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

[cffi-devel] [Fwd: Re: [Lisp] testing CFFI]
by Kenny Tilton 18 Sep '05

18 Sep '05

Luis, Got this report from a fello Lispnyk (thx, dimitri). Looks to my uninformed eyes as if the x86 and >0.9.3 requirements for callbacks are met. Thoughts? kenny -------- Original Message -------- Subject: Re: [Lisp] testing CFFI Date: Sat, 17 Sep 2005 12:19:59 -0400 From: Dimitry Gashinsky <digash(a)gmail.com> Reply-To: i(a)digash.com To: Kenny Tilton <ktilton(a)nyc.rr.com> CC: LISP NYC <lisp(a)lispnyc.org> References: <432AE74A.9060603(a)nyc.rr.com> Hi Kenny, I just tried running CFFI on Linux X86_64. I found couple of problems. CL-USER(3): :c cffi-tests ; loading system definition from ; /home/dig/src/cffi-luis-050916-0148/cffi-tests.asd into #<PACKAGE "ASDF4061"> ; registering #<SYSTEM CFFI-TESTS {1000F01F11}> as CFFI-TESTS ; loading system definition from /usr/share/common-lisp/systems/rt.asd into ; #<PACKAGE "ASDF4088"> ; registering #<SYSTEM :RT {10009C6D51}> as RT ; $ cd /home/dig/src/cffi-luis-050916-0148/tests/; make gcc -o libtest.so -shared -lm -Wall -ansi -pedantic libtest.c /usr/lib/gcc/x86_64-pc-linux-gnu/3.4.4/../../../../x86_64-pc-linux-gnu/bin/ld: /tmp/ccsyYRJ7.o: relocation R_X86_64_32 against `a local symbol' can not be used when making a shared object; recompile with -fPIC /tmp/ccsyYRJ7.o: could not read symbols: Bad value collect2: ld returned 1 exit status make: *** [libtest.so] Error 1 debugger invoked on a ASDF:OPERATION-ERROR in thread #<THREAD "initial thread" {1000003981}>: erred while invoking #<COMPILE-OP NIL {100105BA71}> on #<C-TEST-LIB "libtest" {10011559F1}> 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:COMPILE-OP NIL {100105BA71}> on #<CFFI-TESTS-SYSTEM::C-TEST-LIB "libtest" {10011559F1}>. 1: [ACCEPT] Continue, treating #<ASDF:COMPILE-OP NIL {100105BA71}> on #<CFFI-TESTS-SYSTEM::C-TEST-LIB "libtest" {10011559F1}> as having been successful. 2: [ABORT ] Exit debugger, returning to top level. ((SB-PCL::FAST-METHOD ASDF:PERFORM (ASDF:COMPILE-OP CFFI-TESTS-SYSTEM::C-TEST-LIB)) #<unused argument> #<unused argument> #<ASDF:COMPILE-OP NIL {100105BA71}> #<CFFI-TESTS-SYSTEM::C-TEST-LIB "libtest" {10011559F1}>) source: (ERROR 'ASDF:OPERATION-ERROR :COMPONENT CFFI-TESTS-SYSTEM::C :OPERATION CFFI-TESTS-SYSTEM::O) 0] The first problem coud be easily resolved by adding -fPIC to LDFLAGS. ; file: /home/dig/src/cffi-luis-050916-0148/tests/callbacks.lisp ; in: DEFCALLBACK SUM-CHAR ; (CFFI:DEFCALLBACK CFFI-TESTS::SUM-CHAR ; :CHAR ; ((CFFI-TESTS::A :CHAR) (CFFI-TESTS::B :CHAR)) ; "Test if the named block is present and the docstring too." ; (RETURN-FROM CFFI-TESTS::SUM-CHAR ; (+ CFFI-TESTS::A CFFI-TESTS::B))) ; --> PROGN CFFI-SYS:%DEFCALLBACK SETF LET* LET* MULTIPLE-VALUE-BIND LET ; --> ALIEN-SAP LOCALLY PROGN ALIEN-SAP SB-ALIEN::ALIEN-LAMBDA ; ==> ; (SB-ALIEN-INTERNALS:ALIEN-CALLBACK (FUNCTION CHAR CHAR CHAR) ; (LAMBDA (CFFI-TESTS::A CFFI-TESTS::B) ; (CFFI::INVERSE-TRANSLATE-OBJECTS ; (CFFI-TESTS::A CFFI-TESTS::B) ; (:CHAR :CHAR) ; :CHAR ; (BLOCK CFFI-TESTS::SUM-CHAR ; (RETURN-FROM CFFI-TESTS::SUM-CHAR ; #))))) ; ; caught ERROR: ; (in macroexpansion of (SB-ALIEN-INTERNALS:ALIEN-CALLBACK # #)) ; (hint: For more precise location, try *BREAK-ON-SIGNALS*.) ; The function SB-ALIEN-INTERNALS:ALIEN-CALLBACK-ACCESSOR-FORM is undefined. This seems that my SBCL does not support ALIEN-CALLBACK-ACCESSOR-FORM. Here is my features. This is SBCL 0.9.4, an implementation of ANSI Common Lisp. ... (:ASDF :SB-THREAD :ANSI-CL :COMMON-LISP :SBCL :UNIX :SB-DOC :SB-PACKAGE-LOCKS :SB-UNICODE :IEEE-FLOATING-POINT :X86-64 :ELF :LINUX :GENCGC :STACK-GROWS-DOWNWARD-NOT-UPWARD :C-STACK-IS-CONTROL-STACK :LINKAGE-TABLE :STACK-ALLOCATABLE-CLOSURES :OS-PROVIDES-DLOPEN :OS-PROVIDES-DLADDR) If I accept callback failure and continue with testing all the tests pass! I guess there is no tests for CALLBACK yet. Hope this helps, Dig On 9/16/05, Kenny Tilton <ktilton(a)nyc.rr.com> wrote: > Thought I would make it a little easier: > > 1. Download latest CFFI-LUIS from here (last is latest): > > http://common-lisp.net/project/cffi/tarballs/ > > 2. Download RT from here: > > http://files.b9.com/rt/ > > 3: Adjust the following to match where you put stuff: > > #-asdf > (load "/0devtools/asdf.lisp") > > (progn > #+lispworks > (setf hcl::*handle-existing-defpackage* (list :add)) > > (push (make-pathname :directory '(:absolute "0devtools" "cffi")) > asdf:*central-registry*) > > (push (make-pathname :directory '(:absolute "0devtools" "rt")) > asdf:*central-registry*) > > (asdf:operate 'asdf:load-op 'cffi-tests)) > > Ok, now you can run the tests: > > (cffi-tests::do-tests) ;; runs all tests trapping errors, reporting > only number of failures > > To see what the first failure (if any) is: > > (let ((cffi-tests::*catch-errors* nil)) > (cffi-tests::do-tests)) > > Thx. > > -- > Kenny > > Why Lisp? http://wiki.alu.org/RtL_Highlight_Film > > "I've wrestled with reality for 35 years, Doctor, and I'm happy to state I finally won out over it." > Elwood P. Dowd, "Harvey", 1950 > > > > _______________________________________________ > Lisp mailing list > Lisp(a)lispnyc.org > http://www.lispnyc.org/mailman/listinfo.cgi/lisp > -- ((LAMBDA (LAMBDA) `(,LAMBDA ',LAMBDA)) '(LAMBDA (LAMBDA) `(,LAMBDA ',LAMBDA))) -- Kenny Why Lisp? http://wiki.alu.org/RtL_Highlight_Film "I've wrestled with reality for 35 years, Doctor, and I'm happy to state I finally won out over it." Elwood P. Dowd, "Harvey", 1950

