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

[cffi-devel] requesting help debugging error in libclang bindings
by Eric Schulte 01 Jul '12

01 Jul '12

Hello, I'm working to write CFFI binding for libclang [1], I've used swig to generate CFFI declarations of all libclang bindings, and I'm hoping to wrap those in a nice CLOS-based interface (my code is up here [2]). However I've run into an error which I'm having trouble getting past. Although the following simple C code [3] compiles and runs w/o problem. A direct translation into lisp [4] using the swig-generated cffi bindings throws this error [5]. For some reason the `clang_getTranslationUnitCursor' function always throws an error. FWIW I'll include the original C function headers [6] and the swig-generate CFFI declarations [7] below. I'm using SBCL version 1.0.57 and CFFI version 0.10.6 installed with quicklisp. Any idea what could be causing this error? Could the problem lie in SBCL, in CFFI, or possibly in some difference in state between the C world and the lisp world? Thanks, Footnotes: [1] http://clang.llvm.org/doxygen/group__CINDEX.html [2] https://github.com/eschulte/cl-libclang [3] hello-cursor.c // -*- C -*- #include "clang-c/Index.h" #include <stdio.h> int main(int argc, char *argv[]) { // declare CXIndex Idx; CXTranslationUnit TU; CXCursor CU; enum CXCursorKind Kind; // initialize Idx = clang_createIndex(0, 0); TU = clang_createTranslationUnit(Idx, "hello.ast"); CU = clang_getTranslationUnitCursor(TU); Kind = clang_getCursorKind(CU); // is the cursor null if(clang_Cursor_isNull(CU)) return 1; else printf("cursor is not null\n"); // cursor Kind (should be 300 for 'root') if(Kind != 300) return 1; else printf("kind of cursor is %d\n", Kind); // cleanup clang_disposeTranslationUnit(TU); clang_disposeIndex(Idx); return 0; } [4] hello-cursor.lisp ;; -*- lisp -*- (require :libclang) (in-package :libclang) (defvar idx nil "CXIndex") (defvar tu nil "CXTranslationUnit") (defvar cu nil "CXCursor") (defvar kind nil "CXCursorKind") (setf idx (clang_createIndex 0 0)) (setf tu (clang_createTranslationUnit idx "etc/hello.ast")) (setf cu (clang_getTranslationUnitCursor tu)) ;; we never get this far (setf kind (clang_getCursorKind cu)) [5] error running with SBCL 1.0.57 CORRUPTION WARNING in SBCL pid 30785(tid 140737353975552): Memory fault at 7841 (pc=0x7ffff57edb4a, sp=0x7ffff6e37508) The integrity of this image is possibly compromised. Continuing with fingers crossed. debugger invoked on a SB-SYS:MEMORY-FAULT-ERROR in thread #<THREAD "main thread" RUNNING {1002998D93}>: Unhandled memory fault at #x3E800007841. Type HELP for debugger help, or (SB-EXT:QUIT) to exit from SBCL. restarts (invokable by number or by possibly-abbreviated name): 0: [ABORT] Exit debugger, returning to top level. [6] from Index.h // -*- C -*- CINDEX_LINKAGE CXIndex clang_createIndex(int excludeDeclarationsFromPCH, int displayDiagnostics); // ... CINDEX_LINKAGE CXTranslationUnit clang_createTranslationUnit(CXIndex, const char *ast_filename); // ... CINDEX_LINKAGE CXCursor clang_getTranslationUnitCursor(CXTranslationUnit); [7] from libclang-swig.lisp ;; -*- lisp -*- (cffi:defcfun ("clang_createIndex" clang_createIndex) :pointer (excludeDeclarationsFromPCH :int) (displayDiagnostics :int)) ;; ... (cffi:defcfun ("clang_createTranslationUnit" clang_createTranslationUnit) :pointer (arg0 :pointer) (ast_filename :string)) ;; ... (cffi:defcstruct CXCursor (kind CXCursorKind) (xdata :int) (data :pointer)) ;; ... (cffi:defcfun ("clang_getTranslationUnitCursor" clang_getTranslationUnitCursor) CXCursor (arg0 :pointer)) -- Eric Schulte http://cs.unm.edu/~eschulte

