<p dir="ltr">Right, makes sense; presumably Alpha would've tickled quite a few more concurrency issues by now had it been supported :).</p>
<p dir="ltr">sent from my phone</p>
<div class="gmail_quote">On May 13, 2015 6:34 PM, "Thomas Schatzl" <<a href="mailto:thomas.schatzl@oracle.com">thomas.schatzl@oracle.com</a>> wrote:<br type="attribution"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi,<br>
<br>
On Wed, 2015-05-13 at 18:27 -0400, Vitaly Davidovich wrote:<br>
> You're right Thomas, I was mistakenly looking at<br>
> release_store_ptr_*fence* - all good then.<br>
><br>
> Yes, it sounds like it would break on alpha (based on your<br>
> description, I haven't looked at the code), but that's not a supported<br>
> platform anyway right? Otherwise, you'd probably have to introduce a<br>
> new read barrier type for data dependence that's a nop for all but<br>
> alpha so as to not penalize the others.<br>
<br>
No, we do not support Alpha :)<br>
<br>
I do not think there ever has been a Hotspot for Alpha.<br>
<br>
Thanks,<br>
Thomas<br>
<br>
<br>
</blockquote></div>