You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@harmony.apache.org by Salikh Zakirov <Sa...@Intel.com> on 2006/07/17 14:50:19 UTC
[drlvm] Status of drlvm
Hi,
Can anyone share their experience on running DRLVM built from SVN?
On my workstations, it requires several fixes in order to run properly:
HARMONY-898 'workaround to get correct hythr.dll' (both for Linux and Windows)
HARMONY-853 'DRLVM+classlib segfaults in hyzlib' (on Linux)
Without HARMONY-898, DRLVM segfaults after reading NULL from TLS.
Without HARMONY-853, DRLVM segfaults in libhyzlib.so on Linux.
I would suggest that the patches from HARMONY-898 and HARMONY-853 be applied,
because DRLVM doesn't work at all for me without these fixes.
Any other experiences? Is anyone besides me running DRLVM from SVN?
---------------------------------------------------------------------
Terms of use : http://incubator.apache.org/harmony/mailing.html
To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
For additional commands, e-mail: harmony-dev-help@incubator.apache.org
Re: [drlvm] Status of drlvm
Posted by Alexey Petrenko <al...@gmail.com>.
Salikh,
I'm using drlvm built from svn.
I was not quite sure did I apply any patches or not so I've made a
diff between my repository and Harmony svn... The only difference
shown is the class lib path in build.xml.
So it works OK on my Windows machine without any patches.
SY, Alexey
2006/7/17, Salikh Zakirov <Sa...@intel.com>:
> Hi,
>
> Can anyone share their experience on running DRLVM built from SVN?
>
> On my workstations, it requires several fixes in order to run properly:
>
> HARMONY-898 'workaround to get correct hythr.dll' (both for Linux and Windows)
> HARMONY-853 'DRLVM+classlib segfaults in hyzlib' (on Linux)
>
> Without HARMONY-898, DRLVM segfaults after reading NULL from TLS.
> Without HARMONY-853, DRLVM segfaults in libhyzlib.so on Linux.
>
> I would suggest that the patches from HARMONY-898 and HARMONY-853 be applied,
> because DRLVM doesn't work at all for me without these fixes.
>
> Any other experiences? Is anyone besides me running DRLVM from SVN?
>
>
> ---------------------------------------------------------------------
> Terms of use : http://incubator.apache.org/harmony/mailing.html
> To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
> For additional commands, e-mail: harmony-dev-help@incubator.apache.org
>
>
--
Alexey A. Petrenko
Intel Middleware Products Division
---------------------------------------------------------------------
Terms of use : http://incubator.apache.org/harmony/mailing.html
To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
For additional commands, e-mail: harmony-dev-help@incubator.apache.org
Re: [drlvm] Status of drlvm
Posted by Andrey Chernyshev <a....@gmail.com>.
Apparently segfault disappears if one replaces the "hyzlib" library
(which is compiled from re-distributed zlib at
modules\archive\src\main\native\zlib_dist) with the "natural" libz
which comes with the Linux installation (I'm using GCC 3.3.4 as well).
It seems there is still a conflict between drlvm which is linked to
libz.so and classlib which is linked to libhyzlib.so even if the
workaround HARMONY-853 is applied.
To use the natural zlib on Linux, I did the following:
- removed $(DLLPATH)libhyzlib.so entry in MDLLIBFILES across
classlib's makefiles
- added LDFLAGS = -lz to archive/linux/makefile.
- changed #define HY_ZIP_DLL_NAME to be "z" in
archive/src/main/native/include/shared/zipsup.h
Do people believe this is a right way to go?
Note that for Windows, we'll have either to :
- keep zlib compiled from zlib_dist, or
- download the binaries via "ant fetch-depends" similarly to the drlvm
build (it can be taken from http://www.zlib.net/zlib123-dll.zip).
What will be the preferences? I would stick to the second approach
since recompiling zlib seems to be an odd work.
Thanks,
Andrey.
On 7/19/06, Nataly Naumova <na...@gmail.com> wrote:
> Vladimir,
>
> I've applied both patches, but `segfault` is reproducible nevertheless.
> gdb says that:
> (gdb) run Hello
> Starting program:
> /export/lab_drlbuild/test_tmp/20060719/trunk/build/lnx_ia32_gcc_debug/deploy/jre/bin/ij
> Hello
> [Thread debugging using libthread_db enabled]
> [New Thread 1083593376 (LWP 11375)]
> [New Thread 1425222576 (LWP 11477)]
> [New Thread 1427856304 (LWP 11479)]
> [New Thread 1430490032 (LWP 11480)]
> [New Thread 1433123760 (LWP 11482)]
> [New Thread 1446890416 (LWP 11489)]
> [New Thread 1450089392 (LWP 11508)]
> [New Thread 1453329328 (LWP 11509)]
>
> Program received signal SIGSEGV, Segmentation fault.
> [Switching to Thread 1453329328 (LWP 11509)]
> 0x40783aa2 in malloc_consolidate () from /lib/tls/libc.so.6
>
> Thamks.
> --
> Nataly Naumova,
> Intel Middleware Products Division
>
> On 7/18/06, Vladimir Gorr <vv...@gmail.com> wrote:
> > Nataly,
> >
> > I believe you need to accept both patches mentioned by Silikh in the initial
> > e-mail of this thread,
> > namely, H-898 & H-853. Could you please check if the 'seg fault' issue
> > disappears for gcc 3.3.4
> > as well?
> >
> > Thanks,
> > Vladimir.
> >
> > On 7/18/06, Nataly Naumova <na...@gmail.com> wrote:
> > >
> > > Vladimir and all,
> > > I've compiled with gcc 3.3.4 and received `seg faults` since July,08
> > > till now. I'm not applying any patches, and take just *pure* drlvm
> > > from SVN.
> > > DRLVM build for July,05 is fine, then the build was broken (on linux).
> > > Maybe we should look through the commits since July,05 till July 08 ?
> > > There were not too many commits there.
> > >
> > > Thanks.
> > >
> > > --
> > > Nataly Naumova,
> > > Intel Middleware Products Division
> > >
> > > On 7/18/06, Vladimir Gorr <vv...@gmail.com> wrote:
> > > > The reason of the 'seg fault' issue is clear for me now.
> > > > I see this issue occurs when I used the gcc of 3.4.2 version.
> > > > All works fine for the *gcc (GCC) 3.3.3 (SuSE Linux).*
> > > > Could anybody else confirm or deny this?
> > > >
> > > > Thanks in advance,
> > > > Vladimir.
> > > >
> > > > On 7/18/06, Vladimir Gorr <vv...@gmail.com> wrote:
> > > > >
> > > > > Ignore please my previous e-mail. I was slightly mistaken :-(.
> > > > >
> > > > > Thanks,
> > > > > Vladimir.
> > > > >
> > > > >
> > > > > On 7/18/06, Vladimir Gorr <vv...@gmail.com> wrote:
> > > > > >
> > > > > > Probably I've understood the reason of the segmentation fault on
> > > Linux
> > > > > > platform.
> > > > > >
> > > > > > I almost sure it is related with the build system. An argument why I
> > > > > > think so is the following.
> > > > > >
> > > > > > The basic_istringstream() class from the /usr/include/g++/sstream
> > > > > > include should be used
> > > > > >
> > > > > > to construct the configuration environment for the execution manager
> > > > > > (vm/em/src/DrlEMIml.cpp file).
> > > > > >
> > > > > > However the recent build uses this class from the
> > > > > > /usr/lib/libstdc++.so.5 library instead and as a result
> > > > > >
> > > > > > DRLVM fails. I'm going to continue this investigation and prepare a
> > > > > > patch to fix this issue
> > > > > >
> > > > > > if someone doesn't already know how. Volunteers?
> > > > > >
> > > > > > Thanks,
> > > > > > Vladimir.
> > > > > >
> > > > > >
> > > > > > On 7/18/06, Andrey Chernyshev < a.y.chernyshev@gmail.com> wrote:
> > > > > > >
> > > > > > > On 7/17/06, Vladimir Gorr < vvgorr@gmail.com> wrote:
> > > > > > > > On 7/17/06, Salikh Zakirov < Salikh.Zakirov@intel.com> wrote:
> > > > > > > > >
> > > > > > > > > Hi,
> > > > > > > > >
> > > > > > > > > Can anyone share their experience on running DRLVM built from
> > > SVN?