2 1

[cffi-devel] cffi-libffi funcall path and :string
by Ryan Pavlik 07 Jun '12

07 Jun '12

For functions using the libffi foreign-funcall path, functions defined with :string parameters still fail because TRANSLATE-INTO-FOREIGN-MEMORY is not defined on FOREIGN-STRING-TYPE. E.g., (defcfun ("xcb_intern_atom" xcb-intern-atom) (:struct xcb-intern-atom-cookie-t) (c :pointer) (only_if_exists :unsigned-char) (name_len :unsigned-short) (name :string)) Calling this fails: There is no applicable method for the generic function #<STANDARD-GENERIC-FUNCTION TRANSLATE-INTO-FOREIGN-MEMORY (5)> when called with arguments (#.(SB-SYS:INT-SAP #X7FFFD8000DC0) #<CFFI::FOREIGN-STRING-TYPE :UTF-8> #.(SB-SYS:INT-SAP #X7FFFECDB7FD8)). [Condition of type SIMPLE-ERROR] This is using unmodified master. -- Ryan Pavlik <rpavlik(a)gmail.com> 425.220.9585

3 5

[cffi-devel] problem with include-path in cffi-grovel on windows
by Karl Heinrichmeyer 19 May '12

19 May '12

Hello, The groveler has a problem with the include path when calling the c-compiler on windows. This is because it uses the (directory-namestring) function to get the directory where cffi is installed and thus the device specifier is missing. I reported a bug and attached a patch to it. The patch is using a small lib i wrote to solve; I couldn't find an easy way to do it with ANSI CL (this may be my fault). The patch seem to solve the problem (windows7-64 and ccl-1.8), but I couldn't test it to a great extend. This is because I found the but by trying to use osicat which now seems to be broken with the latest git version. Is there any package that uses the groveler and should work on windows so that I can test the fix? I would love to hear your opinions on the patch and the lib and hope it will be helpful. Best regards, Karl

2 3

[cffi-devel] Round trips through parse-type, unparse-type
by Liam Healy 11 May '12

11 May '12

I just pushed fixes to make the FSBV tests pass. One fix was to the recursive call to libffi-type-pointer that took the slot-type directly, already unparsed, and tried to unparse it. I fixed that by parsing it again. In the course of debugging this, I found this comment in defclass foreign-struct-slot ;; FIXME: the type should probably be parsed? I briefly investigated doing this, and decided to take the expedient route for now. However, this comment got me to thinking that there are a lot of (unnecessary) round trips through parse-type/unparse-type. It seems to me that they could be eliminated or reduced. Perhaps as a design philosophy, we should only have parsed types internally, and then we won't need unparse-type at all? Changing this would be a lot of work, I realize. Liam

