On 10/2/2013 8:23 AM, Christian Tornqvist wrote:
Hi Lois,
Looks good, might be worth clarifying that this disables
CheckUnhandledOops for both 32 and 64-bit Windows.
Hi Christian,
Thank you for the review. I will update the the bug with this
information and include it in the summary for the fix.
Lois
Thanks,
Christian
*From:*hotspot-runtime-dev-boun...@openjdk.java.net
[mailto:hotspot-runtime-dev-boun...@openjdk.java.net] *On Behalf Of
*Lois Foltan
*Sent:* Tuesday, October 1, 2013 4:51 PM
*To:* hotspot-runtime-...@openjdk.java.net; build-dev@openjdk.java.net
*Subject:* RFR (XS) JDK-8025569: -XX:+CheckUnhandledOops crashes on
Windows
Please review the following fix:
Webrev:
http://cr.openjdk.java.net/~coleenp/bug_jdk8025569/
<http://cr.openjdk.java.net/%7Ecoleenp/bug_jdk8025569/>
Bug: -XX:+CheckUnhandledOops crashes on Windows
https://bugs.openjdk.java.net/browse/JDK-8025569
Summary of fix:
Until JDK-8024364,
<https://bugs.openjdk.java.net/browse/JDK-8024364> (Win/x64:
os::current_frame() and os::fetch_frame_from_context() returns wrong
frame pointer),
is fixed, -XX:+CheckUnhandledOops can not be supported for Windows
fastdebug builds due to the need to have
a legitimate frame returned from os::current_frame() early on in
universe_post_init() to register & deregister oops.
Tests:
JTREG on Windows (in progress)
Thank you, Lois
