RFR: 8316422: TestIntegerUnsignedDivMod.java triggers "invalid layout" assert in FrameValues::validate [v2]
Tobias Hartmann
thartmann at openjdk.org
Wed Nov 8 09:24:02 UTC 2023
On Thu, 2 Nov 2023 04:48:29 GMT, Dean Long <dlong at openjdk.org> wrote:
>> Don't truncate caller exception state stack in c1, even if no handlers are found. This makes c1 follow the example of c2. See bug for details.
>
> Dean Long has updated the pull request incrementally with two additional commits since the last revision:
>
> - add test
> - fix assert
Looks reasonable to me but someone with more expertise in C1 and exception handling (@veresov, @tkrodriguez ?) should review this.
test/hotspot/jtreg/compiler/exceptions/TestDeoptExceptionState.java line 28:
> 26: * @bug 8316422
> 27: * @summary Test exception state used for deoptimization.
> 28: * @run main/othervm -XX:+VerifyStack -Xcomp -XX:TieredStopAtLevel=1 -XX:+DeoptimizeALot
`VerifyStack` and `DeoptimizeALot` are debug flags.
-------------
Marked as reviewed by thartmann (Reviewer).
PR Review: https://git.openjdk.org/jdk/pull/16297#pullrequestreview-1719765767
PR Review Comment: https://git.openjdk.org/jdk/pull/16297#discussion_r1386251158
More information about the hotspot-compiler-dev
mailing list