- armedbear-devel - mailman3.common-lisp.net

Re: [armedbear-devel] XCL and ASDF
by Faré 13 Apr '11

13 Apr '11

> Thanks! I've committed asdf.lisp 2.014.2 to XCL's tree on github. Thanks! > Fixed on github. (quit :status 42) or (exit :status 42) should work now. Excellent. >> * Your make-pathname erroneously merges directories as in >> merge-pathnames. It shouldn't. > > Can you give an example of this misbehavior, and what the correct > behavior should be? > for i in sbcl ccl clisp ecl ccl abcl xcl gclcvs allegro scl ; do echo $i ; cl-launch -l $i -s asdf -f ~/bug/xcl-make-pathname.lisp; done 2> /dev/null |& tee /tmp/foo Everyone correctly replies #p"" except abcl and xcl that reply #p"/home/fare/". the bug file contains: (defparameter *h* "/home/fare/") (defparameter *make* (make-pathname :directory nil :defaults *h*)) (format t "~S~%" *make*) I wonder why I didn't detect the bug in abcl earlier. Cc'ing the abcl mailing-list. Also, it's very annoying for cl-launch that xcl doesn't have a --no-inform mode. Even more annoying when some output goes to standard-output *after* I print my stuff and call quit. Can you fix this? [ François-René ÐVB Rideau | Reflection&Cybernethics | http://fare.tunes.org ] In a centralized society social status is a zero-sum game along a single hierarchy. That's why politics can never help the weak, only swap roles.

2 2

[armedbear-devel] Function.execute and in-package
by Pablo Duboue 12 Apr '11

12 Apr '11

Hello, How can I issue a Function.execute so that it executes in the context of a given package? That is, I have an instance of the Function class, obtained from a .getSymbolFunction() call on a findAccessibleSymbol over a Package instance, per the examples in the source tree. This function call executes fine, but it runs on package "CL". I need it to run on package, say, "FOO". I have tried a number of things, to no avail: * Issuing an interpreter.evaluate("(in-package "FOO")") beforehand. That seemed very reasonable, but doesn't work. Looking into the code, it seems related to packages being special bindings and said bindings being reset between calls to evaluate. * Fiddling with the thread.markSpecialBindings calls which seem to reset in-package bindings in Lisp, Interpreter and Function. I haven't had much success there, am I on the right track at all? * Trying to use other calls instead of function.execute, which include an environment and a thread, that was very cumbersome (and didn't work either). Might this work at all...? * Using the JSR-223 interface. That did work, but as I'm planning to put this in a Web service, I don't want any interpreter on the way (for security reasons). By the way, the JSR-223 page is the only instance of "in-package" in the whole wiki. (The mailing list search function for 'in-package' split it as two words and return half the mailing list as output.) Any pointers will be appreciated and I'll add the information to the wiki afterward (suggesting where to put it will also be appreciated). You can also reply on #abcl, my nick there is DrDub. Gracias! P.

2 3

[armedbear-devel] ABCL and XCVB
by Faré 11 Apr '11

11 Apr '11

Dear ABCL developers, did you ever get to look at the make-pathname bug? for i in sbcl ccl clisp scl allegro ecl gclcvs abcl xcl ; do cl-launch -l $i -iw '(make-pathname :directory nil :defaults #p"/home/fare/")' ; done Also, I'd like to support ABCL in XCVB, and for that I need something more powerful than run-shell-command, since I need to be able to grab the output. I support I could write to a temporary file with a shell redirect, but it's dirty and racy and requires figuring out where the temporary files are to be located. Could you implement a variant of run-program? It might require some trickery to support both Unix and Windows, but it would be very useful, probably beyond just XCVB. [ François-René ÐVB Rideau | Reflection&Cybernethics | http://fare.tunes.org ] You may easily play a joke on a man who likes to argue — agree with him. — Edgar Waston Howe

4 4

[armedbear-devel] Fwd: ASDF 2.014 released
by Faré 28 Mar '11

28 Mar '11

[Note: I have skipped ECL and Allegro maintainers from the notification, since I know they subscribe to the announcement. You too can subscribe to the asdf-announce mailing-list or its gmane mirror gmane.lisp.asdf.announce and ask me to take your list off the Cc:] ---------- Forwarded message ---------- From: Faré <fahree(a)gmail.com> Date: 28 March 2011 14:08 Subject: ASDF 2.014 released To: asdf-announce(a)common-lisp.net, ASDF-devel <asdf-devel(a)common-lisp.net> Dear ASDF users, I just blessed ASDF 2.013.7 as 2.014. As compared to 2.013, it includes a significant quality item that makes it essential for everyone who distributes ASDF without an implementation (e.g. Quicklisp). It also includes better SCL support, cosmetic improvements, and robustification of some corner cases. Tests have been added to prevent the regressions observed in the 2.013 release. * Fixed bug preventing upgrade from 2.000 to 2.008 (test added) * Better support for URL pathnames on SCL (test added) * Robustified defsystem-depends-on * Cosmetic changes: * better more portably readable error messages. * remove forward references, again (test added) * canonicalize system names earlier (slight robustness improvement) As usual, I invite people who distribute ASDF to update the version they ship, be it with their implementations, with their software distribution, etc. Happy Happy Joy Joy [ François-René ÐVB Rideau | Reflection&Cybernethics | http://fare.tunes.org ] "Don't worry about people stealing your ideas. If your ideas are any good, you'll have to ram them down people's throats." — Howard Aiken

