[kaffe] [Debian bug report] Kaffe JVMPI shutdown problem on ppc
Hi all, we received a kaffe jvmpi bug report for powerpc in debian. For the thread up to now and followups see: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=335622 JMP seems to work the same as on my x86 box except that during the vm shutdown on ppc of the bug reporter an internal error in kaffe occures [...] teardown called, freeing jmp-data.. Internal error: caught an unexpected exception. Please check your CLASSPATH and your installation. java/lang/NullPointerException Abandon [...] The bug reporter says that this error is reproducible every time on his box. Regards, Wolfgang ___ kaffe mailing list kaffe@kaffe.org http://kaffe.org/cgi-bin/mailman/listinfo/kaffe
Re: [kaffe] JVMPI status?
Hi Jon, Jon Nall wrote: dalibor, i don't think this is a libtool bug. i think the solution is to remove it from libltdl/acinclude.m4:6372. i don't have up to date autoconf/automake, so i can't test it, but manually removing it from libltdl/configure:21057 worked. the libltdl directory comes from GNU libtool: it is created by running libtoolize from libtool 1.5.2. I'll forward your bug report and fix to the libtool developers, it's a very good catch. i'm attaching a patch that should resolve both the malloc issue and the assert i mentioned in my previous email. thanks, I've checked it in. I've adapted the libtool patch slightly to fit into our build system. current status is that i can get EJP to load and run successfully on small programs. i tried it on a larger program and got an internal error in EJP (an error that doesn't show up when i run against the sun JDK). so i'll debug that one next. thanks for doing that! cheers, dalibor topic ___ kaffe mailing list [EMAIL PROTECTED] http://kaffe.org/cgi-bin/mailman/listinfo/kaffe
Re: [kaffe] JVMPI status?
Hi Jon, Jon Nall wrote: Dalibor Topic wrote: yep, I get the same problem with jprofiler-1.1.1. It seems to be some weird memory allocation issue. I haven;t tried tracking it down. i figured out why it's happening, though i'm not familiar enough with the project (yet) to suggest the proper fix. the issue is that in function argz_create_sep() at libltdl/ltdl.c:627 argz is LT_DLMALLOC'ed. and in foreach_dirinpath() at libltdl/ltdl.c:2705 is LT_DLFREE'ed. however, configure found argz_create_sep() on my system defined in /usr/include/argz.h and set the following in libltdl/config.h: #define HAVE_ARGZ_CREATE_SEP 1. wow, that was quick. Thanks a lot for looking into it. this, i assume, causes the external argz_create_sep call which just uses plain old malloc(). send that to LT_DLFREE and chaos ensues. i hand-modified libltdl/config.h to say #undef HAVE_ARGZ_CREATE_SEP, recompiled and can now run ejp without error. sounds reasonable. i've tried both of the following successfully: 1. never use the external argz_create_sep() call 2. add an ifdef in foreach_dirinpath() that calls normal free() for argz. with either of these fixes, i still get a core dump: kaffe-bin: jvmpi_kaffe.c:55: jvmpiPostEvent: Assertion `ev-event_type JVMPI_EVENT_COUNT' failed. if someone can tell me which of 1 or 2 above to pursue, i'll create a patch and get that assert debugged. since there is no way to guarantee that the external argz_create_sep() will use malloc/free (afaik), it may be better to do 1, in my opinion. And since it's a bug in libtool, please send a bug report upstream to the libtool developers, as well. cheers, dalibor topic ___ kaffe mailing list [EMAIL PROTECTED] http://kaffe.org/cgi-bin/mailman/listinfo/kaffe
Re: [kaffe] JVMPI status?
since there is no way to guarantee that the external argz_create_sep()
will use malloc/free (afaik), it may be better to do 1, in my opinion.
And since it's a bug in libtool, please send a bug report upstream to
the libtool developers, as well.
dalibor,
i don't think this is a libtool bug. i think the solution is to remove
it from libltdl/acinclude.m4:6372. i don't have up to date
autoconf/automake, so i can't test it, but manually removing it from
libltdl/configure:21057 worked.
i'm attaching a patch that should resolve both the malloc issue and the
assert i mentioned in my previous email.
current status is that i can get EJP to load and run successfully on
small programs. i tried it on a larger program and got an internal error
in EJP (an error that doesn't show up when i run against the sun JDK).
so i'll debug that one next.
nall.
