Date: Fri, 20 May 2016 08:51:56 +0000 From: bugzilla-noreply@freebsd.org To: java@FreeBSD.org Subject: [Bug 209599] SIGSEGV in regression test suite on java/openjdk8 Message-ID: <bug-209599-8522-9a39mo4z8T@https.bugs.freebsd.org/bugzilla/> In-Reply-To: <bug-209599-8522@https.bugs.freebsd.org/bugzilla/> References: <bug-209599-8522@https.bugs.freebsd.org/bugzilla/>
index | next in thread | previous in thread | raw e-mail
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=209599 --- Comment #12 from Andrew Smith <iamasmith.home@gmail.com> --- Reported as a bug that didn't effect Solaris or Windows but did effect Linux at the time. (Bug ID referenced in the unit test code) http://bugs.java.com/bugdatabase/view_bug.do?bug_id=6934977 The useful bit is right at the end. " EVALUATION The issue here is that the load method is touching each page in a native method rather rather than using the Unsafe API. This means the sig handler doesn't know to unwind the stack." Noting also that I ran this unit test under FreeBSD OpenJDK7 independently and it also produces a SIGSEGV. -- You are receiving this mail because: You are the assignee for the bug.help
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-209599-8522-9a39mo4z8T>