2 3

[cffi-devel] Cannot load cffi-tests
by Liam Healy 04 May '12

04 May '12

I cannot load cffi-tests (from the master branch). I am on Ubuntu 11.10 amd64. I can't even make sense of what the error is; it can't compile libtest.c? Liam This is SBCL 1.0.51.34-43a5265, an implementation of ANSI Common Lisp. More information about SBCL is available at <http://www.sbcl.org/>. SBCL is free software, provided as is, with absolutely no warranty. It is mostly in the public domain; some portions are provided under BSD-style licenses. See the CREDITS and COPYING files in the distribution for more information. CL-USER(1): (ql:quickload "cffi-tests") To load "cffi-tests": Load 1 ASDF system: cffi-tests ; Loading "cffi-tests" .................................................. [package cffi-features] debugger invoked on a ASDF:OPERATION-ERROR in thread #<THREAD "initial thread" RUNNING {1002F81371}>: Error while invoking #<COMPILE-OP (:VERBOSE NIL) {1003900621}> on #<C-TEST-LIB "cffi-tests" "tests" "libtest"> 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 compiling #<C-TEST-LIB "cffi-tests" "tests" "libtest">. 1: [ACCEPT] Continue, treating compiling #<C-TEST-LIB "cffi-tests" "tests" "libtest"> as having been successful. 2: [ABORT ] Give up on "cffi-tests" 3: Exit debugger, returning to top level. ((SB-PCL::FAST-METHOD ASDF:PERFORM (ASDF:COMPILE-OP CFFI-TESTS-SYSTEM::C-TEST-LIB)) #<unavailable argument> #<unavailable argument> #<ASDF:COMPILE-OP (:VERBOSE NIL) {1003900621}> #<CFFI-TESTS-SYSTEM::C-TEST-LIB "cffi-tests" "tests" "libtest">) 0] :ba 0: ((SB-PCL::FAST-METHOD ASDF:PERFORM (ASDF:COMPILE-OP CFFI-TESTS-SYSTEM::C-TEST-LIB)) #<unavailable argument> #<unavailable argument> #<ASDF:COMPILE-OP (:VERBOSE NIL) {1003900621}> #<CFFI-TESTS-SYSTEM::C-TEST-LIB "cffi-tests" "tests" "libtest">) 1: ((SB-PCL::EMF ASDF:PERFORM) #<unavailable argument> #<unavailable argument> #<ASDF:COMPILE-OP (:VERBOSE NIL) {1003900621}> #<CFFI-TESTS-SYSTEM::C-TEST-LIB "cffi-tests" "tests" "libtest">) 2: ((LAMBDA ())) 3: ((FLET SB-THREAD::WITH-RECURSIVE-LOCK-THUNK)) 4: ((FLET #:WITHOUT-INTERRUPTS-BODY-[CALL-WITH-RECURSIVE-LOCK]333)) 5: (SB-THREAD::CALL-WITH-RECURSIVE-LOCK #<CLOSURE (FLET SB-THREAD::WITH-RECURSIVE-LOCK-THUNK) {7FFFF69AEC09}> #<SB-THREAD:MUTEX "World Lock" owner: #<SB-THREAD:THREAD "initial thread" RUNNING {1002F81371}>>) 6: ((FLET SB-C::WITH-IT)) 7: ((SB-PCL::FAST-METHOD ASDF::PERFORM-PLAN (LIST)) #<unused argument> #<unused argument> ((#<ASDF:LOAD-OP (:VERBOSE NIL) {10041368E1}> . #<ASDF:CL-SOURCE-FILE "alexandria" "package">) (#<ASDF:LOAD-OP (:VERBOSE NIL) {10041368E1}> . #<ASDF:CL-SOURCE-FILE "alexandria" "definitions">) (#<ASDF:LOAD-OP (:VERBOSE NIL) {10041368E1}> . #<ASDF:CL-SOURCE-FILE "alexandria" "binding">) (#<ASDF:LOAD-OP (:VERBOSE NIL) {10041368E1}> . #<ASDF:CL-SOURCE-FILE "alexandria" "strings">) (#<ASDF:LOAD-OP (:VERBOSE NIL) {10041368E1}> . #<ASDF:CL-SOURCE-FILE "alexandria" "conditions">) (#<ASDF:LOAD-OP (:VERBOSE NIL) {10041368E1}> . #<ASDF:CL-SOURCE-FILE "alexandria" "hash-tables">) (#<ASDF:LOAD-OP (:VERBOSE NIL) {10041368E1}> . #<ASDF:CL-SOURCE-FILE "alexandria" "symbols">) (#<ASDF:LOAD-OP (:VERBOSE NIL) {10041368E1}> . #<ASDF:CL-SOURCE-FILE "alexandria" "macros">) (#<ASDF:LOAD-OP (:VERBOSE NIL) {10041368E1}> . #<ASDF:CL-SOURCE-FILE "alexandria" "functions">) (#<ASDF:LOAD-OP (:VERBOSE NIL) {10041368E1}> . #<ASDF:CL-SOURCE-FILE "alexandria" "lists">) (#<ASDF:LOAD-OP (:VERBOSE NIL) {10041368E1}> . #<ASDF:CL-SOURCE-FILE "alexandria" "io">) (#<ASDF:LOAD-OP (:VERBOSE NIL) {10041368E1}> . #<ASDF:CL-SOURCE-FILE "alexandria" "control-flow">) ...)) 