8225716: G1 GC: Undefined behaviour in G1BlockOffsetTablePart::block_at_or_preceding

[Aleksey Shipilev]

On 6/17/19 1:32 PM, Andrew Haley wrote:
> This is a crash caused by a bug that I have only ever seen on 32-bit
> x86 GCC.  However, it is possible that it could manifest with other
> compilers and other targets.
> 
> https://bugs.openjdk.java.net/browse/JDK-8225716
> 
> Should we try to get this in to the July CPU? If we don't, people
> targeting 32-bit x86 may have to apply their own patches.

I think it is a fairly small patch, and it fixes the UB, so it makes sense to get it sooner.

-Aleksey