2 1

[cffi-devel] CLisp, defcfun, and load-library timing
by Kenny Tilton 17 Sep '05

17 Sep '05

FYI, Loading opengl32.dll manually before building did get me past the glflush and other defcfuns. But my questions stand about whether it has to be this way, or if CLisp can hold off looking up the function until actually called. -- Kenny Why Lisp? http://wiki.alu.org/RtL_Highlight_Film "I've wrestled with reality for 35 years, Doctor, and I'm happy to state I finally won out over it." Elwood P. Dowd, "Harvey", 1950

1 0

[cffi-devel] does a library have to be loaded before DEFCFUN?
by Kenny Tilton 17 Sep '05

17 Sep '05

Background: I usually load libraries as part of /run-time/ initialization. I know a lot of people just load the libraries at some point during development and then just forget about it. The good news is that my opengl framework (including Freeglut hence callbacks) now runs using CFFI on both ACL and LW on win32. On CLisp, however, I get : xxx - FFI::FOREIGN-LIBRARY-FUNCTION: no dynamic object named "glFlush" in library :default. Hmmm. That ":default" has me curious. How will CLisp find glFlush in anything other than gl32.dll? If we specified a module/library, would CLisp wait until it was called before trying to resolve it? Other? -- Kenny Why Lisp? http://wiki.alu.org/RtL_Highlight_Film "I've wrestled with reality for 35 years, Doctor, and I'm happy to state I finally won out over it." Elwood P. Dowd, "Harvey", 1950

1 0

[cffi-devel] patch to uffi compat?
by Kenny Tilton 17 Sep '05

17 Sep '05

I am using the cffi-uffi-compat package to see if I can quickly port my opengl framework to CFFI. A simple deref-array on type :float died on no element-type method for a built-in type. I added: (defmethod element-type ((other cffi::foreign-type)) (cffi::name other)) Just a guess. -- Kenny Why Lisp? http://wiki.alu.org/RtL_Highlight_Film "I've wrestled with reality for 35 years, Doctor, and I'm happy to state I finally won out over it." Elwood P. Dowd, "Harvey", 1950

2 2

[cffi-devel] Good news on the Corman Lisp front, and other changes
by Luis Oliveira 15 Sep '05