> > > > > > > > >
> > > > > > > > > On my workstations, it requires several fixes in order to run
> > > > > > > properly:
> > > > > > > > >
> > > > > > > > > HARMONY-898 'workaround to get correct hythr.dll' (both for
> > > Linux
> > > > > > > and
> > > > > > > > > Windows)
> > > > > > > > > HARMONY-853 'DRLVM+classlib segfaults in hyzlib' (on Linux)
> > > > > > > > >
> > > > > > > > > Without HARMONY-898, DRLVM segfaults after reading NULL from
> > > TLS.
> > > > > > > > > Without HARMONY-853, DRLVM segfaults in libhyzlib.so on Linux.
> > > > > > > > >
> > > > > > > > > I would suggest that the patches from HARMONY-898 and
> > > HARMONY-853
> > > > > > > be
> > > > > > > > > applied,
> > > > > > > > > because DRLVM doesn't work at all for me without these fixes.
> > > > > > > > >
> > > > > > > > > Any other experiences? Is anyone besides me running DRLVM from
> > > > > > > SVN?
> > > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > > I've built the DRLVM on Linux for the following cases:
> > > > > > > >
> > > > > > > > 1. w/o patches listed above;
> > > > > > > >
> > > > > > > > 2. w/ H-898 patch;
> > > > > > > >
> > > > > > > > 3. w/ H-898 & H-853 patches.
> > > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > > 'Segmentation fault' occurs for all cases, unfortunately.
> > > > > > > >
> > > > > > >
> > > > > > > I was able to run most of smoke tests on Linux a few days ago, but
> > > now
> > > > > > >
> > > > > > > I'm having the same segfault failure.
> > > > > > > Some observations are:
> > > > > > >
> > > > > > > - In the interpreted mode (-Xint option) it reports
> > > > > > > ClassNotFoundException, so
> > > > > > > it seems VM can no longer load the classes from jars on Linux. I
> > > was
> > > > > > > able to run
> > > > > > > a single "Hello" class though.
> > > > > > >
> > > > > > > - Segfault actually happens during VM shutdown, after the main()
> > > is
> > > > > > > already completed with CNFE error. I wasn't able to find the exact
> > > > > > > point of segfault though - it seems it happens somewhere in a
> > > jitted
> > > > > > > code, gdb doesn't show a much of useful info in that case:
> > > > > > >
> > > > > > > #0 0x40788aa2 in malloc_consolidate () from /lib/tls/libc.so.6
> > > > > > > #1 0x00000010 in ?? ()
> > > > > > > #2 0x4083c7c8 in main_arena () from /lib/tls/libc.so.6
> > > > > > > #3 0x00000000 in ?? ()
> > > > > > > #4 0x4083c7b4 in main_arena () from /lib/tls/libc.so.6
> > > > > > > #5 0x4083c798 in main_arena () from /lib/tls/libc.so.6
> > > > > > > #6 0x4083c780 in __malloc_initialize_hook () from
> > > /lib/tls/libc.so.6
> > > > > > > #7 0x4083aff4 in ?? () from /lib/tls/libc.so.6
> > > > > > > #8 0x0855cdd0 in ?? ()
> > > > > > > #9 0x00037230 in ?? ()
> > > > > > > #10 0x56500158 in ?? ()
> > > > > > > #11 0x40789bf6 in _int_free () from /lib/tls/libc.so.6
> > > > > > > Previous frame inner to this frame (corrupt stack?)
> > > > > > >
> > > > > > > I'm continuing investigation, the same behavior is observed on
> > > Linux
> > > > > > > if run with classlib's launcher.
> > > > > > >
> > > > > > > Thanks,
> > > > > > > Andrey.
> > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > > Thanks,
> > > > > > > >
> > > > > > > > Vladimir.
> > > > > > > >
> > > > > > > >
> > > > > > >
> > > ---------------------------------------------------------------------
> > > > > > > > > Terms of use :
> > > http://incubator.apache.org/harmony/mailing.html
> > > > > > > > > To unsubscribe, e-mail:
> > > > > > > harmony-dev-unsubscribe@incubator.apache.org
> > > > > > > > > For additional commands, e-mail:
> > > > > > > harmony-dev-help@incubator.apache.org
> > > > > > > > >
> > > > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > --
> > > > > > > Andrey Chernyshev
> > > > > > > Intel Middleware Products Division
> > > > > > >
> > > > > > >
> > > ---------------------------------------------------------------------
> > > > > > > Terms of use : http://incubator.apache.org/harmony/mailing.html
> > > > > > > To unsubscribe, e-mail:
> > > harmony-dev-unsubscribe@incubator.apache.org
> > > > > > > For additional commands, e-mail:
> > > harmony-dev-help@incubator.apache.org
> > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > >
> > > >
> > >
> > > ---------------------------------------------------------------------
> > > Terms of use : http://incubator.apache.org/harmony/mailing.html
> > > To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
> > > For additional commands, e-mail: harmony-dev-help@incubator.apache.org
> > >
> > >
> >
> >
>
> ---------------------------------------------------------------------
> Terms of use : http://incubator.apache.org/harmony/mailing.html
> To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
> For additional commands, e-mail: harmony-dev-help@incubator.apache.org
>
>
--
Andrey Chernyshev
Intel Middleware Products Division
---------------------------------------------------------------------
Terms of use : http://incubator.apache.org/harmony/mailing.html
To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
For additional commands, e-mail: harmony-dev-help@incubator.apache.org
Re: [drlvm] Status of drlvm
Posted by Nataly Naumova <na...@gmail.com>.
Vladimir,
I've applied both patches, but `segfault` is reproducible nevertheless.
gdb says that:
(gdb) run Hello
Starting program:
/export/lab_drlbuild/test_tmp/20060719/trunk/build/lnx_ia32_gcc_debug/deploy/jre/bin/ij
Hello
[Thread debugging using libthread_db enabled]
[New Thread 1083593376 (LWP 11375)]
[New Thread 1425222576 (LWP 11477)]
[New Thread 1427856304 (LWP 11479)]
[New Thread 1430490032 (LWP 11480)]
[New Thread 1433123760 (LWP 11482)]
[New Thread 1446890416 (LWP 11489)]
[New Thread 1450089392 (LWP 11508)]
[New Thread 1453329328 (LWP 11509)]
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 1453329328 (LWP 11509)]
0x40783aa2 in malloc_consolidate () from /lib/tls/libc.so.6
Thamks.
--
Nataly Naumova,
Intel Middleware Products Division
On 7/18/06, Vladimir Gorr <vv...@gmail.com> wrote:
> Nataly,
>
> I believe you need to accept both patches mentioned by Silikh in the initial
> e-mail of this thread,
> namely, H-898 & H-853. Could you please check if the 'seg fault' issue
> disappears for gcc 3.3.4
> as well?
>
> Thanks,
> Vladimir.
>
> On 7/18/06, Nataly Naumova <na...@gmail.com> wrote:
> >
> > Vladimir and all,
> > I've compiled with gcc 3.3.4 and received `seg faults` since July,08
> > till now. I'm not applying any patches, and take just *pure* drlvm
> > from SVN.
> > DRLVM build for July,05 is fine, then the build was broken (on linux).
> > Maybe we should look through the commits since July,05 till July 08 ?
> > There were not too many commits there.
> >
> > Thanks.
> >
> > --
> > Nataly Naumova,
> > Intel Middleware Products Division
> >
> > On 7/18/06, Vladimir Gorr <vv...@gmail.com> wrote:
> > > The reason of the 'seg fault' issue is clear for me now.
> > > I see this issue occurs when I used the gcc of 3.4.2 version.
> > > All works fine for the *gcc (GCC) 3.3.3 (SuSE Linux).*
> > > Could anybody else confirm or deny this?
> > >
> > > Thanks in advance,
> > > Vladimir.
