Hi,
Continuing from a previous discussion I was having directly with Luigi Panzeri about this issue. I was having problems with unknown foreign type errors.
I'm running Leopard (10.5.1) on a PowerBook G4 under SBCL v.1.0.2
I've quoted his initial response. He suggested I ask here to see if someone else has the solution, though my initial problem was solved by his suggestion. I've run into a new problem though.
On Jan 18, 2008 10:50 AM, Luigi Panzeri matley@muppetslab.org wrote:
asdf-install should work fine. I have to admit that i have not tested cl-objc on a powerpc with leopard yet, though i don't think it is related to that. Maybe if you send email to cl-objc mailing list [1] someone else can also help you.
At any rate, I think that the problem is due to an older version of cffi you are using. Try to use the darcs repository [2] or a more recent tarball, as in older version ( < April 2007 IIRC) the interface for defining foreign type is changed.
This was the source of the type error messages, but when I fixed that and tried again, I got these errors from SBCL:
==== START OF LOG ==== Compiling CLOS bindings for Appkit framework in /usr/local/lib/sbcl/site/cl-objc/src/frameworks/Appkit-CLOS.lisp Control stack guard page temporarily disabled: proceed with caution
debugger invoked on a SB-KERNEL::CONTROL-STACK-EXHAUSTED: Control stack exhausted (no more space for function call frames). This is probably due to heavily nested or infinitely recursive function calls, or a tail call that SBCL cannot or has not optimized away.
Type HELP for debugger help, or (SB-EXT:QUIT) to exit from SBCL.
restarts (invokable by number or by possibly-abbreviated name): 0: [RETRY ] Retry performing #<ASDF:LOAD-OP NIL {11ABD101}> on #<ASDF:CL-SOURCE-FILE "generate-frameworks-bindings" {11AC19A9}>. 1: [ACCEPT] Continue, treating #<ASDF:LOAD-OP NIL {11ABD101}> on #<ASDF:CL-SOURCE-FILE "generate-frameworks-bindings" {11AC19A9}> as having been successful. 2: [RETRY ] Retry installation 3: [ABORT ] Exit debugger, returning to top level. ==== END OF LOG ====
Retrying does nothing to help with this problem, and even eventually leads to bad memory accesses on the part of SBCL, and an eventual seg fault.
Oh yeah, I'm using ASDF-Install to install Cl-ObjC by running the following in SBCL:
(require 'asdf) (require 'asdf-install) (asdf-install:install 'cl-objc)
Anyone have any idea what might be the problem?
Thanks, Jeremy Shoemaker
On Jan 20, 2008, at 7:31 PM, Jeremy Shoemaker wrote:
Anyone have any idea what might be the problem?
I have never been able to get Luigi Panzeri's cl-objc to run successfully under sbcl on an intel mac (neither 32 nor 64 bit) running Leopard. I always run into some error or other before it finishes compiling.
I haven't looked at the source much at all, but does cl-objc take any cognisance of the fact that the interface to the objective-c runtime is different in Leopard than it was in Tiger?
regards,
Ralph
Raffael Cavallaro, Ph.D. raffaelcavallaro@mac.com
On Jan 21, 2008 12:02 AM, Raffael Cavallaro raffaelcavallaro@mac.com wrote:
On Jan 20, 2008, at 7:31 PM, Jeremy Shoemaker wrote:
Anyone have any idea what might be the problem?
I have never been able to get Luigi Panzeri's cl-objc to run successfully under sbcl on an intel mac (neither 32 nor 64 bit) running Leopard.
Well, I'm running on a PowerPC, but I don't think that would make a big enough difference. Could be wrong though.
I always run into some error or other before it finishes compiling.
I haven't looked at the source much at all, but does cl-objc take any cognisance of the fact that the interface to the objective-c runtime is different in Leopard than it was in Tiger?
If that's true then that might be the problem, though I don't know why it would be causing the errors I'm getting. If I knew more about how the lisp code is interfacing with objective-c code, I might try to solve the problem myself.
regards,
Ralph
Raffael Cavallaro, Ph.D. raffaelcavallaro@mac.com
cl-objc-devel mailing list cl-objc-devel@common-lisp.net http://common-lisp.net/cgi-bin/mailman/listinfo/cl-objc-devel
Going through the list archives I noticed that my problem has been reported previously. Its basically a stack overflow so its the same thing that Jianshi Huang reported on Dec 1, 2007.
I'll just wait, like he is, for the next release.
Thanks, Jeremy Shoemaker