- cffi-devel - mailman3.common-lisp.net

[cffi-devel] Program hanging on foreign-funcall
by Jakob Reschke 04 Mar '10

04 Mar '10

Hello, I'm writing a binding to Agar (http://libagar.com) an SDL GUI library (among other things), but I'm facing a problem: AG_Editable objects can be bound to a text buffer so the text entered into the editable widget is written into that buffer and vice versa. The buffer can be set with void AG_EditableSetString(AG_Editable *ed, char *text); The following C Program works fine: #include <agar/core.h> #include <agar/gui.h> int main(int argc, char **argv) { AG_InitCore("myprog", 0); AG_InitVideo(320, 240, 32, 0); char buffer[100]; /* text buffer */ AG_Window *win = AG_WindowNew(0); AG_Textbox *tb = AG_TextboxNew(win, 0, "testbox:"); AG_Editable *ed = tb->ed; AG_WindowShow(win); AG_EditableBindUTF8(ed, &buffer, 100); /* no hang here */ AG_EditableSetString(ed, "foobar"); AG_EventLoop(); return 0; } Whereas the following CFFI Lisp, which is supposed to be equivalent, at the REPL does not: (with-core-and-video (cffi:with-foreign-object (buffer :char 100) (let* ((win (cffi::foreign-funcall "AG_WindowNew" :int 0 :pointer)) (tb (cffi::foreign-funcall "AG_TextboxNew" :pointer win :int 0 :string "box:" :pointer)) (ed (cffi::foreign-slot-value tb 'ag-cffi::textbox 'ag-cffi::editable))) (ag:window-show win) ;; hangs here: (cffi::foreign-funcall "AG_EditableBindUTF8" :pointer ed :pointer buffer :int 100 :void) (cffi::foreign-funcall "AG_EditableSetString" :pointer ed :string "foobar" :void) (cffi::foreign-funcall "AG_EventLoop_FixedFPS" :void)))) The program hangs at the line with the according comment. The window shows up but doesn't get into the event loop, there are no error conditions and if I interrupt in Slime with C-c C-c (which works), the backtrace always reads "bogus stack frame" > "foreign function" > "foreign function" > (lambda ()) > (sb-int:simple-eval-in-lexenv ...) > ... What could cause this and how do I fix it? Or did I do something wrong? Best regards, Jakob

1 1

[cffi-devel] [PATCH] [RFC] Unloading libraries for delivery
by Francisco Vides Fernández 27 Feb '10

27 Feb '10

Hello! I'm having the very same trouble that is described in the following messages: http://common-lisp.net/pipermail/cffi-devel/2009-January/003048.html http://common-lisp.net/pipermail/cffi-devel/2009-January/003049.html http://common-lisp.net/pipermail/cffi-devel/2009-January/003050.html http://common-lisp.net/pipermail/cffi-devel/2009-July/003196.html I need to deliver an image that contains some foreign libraries. Now I'm going to test John Fremlin's solution, I think that will do for now, but: Will this solution or something equivalent be included in CFFI trunk? Thanks for such a nice piece of work! +----------------- | Francisco Vides Fernández <fvides(a)dedaloingenieros.com> | Gerente | Teléfono fijo: 952 60 29 59 | Fax: 952 60 29 59 | Dédalo Ingenieros http://www.dedaloingenieros.com/ | PGP: http://pgp.rediris.es:11371/pks/lookup?op=index&search=0xB1299C15 +------

1 0

[cffi-devel] New patches: 19-Feb-2010
by Luis Oliveira 20 Feb '10

20 Feb '10

Fri Feb 19 12:33:53 EST 2010 Stelian Ionescu <sionescu(a)cddr.org> * Remove unused C macro SIGNED_ from common.h M ./grovel/common.h -1 An updated tarball of CFFI's source can be downloaded here: http://common-lisp.net/project/cffi/tarballs/cffi-100219.tar.gz Darcsweb URL: http://common-lisp.net/cgi-bin/darcsweb/darcsweb.cgi?r=cffi-cffi;a=summary