> > >
> > > On 7/18/06, Vladimir Gorr <vv...@gmail.com> wrote:
> > > >
> > > > Ignore please my previous e-mail. I was slightly mistaken :-(.
> > > >
> > > > Thanks,
> > > > Vladimir.
> > > >
> > > >
> > > > On 7/18/06, Vladimir Gorr <vv...@gmail.com> wrote:
> > > > >
> > > > > Probably I've understood the reason of the segmentation fault on
> > Linux
> > > > > platform.
> > > > >
> > > > > I almost sure it is related with the build system. An argument why I
> > > > > think so is the following.
> > > > >
> > > > > The basic_istringstream() class from the /usr/include/g++/sstream
> > > > > include should be used
> > > > >
> > > > > to construct the configuration environment for the execution manager
> > > > > (vm/em/src/DrlEMIml.cpp file).
> > > > >
> > > > > However the recent build uses this class from the
> > > > > /usr/lib/libstdc++.so.5 library instead and as a result
> > > > >
> > > > > DRLVM fails. I'm going to continue this investigation and prepare a
> > > > > patch to fix this issue
> > > > >
> > > > > if someone doesn't already know how. Volunteers?
> > > > >
> > > > > Thanks,
> > > > > Vladimir.
> > > > >
> > > > >
> > > > > On 7/18/06, Andrey Chernyshev < a.y.chernyshev@gmail.com> wrote:
> > > > > >
> > > > > > On 7/17/06, Vladimir Gorr < vvgorr@gmail.com> wrote:
> > > > > > > On 7/17/06, Salikh Zakirov < Salikh.Zakirov@intel.com> wrote:
> > > > > > > >
> > > > > > > > Hi,
> > > > > > > >
> > > > > > > > Can anyone share their experience on running DRLVM built from
> > SVN?
> > > > > > > >
> > > > > > > > On my workstations, it requires several fixes in order to run
> > > > > > properly:
> > > > > > > >
> > > > > > > > HARMONY-898 'workaround to get correct hythr.dll' (both for
> > Linux
> > > > > > and
> > > > > > > > Windows)
> > > > > > > > HARMONY-853 'DRLVM+classlib segfaults in hyzlib' (on Linux)
> > > > > > > >
> > > > > > > > Without HARMONY-898, DRLVM segfaults after reading NULL from
> > TLS.
> > > > > > > > Without HARMONY-853, DRLVM segfaults in libhyzlib.so on Linux.
> > > > > > > >
> > > > > > > > I would suggest that the patches from HARMONY-898 and
> > HARMONY-853
> > > > > > be
> > > > > > > > applied,
> > > > > > > > because DRLVM doesn't work at all for me without these fixes.
> > > > > > > >
> > > > > > > > Any other experiences? Is anyone besides me running DRLVM from
> > > > > > SVN?
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > I've built the DRLVM on Linux for the following cases:
> > > > > > >
> > > > > > > 1. w/o patches listed above;
> > > > > > >
> > > > > > > 2. w/ H-898 patch;
> > > > > > >
> > > > > > > 3. w/ H-898 & H-853 patches.
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > 'Segmentation fault' occurs for all cases, unfortunately.
> > > > > > >
> > > > > >
> > > > > > I was able to run most of smoke tests on Linux a few days ago, but
> > now
> > > > > >
> > > > > > I'm having the same segfault failure.
> > > > > > Some observations are:
> > > > > >
> > > > > > - In the interpreted mode (-Xint option) it reports
> > > > > > ClassNotFoundException, so
> > > > > > it seems VM can no longer load the classes from jars on Linux. I
> > was
> > > > > > able to run
> > > > > > a single "Hello" class though.
> > > > > >
> > > > > > - Segfault actually happens during VM shutdown, after the main()
> > is
> > > > > > already completed with CNFE error. I wasn't able to find the exact
> > > > > > point of segfault though - it seems it happens somewhere in a
> > jitted
> > > > > > code, gdb doesn't show a much of useful info in that case:
> > > > > >
> > > > > > #0 0x40788aa2 in malloc_consolidate () from /lib/tls/libc.so.6
> > > > > > #1 0x00000010 in ?? ()
> > > > > > #2 0x4083c7c8 in main_arena () from /lib/tls/libc.so.6
> > > > > > #3 0x00000000 in ?? ()
> > > > > > #4 0x4083c7b4 in main_arena () from /lib/tls/libc.so.6
> > > > > > #5 0x4083c798 in main_arena () from /lib/tls/libc.so.6
> > > > > > #6 0x4083c780 in __malloc_initialize_hook () from
> > /lib/tls/libc.so.6
> > > > > > #7 0x4083aff4 in ?? () from /lib/tls/libc.so.6
> > > > > > #8 0x0855cdd0 in ?? ()
> > > > > > #9 0x00037230 in ?? ()
> > > > > > #10 0x56500158 in ?? ()
> > > > > > #11 0x40789bf6 in _int_free () from /lib/tls/libc.so.6
> > > > > > Previous frame inner to this frame (corrupt stack?)
> > > > > >
> > > > > > I'm continuing investigation, the same behavior is observed on
> > Linux
> > > > > > if run with classlib's launcher.
> > > > > >
> > > > > > Thanks,
> > > > > > Andrey.
> > > > > >
> > > > > > >
> > > > > > >
> > > > > > > Thanks,
> > > > > > >
> > > > > > > Vladimir.
> > > > > > >
> > > > > > >
> > > > > >
> > ---------------------------------------------------------------------
> > > > > > > > Terms of use :
> > http://incubator.apache.org/harmony/mailing.html
> > > > > > > > To unsubscribe, e-mail:
> > > > > > harmony-dev-unsubscribe@incubator.apache.org
> > > > > > > > For additional commands, e-mail:
> > > > > > harmony-dev-help@incubator.apache.org
> > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > > >
> > > > > >
> > > > > >
> > > > > > --
> > > > > > Andrey Chernyshev
> > > > > > Intel Middleware Products Division
> > > > > >
> > > > > >
> > ---------------------------------------------------------------------
> > > > > > Terms of use : http://incubator.apache.org/harmony/mailing.html
> > > > > > To unsubscribe, e-mail:
> > harmony-dev-unsubscribe@incubator.apache.org
> > > > > > For additional commands, e-mail:
> > harmony-dev-help@incubator.apache.org
> > > > > >
> > > > > >
> > > > >
> > > >
> > >
> > >
> >
> > ---------------------------------------------------------------------
> > Terms of use : http://incubator.apache.org/harmony/mailing.html
> > To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
> > For additional commands, e-mail: harmony-dev-help@incubator.apache.org
> >
> >
>
>
---------------------------------------------------------------------
Terms of use : http://incubator.apache.org/harmony/mailing.html
To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
For additional commands, e-mail: harmony-dev-help@incubator.apache.org
Re: [drlvm] Status of drlvm
Posted by Vladimir Gorr <vv...@gmail.com>.
Nataly,
I believe you need to accept both patches mentioned by Silikh in the initial
e-mail of this thread,
namely, H-898 & H-853. Could you please check if the 'seg fault' issue
disappears for gcc 3.3.4
as well?
Thanks,
Vladimir.
On 7/18/06, Nataly Naumova <na...@gmail.com> wrote:
>
> Vladimir and all,
> I've compiled with gcc 3.3.4 and received `seg faults` since July,08
> till now. I'm not applying any patches, and take just *pure* drlvm
> from SVN.
> DRLVM build for July,05 is fine, then the build was broken (on linux).
> Maybe we should look through the commits since July,05 till July 08 ?
> There were not too many commits there.
>
> Thanks.
>
> --
> Nataly Naumova,
> Intel Middleware Products Division
>
> On 7/18/06, Vladimir Gorr <vv...@gmail.com> wrote:
> > The reason of the 'seg fault' issue is clear for me now.
> > I see this issue occurs when I used the gcc of 3.4.2 version.
> > All works fine for the *gcc (GCC) 3.3.3 (SuSE Linux).*
> > Could anybody else confirm or deny this?
