Hi Christian,

On Tue, 2005-10-25 at 01:10 +0200, Christian Thalinger wrote:
> Just noticed that the default implementation in GNU classpath has
> problems with gnu.testlet.java.lang.Thread.sleep.  This patch should fix
> the issue, without breaking anything.
>
> 2005-10-25  Christian Thalinger  <[EMAIL PROTECTED]>
> 
>       * vm/reference/java/lang/VMThread.java (sleep): Don't round     
>       ms and pass ns to Object.wait, fixes mauve test.

The round up probably comes from join() which uses a similar way to
wait. And according to the bug pointed to in the comment apparently some
VMs mishandle a zero ms argument (we also had that bug in the past).

> -             vt.wait(ms);
> +             vt.wait(ms, ns);
>               now = System.currentTimeMillis();
>               if (now >= end)
>                 break;

Should we add a ns = 0 after wait() returns here? Whether or not you
think that change is necessary, please feel free to commit this.

Thanks,

Mark

Attachment: signature.asc
Description: This is a digitally signed message part

_______________________________________________
Classpath-patches mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/classpath-patches

Reply via email to