1 0

[cffi-devel] Fix CFFI-UFFI-compat on ECL
by Michael Raskin 01 Feb '10

01 Feb '10

Hello. I am trying to simplify ASDF-Install'ing Cells-GTK on different platforms. When trying with ECL, I find the following problem: CFFI defines uffi-compat package with "uffi" alias. Comment even expresses doubt whether it is a good idea. ECL has its own UFFI. So CFFI should either not declare such nickname for ECL or remove the alias from "FFI" package. I think the second is a better (although marginally more complex) behavior, because it increases portability and reduces number of people who notice implementation quirks of UFFI package. I include sample patch to remove this alias. I hope something like that will get into the next release.. Thanks Michael Raskin --- /home/raskin/.sbcl/site/cffi_0.10.5/uffi-compat/uffi-compat.lisp 2009-06-16 20:20:08.000000000 +0400 +++ /home/raskin/.ecl/site/cffi_0.10.5/uffi-compat/uffi-compat.lisp 2010-01-29 10:58:24.000000000 +0300 @@ -28,6 +28,10 @@ ;;; Code borrowed from UFFI is Copyright (c) Kevin M. Rosenberg. +#+ecl (eval-when (:compile-toplevel :execute :load-toplevel) + (rename-package :ffi :ffi + (remove "UFFI" (package-nicknames :ffi) :test 'equal))) + (defpackage #:cffi-uffi-compat (:nicknames #:uffi) ;; is this a good idea? (:use #:cl)

3 7

[cffi-devel] cffi-grovel malforms ioctl requests
by Vitaly Mayatskikh 28 Jan '10

28 Jan '10

Hi! ioctl() requires `request' parameter to be of type `int'. CFFI processes constants as they are of type (u)int64_t. This malforms some requests, for example Video4Linux2 VIDIOC_* requests. This patch adds type :int to constant groveller to solve such problems. --- grovel/grovel.lisp~ 2009-06-16 18:20:08.000000000 +0200 +++ grovel/grovel.lisp 2009-12-21 12:15:14.612484158 +0100 @@ -468,6 +468,8 @@ char* print_double_for_lisp(double n) (c-print-symbol out lisp-name t) (c-format out " ") (ecase type + (:int + (format out "~& fprintf(output, \"%i\", (int32_t) ~A);" c-name)) (integer (format out "~& if(SIGNED64P(~A))~%" c-name) (format out " fprintf(output, \"%lli\", (int64_t) ~A);" c-name) -- wbr, Vitaly

1 0

[cffi-devel] %load-foreign-library on SBCL
by Elliott Slaughter 27 Jan '10

27 Jan '10

Hi, I have a question about the implementation of %load-foreign-library on SBCL. It seems that the call to load-shared-object remembers the absolute pathname to the library that is loaded unless :dont-save t is used. This causes portability issues for my application, which can't load from the same path when used on different systems. Is there any reason to choose this behavior? Of course, with :dont-save t, the user then has to call use-foreign-library again when starting up from the saved image, but I find this preferable to the crash I receive otherwise. An example patch is included below. Thanks. --- src/cffi-sbcl.lisp~ 2009-04-14 19:33:40.000000000 -0700 +++ src/cffi-sbcl.lisp 2009-04-03 15:28:37.000000000 -0700 @@ -324,7 +324,7 @@ (defun %load-foreign-library (name path) "Load a foreign library." (declare (ignore name)) - (load-shared-object path)) + (load-shared-object path :dont-save t)) ;;; SBCL 1.0.21.15 renamed SB-ALIEN::SHARED-OBJECT-FILE but introduced ;;; SB-ALIEN:UNLOAD-SHARED-OBJECT which we can use instead. -- Elliott Slaughter "Don't worry about what anybody else is going to do. The best way to predict the future is to invent it." - Alan Kay

3 10

[cffi-devel] translate-from-foreign not being called ....
by Frank Goenninger 26 Jan '10

26 Jan '10

Hey all! I can't seem to figure out why my translate-from-foreign method is not getting called: (defconstant +PIB-OK+ 0) (defctype tPIBAMQPrc :int) (defctype tsPIBAMQPContextPtr :pointer) (defmethod translate-from-foreign (value (type (eql 'tPIBAMQPrc))) (if (not (= value +PIB-OK+)) (error "*** PIB AMQP error: ~d." value)) value) Having a foreign function like (defcfun ("nPIBAMQPConnect" %mq-connect) tPIBAMQPrc (mq-context-ptr tsPIBAMQPContextPtr)) return values of > 0 does not trigger the above-mentioned method ... I haven't looked into CFFI internals yet. TIA ! Regards Frank