> >
> > Thanks in advance,
> > Vladimir.
> >
> > On 7/18/06, Vladimir Gorr <vv...@gmail.com> wrote:
> > >
> > > Ignore please my previous e-mail. I was slightly mistaken :-(.
> > >
> > > Thanks,
> > > Vladimir.
> > >
> > >
> > > On 7/18/06, Vladimir Gorr <vv...@gmail.com> wrote:
> > > >
> > > > Probably I've understood the reason of the segmentation fault on
> Linux
> > > > platform.
> > > >
> > > > I almost sure it is related with the build system. An argument why I
> > > > think so is the following.
> > > >
> > > > The basic_istringstream() class from the /usr/include/g++/sstream
> > > > include should be used
> > > >
> > > > to construct the configuration environment for the execution manager
> > > > (vm/em/src/DrlEMIml.cpp file).
> > > >
> > > > However the recent build uses this class from the
> > > > /usr/lib/libstdc++.so.5 library instead and as a result
> > > >
> > > > DRLVM fails. I'm going to continue this investigation and prepare a
> > > > patch to fix this issue
> > > >
> > > > if someone doesn't already know how. Volunteers?
> > > >
> > > > Thanks,
> > > > Vladimir.
> > > >
> > > >
> > > > On 7/18/06, Andrey Chernyshev < a.y.chernyshev@gmail.com> wrote:
> > > > >
> > > > > On 7/17/06, Vladimir Gorr < vvgorr@gmail.com> wrote:
> > > > > > On 7/17/06, Salikh Zakirov < Salikh.Zakirov@intel.com> wrote:
> > > > > > >
> > > > > > > Hi,
> > > > > > >
> > > > > > > Can anyone share their experience on running DRLVM built from
> SVN?
> > > > > > >
> > > > > > > On my workstations, it requires several fixes in order to run
> > > > > properly:
> > > > > > >
> > > > > > > HARMONY-898 'workaround to get correct hythr.dll' (both for
> Linux
> > > > > and
> > > > > > > Windows)
> > > > > > > HARMONY-853 'DRLVM+classlib segfaults in hyzlib' (on Linux)
> > > > > > >
> > > > > > > Without HARMONY-898, DRLVM segfaults after reading NULL from
> TLS.
> > > > > > > Without HARMONY-853, DRLVM segfaults in libhyzlib.so on Linux.
> > > > > > >
> > > > > > > I would suggest that the patches from HARMONY-898 and
> HARMONY-853
> > > > > be
> > > > > > > applied,
> > > > > > > because DRLVM doesn't work at all for me without these fixes.
> > > > > > >
> > > > > > > Any other experiences? Is anyone besides me running DRLVM from
> > > > > SVN?
> > > > > >
> > > > > >
> > > > > >
> > > > > > I've built the DRLVM on Linux for the following cases:
> > > > > >
> > > > > > 1. w/o patches listed above;
> > > > > >
> > > > > > 2. w/ H-898 patch;
> > > > > >
> > > > > > 3. w/ H-898 & H-853 patches.
> > > > > >
> > > > > >
> > > > > >
> > > > > > 'Segmentation fault' occurs for all cases, unfortunately.
> > > > > >
> > > > >
> > > > > I was able to run most of smoke tests on Linux a few days ago, but
> now
> > > > >
> > > > > I'm having the same segfault failure.
> > > > > Some observations are:
> > > > >
> > > > > - In the interpreted mode (-Xint option) it reports
> > > > > ClassNotFoundException, so
> > > > > it seems VM can no longer load the classes from jars on Linux. I
> was
> > > > > able to run
> > > > > a single "Hello" class though.
> > > > >
> > > > > - Segfault actually happens during VM shutdown, after the main()
> is
> > > > > already completed with CNFE error. I wasn't able to find the exact
> > > > > point of segfault though - it seems it happens somewhere in a
> jitted
> > > > > code, gdb doesn't show a much of useful info in that case:
> > > > >
> > > > > #0 0x40788aa2 in malloc_consolidate () from /lib/tls/libc.so.6
> > > > > #1 0x00000010 in ?? ()
> > > > > #2 0x4083c7c8 in main_arena () from /lib/tls/libc.so.6
> > > > > #3 0x00000000 in ?? ()
> > > > > #4 0x4083c7b4 in main_arena () from /lib/tls/libc.so.6
> > > > > #5 0x4083c798 in main_arena () from /lib/tls/libc.so.6
> > > > > #6 0x4083c780 in __malloc_initialize_hook () from
> /lib/tls/libc.so.6
> > > > > #7 0x4083aff4 in ?? () from /lib/tls/libc.so.6
> > > > > #8 0x0855cdd0 in ?? ()
> > > > > #9 0x00037230 in ?? ()
> > > > > #10 0x56500158 in ?? ()
> > > > > #11 0x40789bf6 in _int_free () from /lib/tls/libc.so.6
> > > > > Previous frame inner to this frame (corrupt stack?)
> > > > >
> > > > > I'm continuing investigation, the same behavior is observed on
> Linux
> > > > > if run with classlib's launcher.
> > > > >
> > > > > Thanks,
> > > > > Andrey.
> > > > >
> > > > > >
> > > > > >
> > > > > > Thanks,
> > > > > >
> > > > > > Vladimir.
> > > > > >
> > > > > >
> > > > >
> ---------------------------------------------------------------------
> > > > > > > Terms of use :
> http://incubator.apache.org/harmony/mailing.html
> > > > > > > To unsubscribe, e-mail:
> > > > > harmony-dev-unsubscribe@incubator.apache.org
> > > > > > > For additional commands, e-mail:
> > > > > harmony-dev-help@incubator.apache.org
> > > > > > >
> > > > > > >
> > > > > >
> > > > > >
> > > > >
> > > > >
> > > > > --
> > > > > Andrey Chernyshev
> > > > > Intel Middleware Products Division
> > > > >
> > > > >
> ---------------------------------------------------------------------
> > > > > Terms of use : http://incubator.apache.org/harmony/mailing.html
> > > > > To unsubscribe, e-mail:
> harmony-dev-unsubscribe@incubator.apache.org
> > > > > For additional commands, e-mail:
> harmony-dev-help@incubator.apache.org
> > > > >
> > > > >
> > > >
> > >
> >
> >
>
> ---------------------------------------------------------------------
> Terms of use : http://incubator.apache.org/harmony/mailing.html
> To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
> For additional commands, e-mail: harmony-dev-help@incubator.apache.org
>
>
Re: [drlvm] Status of drlvm
Posted by Nataly Naumova <na...@gmail.com>.
Vladimir and all,
I've compiled with gcc 3.3.4 and received `seg faults` since July,08
till now. I'm not applying any patches, and take just *pure* drlvm
from SVN.
DRLVM build for July,05 is fine, then the build was broken (on linux).
Maybe we should look through the commits since July,05 till July 08 ?
There were not too many commits there.
Thanks.
--
Nataly Naumova,
Intel Middleware Products Division
On 7/18/06, Vladimir Gorr <vv...@gmail.com> wrote:
> The reason of the 'seg fault' issue is clear for me now.
> I see this issue occurs when I used the gcc of 3.4.2 version.
> All works fine for the *gcc (GCC) 3.3.3 (SuSE Linux).*
> Could anybody else confirm or deny this?
>
> Thanks in advance,
> Vladimir.
>
> On 7/18/06, Vladimir Gorr <vv...@gmail.com> wrote:
> >
> > Ignore please my previous e-mail. I was slightly mistaken :-(.
> >
> > Thanks,
> > Vladimir.
> >
> >
> > On 7/18/06, Vladimir Gorr <vv...@gmail.com> wrote:
> > >
> > > Probably I've understood the reason of the segmentation fault on Linux
> > > platform.
> > >
> > > I almost sure it is related with the build system. An argument why I
> > > think so is the following.
> > >
> > > The basic_istringstream() class from the /usr/include/g++/sstream
> > > include should be used
> > >
> > > to construct the configuration environment for the execution manager
> > > (vm/em/src/DrlEMIml.cpp file).
