bugs/171580 # ./171580_test "GList" vs mowgli_list_t test. GList is in quotes because it's actually a queue, and thus is expected to be O(n). Scores are measured in seconds required to finish the job. 100 iterations: GList: 0.0002679825 Mowgli: 0.0000350475 1000 iterations: GList: 0.0035519600 Mowgli: 0.0003941059 10000 iterations: GList: 0.3277170658 Mowgli: 0.0042901039 25000 iterations: GList: 2.0604171753 Mowgli: 0.0132899284 50000 iterations: GList: 8.3479809761 Mowgli: 0.0364811420 100000 iterations: GList: 50.1893849373 Mowgli: 0.1108620167 =dev-libs/libmowgli-0.1.2 marked ~hppa.
Created attachment 113978 [details] Test/Benchmark for dev-utils/libmowgli Testin on amd64 yielded a segfault during bench(100) First try to fix: mowgli_list_t l = mowgli_list_create(); ... mowgli_list_free(l); <--segfaulted like this: Program received signal SIGSEGV, Segmentation fault. 0x00002aec0b7664bd in mowgli_node_move (m=0x511180, oldlist=0x511178, newlist=0x511160) at mowgli_list.c:271 271 m->prev->next = m->next; (gdb) bt #0 0x00002aec0b7664bd in mowgli_node_move (m=0x511180, oldlist=0x511178, newlist=0x511160) at mowgli_list.c:271 #1 0x00002aec0b7655e1 in mowgli_heap_free (heap=0x5110d0, data=0x5111a0) at mowgli_heap.c:198 #2 0x00002aec0b7665cd in mowgli_list_free (l=0x5111a0) at mowgli_list.c:308 #3 0x0000000000400a45 in bench (iterations=100) at mowgli-test.c:29 #4 0x0000000000400a83 in main (argc=1, argv=0x7fff9f4621d8) at mowgli-test.c:42 Next try: keep the mowgli_list_t on the stack, but clear it like this: mowgli_list_t l = { 0 }; This finally works, but it's not the clearest way. I'd like to hear from the author why the mowgli_list_free() call segfaults. Won't keyword until this is cleared. I could fix it by clearing the allocated mowgli_list_t li
Just ignore the last line of the previous post please. :-)
(In reply to comment #9) > Created an attachment (id=113978) [edit] > Test/Benchmark for dev-utils/libmowgli > > Testin on amd64 yielded a segfault during bench(100) > > First try to fix: > mowgli_list_t l = mowgli_list_create(); > ... > mowgli_list_free(l); <--segfaulted like this: > Program received signal SIGSEGV, Segmentation fault. > 0x00002aec0b7664bd in mowgli_node_move (m=0x511180, oldlist=0x511178, > newlist=0x511160) at mowgli_list.c:271 > 271 m->prev->next = m->next; > (gdb) bt > #0 0x00002aec0b7664bd in mowgli_node_move (m=0x511180, oldlist=0x511178, > newlist=0x511160) at mowgli_list.c:271 > #1 0x00002aec0b7655e1 in mowgli_heap_free (heap=0x5110d0, data=0x5111a0) at > mowgli_heap.c:198 > #2 0x00002aec0b7665cd in mowgli_list_free (l=0x5111a0) at mowgli_list.c:308 > #3 0x0000000000400a45 in bench (iterations=100) at mowgli-test.c:29 > #4 0x0000000000400a83 in main (argc=1, argv=0x7fff9f4621d8) at > mowgli-test.c:42 > > > Next try: keep the mowgli_list_t on the stack, but clear it like this: > mowgli_list_t l = { 0 }; > This finally works, but it's not the clearest way. I'd like to hear from the > author why the mowgli_list_free() call segfaults. Won't keyword until this is > cleared. > > I could fix it by clearing the allocated mowgli_list_t li > There was a bug in the heap allocator. It has been fixed in 0.1.3.
