On Wed, 11 Jun 2025 02:05:33 GMT, Sergey Bylokhov <s...@openjdk.org> wrote:
>> src/java.desktop/windows/native/libawt/windows/awt_Window.cpp line 2507: >> >>> 2505: // before our java peer - if we're dispose()d, for instance. >>> 2506: // Return the default screen. >>> 2507: JNI_CHECK_PEER_GOTO(self, ret); >> >> I suggest reordering it slightly - this pattern is commonly used in most >> cases where JNI_CHECK_PEER_GOTO is used: >> >> result...; >> AwtWindow *window = NULL; >> >> PDATA pData; >> JNI_CHECK_PEER_GOTO(self, ret); >> window = (AwtWindow *)pData; > > I still suggest this order. > > Also, please add the new bug ID to the GetGraphicsStressTest. > btw I'm still looking into the new test - it always passes for me. Can we > tweak it to reproduce the bug more reliably? I cant reproduce it locally. I can see it only on CI 20/20 with the new test and 2/20 with the old one ------------- PR Review Comment: https://git.openjdk.org/jdk/pull/25619#discussion_r2139064354