> > >
> > > However the recent build uses this class from the
> > > /usr/lib/libstdc++.so.5 library instead and as a result
> > >
> > > DRLVM fails. I'm going to continue this investigation and prepare a
> > > patch to fix this issue
> > >
> > > if someone doesn't already know how. Volunteers?
> > >
> > > Thanks,
> > > Vladimir.
> > >
> > >
> > > On 7/18/06, Andrey Chernyshev < a.y.chernyshev@gmail.com> wrote:
> > > >
> > > > On 7/17/06, Vladimir Gorr < vvgorr@gmail.com> wrote:
> > > > > On 7/17/06, Salikh Zakirov < Salikh.Zakirov@intel.com> wrote:
> > > > > >
> > > > > > Hi,
> > > > > >
> > > > > > Can anyone share their experience on running DRLVM built from SVN?
> > > > > >
> > > > > > On my workstations, it requires several fixes in order to run
> > > > properly:
> > > > > >
> > > > > > HARMONY-898 'workaround to get correct hythr.dll' (both for Linux
> > > > and
> > > > > > Windows)
> > > > > > HARMONY-853 'DRLVM+classlib segfaults in hyzlib' (on Linux)
> > > > > >
> > > > > > Without HARMONY-898, DRLVM segfaults after reading NULL from TLS.
> > > > > > Without HARMONY-853, DRLVM segfaults in libhyzlib.so on Linux.
> > > > > >
> > > > > > I would suggest that the patches from HARMONY-898 and HARMONY-853
> > > > be
> > > > > > applied,
> > > > > > because DRLVM doesn't work at all for me without these fixes.
> > > > > >
> > > > > > Any other experiences? Is anyone besides me running DRLVM from
> > > > SVN?
> > > > >
> > > > >
> > > > >
> > > > > I've built the DRLVM on Linux for the following cases:
> > > > >
> > > > > 1. w/o patches listed above;
> > > > >
> > > > > 2. w/ H-898 patch;
> > > > >
> > > > > 3. w/ H-898 & H-853 patches.
> > > > >
> > > > >
> > > > >
> > > > > 'Segmentation fault' occurs for all cases, unfortunately.
> > > > >
> > > >
> > > > I was able to run most of smoke tests on Linux a few days ago, but now
> > > >
> > > > I'm having the same segfault failure.
> > > > Some observations are:
> > > >
> > > > - In the interpreted mode (-Xint option) it reports
> > > > ClassNotFoundException, so
> > > > it seems VM can no longer load the classes from jars on Linux. I was
> > > > able to run
> > > > a single "Hello" class though.
> > > >
> > > > - Segfault actually happens during VM shutdown, after the main() is
> > > > already completed with CNFE error. I wasn't able to find the exact
> > > > point of segfault though - it seems it happens somewhere in a jitted
> > > > code, gdb doesn't show a much of useful info in that case:
> > > >
> > > > #0 0x40788aa2 in malloc_consolidate () from /lib/tls/libc.so.6
> > > > #1 0x00000010 in ?? ()
> > > > #2 0x4083c7c8 in main_arena () from /lib/tls/libc.so.6
> > > > #3 0x00000000 in ?? ()
> > > > #4 0x4083c7b4 in main_arena () from /lib/tls/libc.so.6
> > > > #5 0x4083c798 in main_arena () from /lib/tls/libc.so.6
> > > > #6 0x4083c780 in __malloc_initialize_hook () from /lib/tls/libc.so.6
> > > > #7 0x4083aff4 in ?? () from /lib/tls/libc.so.6
> > > > #8 0x0855cdd0 in ?? ()
> > > > #9 0x00037230 in ?? ()
> > > > #10 0x56500158 in ?? ()
> > > > #11 0x40789bf6 in _int_free () from /lib/tls/libc.so.6
> > > > Previous frame inner to this frame (corrupt stack?)
> > > >
> > > > I'm continuing investigation, the same behavior is observed on Linux
> > > > if run with classlib's launcher.
> > > >
> > > > Thanks,
> > > > Andrey.
> > > >
> > > > >
> > > > >
> > > > > Thanks,
> > > > >
> > > > > Vladimir.
> > > > >
> > > > >
> > > > ---------------------------------------------------------------------
> > > > > > Terms of use : http://incubator.apache.org/harmony/mailing.html
> > > > > > To unsubscribe, e-mail:
> > > > harmony-dev-unsubscribe@incubator.apache.org
> > > > > > For additional commands, e-mail:
> > > > harmony-dev-help@incubator.apache.org
> > > > > >
> > > > > >
> > > > >
> > > > >
> > > >
> > > >
> > > > --
> > > > Andrey Chernyshev
> > > > Intel Middleware Products Division
> > > >
> > > > ---------------------------------------------------------------------
> > > > Terms of use : http://incubator.apache.org/harmony/mailing.html
> > > > To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
> > > > For additional commands, e-mail: harmony-dev-help@incubator.apache.org
> > > >
> > > >
> > >
> >
>
>
---------------------------------------------------------------------
Terms of use : http://incubator.apache.org/harmony/mailing.html
To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
For additional commands, e-mail: harmony-dev-help@incubator.apache.org
Re: [drlvm] Status of drlvm
Posted by Vladimir Gorr <vv...@gmail.com>.
The reason of the 'seg fault' issue is clear for me now.
I see this issue occurs when I used the gcc of 3.4.2 version.
All works fine for the *gcc (GCC) 3.3.3 (SuSE Linux).*
Could anybody else confirm or deny this?
Thanks in advance,
Vladimir.
On 7/18/06, Vladimir Gorr <vv...@gmail.com> wrote:
>
> Ignore please my previous e-mail. I was slightly mistaken :-(.
>
> Thanks,
> Vladimir.
>
>
> On 7/18/06, Vladimir Gorr <vv...@gmail.com> wrote:
> >
> > Probably I've understood the reason of the segmentation fault on Linux
> > platform.
> >
> > I almost sure it is related with the build system. An argument why I
> > think so is the following.
> >
> > The basic_istringstream() class from the /usr/include/g++/sstream
> > include should be used
> >
> > to construct the configuration environment for the execution manager
> > (vm/em/src/DrlEMIml.cpp file).
> >
> > However the recent build uses this class from the
> > /usr/lib/libstdc++.so.5 library instead and as a result
> >
> > DRLVM fails. I'm going to continue this investigation and prepare a
> > patch to fix this issue
> >
> > if someone doesn't already know how. Volunteers?
> >
> > Thanks,
> > Vladimir.
> >
> >
> > On 7/18/06, Andrey Chernyshev < a.y.chernyshev@gmail.com> wrote:
> > >
> > > On 7/17/06, Vladimir Gorr < vvgorr@gmail.com> wrote:
> > > > On 7/17/06, Salikh Zakirov < Salikh.Zakirov@intel.com> wrote:
> > > > >
> > > > > Hi,
> > > > >
> > > > > Can anyone share their experience on running DRLVM built from SVN?
> > > > >
> > > > > On my workstations, it requires several fixes in order to run
> > > properly:
> > > > >
> > > > > HARMONY-898 'workaround to get correct hythr.dll' (both for Linux
> > > and
> > > > > Windows)
> > > > > HARMONY-853 'DRLVM+classlib segfaults in hyzlib' (on Linux)
> > > > >
> > > > > Without HARMONY-898, DRLVM segfaults after reading NULL from TLS.
> > > > > Without HARMONY-853, DRLVM segfaults in libhyzlib.so on Linux.
> > > > >
> > > > > I would suggest that the patches from HARMONY-898 and HARMONY-853
> > > be
> > > > > applied,
> > > > > because DRLVM doesn't work at all for me without these fixes.
> > > > >
> > > > > Any other experiences? Is anyone besides me running DRLVM from
> > > SVN?
> > > >
> > > >
> > > >
> > > > I've built the DRLVM on Linux for the following cases:
> > > >
> > > > 1. w/o patches listed above;
> > > >
> > > > 2. w/ H-898 patch;
> > > >
> > > > 3. w/ H-898 & H-853 patches.
