On Wed, 11 Mar 2026 04:32:40 GMT, David Holmes <[email protected]> wrote:
>> That's OK. This code is only excluding a monitorexit when there is a >> matching monitorenter. The monitorenter is where the null pointer would be >> detected and is not excluded. > > Can't I write bytecode that would invoke monitorexit directly with null on > the stack? Though I guess we don't really care what happens in that case as > it is broken bytecode. In that case, "monitor matching" that the comment below is talking about would have failed. But this is all existing behavior that this PR does not change. ------------- PR Review Comment: https://git.openjdk.org/jdk/pull/30171#discussion_r2916589267