2 1

[armedbear-devel] Improvements to our Gray streams implementation
by Mark Evenson 28 Mar '11

28 Mar '11

As part of the long continuing road to Hunchentoot, I am pleased to announce that with [r13255][1], ABCL can now pass the flexi-streams-1.0.7 test suite as long as [trivial-gray-streams is patched to recognize the new FILE-POSITION generic][2]. [1]; http://trac.common-lisp.net/armedbear/changeset/13255 [2]: http://detroit.slack.net/~evenson/abcl/trivial-gray-streams-abcl-20110320a.… -- "A screaming comes across the sky. It has happened before, but there is nothing to compare to it now."

1 1

[armedbear-devel] Patches for metatilities-base and lift
by Mark Evenson 22 Mar '11

22 Mar '11

Patches to get [metatilities-base][1] and [lift][2] as distributed by Quicklisp MAR2011. We really need to either a) make our own Quicklisp distribution or (my preference) b) make a QUICKLISP-PATCH system to apply patches as QL:QUICKLOAD happens. [1]: http://detroit.slack.net/~evenson/abcl/metatilities-base-20101006-darcs-abc… [2]: http://detroit.slack.net/~evenson/abcl/lift-20110320-git-abcl-20110322.diff -- "A screaming comes across the sky. It has happened before, but there is nothing to compare to it now."

3 8

Re: [armedbear-devel] [Bordeaux-threads-devel] [Updated4] Patch for ABCL against BORDEAUX-THREADS HEAD
by Chun Tian (binghe) 21 Mar '11

21 Mar '11

Hi, Mark I'm interesting in your work. First, I'd like to know, how to do a "condition variable wait with timeout" with your exist work, Second, do you think it's possible to make ABCL supporting some atomic operations, something like ATOMIC-INCF or ATOMIC-PUSH, this can be very useful to implement some lockless algorithms. Regards, Chun Tian (binghe) 在 2011-3-18，15:50， Mark Evenson 写道： > On 3/2/11 8:50 PM, Martin Simmons wrote: > >> I think that the implementation of condition-wait has some major problems: > > Thanks very much for the comments. > > I have an [updated patch][1] that attempts to address these issues for > further review for inclusion in BORDEAUX-THREADS. > > [1]: > http://detroit.slack.net/~evenson/abcl/bordeaux-threads-abcl-20110318a.diff > >> 1) If no other threads are trying to claim the lock, then condition-wait will >> return immediately rather than waiting. > > I couldn't reproduce this behavior. In ABCL, the THREADS:OBJECT-WAIT > implementation is mainly a simple wrapping of the underlying > [java.lang.Object.wait()][2] which clearly should wait. > > [2]: > http://download.oracle.com/javase/6/docs/api/java/lang/Object.html#wait%28%… > >> 2) If two threads are waiting in condition-wait and two other threads call >> condition-notify, then it is possible that only one thread will return from >> condition-wait because the call to acquire-lock in one of them might return >> nil causing it to wait again. > > The DO looping construct has been replaced with a simpler subsequent > BT:ACQUIRE-LOCK that should solve this bug. > >> 3) If condition-notify is called by one thread when a waiting thread is just >> about to enter the threads:synchronized-on form (but before it gets >> synchronized), then the notify will be lost. This happens because the >> underlying threading primitives have no "memory" of calls to notify when >> nothing it waiting (which is also the expected semantics for POSIX condition >> variables BTW). > > THREADS:SYNCHRONIZED-ON is now the first form executed by > BT:CONDITION-WAIT so the window here is considerably narrowed. > >> Also, I think condition-notify should be using threads:object-notify rather >> than threads:object-notify-all. > > We now use THREADS:OBJECT-NOTIFY. > > > > > -- > "A screaming comes across the sky. It has happened before, but there > is nothing to compare to it now." > > > _______________________________________________ > Bordeaux-threads-devel mailing list > Bordeaux-threads-devel(a)common-lisp.net > http://common-lisp.net/cgi-bin/mailman/listinfo/bordeaux-threads-devel