Index: kaffe/kaffe/jvmpi/jvmpi_kaffe.c
===
RCS file: /cvs/kaffe/kaffe/kaffe/jvmpi/jvmpi_kaffe.c,v
retrieving revision 1.4
diff -u -r1.4 jvmpi_kaffe.c
--- kaffe/kaffe/jvmpi/jvmpi_kaffe.c 11 Mar 2004 16:00:08 - 1.4
+++ kaffe/kaffe/jvmpi/jvmpi_kaffe.c 11 Mar 2004 19:35:30 -
@@ -52,7 +52,7 @@
{
assert(ev != NULL);
assert(ev-event_type = 0);
- assert(ev-event_type JVMPI_EVENT_COUNT);
+ assert((ev-event_type ~JVMPI_REQUESTED_EVENT) JVMPI_EVENT_COUNT);
ev-env_id = THREAD_JNIENV();
switch( ev-event_type )
Index: kaffe/libltdl/acinclude.m4
===
RCS file: /cvs/kaffe/kaffe/libltdl/acinclude.m4,v
retrieving revision 1.25
diff -u -r1.25 acinclude.m4
--- kaffe/libltdl/acinclude.m4 27 Jan 2004 11:08:43 - 1.25
+++ kaffe/libltdl/acinclude.m4 11 Mar 2004 19:35:32 -
@@ -6369,5 +6369,5 @@
# include argz.h
#endif])
-AC_CHECK_FUNCS([argz_append argz_create_sep argz_insert argz_next argz_stringify])
+AC_CHECK_FUNCS([argz_append argz_insert argz_next argz_stringify])
])# AC_LTDL_FUNC_ARGZ
Index: kaffe/libltdl/configure
===
RCS file: /cvs/kaffe/kaffe/libltdl/configure,v
retrieving revision 1.54
diff -u -r1.54 configure
--- kaffe/libltdl/configure 11 Mar 2004 11:56:48 - 1.54
+++ kaffe/libltdl/configure 11 Mar 2004 19:35:38 -
@@ -21054,7 +21054,7 @@
-for ac_func in argz_append argz_create_sep argz_insert argz_next argz_stringify
+for ac_func in argz_append argz_insert argz_next argz_stringify
do
as_ac_var=`echo ac_cv_func_$ac_func | $as_tr_sh`
echo $as_me:$LINENO: checking for $ac_func 5
[kaffe] JVMPI status?
hello. i'm curious about the status of kaffe's JVMPI implementation. could someone clue me in as to where it's at? also, i have a patchset for 1.1.4 that fixes the compile when --enable-jvmpi is set in configure. i can post it if there's interest. nall. ___ kaffe mailing list [EMAIL PROTECTED] http://kaffe.org/cgi-bin/mailman/listinfo/kaffe
Re: [kaffe] JVMPI status?
tim,
thanks for the reply. i'm attaching the diff to this mail. the only bit
i'm not sure about are the changes in jvmpi_kaffe.c. it looks like
someone changed the definition of java_lang_String structs. the other
diffs are just typos.
also, JVMPI functionality still seems broken, even with this patch. i
get the stack trace attached below. i don't have time to look at it now.
i might get to it this week, though.
nall.
[EMAIL PROTECTED] ~/kaffe-1.1.4]$ java -Xruntracer.so foo
GNU gdb Red Hat Linux (5.2-2)
Copyright 2002 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 i386-redhat-linux...
(gdb) run
Starting program: /home/nall/JDKs/kaffe/jre/bin/kaffe-bin -Xruntracer.so foo
loadNativeLib(libtracer.so)
Program received signal SIGFPE, Arithmetic exception.
