You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@harmony.apache.org by Weldon Washburn <we...@gmail.com> on 2007/03/27 00:12:48 UTC
[drlvm][threading] harmony-3065 problems
Unfortunately, h3065 patch fails the jitrino.OPT tests on windowsxp. Since
this is a performance patch and has failed three or four attempts to commit,
how about recoding this patch as "later"? This will give us a chance to
focus on basic stability problems. Once the system is stable, we can go
back and fix the performance issues.
--
Weldon Washburn
Intel Enterprise Solutions Software Division
Re: [drlvm][threading] harmony-3065 problems
Posted by Weldon Washburn <we...@gmail.com>.
H3065 has been committed. I ran "build test" a second time on win32 and it
passed this time. It seems there is a stability bug in java.lang.Thread.
We need to focus on stability soon.
On 3/27/07, Weldon Washburn <we...@gmail.com> wrote:
>
> Peter,
> I will rerun "build test". I only ran it once so far. Maybe the bug is
> intermittant?? Below is the console output. It should be the same info as
> in the /reports directories. For some strange reason, I am having problems
> finding the correct index.html. Maybe it moved?
>
>
> On 3/27/07, Weldon Washburn <we...@gmail.com> wrote:
> >
> > I will track down the logs right now. I have no problem continuing work
> > on h3065 (I thought you might want an easy excuse to set it aside.)
>
>
> [junit] Tests run: 48, Failures: 0, Errors: 0, Time elapsed: 14.371 sec
> [junit] Tests run: 3, Failures: 0, Errors: 0, Time elapsed: 0.771 sec
> [junit] Tests run: 91, Failures: 1, Errors: 0, Time elapsed: 44.916 sec
> [junit] TEST java.lang.ThreadTest FAILED
> [junit] Tests run: 3, Failures: 0, Errors: 0, Time elapsed: 0.361 sec
> [junit] Tests run: 16, Failures: 0, Errors: 0, Time elapsed: 1.342 sec
>
>
>
> report-kernel-test:
> [junitreport] Transform time: 20961ms
> [junitreport] Transform time: 23116ms
> [junitreport] Transform time: 19377ms
>
> check-kernel-test:
> [echo] ***************************
> [echo] *** Kernel tests status ***
> [echo] ***************************
>
> [echo] Client mode JIT (default) PASSED
> [echo] jitrino.OPT FAILED
> [echo] interpreter PASSED
>
> [echo] Please find detailed results under
> C:\t_harmony\trunk\working_vm\bui
> ld\win_ia32_msvc_debug\semis\kernel.tests/reports directory.
>
> BUILD FAILED
> C:\t_harmony\trunk\working_vm\build\make\build.xml:373: The following
> error occu
> rred while executing this line:
> C:\t_harmony\trunk\working_vm\build\make\build_component.xml:72: The
> following e
> rror occurred while executing this line:
>
> C:\t_harmony\trunk\working_vm\build\win_ia32_msvc_debug\semis\build\targets\kern
> el.test.xml:52: There were test failures.
>
> Total time: 104 minutes 32 seconds
> C:\t_harmony\trunk\working_vm\build>
>
>
>
>
>
> On 3/27/07, Peter Novodvorsky <peter.novodvorsky@gmail.com > wrote:
> > >
> > > It would be very disappointing to leave it, because me and Pavel
> > > Afremov already gave too much time to this patch. However, in general,
> > >
> > > I agree that we should put off performance patches and concentrate on
> > > architecture and stability.
> > >
> > > Weldon, could you please send us the log with and error, because I run
> > > tests with patch from H-3065 applied and see no problems.
> > >
> > > Thanks,
> > > Peter.
> > >
> > > On 3/27/07, Weldon Washburn <we...@gmail.com> wrote:
> > > > Unfortunately, h3065 patch fails the jitrino.OPT tests on
> > > windowsxp. Since
> > > > this is a performance patch and has failed three or four attempts to
> > > commit,
> > > > how about recoding this patch as "later"? This will give us a
> > > chance to
> > > > focus on basic stability problems. Once the system is stable, we
> > > can go
> > > > back and fix the performance issues.
> > > >
> > > > --
> > > > Weldon Washburn
> > > > Intel Enterprise Solutions Software Division
> > > >
> > >
> >
> >
> >
> > --
> > Weldon Washburn
> > Intel Enterprise Solutions Software Division
> >
>
>
>
> --
> Weldon Washburn
> Intel Enterprise Solutions Software Division
>
--
Weldon Washburn
Re: [drlvm][threading] harmony-3065 problems
Posted by Weldon Washburn <we...@gmail.com>.
Peter,
I will rerun "build test". I only ran it once so far. Maybe the bug is
intermittant?? Below is the console output. It should be the same info as
in the /reports directories. For some strange reason, I am having problems
finding the correct index.html. Maybe it moved?
On 3/27/07, Weldon Washburn <we...@gmail.com> wrote:
>
> I will track down the logs right now. I have no problem continuing work
> on h3065 (I thought you might want an easy excuse to set it aside.)