15 Sep '05

Hello, I have fixed a lame bug in cffi-corman.lisp and know it passes 64 out of 71 tests (this includes the callbacks tests), which is pretty good. At least 6 of these 7 failures should be easy to fix and I'll be asking Roger Corman for help once I write some test cases by translation the cffi declarations to native corman lisp FFI. That's the highlight of the following changes :-) Thu Sep 15 05:34:23 WEST 2005 Luis Oliveira <loliveira(a)common-lisp.net> * Corman Lisp update and note added to the manual. - Added note to the User Manual that it reflects what's in cffi-luis, not the main branch. - Updated cffi-corman.lisp and fixed a critical bug. Now only 7 tests fail. Thu Sep 15 02:30:27 WEST 2005 Luis Oliveira <loliveira(a)common-lisp.net> * Removed test, tweaked tests/Makefile - Commented out the test STRUCT.ALIGNMENT.5 because an empty struct is not valid standard C and some compilers, namely VC++, won't compile it. - Add -ansi -pedantic to the compiler flags to catch things like this sooner next time. Thu Sep 15 01:14:14 WEST 2005 Luis Oliveira <loliveira(a)common-lisp.net> * Missing DLLEXPORTs in libtest.c Wed Sep 14 21:32:04 WEST 2005 Luis Oliveira <loliveira(a)common-lisp.net> * make clean -> rm *.fsl too. Wed Sep 14 01:25:07 WEST 2005 Luis Oliveira <loliveira(a)common-lisp.net> * Documentation and new test - Added detailed node listing to the User Manual. - New regression test: WITH-FOREIGN-PTR.EVALX2 (checks if the size argument is not evaluated twice by mistake) - Also removed an old comment-out test function from cffi-lispworks.lisp -- Luís Oliveira http://student.dei.uc.pt/~lmoliv/ Equipa Portuguesa do Translation Project http://www.iro.umontreal.ca/translation/registry.cgi?team=pt

2 3

[cffi-devel] next review steps
by Kenny Tilton 15 Sep '05

15 Sep '05

ok, back in action. my current task is to take my working acl/win32 openal demo (using CFFI bindings generated by vzn) and try running it under; Lispworks/win32 CLisp/win32 Corman Lisp/win32 other? Armed bear? ECL? GCL? and then: OpenMCL/OSX I may ask for others to try the code on various *nix platforms. Then I will review the documentation and test suite for completeness and make my CFFI-Luis evaluation (tho with time I would like to get a GLUT demo running to test callbacks). After that I will return to VZN and see if Rayiner has my OpenGL bindings problem resolved. I will also look at FTGL, a C++ library, then do my VZN evaluation. Did you gentlemen prepare student evaluations and forward them to Heow? I would like to see those so I know what you /think/ is there and so we can resolve any differences before getting to Google. Also, James, did you see my request for input on Luis's work? You have been the true mentor on that project, so I value your input. cheers, kenny

2 5

[cffi-devel] <sigh> libtest.dll anyone?
by Kenny Tilton 15 Sep '05

15 Sep '05

VC++ considers: struct empty_struct {}; ...to be an error. Anyone got libtest.dll? Or should I break down and try Cygwin? Is that what other folks do? Maybe I missed a switch for the C++ compiler? -- Kenny Why Lisp? http://wiki.alu.org/RtL_Highlight_Film "I've wrestled with reality for 35 years, Doctor, and I'm happy to state I finally won out over it." Elwood P. Dowd, "Harvey", 1950

2 2

[cffi-devel] cffi tests vis a vis verrazano
by Kenny Tilton 14 Sep '05

14 Sep '05

Hmmm, no libctest.h for vzn to eat? Or am I missing something? Can vzn eat libctest.c? I can see how CFFI needs to be testable independent of the existence of vzn, but it would be nice if the test suite can be /also/ approached via vzn-generated bindings. no big deal. -- Kenny Why Lisp? http://wiki.alu.org/RtL_Highlight_Film "I've wrestled with reality for 35 years, Doctor, and I'm happy to state I finally won out over it." Elwood P. Dowd, "Harvey", 1950

1 0

[cffi-devel] belated mentoring
by Kenny Tilton 13 Sep '05

13 Sep '05

Just noticed foreign-alloc (while looking for foreign-allocate <g>), which reminds me of Tilton's Law of Abbreviation: Only abbreviate names (a) longer than six characters (b) where a reasonable abbreviation saves more than half the characters. Exceptions are made where a well-known abbreviation exists, such as serial-no instead of serial-number (number having only six characters). The nice thing about a strict abbreviations rule is the elimination of coding anxiety / guesswork / aproposing. -- Kenny Why Lisp? http://wiki.alu.org/RtL_Highlight_Film "I've wrestled with reality for 35 years, Doctor, and I'm happy to state I finally won out over it." Elwood P. Dowd, "Harvey", 1950

2 1

Re: [cffi-devel] next review steps
by Kenny Tilton 13 Sep '05

13 Sep '05

Frank Goenninger - PRION Consulting wrote: > Kenny, > > here it is (hoping I am the first one to respond)... ;-) You are, thanks, now get some sleep. :) kt

1 0