LOR (vm object - swap_pager swhash)
Jon Noack
noackjr at alumni.rice.edu
Thu Jul 1 04:04:11 PDT 2004
On 07/01/04 05:47, Jon Noack wrote:
> On 07/01/04 05:35, Daniel Lang wrote:
>> Jon Noack wrote on Thu, Jul 01, 2004 at 04:49:33AM -0500:
>> [..]
>>
>>> Yes, but I think the idea was to get it put on the FreeBSD web site
>>> first (Bjoern's been doing a great job, but it simplifies things a
>>> lot to have it on the FreeBSD web site). I started working on this,
>>> but <insert lame excuse> [1].
>> [..]
>>
>> How about just creating a redirect from the FreeBSD.org website?
>>
>> So that in the lor-code the message will refer to s.th. like
>>
>> http://www.freebsd.org/doc/bla/blubb/lor.html
>>
>> and lor.html just contains a META redirect tag to Bjoern's site?
>>
>> So the message can be included right now, without mentioning
>> Bjoern's site in the code. If Bjoern's page (or the content)
>> is actually moved to the FreeBSD webserver, the redirect
>> can simply be removed.
>
>
> *banging head on wall* Perhaps when this infernal banging stops I'll be
> able to come up with great ideas like this.
>
> Perhaps the best bet is to add a FAQ. The FAQ would point to Bjoern's
> site and the LOR-message would point to the FAQ entry.
Again, my docproj foo is weak, but see the attached FAQ.
Jon
-------------- next part --------------
Index: book.sgml
===================================================================
RCS file: /home/ncvs/doc/en_US.ISO8859-1/books/faq/book.sgml,v
retrieving revision 1.626
diff -r1.626 book.sgml
4381a4382,4426
>
> <qandaentry>
> <question id="lock-order-reversal">
> <para>What is a <errorname>lock order reversal</errorname>?</para>
> </question>
>
> <answer>
> <para>&a.rwatson; answered this question very succinctly on
> the freebsd-current list in a thread entitled <quote><ulink
> url="http://docs.freebsd.org/cgi/getmsg.cgi?fetch=65165+0+/usr/local/www/db/text/2003/freebsd-current/20031221.freebsd-current">lock
> order reversals - what do they mean?</ulink></quote>
>
> <blockquote>
> <attribution>&a.rwatson; on freebsd-current, December 14,
> 2003</attribution>
>
> <para>These warnings are generated by Witness, a run-time lock
> diagnostic system found in FreeBSD 5-CURRENT kernels (but
> removed in releases). You can read more about Witness in the
> WITNESS(4) man page, which talks about its capabilities. Among
> other things, Witness performs run-time lock order verification
> using a combination of hard coded lock orders, and run-time
> detected lock orders, and generates console warnings when lock
> orders are violated. The intent of this is to detect the
> potential for deadlocks due to lock order violations; it's worth
> observing that Witness is actually slightly conservative, and so
> it's possible to get false positives. In the event that Witness
> is accurately reporting a lock order problem, it's basically
> saying "If you were unlucky, a deadlock would have happened
> here". There are a couple of "well known" false positives,
> which we need to do a better job of documenting to prevent
> spurious reports. The non-well-known ones typically correspond
> to bugs in newly added locking, as lock order reversals usually
> get fixed pretty quickly because Witness is busy generating
> warnings :-).</para>
>
> <note>
> <para>See <ulink
> url="http://sources.zabbadoz.net/freebsd/lor.html">Bjoern
> Zeeb's lock order reversal page</ulink> for the status of
> known lock order reversals.</para>
> </note>
> </blockquote>
> </answer>
> </qandaentry>
More information about the freebsd-current
mailing list