[junit] Tests run: 48, Failures: 0, Errors: 0, Time elapsed: 14.371 sec
[junit] Tests run: 3, Failures: 0, Errors: 0, Time elapsed: 0.771 sec
[junit] Tests run: 91, Failures: 1, Errors: 0, Time elapsed: 44.916 sec
[junit] TEST java.lang.ThreadTest FAILED
[junit] Tests run: 3, Failures: 0, Errors: 0, Time elapsed: 0.361 sec
[junit] Tests run: 16, Failures: 0, Errors: 0, Time elapsed: 1.342 sec
report-kernel-test:
[junitreport] Transform time: 20961ms
[junitreport] Transform time: 23116ms
[junitreport] Transform time: 19377ms
check-kernel-test:
[echo] ***************************
[echo] *** Kernel tests status ***
[echo] ***************************
[echo] Client mode JIT (default) PASSED
[echo] jitrino.OPT FAILED
[echo] interpreter PASSED
[echo] Please find detailed results under
C:\t_harmony\trunk\working_vm\bui
ld\win_ia32_msvc_debug\semis\kernel.tests/reports directory.
BUILD FAILED
C:\t_harmony\trunk\working_vm\build\make\build.xml:373: The following error
occu
rred while executing this line:
C:\t_harmony\trunk\working_vm\build\make\build_component.xml:72: The
following e
rror occurred while executing this line:
C:\t_harmony\trunk\working_vm\build\win_ia32_msvc_debug\semis\build\targets\kern
el.test.xml:52: There were test failures.
Total time: 104 minutes 32 seconds
C:\t_harmony\trunk\working_vm\build>
On 3/27/07, Peter Novodvorsky <pe...@gmail.com> wrote:
> >
> > It would be very disappointing to leave it, because me and Pavel
> > Afremov already gave too much time to this patch. However, in general,
> > I agree that we should put off performance patches and concentrate on
> > architecture and stability.
> >
> > Weldon, could you please send us the log with and error, because I run
> > tests with patch from H-3065 applied and see no problems.
> >
> > Thanks,
> > Peter.
> >
> > On 3/27/07, Weldon Washburn <we...@gmail.com> wrote:
> > > Unfortunately, h3065 patch fails the jitrino.OPT tests on
> > windowsxp. Since
> > > this is a performance patch and has failed three or four attempts to
> > commit,
> > > how about recoding this patch as "later"? This will give us a chance
> > to
> > > focus on basic stability problems. Once the system is stable, we can
> > go
> > > back and fix the performance issues.
> > >
> > > --
> > > Weldon Washburn
> > > Intel Enterprise Solutions Software Division
> > >
> >
>
>
>
> --
> Weldon Washburn
> Intel Enterprise Solutions Software Division
>
--
Weldon Washburn
Intel Enterprise Solutions Software Division
Re: [drlvm][threading] harmony-3065 problems
Posted by Weldon Washburn <we...@gmail.com>.
I will track down the logs right now. I have no problem continuing work on
h3065 (I thought you might want an easy excuse to set it aside.)
On 3/27/07, Peter Novodvorsky <pe...@gmail.com> wrote:
>
> It would be very disappointing to leave it, because me and Pavel
> Afremov already gave too much time to this patch. However, in general,
> I agree that we should put off performance patches and concentrate on
> architecture and stability.
>
> Weldon, could you please send us the log with and error, because I run
> tests with patch from H-3065 applied and see no problems.
>
> Thanks,
> Peter.
>
> On 3/27/07, Weldon Washburn <we...@gmail.com> wrote:
> > Unfortunately, h3065 patch fails the jitrino.OPT tests on
> windowsxp. Since
> > this is a performance patch and has failed three or four attempts to
> commit,
> > how about recoding this patch as "later"? This will give us a chance to
> > focus on basic stability problems. Once the system is stable, we can go
> > back and fix the performance issues.
> >
> > --
> > Weldon Washburn
> > Intel Enterprise Solutions Software Division
> >
>
--
Weldon Washburn
Intel Enterprise Solutions Software Division
Re: [drlvm][threading] harmony-3065 problems
Posted by Peter Novodvorsky <pe...@gmail.com>.
It would be very disappointing to leave it, because me and Pavel
Afremov already gave too much time to this patch. However, in general,
I agree that we should put off performance patches and concentrate on
architecture and stability.
Weldon, could you please send us the log with and error, because I run
tests with patch from H-3065 applied and see no problems.
Thanks,
Peter.
On 3/27/07, Weldon Washburn <we...@gmail.com> wrote:
> Unfortunately, h3065 patch fails the jitrino.OPT tests on windowsxp. Since
> this is a performance patch and has failed three or four attempts to commit,
> how about recoding this patch as "later"? This will give us a chance to
> focus on basic stability problems. Once the system is stable, we can go
> back and fix the performance issues.
>
> --
> Weldon Washburn
> Intel Enterprise Solutions Software Division
>