2 2

[cffi-devel] Strange... Help!
by Frank Goenninger 18 Jan '10

18 Jan '10

Hi again - I have the following C function definition (from RabbitMQ C client): typedef enum amqp_response_type_enum_ { AMQP_RESPONSE_NONE = 0, AMQP_RESPONSE_NORMAL, AMQP_RESPONSE_LIBRARY_EXCEPTION, AMQP_RESPONSE_SERVER_EXCEPTION } amqp_response_type_enum; typedef struct amqp_rpc_reply_t_ { amqp_response_type_enum reply_type; amqp_method_t reply; int library_errno; } amqp_rpc_reply_t; /* Opaque struct. */ typedef struct amqp_connection_state_t_ *amqp_connection_state_t; typedef enum amqp_sasl_method_enum_ { AMQP_SASL_METHOD_PLAIN = 0 } amqp_sasl_method_enum; extern amqp_rpc_reply_t amqp_login(amqp_connection_state_t state, char const *vhost, int channel_max, int frame_max, int heartbeat, amqp_sasl_method_enum sasl_method, ...); which I hand-translated to the following CFFI statements: (defcenum amqp-sasl-method-enum (:AMQP-SASL-METHOD-PLAIN 0)) (defcenum amqp-response-type-enum (:AMQP-RESPONSE-NONE 0) :AMQP-RESPONSE-NORMAL :AMQP-RESPONSE-LIBRARY-EXCEPTION :AMQP-RESPONSE-SERVER-EXCEPTION) (defcstruct amqp-rpc-reply-struct (reply-type amqp-response-type-enum) (reply amqp-method-t) (library-errno :int)) (defctype amqp-rpc-reply-t amqp-rpc-reply-struct) (defctype amqp-connection-state-t :pointer) (defcfun ("amqp_login" %amqp-login) amqp-rpc-reply-t (state amqp-connection-state-t) (vhost :string) (channel-max :int) (frame-max :int) (heartbeat :int) (sasl-method amqp-sasl-method-enum) &rest) ;; varargs ! You still there ? ;-) Good. Now for my "issue": Calling the C function amqp_login function as follows: (defparameter *hostname* "192.168.2.121") (defparameter *port* 5672) (defparameter *user* "ccag.pib") (defparameter *password* "ccag.pib") (defparameter *vhost* "/ccag/pib") (let* ((conn (%amqp-new-connection)) (sockfd (%amqp-open-socket *hostname* *port*))) (%amqp-login conn *vhost* 0 131072 0 :AMQP-SASL-METHOD-PLAIN :string *user* :string *password*) ... results in the following output from the C function amqp_login (which I instrumented): *** AMQP_LOGIN sees: -> state = 0x64034010 -> vhost = (null) -> channel_max = 131072 -> frame_max = 0 -> heartbeat = 0 -> sasl_method = 1677934672 Hmmmmmm - let's see: vhost .........: Not good. Should be "/ccag/pib". channel_max ...: Not good. Should be 0. frame_max .....: Not good. Should be 131072. sasl_method ...: Not good. Should be 0. What am I doing wrong here? By now I'm trying to figure this out for more than 10 hours and I am at a loss here.... Any help/ ideas really appreciated! TIA!!! Frank

2 2

[cffi-devel] :uint64 ... how to handle on non-supporting platform ?
by Frank Goenninger 16 Jan '10

16 Jan '10

Hi - thanks to Louis for answering my last question about function pointers. I now have a new challenge: How to handle :uint64 when I using AllegroCL 8.1 express edition on Mac OS X when this implementation is not supporting :unsigned-long-long ? I do have: (defcfun ("amqp_basic_ack" %amqp-basic-ack) :int (state amqp-connection-state-t) (channel amqp-channel-t) (delivery-tag :uint64) ;;; <<<<<< ---- this is my problem ... (multiple amqp-boolean-t)) Could I get around this with a clever padding of the 'multiple' argument ? Thanks for any advice! Cheers Frank

2 2

[cffi-devel] Lispworks FLI to CFFI translation question
by Frank Goenninger 15 Jan '10

15 Jan '10

Dear CFFI Gurus: I do have a Lispworks FLI definition like this: (fli:define-foreign-type pfunc (&rest fargs) `(:ptr (:function ,@fargs))) How would I translate this into a suitable CFFI definition? And why/how? Thanks in advance!! Cheers Frank

2 1