harmony-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Alexey Varlamov" <alexey.v.varla...@gmail.com>
Subject Re: [build] Use Sun 5.0_8 or Eclipse Compiler for automated builds
Date Thu, 07 Sep 2006 15:29:31 GMT
2006/9/7, Geir Magnusson Jr. <geir@pobox.com>:
>
>
> Alexey Varlamov wrote:
> > 2006/9/7, Oliver Deakin <oliver.deakin@googlemail.com>:
> >> Alexey Varlamov wrote:
> >> > 2006/9/7, Oliver Deakin <oliver.deakin@googlemail.com>:
> >> >>
> >> >>
> >> >> Nathan Beyer wrote:
> >> >> >
> >> >> >> -----Original Message-----
> >> >> >> From: Richard Liang [mailto:richard.liangyx@gmail.com]
> >> >> >>
> >> >> >>  Sun compiler (5.0_8) also has some unexpected behavior. See[1]
> >> >> >>
> >> >> >> [1]http://mail-archives.apache.org/mod_mbox/incubator-harmony-
> >> >> >> dev/200608.mbox/%3c44F2DADE.5060900@gmail.com%3e
> >> >> >>
> >> >> >>
> >> >> >
> >> >> > I've found that a second or third "ant build" takes care of it.
> >> >> What I've
> >> >> > noticed though is that the failures are all linking to elements
> >> >> from classes
> >> >> > in luni-kernel. I'm looking at it, but I think there are some
> >> >> issues with
> >> >> > what's put on the bootclasspath. I've already found one discrepancy
> >> >> between
> >> >> > some of our 'luni-kernel' stubs and the spec. The Method.invoke()
> >> >> uses a
> >> >> > vararg and our stub didn't have this.
> >> >> >
> >> >>
> >> >> I spotted this as well, was preparing a patch for it - unless you want
> >> >> to just go ahead and
> >> >> fix it?
> >> >
> >> > Could you please restore j.l.Thread.setInterruptAction() & co, as well?
> >>
> >> Isn't setInterruptAction() a private method? Why do we need this in the
> >> luni-kernel
> >> stubs?
> >
> > As long as luni-kernel stubs actually document VMI, we should keep all
> > related methods there (until someone can suggest better place, which I
> > doubt - you know, the code is best doc itslef and any separate docs
> > become outdated sooner or later). Real-life example: I noted
> > NoSuchMethodException in DRLVM trace logs, and was quite puzzled what
> > it is about - so had to browse dev-list history...
>
> When I first saw this suggestion, I too thought that it shouldn't go in
> Thread.
Yep, I let the original discussion slip and still not comfortable
about current state. I especially dislike using brute force as
reflection to overcome design flaws...
I will try to look into this closer, when time permits.

> But where is it used or documented as part of VMI?  I couldn't find it...
Used in java.nio.channels.spi.AbstractInterruptibleChannel, the only
documentation was in javadocs to j.l.Thread.setInterruptAction().

>
> geir
>
> >
> >>
> >> As far as I can see, that is the only method that has been removed
> >> recently from the
> >> Thread class stub - were there any others that I have missed that are
> >> required?
> > Simply reapply luni-kernel part of HARMONY-635. Thank you!
> >
> > --
> > Regards,
> > Alexey
> >
> > ---------------------------------------------------------------------
> > 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


Mime
View raw message