0x4003af10 in gcFree (gcif=0x400bf4d4, mem=0x0) at mem/gc-incremental.c:1249
1249idx = GCMEM2IDX(info, unit);
(gdb) where
#0 0x4003af10 in gcFree (gcif=0x400bf4d4, mem=0x0) at
mem/gc-incremental.c:1249
#1 0x4003b61d in jfree (mem=0x85ef018) at gc.c:34
#2 0x4003764f in kdlfree (ptr=0x85ef018) at
../../include/kaffe/jmalloc.h:37
#3 0x40078ee3 in foreach_dirinpath (search_path=0x85ef050
/home/nall/ejp/tracer/lib, base_name=0x8394714 libtracer.so,
func=0x40079220 find_handle_callback, data1=0xbfffd848, data2=0x0)
at ltdl.c:2712
#4 0x400792d6 in find_handle (search_path=0x0, base_name=0x85ef018 o,
handle=0xbfffd848) at ltdl.c:2791
#5 0x4007987f in try_dlopen (phandle=0xbfffd874, filename=0xbfffe36e
libtracer.so) at ltdl.c:3280
#6 0x40079fdf in lt_dlopen (filename=0x85ef018 o) at ltdl.c:3411
#7 0x4007a0d5 in lt_dlopenext (filename=0xbfffe36e libtracer.so) at
ltdl.c:3463
#8 0x40036dce in loadNativeLibrary2 (path=0xbfffe36e libtracer.so,
default_refs=1, errbuf=0xbfffd920 , errsiz=512) at external.c:293
#9 0x40036cd6 in loadNativeLibrary (lib=0x85ef018 o, errbuf=0x85ef018
o, errsiz=140439576) at external.c:232
#10 0x08048fff in main (argc=1, argv=0xbfffdba4) at main.c:169
#11 0x4012d336 in __libc_start_main (main=0x8048ea0 main, argc=3,
ubp_av=0xbfffdba4, init=0x8048bb8 _init, fini=0x804a710 _fini,
rtld_fini=0x4000d2fc _dl_fini, stack_end=0xbfffdb9c) at
../sysdeps/generic/libc-start.c:129
diff -u -r kaffe-1.1.4-orig/kaffe/jvmpi/jvmpi_kaffe.c
kaffe-1.1.4/kaffe/jvmpi/jvmpi_kaffe.c
--- kaffe-1.1.4-orig/kaffe/jvmpi/jvmpi_kaffe.c 2003-10-11 15:45:48.0 -0500
+++ kaffe-1.1.4/kaffe/jvmpi/jvmpi_kaffe.c 2004-03-09 14:57:33.0 -0600
@@ -182,8 +182,8 @@
assert(tid != NULL);
ev-event_type = JVMPI_EVENT_THREAD_START;
- if( (name = stringCharArray2Java(unhand_char_array(tid-name),
-obj_length(tid-name))) != NULL )
+ if( (name = stringCharArray2Java(unhand_char_array(tid-name-value),
+tid-name-count)) != NULL )
{
ev-u.thread_start.thread_name = stringJava2C(name);
}
diff -u -r kaffe-1.1.4-orig/kaffe/jvmpi/Makefile.am kaffe-1.1.4/kaffe/jvmpi/Makefile.am
--- kaffe-1.1.4-orig/kaffe/jvmpi/Makefile.am2004-01-14 20:29:29.0 -0600
+++ kaffe-1.1.4/kaffe/jvmpi/Makefile.am 2004-03-09 14:56:58.0 -0600
@@ -5,7 +5,7 @@
libkaffejvmpi_la_LDFLAGS = \
-export-dynamic \
- -release $(PACKAGE_PERSION)
+ -release $(PACKAGE_VERSION)
libkaffejvmpi_la_DEPENDENCIES = \
$(LIBLTDL) \
diff -u -r kaffe-1.1.4-orig/kaffe/jvmpi/Makefile.in kaffe-1.1.4/kaffe/jvmpi/Makefile.in
--- kaffe-1.1.4-orig/kaffe/jvmpi/Makefile.in2004-02-18 15:18:56.0 -0600
+++ kaffe-1.1.4/kaffe/jvmpi/Makefile.in 2004-03-09 14:57:07.0 -0600
@@ -314,7 +314,7 @@
lib_LTLIBRARIES = libkaffejvmpi.la
libkaffejvmpi_la_LDFLAGS = \
-export-dynamic \
- -release $(PACKAGE_PERSION)
+ -release $(PACKAGE_VERSION)
libkaffejvmpi_la_DEPENDENCIES = \
$(LIBLTDL) \
diff -u -r kaffe-1.1.4-orig/kaffe/kaffevm/systems/unix-jthreads/jthread.c
kaffe-1.1.4/kaffe/kaffevm/systems/unix-jthreads/jthread.c
--- kaffe-1.1.4-orig/kaffe/kaffevm/systems/unix-jthreads/jthread.c 2004-02-01
16:14:54.0 -0600
+++ kaffe-1.1.4/kaffe/kaffevm/systems/unix-jthreads/jthread.c 2004-03-09
14:58:48.0 -0600
@@ -2602,7 +2602,7 @@
/*
* Check if thread is interrupted.