> > > >
> > > >
> > > >
> > > > 'Segmentation fault' occurs for all cases, unfortunately.
> > > >
> > >
> > > I was able to run most of smoke tests on Linux a few days ago, but now
> > >
> > > I'm having the same segfault failure.
> > > Some observations are:
> > >
> > > - In the interpreted mode (-Xint option) it reports
> > > ClassNotFoundException, so
> > > it seems VM can no longer load the classes from jars on Linux. I was
> > > able to run
> > > a single "Hello" class though.
> > >
> > > - Segfault actually happens during VM shutdown, after the main() is
> > > already completed with CNFE error. I wasn't able to find the exact
> > > point of segfault though - it seems it happens somewhere in a jitted
> > > code, gdb doesn't show a much of useful info in that case:
> > >
> > > #0 0x40788aa2 in malloc_consolidate () from /lib/tls/libc.so.6
> > > #1 0x00000010 in ?? ()
> > > #2 0x4083c7c8 in main_arena () from /lib/tls/libc.so.6
> > > #3 0x00000000 in ?? ()
> > > #4 0x4083c7b4 in main_arena () from /lib/tls/libc.so.6
> > > #5 0x4083c798 in main_arena () from /lib/tls/libc.so.6
> > > #6 0x4083c780 in __malloc_initialize_hook () from /lib/tls/libc.so.6
> > > #7 0x4083aff4 in ?? () from /lib/tls/libc.so.6
> > > #8 0x0855cdd0 in ?? ()
> > > #9 0x00037230 in ?? ()
> > > #10 0x56500158 in ?? ()
> > > #11 0x40789bf6 in _int_free () from /lib/tls/libc.so.6
> > > Previous frame inner to this frame (corrupt stack?)
> > >
> > > I'm continuing investigation, the same behavior is observed on Linux
> > > if run with classlib's launcher.
> > >
> > > Thanks,
> > > Andrey.
> > >
> > > >
> > > >
> > > > Thanks,
> > > >
> > > > Vladimir.
> > > >
> > > >
> > > ---------------------------------------------------------------------
> > > > > Terms of use : http://incubator.apache.org/harmony/mailing.html
> > > > > To unsubscribe, e-mail:
> > > harmony-dev-unsubscribe@incubator.apache.org
> > > > > For additional commands, e-mail:
> > > harmony-dev-help@incubator.apache.org
> > > > >
> > > > >
> > > >
> > > >
> > >
> > >
> > > --
> > > Andrey Chernyshev
> > > Intel Middleware Products Division
> > >
> > > ---------------------------------------------------------------------
> > > Terms of use : http://incubator.apache.org/harmony/mailing.html
> > > To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
> > > For additional commands, e-mail: harmony-dev-help@incubator.apache.org
> > >
> > >
> >
>
Re: [drlvm] Status of drlvm
Posted by Vladimir Gorr <vv...@gmail.com>.
Ignore please my previous e-mail. I was slightly mistaken :-(.
Thanks,
Vladimir.
On 7/18/06, Vladimir Gorr <vv...@gmail.com> wrote:
>
> Probably I've understood the reason of the segmentation fault on Linux
> platform.
>
> I almost sure it is related with the build system. An argument why I think
> so is the following.
>
> The basic_istringstream() class from the /usr/include/g++/sstream include
> should be used
>
> to construct the configuration environment for the execution manager
> (vm/em/src/DrlEMIml.cpp file).
>
> However the recent build uses this class from the /usr/lib/libstdc++.so.5
> library instead and as a result
>
> DRLVM fails. I'm going to continue this investigation and prepare a patch
> to fix this issue
>
> if someone doesn't already know how. Volunteers?
>
> Thanks,
> Vladimir.
>
>
> On 7/18/06, Andrey Chernyshev <a....@gmail.com> wrote:
> >
> > On 7/17/06, Vladimir Gorr <vv...@gmail.com> wrote:
> > > On 7/17/06, Salikh Zakirov < Salikh.Zakirov@intel.com> wrote:
> > > >
> > > > Hi,
> > > >
> > > > Can anyone share their experience on running DRLVM built from SVN?
> > > >
> > > > On my workstations, it requires several fixes in order to run
> > properly:
> > > >
> > > > HARMONY-898 'workaround to get correct hythr.dll' (both for Linux
> > and
> > > > Windows)
> > > > HARMONY-853 'DRLVM+classlib segfaults in hyzlib' (on Linux)
> > > >
> > > > Without HARMONY-898, DRLVM segfaults after reading NULL from TLS.
> > > > Without HARMONY-853, DRLVM segfaults in libhyzlib.so on Linux.
> > > >
> > > > I would suggest that the patches from HARMONY-898 and HARMONY-853 be
> >
> > > > applied,
> > > > because DRLVM doesn't work at all for me without these fixes.
> > > >
> > > > Any other experiences? Is anyone besides me running DRLVM from SVN?
> > >
> > >
> > >
> > > I've built the DRLVM on Linux for the following cases:
> > >
> > > 1. w/o patches listed above;
> > >
> > > 2. w/ H-898 patch;
> > >
> > > 3. w/ H-898 & H-853 patches.
> > >
> > >
> > >
> > > 'Segmentation fault' occurs for all cases, unfortunately.
> > >
> >
> > I was able to run most of smoke tests on Linux a few days ago, but now
> > I'm having the same segfault failure.
> > Some observations are:
> >
> > - In the interpreted mode (-Xint option) it reports
> > ClassNotFoundException, so
> > it seems VM can no longer load the classes from jars on Linux. I was
> > able to run
> > a single "Hello" class though.
> >
> > - Segfault actually happens during VM shutdown, after the main() is
> > already completed with CNFE error. I wasn't able to find the exact
> > point of segfault though - it seems it happens somewhere in a jitted
> > code, gdb doesn't show a much of useful info in that case:
> >
> > #0 0x40788aa2 in malloc_consolidate () from /lib/tls/libc.so.6
> > #1 0x00000010 in ?? ()
> > #2 0x4083c7c8 in main_arena () from /lib/tls/libc.so.6
> > #3 0x00000000 in ?? ()
> > #4 0x4083c7b4 in main_arena () from /lib/tls/libc.so.6
> > #5 0x4083c798 in main_arena () from /lib/tls/libc.so.6
> > #6 0x4083c780 in __malloc_initialize_hook () from /lib/tls/libc.so.6
> > #7 0x4083aff4 in ?? () from /lib/tls/libc.so.6
> > #8 0x0855cdd0 in ?? ()
> > #9 0x00037230 in ?? ()
> > #10 0x56500158 in ?? ()
> > #11 0x40789bf6 in _int_free () from /lib/tls/libc.so.6
> > Previous frame inner to this frame (corrupt stack?)
> >
> > I'm continuing investigation, the same behavior is observed on Linux
> > if run with classlib's launcher.
> >
> > Thanks,
> > Andrey.
> >
> > >
> > >
> > > Thanks,
> > >
> > > Vladimir.
> > >
> > > ---------------------------------------------------------------------
> > > > Terms of use : http://incubator.apache.org/harmony/mailing.html
> > > > To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
> > > > For additional commands, e-mail:
> > harmony-dev-help@incubator.apache.org
> > > >
> > > >
> > >
> > >
> >
> >
> > --
> > Andrey Chernyshev
> > Intel Middleware Products Division
> >
> > ---------------------------------------------------------------------
> > Terms of use : http://incubator.apache.org/harmony/mailing.html
> > To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
> > For additional commands, e-mail: harmony-dev-help@incubator.apache.org
> >
> >
>
Re: [drlvm] Status of drlvm
Posted by Vladimir Gorr <vv...@gmail.com>.
Probably I've understood the reason of the segmentation fault on Linux
platform.
I almost sure it is related with the build system. An argument why I think
so is the following.
The basic_istringstream() class from the /usr/include/g++/sstream include
should be used
to construct the configuration environment for the execution manager
(vm/em/src/DrlEMIml.cpp file).
However the recent build uses this class from the /usr/lib/libstdc++.so.5
library instead and as a result
DRLVM fails. I'm going to continue this investigation and prepare a patch to
fix this issue
if someone doesn't already know how. Volunteers?
