[prev in list] [next in list] [prev in thread] [next in thread] 

List:       openjdk-serviceability-dev
Subject:    Re: RFR(XXS): 8214105: Invalid bit tests in jtreg
From:       Severin Gehwolf <sgehwolf () redhat ! com>
Date:       2018-11-22 9:03:22
Message-ID: e714b4ba5d1c798cb1bc827d3b6202f06011485f.camel () redhat ! com
[Download RAW message or body]

On Wed, 2018-11-21 at 09:20 +1000, David Holmes wrote:
> On 21/11/2018 9:04 am, JC Beyler wrote:
> > Also +1 for the fix,
> > 
> > If the submit repo is enough for testing, I can do the legwork to test 
> > it and push it once it passes,
> 
> Not sure if submit-repo will do much JVM TI testing ... I think we need 
> tier 3 for JVM TI and pretty sure submit-repo is only tier 1.
> 
> But thanks for the offer.

Thanks everyone. I'll sponsor this for Simon.

Thanks,
Severin

> David
> 
> > Jc
> > 
> > Ps: same for the other one he submitted
> > 
> > On Tue, Nov 20, 2018 at 2:58 PM David Holmes <david.holmes@oracle.com 
> > <mailto:david.holmes@oracle.com>> wrote:
> > 
> >     +1 on the fix.
> > 
> >     Simon is neither Committer nor Author so will need a sponsor.
> > 
> >     Thanks,
> >     David
> > 
> >     On 21/11/2018 4:36 am, serguei.spitsyn@oracle.com
> >     <mailto:serguei.spitsyn@oracle.com> wrote:
> >      > Hi Simon,
> >      >
> >      > The fix looks good.
> >      > Thank you for taking care about it!
> >      >
> >      > Questions:
> >      >    - Do you have an Author status?
> >      >    - You probably need a sponsor for this, do you?
> >      >
> >      > Thanks,
> >      > Serguei
> >      >
> >      >
> >      > On 11/20/18 06:34, Simon Tooke wrote:
> >      >> While compiling the JDK with GCC 8.1, I discovered an invalid
> >     bit test
> >      >> in
> >      >>
> >     test/hotspot/jtreg/serviceability/jvmti/StartPhase/AllowedFunctions/libAllowedFunctions.c.
> > 
> >      >>
> >      >>
> >      >>
> >      >>      (status & JVMTI_CLASS_STATUS_INITIALIZED) == 1
> >      >>
> >      >> Which only has a chance of being true if
> >     JVMTI_CLASS_STATUS_INITIALIZED
> >      >> has a value 1 (its actual value is 4, but that's beside the point).
> >      >> My proposed fix is to test for != 0 instead.  I chose this
> >     instead of
> >      >> testing for equality to JVMTI_CLASS_STATUS_INITIALIZED purely for
> >      >> cosmetic reasons.
> >      >>
> >      >> Bug: https://bugs.openjdk.java.net/browse/JDK-8214105
> >      >> webrev:
> >      >>
> >     http://cr.openjdk.java.net/~sgehwolf/webrevs/stooke/JDK-8214105/01/webrev/
> > 
> >      >>
> >      >>
> >      >> Please let me know what you think.
> >      >>
> >      >> Thanks,
> >      >> -Simon
> >      >>
> >      >>
> >      >
> > 
> > 
> > 
> > -- 
> > 
> > Thanks,
> > Jc

[prev in list] [next in list] [prev in thread] [next in thread]