[Bug 3533] HotSpot generates code with unaligned stack, crashes on SSE operations

[bugzilla-daemon at icedtea.classpath.org]

https://icedtea.classpath.org/bugzilla/show_bug.cgi?id=3533

--- Comment #7 from Maciej S. Szmigiero <mail at maciej.szmigiero.name> ---
(In reply to Andrew John Hughes from comment #6)
> (In reply to Maciej S. Szmigiero from comment #5)
> > (In reply to Andrew John Hughes from comment #4)
> > > Looks like this issue:
> > > 
> > > http://mail.openjdk.java.net/pipermail/hotspot-dev/2018-February/030076.html
> > 
> > As I wrote in the Gentoo bug: it is not this issue.
> > 
> 
> Have you actually tested with this patch?

I guess you mean this patch:
http://cr.openjdk.java.net/~aph/8197429-4/

I have tested it now, and it does not help as a replacement for
"-mincoming-stack-boundary=2".

> 
> > Also, the "icedtea-align-fix.patch" patch that I had posted earlier is not
> > enough, there are more entry points affected.
> > The current, working workaround is to compile icedtea with
> > "-mincoming-stack-boundary=2" {C,CXX}FLAG.
> > 
> > Oracle binaries are affected, too, I would open an OpenJDK bug instead of
> > this icedtea one, but OpenJDK accounts required to do so are not available to
> > the general public.
> 
> Well, yes, because you're not a developer.
> 
> You can open one with https://bugs.java.com/ as I mentioned in the Gentoo
> bug.

I've reported it there now.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.openjdk.java.net/pipermail/distro-pkg-dev/attachments/20180317/d72ebc50/attachment.html>