*/
-int jthread_is_intrerrupted(jthread_t jt)
+int jthread_is_interrupted(jthread_t jt)
{
return (jt-flags THREAD_FLAGS_INTERRUPTED);
}
Re: [kaffe] jvmpi
hi, I've checked in some JVMPI stuff. Its not completely done yet, but its a pretty good start. I'll try and get it finish RSN. Hopefully, i didn't break anything... Real soon now indeed... The last checkin gets things a bit farther, atleast for the jit3/unix-jthreads configuration. I can run the tracer for the Extensible Java Profiler and get reasonable output for method traces, but the class loader trace does not seem to work. You can get EJP here: http://ejp.sourceforge.net/ I've only run the tracer library in kaffe though, I ran the presenter interface using jdk1.4. thanks, tim ___ kaffe mailing list [EMAIL PROTECTED] http://kaffe.org/cgi-bin/mailman/listinfo/kaffe
Re: [kaffe] jvmpi
hi Tim, hi, I've checked in some JVMPI stuff. Its not completely done yet, but its a pretty good start. I'll try and get it finish RSN. Hopefully, i didn't break anything... thanks! I've tried to build kaffe with jvmpi enabled, but it breaks during linking because a function is missing. I've modified the Makefile.am to link in libkaffevm, and I thought about exporting the missing function, but you seem to use it in jvmpi_kaffe.c with a different signature, so I'm not sure how to get JVMPI to build ;) Whats the function? Oh, and do you know of any JVMPI applications that would work with your implementation and kaffe? I was mainly just gonna try and get sun's hprof to work. I actually have some more of it written in my local tree, but I still need to implement the dump stuff. I'll try to get to that over the weekend too. cheers, dalibor topic tim ___ kaffe mailing list [EMAIL PROTECTED] http://kaffe.org/cgi-bin/mailman/listinfo/kaffe
Re: [kaffe] jvmpi
hi Tim, Timothy Stack wrote: hi, I've checked in some JVMPI stuff. Its not completely done yet, but its a pretty good start. I'll try and get it finish RSN. Hopefully, i didn't break anything... thanks! I've tried to build kaffe with jvmpi enabled, but it breaks during linking because a function is missing. I've modified the Makefile.am to link in libkaffevm, and I thought about exporting the missing function, but you seem to use it in jvmpi_kaffe.c with a different signature, so I'm not sure how to get JVMPI to build ;) Oh, and do you know of any JVMPI applications that would work with your implementation and kaffe? cheers, dalibor topic ___ kaffe mailing list [EMAIL PROTECTED] http://kaffe.org/cgi-bin/mailman/listinfo/kaffe
Re: [kaffe] jvmpi
On Fri, 8 Aug 2003 10:24:35 -0600 (MDT) Timothy Stack [EMAIL PROTECTED] wrote: Hi, I've checked in some JVMPI stuff. Its not completely done yet, but its a pretty good start. I'll try and get it finish RSN. Hopefully, i didn't break anything... thanks! I've tried to build kaffe with jvmpi enabled, but it breaks during linking because a function is missing. I've modified the Makefile.am to link in libkaffevm, and I thought about exporting the missing function, but you seem to use it in jvmpi_kaffe.c with a different signature, so I'm not sure how to get JVMPI to build ;) Whats the function? part of this might be actually my fault. While moving the engine dependent part of the stacktrace handling into different directories, I also modified the handling of stacktraces themself a little bit, but forgot to check whether jvmpi was affected or not :( To make it short, you don't have to call stackTraceFindMethod() anymore in order to determine the struct _methods* of a stackframe, buildStackTrace() already initializes that field. Sorry, Helmer ___ kaffe mailing list [EMAIL PROTECTED] http://kaffe.org/cgi-bin/mailman/listinfo/kaffe
Re: [kaffe] jvmpi
On Sat, 26 Jul 2003 11:10:46 -0600 Timothy Stack [EMAIL PROTECTED] wrote: Hi, I've checked in some JVMPI stuff. Its not completely done yet, but its a pretty good start. I'll try and get it finish RSN. Hopefully, i didn't break anything... you seem to have missed intrp and jit ;) Will check in fixes for those later today. Greetings, Helmer ___ kaffe mailing list [EMAIL PROTECTED] http://kaffe.org/cgi-bin/mailman/listinfo/kaffe
