April 2012 - bordeaux-threads-devel - mailman3.common-lisp.net

[Bordeaux-threads-devel] CONDITION-WAIT with timeout
by Thomas Munro 14 Sep '13

14 Sep '13

Hi Are there any plans to support a timeout for bordeaux-threads:condition-wait? The implementations I'm aware of support this threading primitive: * ABCL could presumably do it since Java Object.wait can take a timeout * SBCL supports it using sb-thread:condition-wait with keyword :timeout [1] * anything with POSIX threads underneath could use pthread_cond_timedwait The main variation seems to be between those implementations that use an absolute time and those that use a relative time (how long to wait in milliseconds for example). Since I think I need this feature I might try to write a patch to do this for ABCL and SBCL if there is nothing already in the works (unless there is some technical problem I'm not seeing?) Thanks Thomas Munro [1] http://random-state.net/log/3523852985.html

3 6

[Bordeaux-threads-devel] Deprecating recursive locks
by Stelian Ionescu 03 May '12

03 May '12

They're an ugly hack, as described their inventor David Butenhof (http://tinyurl.com/butenhof-recursive-mutexes) and aren't supported by all lisps: Allegro, Clozure, CMUCL, SBCL(it has a kind of recursive lock, but it's not a posix recursive lock), so I was thinking of making a 1.0 release in which to remove them altogether. What do you think ? -- Stelian Ionescu a.k.a. fe[nl]ix Quidquid latine dictum sit, altum videtur. http://common-lisp.net/project/iolib

7 15

[Bordeaux-threads-devel] [patch] LispWorks 6.x MP updates + EOS
by Kamil Shakirov 21 Apr '12

21 Apr '12

Hello, The attached patches do not add or fix anything, just clean up the lispworks implementation a bit. ;-) 1) bordeaux-threads-0.8.1+lw6x.patch This patch updates lispworks implementation reflecting all recent changes in LispWorks 6.x MP API. Some MP calls are still available in 6.x but considered deprecated and removed from documentation. 1.1) current-thread() changes It is better to use mp:get-current-process() that was introduced in LispWorks 5.1 instead of mp:*current-process* that can be bound to another process. 1.2) threadp() changes MP provides its own predicates for processes. 1.3) acquire-recursive-lock() and release-recursive-lock() changes I am getting warnings when compiling bordeaux-threads, so why not 'declaim' these calls 'inline' instead (fixed in the patch) ;;;*** Warning in BORDEAUX-THREADS:ACQUIRE-RECURSIVE-LOCK: Inline expansion for BORDEAUX-THREADS:ACQUIRE-LOCK not found ;;;*** Warning in BORDEAUX-THREADS:ACQUIRE-RECURSIVE-LOCK: Inline expansion for BORDEAUX-THREADS:ACQUIRE-LOCK not found ; BORDEAUX-THREADS:ACQUIRE-RECURSIVE-LOCK ;;;*** Warning in BORDEAUX-THREADS:RELEASE-RECURSIVE-LOCK: Inline expansion for BORDEAUX-THREADS:RELEASE-LOCK not found ;;;*** Warning in BORDEAUX-THREADS:RELEASE-RECURSIVE-LOCK: Inline expansion for BORDEAUX-THREADS:RELEASE-LOCK not found 1.4) join-thread() changes LispWorks 6.0 introduces its own mp:process-join() function. 2) bordeaux-threads-0.8.1+eos.patch This patch replaces FiveAM with Eos. Eos has no dependencies and backward compatible with FIveAM API. I couldn't manage to load FiveAM in LW because of arnesi loading errors. All tests work fine with Eos (tested with recent Clozure CL, SBCL and LW 6.1). Thank you for your time. ;-) -- --ska

3 4

[Bordeaux-threads-devel] Upcoming release 0.8.2
by Stelian Ionescu 17 Apr '12

17 Apr '12

I'm planning to make a new release, so if anybody has any request, please speak up -- Stelian Ionescu a.k.a. fe[nl]ix Quidquid latine dictum sit, altum videtur. http://common-lisp.net/project/iolib

1 0