Should lock order reversals be reported?
kaduk at MIT.EDU
Tue Jun 18 00:33:55 UTC 2013
On Mon, 17 Jun 2013, Damien Fleuriot wrote:
> +1 to the question, reported 2 a while ago, one for PF and one for FS.
> Gleb quickly fixed the PF one, never had a reply from anyone on FS though.
> I still have it with -CURRENT from last week, can report again.
Those FS LORs (from "10-CURRENT r244183 amd64 multiple lock order
reversals") look to be well known false positives.
E.g., number 285 at http://ipv4.sources.zabbadoz.net/freebsd/lor.html
I have seem the other one as well, though it seems to be not in the top 10
(a bit surprising, actually).
More information about the freebsd-current