[GBBopen-developer] Agenda Shell test failed on LispWorks 5.1
Chun Tian (binghe)
binghe.lisp at gmail.com
Tue Apr 15 01:44:18 EDT 2008
Hi again
>
>> I'm new to GBBopen. On LispWorks 5.1 (amd64-linux), the Agenda
>> Shell test cannot pass:
>
> Thanks for the error report. I suspect that multiprocessing has not
> been started in your LispWorks session. I've committed a change to
> GBBopen's Portable Threads interface that warns if multiprocessing
> is not running and added more details to the recursive-lock-attempt
> error message.
You're right that I didn't start multiprocessing in my lispworks
session.
>
>
> Please try rerunning the :agenda-shell-test using the latest SVN
> checkout (first without starting multiprocessing, if that is indeed
> the cause--just to see if the warning and better error messages are
> helpful; then with multiprocessing running if it was not).
Without multiprocessing,
This time I can see a warning:
Warning: You must start multiprocessing on Lispworks by calling
(MP:INITIALIZE-MULTIPROCESSING)
for WITH-TIMEOUT and other thread operations to function properly.
But the error still exist:
; Loading fasl file /home/binghe/Lisp/GBBopen/linux86-lispworks-5.1/
gbbopen/control-shells/test/agenda-shell-test.64ufasl
;;; Starting Agenda Shell Test....
;; Control shell 1 started
Error: A recursive attempt was made by NIL to hold lock #<MP:LOCK
"event-buffer" Unlocked 41507C9123> (held by NIL)
1 (continue) Try loading /home/binghe/Lisp/GBBopen/linux86-
lispworks-5.1/gbbopen/control-shells/test/agenda-shell-test.64ufasl
again.
2 Give up loading /home/binghe/Lisp/GBBopen/linux86-lispworks-5.1/
gbbopen/control-shells/test/agenda-shell-test.64ufasl.
3 Try loading another file instead of /home/binghe/Lisp/GBBopen/
linux86-lispworks-5.1/gbbopen/control-shells/test/agenda-shell-test.
64ufasl.
4 (abort) Return to level 0.
5 Restart top-level loop.
Type :b for backtrace, :c <option number> to proceed, or :? for other
options
With multiprocessing, tests pass well:
CL-USER 17 > (mini-module:compile-module :agenda-shell-
test :propagate :create-dirs)
; Loading fasl file /home/binghe/Lisp/GBBopen/linux86-lispworks-5.1/
tools/portable-threads.64ufasl
;;;
------------------------------------------------------------------------
;;; Portable Threads Interface 2.3
;;;
;;; Developed and supported by the GBBopen Project (http:/
GBBopen.org/)
;;; (See http://GBBopen.org/downloads/LICENSE for license details.)
;;;
------------------------------------------------------------------------
; Loading fasl file /home/binghe/Lisp/GBBopen/linux86-lispworks-5.1/
gbbopen/instances.64ufasl
; Loading fasl file /home/binghe/Lisp/GBBopen/linux86-lispworks-5.1/
gbbopen/find.64ufasl
; Loading fasl file /home/binghe/Lisp/GBBopen/linux86-lispworks-5.1/
gbbopen/control-shells/test/agenda-shell-test.64ufasl
;;; Starting Agenda Shell Test....
;; Control shell 1 started
;; Start-control-shell-ks :activation-predicate executed
;; Start-control-shell-ks :precondition-function executed
;; Start-control-shell-ks :revalidation-predicate executed
;; Start-control-shell-ks :execution-function executed
;; Quiescense ks executed
;; Add-uc-one-to-space-instance-event ks executed
;; Update-uc-one-value-slot-event ks executed
;; Instantiating SPACE-3...
;; Create-uc-one-instance-event ks executed
;; Retriggered KS retriggered
;; Retriggered KS executed
;; Add-uc-one-to-future-space-instance-event ks executed
;; Link-uc-two-event ks executed
;; Obviation KS obviated
;; No executable KSAs remain, exiting control shell
;; Control shell 1 exited: 11 cycles completed
;; Run time: 0.01 seconds
;; Elapsed time: 0 seconds
;;; Agenda Shell Test Completed
:QUIESCENCE
NIL
Thanks.
binghe
>
>
> I've left the event-buffer-lock non-recursive, as I don't think that
> the same thread should be able to grab the lock recursively.
>
--
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.
More information about the GBBopen-developer
mailing list