Thanks,
Vladimir.
On 7/18/06, Andrey Chernyshev <a....@gmail.com> wrote:
>
> On 7/17/06, Vladimir Gorr <vv...@gmail.com> wrote:
> > On 7/17/06, Salikh Zakirov <Sa...@intel.com> wrote:
> > >
> > > Hi,
> > >
> > > Can anyone share their experience on running DRLVM built from SVN?
> > >
> > > On my workstations, it requires several fixes in order to run
> properly:
> > >
> > > HARMONY-898 'workaround to get correct hythr.dll' (both for Linux and
> > > Windows)
> > > HARMONY-853 'DRLVM+classlib segfaults in hyzlib' (on Linux)
> > >
> > > Without HARMONY-898, DRLVM segfaults after reading NULL from TLS.
> > > Without HARMONY-853, DRLVM segfaults in libhyzlib.so on Linux.
> > >
> > > I would suggest that the patches from HARMONY-898 and HARMONY-853 be
> > > applied,
> > > because DRLVM doesn't work at all for me without these fixes.
> > >
> > > Any other experiences? Is anyone besides me running DRLVM from SVN?
> >
> >
> >
> > I've built the DRLVM on Linux for the following cases:
> >
> > 1. w/o patches listed above;
> >
> > 2. w/ H-898 patch;
> >
> > 3. w/ H-898 & H-853 patches.
> >
> >
> >
> > 'Segmentation fault' occurs for all cases, unfortunately.
> >
>
> I was able to run most of smoke tests on Linux a few days ago, but now
> I'm having the same segfault failure.
> Some observations are:
>
> - In the interpreted mode (-Xint option) it reports
> ClassNotFoundException, so
> it seems VM can no longer load the classes from jars on Linux. I was able
> to run
> a single "Hello" class though.
>
> - Segfault actually happens during VM shutdown, after the main() is
> already completed with CNFE error. I wasn't able to find the exact
> point of segfault though - it seems it happens somewhere in a jitted
> code, gdb doesn't show a much of useful info in that case:
>
> #0 0x40788aa2 in malloc_consolidate () from /lib/tls/libc.so.6
> #1 0x00000010 in ?? ()
> #2 0x4083c7c8 in main_arena () from /lib/tls/libc.so.6
> #3 0x00000000 in ?? ()
> #4 0x4083c7b4 in main_arena () from /lib/tls/libc.so.6
> #5 0x4083c798 in main_arena () from /lib/tls/libc.so.6
> #6 0x4083c780 in __malloc_initialize_hook () from /lib/tls/libc.so.6
> #7 0x4083aff4 in ?? () from /lib/tls/libc.so.6
> #8 0x0855cdd0 in ?? ()
> #9 0x00037230 in ?? ()
> #10 0x56500158 in ?? ()
> #11 0x40789bf6 in _int_free () from /lib/tls/libc.so.6
> Previous frame inner to this frame (corrupt stack?)
>
> I'm continuing investigation, the same behavior is observed on Linux
> if run with classlib's launcher.
>
> Thanks,
> Andrey.
>
> >
> >
> > Thanks,
> >
> > Vladimir.
> >
> > ---------------------------------------------------------------------
> > > Terms of use : http://incubator.apache.org/harmony/mailing.html
> > > To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
> > > For additional commands, e-mail: harmony-dev-help@incubator.apache.org
> > >
> > >
> >
> >
>
>
> --
> Andrey Chernyshev
> Intel Middleware Products Division
>
> ---------------------------------------------------------------------
> Terms of use : http://incubator.apache.org/harmony/mailing.html
> To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
> For additional commands, e-mail: harmony-dev-help@incubator.apache.org
>
>
Re: [drlvm] Status of drlvm
Posted by Andrey Chernyshev <a....@gmail.com>.
On 7/17/06, Vladimir Gorr <vv...@gmail.com> wrote:
> On 7/17/06, Salikh Zakirov <Sa...@intel.com> wrote:
> >
> > Hi,
> >
> > Can anyone share their experience on running DRLVM built from SVN?
> >
> > On my workstations, it requires several fixes in order to run properly:
> >
> > HARMONY-898 'workaround to get correct hythr.dll' (both for Linux and
> > Windows)
> > HARMONY-853 'DRLVM+classlib segfaults in hyzlib' (on Linux)
> >
> > Without HARMONY-898, DRLVM segfaults after reading NULL from TLS.
> > Without HARMONY-853, DRLVM segfaults in libhyzlib.so on Linux.
> >
> > I would suggest that the patches from HARMONY-898 and HARMONY-853 be
> > applied,
> > because DRLVM doesn't work at all for me without these fixes.
> >
> > Any other experiences? Is anyone besides me running DRLVM from SVN?
>
>
>
> I've built the DRLVM on Linux for the following cases:
>
> 1. w/o patches listed above;
>
> 2. w/ H-898 patch;
>
> 3. w/ H-898 & H-853 patches.
>
>
>
> 'Segmentation fault' occurs for all cases, unfortunately.
>
I was able to run most of smoke tests on Linux a few days ago, but now
I'm having the same segfault failure.
Some observations are:
- In the interpreted mode (-Xint option) it reports ClassNotFoundException, so
it seems VM can no longer load the classes from jars on Linux. I was able to run
a single "Hello" class though.
- Segfault actually happens during VM shutdown, after the main() is
already completed with CNFE error. I wasn't able to find the exact
point of segfault though - it seems it happens somewhere in a jitted
code, gdb doesn't show a much of useful info in that case:
#0 0x40788aa2 in malloc_consolidate () from /lib/tls/libc.so.6
#1 0x00000010 in ?? ()
#2 0x4083c7c8 in main_arena () from /lib/tls/libc.so.6
#3 0x00000000 in ?? ()
#4 0x4083c7b4 in main_arena () from /lib/tls/libc.so.6
#5 0x4083c798 in main_arena () from /lib/tls/libc.so.6
#6 0x4083c780 in __malloc_initialize_hook () from /lib/tls/libc.so.6
#7 0x4083aff4 in ?? () from /lib/tls/libc.so.6
#8 0x0855cdd0 in ?? ()
#9 0x00037230 in ?? ()
#10 0x56500158 in ?? ()
#11 0x40789bf6 in _int_free () from /lib/tls/libc.so.6
Previous frame inner to this frame (corrupt stack?)
I'm continuing investigation, the same behavior is observed on Linux
if run with classlib's launcher.
Thanks,
Andrey.
>
>
> Thanks,
>
> Vladimir.
>
> ---------------------------------------------------------------------
> > Terms of use : http://incubator.apache.org/harmony/mailing.html
> > To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
> > For additional commands, e-mail: harmony-dev-help@incubator.apache.org
> >
> >
>
>
--
Andrey Chernyshev
Intel Middleware Products Division
---------------------------------------------------------------------
Terms of use : http://incubator.apache.org/harmony/mailing.html
To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
For additional commands, e-mail: harmony-dev-help@incubator.apache.org
Re: [drlvm] Status of drlvm
Posted by Vladimir Gorr <vv...@gmail.com>.
On 7/17/06, Salikh Zakirov <Sa...@intel.com> wrote:
>
> Hi,
>
> Can anyone share their experience on running DRLVM built from SVN?
>
> On my workstations, it requires several fixes in order to run properly:
>
> HARMONY-898 'workaround to get correct hythr.dll' (both for Linux and
> Windows)
> HARMONY-853 'DRLVM+classlib segfaults in hyzlib' (on Linux)
>
> Without HARMONY-898, DRLVM segfaults after reading NULL from TLS.
> Without HARMONY-853, DRLVM segfaults in libhyzlib.so on Linux.
>
> I would suggest that the patches from HARMONY-898 and HARMONY-853 be
> applied,
> because DRLVM doesn't work at all for me without these fixes.
>
> Any other experiences? Is anyone besides me running DRLVM from SVN?
I've built the DRLVM on Linux for the following cases:
1. w/o patches listed above;
2. w/ H-898 patch;
3. w/ H-898 & H-853 patches.
