Vladimir Ivanov wrote:
On 4/27/07, Xiao-Feng Li <[EMAIL PROTECTED]> wrote:
On 4/27/07, Vladimir Ivanov <[EMAIL PROTECTED]> wrote:
> Thanks, it looks OK now.
In my testing, the H3730 has a seg fault in Linux64. Can anyone
repeat? Thanks, xiaofeng
Yes, I add comment to 3730.
But now this test at least compiled :)
Thanks, Vladimir
I believe this test should be excluded for x86_64 because JVMTI in JIT
mode is not implemented for this architecture yet. The test itself is
for a bug in JVMTI in JIT mode, so there is no sense to run this test on
interpreter.
I'm going to update the exclude list soon.
(btw, it's tested with GCv4.1).
> Vladimir
>
> On 4/27/07, Stepan Mishura <[EMAIL PROTECTED]> wrote:
> > Fixed at r532967. Please verify if it works for you.
> >
> > -Stepan.
> >
> > On 4/27/07, Vladimir Ivanov wrote:
> > > Hello everyone,
> > > CC/CI reports failure of reg.test project due to compilation
error on
> > > my Linux x86_64 + gcc version 3.3.3 (SuSE Linux):
> > > ------------------------------------
> > > drlvm/trunk/src/test/regression/H3730/StepM2N.cpp: In
> > > function `void SingleStep(jvmtiEnv*, JNIEnv*, _jobject*,
_jmethodID*, long
> > > long int)':
> > >
/export/cruise/trunk/cc/projects/drlvm/trunk/src/test/regression/H3730/StepM2N.cpp:228:
> > > error: ISO
> > > C++ forbids declaration of `hit_count' with no type/
> > > -----------------------------------
> > >
> > > Note this tests was commited yesterday:
> > > Author: gshimansky
> > > Date: Thu Apr 26 08:03:52 2007
> > > New Revision: 532770
> > >
> > > Could someone fix it?
> > >
> > > Thanks, Vladimir
> > >
> >
> >
> > --
> > Stepan Mishura
> > Intel Enterprise Solutions Software Division
> >
>
--
http://xiao-feng.blogspot.com
--
Gregory