3 3

[armedbear-devel] ASDF 2.013 released
by Faré 20 Mar '11

20 Mar '11

Dear Armed Bear developers, congratulations for your many releases since I last contacted you. I've just released ASDF 2.013. I don't know which version you're using, but there have been numerous small bug fixes and improvements in ASDF. For the sake of sharing those improvements, I think it would be a good time to push ASDF to all Lisp implementations. Could you include it in yours? While abcl is used as part of my automated test suite, I don't often upgrade my abcl copy (currently 0.22.0) -- it would help if there were a debian package. Please make sure ASDF suits your purposes, and contact me with any patch you use for inclusion upstream. PS: I created a mailing-list asdf-announce. Maybe someone amongst you could join it, so I don't need to push my announces to this mailing-list. Or maybe I can just Cc: my announce to this mailing-list and not bother with further explanations? [ François-René ÐVB Rideau | Reflection&Cybernethics | http://fare.tunes.org ] The common argument that crime is caused by poverty is a kind of slander on the poor. — H. L. Mencken ---------- Forwarded message ---------- From: Faré <fahree(a)gmail.com> Date: 20 March 2011 00:29 Subject: ASDF 2.013 released To: asdf-announce(a)common-lisp.net Dear Common Lisp Users, I just blessed ASDF 2.012.12 as 2.013. It includes many small improvements since previous release, for portability, robustness, usability. For the sake of making these improvements available to all users, I invite all vendors and packagers to update their software to include this latest release. Improvements since 2.012: * Fixes and cleanups for CLISP, ECL, SBCL. * Added support for two legacy platforms: (R)MCL, Genera * Cleanup features: now using #+asdf-unix and #+asdf-window * Many cleanups and improvements regarding our portable pathname layer * Fixes a regression in source-registry scanning from 2.011.4 * Remember *source-registry-parameter* and *output-translations-parameter* * Export and document coerce-pathname (f.k.a merge-component-name-type) * allow description and long-description for all components * Several documentation updates Happy hacking, [ François-René ÐVB Rideau | Reflection&Cybernethics | http://fare.tunes.org ] I hold that America, Champion of the World is a betrayal of its humble beginnings as One Nation Underdog.

2 2

Re: [armedbear-devel] [Updated4] Patch for ABCL against BORDEAUX-THREADS HEAD
by Mark Evenson 18 Mar '11

18 Mar '11

On 3/18/11 4:41 PM, Ville Voutilainen wrote: > On 18 March 2011 17:37, Mark Evenson<evenson(a)panix.com> wrote: >> This was the previous implementation, which was the subject of the >> criticism: >> >> (defun condition-wait (condition lock) >> (check-type condition condition-variable) >> (release-lock lock) >> (do () >> ((acquire-lock lock nil)) >> (threads:synchronized-on condition >> (threads:object-wait condition)))) > > If this is directly mapped to a pthread implementation, the loop must be there. I think it is slightly different, as the semantics of BT:CONDITION-WAIT are merely that the lock is always reacquired before returning. The caller of BT:CONDITION-WAIT is required to check for spurious wakeups. > >> I believe the race condition is on the the synchronization on the condition >> variable, not on the mutex lock. > > The condition variable needs to use the mutex lock given, not a separate lock. The condition variable interface is using the mutex lock specified in BT:CONDITION-WAIT. For the ABCL implementation, I'm using the Java monitor associated with the condition variable for the additional synchronization needed to implement BT:CONDITION-NOTIFY which couldn't use the mutex lock without contention problems. But definitely check my logic here, as it has been faulty before. -- "A screaming comes across the sky. It has happened before, but there is nothing to compare to it now."

2 3

Re: [armedbear-devel] [Updated4] Patch for ABCL against BORDEAUX-THREADS HEAD
by Mark Evenson 18 Mar '11

18 Mar '11

On 3/2/11 8:50 PM, Martin Simmons wrote: > I think that the implementation of condition-wait has some major > problems: Thanks very much for the comments. I have an [updated patch][1] that attempts to address these issues for further review for inclusion in BORDEAUX-THREADS. [1]: http://detroit.slack.net/~evenson/abcl/bordeaux-threads-abcl-20110318a.diff > 1) If no other threads are trying to claim the lock, then > condition-wait will return immediately rather than waiting. I couldn't reproduce this behavior. In ABCL, the THREADS:OBJECT-WAIT implementation is mainly a simple wrapping of the underlying [java.lang.Object.wait()][2] which clearly should wait. [2]: http://download.oracle.com/javase/6/docs/api/java/lang/Object.html#wait%28%… > 2) If two threads are waiting in condition-wait and two other threads > call condition-notify, then it is possible that only one thread will > return from condition-wait because the call to acquire-lock in one of > them might return nil causing it to wait again. The DO looping construct has been replaced with a simpler subsequent BT:ACQUIRE-LOCK that should solve this bug. > 3) If condition-notify is called by one thread when a waiting thread > is just about to enter the threads:synchronized-on form (but before > it gets synchronized), then the notify will be lost. This happens > because the underlying threading primitives have no "memory" of > calls to notify when nothing it waiting (which is also the expected > semantics for POSIX condition variables BTW). THREADS:SYNCHRONIZED-ON is now the first form executed by BT:CONDITION-WAIT so the window here is considerably narrowed. > Also, I think condition-notify should be using threads:object-notify > rather than threads:object-notify-all. We now use THREADS:OBJECT-NOTIFY. -- "A screaming comes across the sky. It has happened before, but there is nothing to compare to it now."

3 5