'Segmentation fault' occurs for all cases, unfortunately.
Thanks,
Vladimir.
---------------------------------------------------------------------
> Terms of use : http://incubator.apache.org/harmony/mailing.html
> To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
> For additional commands, e-mail: harmony-dev-help@incubator.apache.org
>
>
Re: [drlvm] Status of drlvm
Posted by Weldon Washburn <we...@gmail.com>.
On 7/17/06, Salikh Zakirov <Sa...@intel.com> wrote:
> Hi,
>
> Can anyone share their experience on running DRLVM built from SVN?
Removing classlib from drlvm build is a big improvement. At this
point, I develop only on windows. No recent experience with the Linux
build. ant seems to have intermittant problems with windows file
paths that have blanks in them. The error messages are very vague.
For example, "set JAVA_HOME=c:\Program Files\jdk..." causes strange
problems.
>
> On my workstations, it requires several fixes in order to run properly:
>
> HARMONY-898 'workaround to get correct hythr.dll' (both for Linux and Windows)
> HARMONY-853 'DRLVM+classlib segfaults in hyzlib' (on Linux)
>
> Without HARMONY-898, DRLVM segfaults after reading NULL from TLS.
> Without HARMONY-853, DRLVM segfaults in libhyzlib.so on Linux.
>
> I would suggest that the patches from HARMONY-898 and HARMONY-853 be applied,
> because DRLVM doesn't work at all for me without these fixes.
I don't see the segfaults. It might be because the initial port of
MMTk does not involve threads.
>
> Any other experiences? Is anyone besides me running DRLVM from SVN?
Getting the debugger to single step through unoptimized jitrino.jet
code shows the build system needs some work. I don't want to plunge
into fixing make files until I have more experience with the build
system but it definitely needs fixing!
>
>
> ---------------------------------------------------------------------
> Terms of use : http://incubator.apache.org/harmony/mailing.html
> To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
> For additional commands, e-mail: harmony-dev-help@incubator.apache.org
>
>
--
Weldon Washburn
Intel Middleware Products Division
---------------------------------------------------------------------
Terms of use : http://incubator.apache.org/harmony/mailing.html
To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
For additional commands, e-mail: harmony-dev-help@incubator.apache.org
Re: [drlvm] Status of drlvm
Posted by Geir Magnusson Jr <ge...@pobox.com>.
Salikh Zakirov wrote:
> Geir Magnusson Jr wrote:
>> I'd like to see us figure out the hythr problem, since 898 is a hack to
>> just work around what seems to be an important issue. can we grind this
>> one out?
>
> A couple of words in defense of HARMONY-898.
> The classlib doesn't depend on a particular hythr implementation.
> In fact, it depends on hythr *interface*, which must be implemented
> somewhere in the system. At the same time, classlib provides its
> own hythr implementation.
>
> DRLVM itself reimplements the subset of hythr interface, thus making
> possible interoperation with classlib, but doesn't uses classlib's hythr implementation.
>
> So, this is not the case of "classlib wants hythr1, and drlvm wants hythr2, and hythr1 and hythr2 conflicts".
> In fact, this is "classlib wants some hythr, and provides hythr1. DLRVM provides hythr2, and conflicts with hythr1".
>
> Thus, IMHO, the issue is not as important as it may seem, because there is no
> issue on "who provides hythr", but there is an issue of competing hythr implementations,
> one of which comes with with DRLVM and works with it, and the other probably works with J9 (I have never checked it myself).
>
This is darn important. We have two implementations of the same thing,
and based on build order, we get differing results. Clearly, one
implementation is *wrong*.
Either :
0) Classlib implements hythread in entirety, and VM uses that.
1) Classlib implements hythread and maybe asks the VM to provide
implementations for some stubs (like we do w/ kernel classes for the
classlib)
2) Classlib depends entirely on the VM to provide.
If there's a difference of opinion on how it should be implemented, or
what functionality should be in there, lets have this out and sooner
than later...
geir
---------------------------------------------------------------------
Terms of use : http://incubator.apache.org/harmony/mailing.html
To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
For additional commands, e-mail: harmony-dev-help@incubator.apache.org
Re: [drlvm] Status of drlvm
Posted by Salikh Zakirov <Sa...@Intel.com>.
Geir Magnusson Jr wrote:
>> Without HARMONY-898, DRLVM segfaults after reading NULL from TLS.
>> Without HARMONY-853, DRLVM segfaults in libhyzlib.so on Linux.
>
> Odd - I don't think I have those problems, since the tests that do run
> pass on my box. Adding a test to show this would be good.
The hythr issue is not predictable, and may or may not work
properly. I think that you've got correct hythr in deploy directory,
and haven't done any rebuilds since then. Or maybe the build
system chooses which hythr to take in some magic manner,
that I cannot grasp.
The easiest way to reproduce hythr issue on Windows (that's how I've come
across it) is to check out and build fresh classlib, and rebuild
DRLVM incrementally.
In any case, having two hythr.dlls copied to a single locations is
definitely a case of race condition with unpredictable result.
> I'd like to see us figure out the hythr problem, since 898 is a hack to
> just work around what seems to be an important issue. can we grind this
> one out?
A couple of words in defense of HARMONY-898.
The classlib doesn't depend on a particular hythr implementation.
In fact, it depends on hythr *interface*, which must be implemented
somewhere in the system. At the same time, classlib provides its
own hythr implementation.
DRLVM itself reimplements the subset of hythr interface, thus making
possible interoperation with classlib, but doesn't uses classlib's hythr implementation.
So, this is not the case of "classlib wants hythr1, and drlvm wants hythr2, and hythr1 and hythr2 conflicts".
In fact, this is "classlib wants some hythr, and provides hythr1. DLRVM provides hythr2, and conflicts with hythr1".
Thus, IMHO, the issue is not as important as it may seem, because there is no
issue on "who provides hythr", but there is an issue of competing hythr implementations,
one of which comes with with DRLVM and works with it, and the other probably works with J9 (I have never checked it myself).
> The zlib seems reasonable, I guess.
Thanks.
---------------------------------------------------------------------
Terms of use : http://incubator.apache.org/harmony/mailing.html
To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
For additional commands, e-mail: harmony-dev-help@incubator.apache.org
Re: [drlvm] Status of drlvm
Posted by Geir Magnusson Jr <ge...@pobox.com>.
Salikh Zakirov wrote:
> Hi,
>
> Can anyone share their experience on running DRLVM built from SVN?
>
> On my workstations, it requires several fixes in order to run properly:
>
> HARMONY-898 'workaround to get correct hythr.dll' (both for Linux and Windows)
> HARMONY-853 'DRLVM+classlib segfaults in hyzlib' (on Linux)
>
> Without HARMONY-898, DRLVM segfaults after reading NULL from TLS.
> Without HARMONY-853, DRLVM segfaults in libhyzlib.so on Linux.
Odd - I don't think I have those problems, since the tests that do run
pass on my box. Adding a test to show this would be good.
I'd like to see us figure out the hythr problem, since 898 is a hack to
just work around what seems to be an important issue. can we grind this
one out?
The zlib seems reasonable, I guess.
>
> I would suggest that the patches from HARMONY-898 and HARMONY-853 be applied,
> because DRLVM doesn't work at all for me without these fixes.
I'll do the second, and maybe you can convince me to do the first :)
But we really should figure this out. If classlib needs one, and drlvm
needs another, we have a problem...
>
> Any other experiences? Is anyone besides me running DRLVM from SVN?
There's no other way to run it since we don't distribute a binary yet.
That would be a good to do this week.
geir
>
>
> ---------------------------------------------------------------------
> Terms of use : http://incubator.apache.org/harmony/mailing.html
> To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
> For additional commands, e-mail: harmony-dev-help@incubator.apache.org
>
>
>
---------------------------------------------------------------------
Terms of use : http://incubator.apache.org/harmony/mailing.html
To unsubscribe, e-mail: harmony-dev-unsubscribe@incubator.apache.org
For additional commands, e-mail: harmony-dev-help@incubator.apache.org