Old Synopsis: 32bit application on FreeBSD-6.3 amd64 gets SIGBUS
New Synopsis: [amd64] 32bit application on FreeBSD-6.3 amd64 gets SIGBUS

Responsible-Changed-From-To: freebsd-bugs->freebsd-amd64
Responsible-Changed-By: linimon
Responsible-Changed-When: Tue Sep 16 02:53:06 UTC 2008
Responsible-Changed-Why:
With a little bit of hesitation, move this one over to the amd64 category,
in hopes that it will be a little more high-profile there.

http://www.freebsd.org/cgi/query-pr.cgi?pr=127397

I can't quickly reproduce this on a roughly week old 6.4-PRERELEASE
with a Turion64x2, though that system has been up for nearly 4 days.

What else, if anything, is running on the system when it fails or
doesn't fail?

Does running lots of (especially large) processes before running your
test reduce the likelihood of the failure occurring?

Off the top of my head this looks like a race condition in the code that
migrates i386 processes between CPUs - possibly a missing or misplaced
memory barrier. Actually tracking this down is likely to be difficult.

--
Peter Jeremy
Please excuse any delays as the result of my ISP's inability to implement
an MTA that is either RFC2821-compliant or matches their claimed behaviour.

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.9 (FreeBSD)

iEYEARECAAYFAkjPbFcACgkQ/opHv/APuIdVtgCgwngCoUtqn0eo+eMwky6q2Xud
9IkAn2/CFsOf6j1cAe6QDIYD2Ch/X+ru
=Jht8
-----END PGP SIGNATURE-----