RFR(S): 8231405: [Shenandoah] guarantee(d != NULL) failed: Null dominator info
Roman Kennke
rkennke at redhat.com
Fri Sep 27 08:37:15 UTC 2019
Looks good to me, thanks!
Roman
> http://cr.openjdk.java.net/~roland/8231405/webrev.00/
>
> A load barrier can sometimes be scheduled right after a call to the
> rethrow stub. The rethrow stub has more projection than a usual call and
> the logic to duplicate load barriers along the exception/fallthrough
> paths can't work. There can't be any dependency between the load barrier
> and the rethrow stub so the fix moves the barrier above the call.
>
> Roland.
>
More information about the hotspot-compiler-dev
mailing list