Created attachment 114021 [details, diff] Fixes a segfault when freeing allocated memory on 64bit little endian systems, by nenolod The author provided the attached patch. Works fine on amd64.
0.1.2 works for me on amd64: odin ~ # ./bench "GList" vs mowgli_list_t test. GList is in quotes because it's actually a queue, and thus is expected to be O(n). Scores are measured in seconds required to finish the job. 100 iterations: GList: 0.0000510216 Mowgli: 0.0000050068 1000 iterations: GList: 0.0009350777 Mowgli: 0.0000832081 10000 iterations: GList: 0.2458400726 Mowgli: 0.0009701252 25000 iterations: GList: 3.1194939613 Mowgli: 0.0046420097 50000 iterations: GList: 15.7011108398 Mowgli: 0.0317502022 100000 iterations: GList: 66.3843708038 Mowgli: 0.1786589622 Portage 2.1.2.2 (default-linux/amd64/2006.1/desktop, gcc-4.1.1, glibc-2.5-r0, 2.6.20-beyond2 x86_64) ================================================================= System uname: 2.6.20-beyond2 x86_64 AMD Athlon(tm) 64 X2 Dual Core Processor 4600+ Gentoo Base System release 1.12.9 Timestamp of tree: Wed, 21 Mar 2007 21:50:01 +0000 ccache version 2.4 [enabled] dev-java/java-config: 1.3.7, 2.0.31 dev-lang/python: 2.4.3-r4 dev-python/pycrypto: 2.0.1-r5 dev-util/ccache: 2.4-r6 sys-apps/sandbox: 1.2.17 sys-devel/autoconf: 2.13, 2.61 sys-devel/automake: 1.4_p6, 1.5, 1.6.3, 1.7.9-r1, 1.8.5-r3, 1.9.6-r2, 1.10 sys-devel/binutils: 2.16.1-r3 sys-devel/gcc-config: 1.3.14 sys-devel/libtool: 1.5.22 virtual/os-headers: 2.6.17-r2 ACCEPT_KEYWORDS="amd64" AUTOCLEAN="yes" CBUILD="x86_64-pc-linux-gnu" CFLAGS="-march=k8 -O2 -pipe -msse3" CHOST="x86_64-pc-linux-gnu" CONFIG_PROTECT="/etc /usr/share/X11/xkb" CONFIG_PROTECT_MASK="/etc/env.d /etc/env.d/java/ /etc/gconf /etc/java-config/vms/ /etc/php/apache1-php5/ext-active/ /etc/php/apache2-php5/ext-active/ /etc/php/cgi-php5/ext-active/ /etc/php/cli-php5/ext-active/ /etc/revdep-rebuild /etc/splash /etc/terminfo" CXXFLAGS="-march=k8 -O2 -pipe -msse3" DISTDIR="/usr/portage/distfiles" FEATURES="autoconfig buildsyspkg ccache collision-protect distlocks metadata-transfer multilib-strict parallel-fetch sandbox sfperms strict test" GENTOO_MIRRORS="ftp://linux.rz.ruhr-uni-bochum.de/gentoo-mirror/ ftp://ftp.uni-erlangen.de/pub/mirrors/gentoo ftp://ftp.join.uni-muenster.de/pub/linux/distributions/gentoo ftp://ftp.wh2.tu-dresden.de/pub/mirrors/gentoo ftp://ftp.join.uni-muenster.de/pub/linux/distributions/gentoo ftp://ftp-stud.fht-esslingen.de/pub/Mirrors/gentoo/ ftp://ftp.gentoo.mesh-solutions.com/gentoo/ ftp://pandemonium.tiscali.de/pub/gentoo/ " LANG="en_US.ISO-8859-15" LC_ALL="en_US.ISO-8859-15" MAKEOPTS="-j3" PKGDIR="/usr/portage/packages" PORTAGE_RSYNC_EXTRA_OPTS="--exclude-from=/etc/portage/rsync_excludes" PORTAGE_RSYNC_OPTS="--recursive --links --safe-links --perms --times --compress --force --whole-file --delete --delete-after --stats --timeout=180 --exclude=/distfiles --exclude=/local --exclude=/packages --filter=H_**/files/digest-*" PORTAGE_TMPDIR="/var/tmp" PORTDIR="/usr/portage" PORTDIR_OVERLAY="/usr/local/portage/overlay" SYNC="rsync://rsync.europe.gentoo.org/gentoo-portage" USE="X a52 aac acpi alsa amd64 amr audiofile berkdb bitmap-fonts bzip2 cairo cdinstall cdr cli cracklib crypt cups dbus dri dts dvd dvdr dvdread eds emboss encode fam firefox fortran gdbm gif gpm gstreamer gtk gtk2 hal iconv jpeg ldap libg++ lirc logrotate mad midi mikmod mp3 mpeg ncurses nls nptl nptlonly offensive ogg opengl pam pcre php png ppds pppd quicktime readline reflection sdl session smp socks5 spl ssl svg symlink tcpd test tiff truetype truetype-fonts type1-fonts unicode v4l vorbis x264 xinerama xorg xv xvid zlib" ALSA_CARDS="emu10k1" ALSA_PCM_PLUGINS="adpcm alaw asym copy dmix dshare dsnoop empty extplug file hooks iec958 ioplug ladspa lfloat linear meter mulaw multi null plug rate route share shm softvol" ELIBC="glibc" INPUT_DEVICES="evdev keyboard" KERNEL="linux" LCD_DEVICES="bayrad cfontz cfontz633 glk hd44780 lb216 lcdm001 mtxorb ncurses text" LIRC_DEVICES="inputlirc" USERLAND="GNU" VIDEO_CARDS="nvidia" Unset: CTARGET, EMERGE_DEFAULT_OPTS, INSTALL_MASK, LDFLAGS, LINGUAS, PORTAGE_COMPRESS, PORTAGE_COMPRESS_FLAGS
~sparc it's got.
Comment on attachment 114021 [details, diff] Fixes a segfault when freeing allocated memory on 64bit little endian systems, by nenolod 0.1.3 is in portage, it already incorporates this patch.
0.1.3 is ~amd64
~ia64 done. Please note: dodoc: README does not exist dodoc: TODO does not exist
Fixed dodoc statement, thanks Raúl.
This is what I get on alpha with the provided testcase: ---8<--- GNU gdb 6.6 Copyright (C) 2006 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type "show copying" to see the conditions. There is absolutely no warranty for GDB. Type "show warranty" for details. This GDB was configured as "alpha-unknown-linux-gnu"... Using host libthread_db library "/lib/libthread_db.so.1". (gdb) r Starting program: /root/bench "GList" vs mowgli_list_t test. GList is in quotes because it's actually a queue, and thus is expected to be O(n). Scores are measured in seconds required to finish the job. 100 iterations: Program received signal SIGSEGV, Segmentation fault. mowgli_node_add (data=0x0, n=0x1200140f0, l=0x11fa21b70) at mowgli_list.c:102 102 mowgli_list.c: No such file or directory. in mowgli_list.c (gdb) bt #0 mowgli_node_add (data=0x0, n=0x1200140f0, l=0x11fa21b70) at mowgli_list.c:102 #1 0x0000000120000c04 in bench (iterations=100) at test-mowgli.c:25 #2 0x0000000120000d60 in main (argc=1, argv=0x11fa21cf8) at test-mowgli.c:41 (gdb) ---8<--- - ferdy
~alpha done, works fine for me
*** Bug 196169 has been marked as a duplicate of this bug. ***
05 Dec 2007; Joshua Kinard <kumba@gentoo.org> libmowgli-0.5.0.ebuild: Marked unstable on mips.
audacious doesn't have arm/s390 keywords. Closing