8: ((LAMBDA ())) 9: (ASDF::CALL-WITH-SYSTEM-DEFINITIONS #<CLOSURE (LAMBDA #) {10038DFA29}>) 10: ((SB-PCL::FAST-METHOD ASDF:OPERATE (T T)) #<unused argument> #<unused argument> ASDF:LOAD-OP "cffi-tests" :VERBOSE NIL) 11: ((SB-PCL::EMF ASDF:OPERATE) #<unused argument> #<unused argument> ASDF:LOAD-OP "cffi-tests" :VERBOSE NIL) 12: (QUICKLISP-CLIENT::CALL-WITH-MACROEXPAND-PROGRESS #<CLOSURE (LAMBDA #) {10038DF939}>) 13: (QUICKLISP-CLIENT::AUTOLOAD-SYSTEM-AND-DEPENDENCIES "cffi-tests" :PROMPT NIL) 14: ((SB-PCL::FAST-METHOD QL-IMPL-UTIL::%CALL-WITH-QUIET-COMPILATION (T T)) #<unavailable argument> #<unavailable argument> #<unavailable argument> #<CLOSURE (FLET QUICKLISP-CLIENT::QL) {10033AD129}>) 15: ((SB-PCL::FAST-METHOD QL-IMPL-UTIL::%CALL-WITH-QUIET-COMPILATION :AROUND (QL-IMPL:SBCL T)) #<unavailable argument> #S(SB-PCL::FAST-METHOD-CALL :FUNCTION #<FUNCTION #> :PV NIL :NEXT-METHOD-CALL NIL :ARG-INFO (2)) #<QL-IMPL:SBCL {10005F8BB1}> #<CLOSURE (FLET QUICKLISP-CLIENT::QL) {10033AD129}>) 16: ((SB-PCL::FAST-METHOD QUICKLISP-CLIENT:QUICKLOAD (T)) #<unused argument> #<unused argument> #<unavailable argument> :PROMPT NIL :VERBOSE NIL) 17: (QL-DIST::CALL-WITH-CONSISTENT-DISTS #<CLOSURE (LAMBDA #) {1003394369}>) 18: (SB-INT:SIMPLE-EVAL-IN-LEXENV (QUICKLISP-CLIENT:QUICKLOAD "cffi-tests") #<NULL-LEXENV>) 19: (EVAL (QUICKLISP-CLIENT:QUICKLOAD "cffi-tests")) 20: (INTERACTIVE-EVAL (QUICKLISP-CLIENT:QUICKLOAD "cffi-tests") :EVAL NIL) 21: (SB-ACLREPL::REP-ONE) 22: (SB-ACLREPL::REPL :BREAK-LEVEL 0 :NOPRINT NIL :INSPECT NIL :CONTINUABLE NIL) 23: ((LAMBDA (SB-ACLREPL::NOPRINT)) NIL) 24: ((LAMBDA ())) 25: (SB-IMPL::%WITH-REBOUND-IO-SYNTAX #<CLOSURE (LAMBDA #) {10032B0079}>) 26: (SB-IMPL::TOPLEVEL-REPL NIL) 27: (SB-IMPL::TOPLEVEL-INIT) 28: ((FLET #:WITHOUT-INTERRUPTS-BODY-[RESTART-LISP]38)) 29: ((LABELS SB-IMPL::RESTART-LISP)) 0]

2 3

[cffi-devel] Next release of CFFI with cffi-libffi
by Liam Healy 01 May '12

01 May '12

The system cffi-libffi allows calling functions with structures passed and returned by value. The cffi-libffi branch has been merged into master. Because bare structure reference is deprecated, the merge affects a lot code using CFFI, not just those calling functions with structures by value. Expect to see many style warnings in code that refers to bare structure names. Please test and report successes/problems. We hope to have a new release of CFFI soon. See blog: http://lhealy.livejournal.com/2012/04/14/ Thank you. Liam

9 25

[cffi-devel] Recent library breakage
by Zach Beane 01 May '12

01 May '12

fnv (from git://github.com/blindglobe/fnv.git) used to build, but as of today I get this full warning: ; caught WARNING: ; Defining a foreign type named :COMPLEX-DOUBLE. This symbol ; belongs to the KEYWORD package and that may interfere with other ; code using CFFI. I'm using CFFI 0.10.7.1 on SBCL. What should be done to fix it? Thanks, Zach

4 5

[cffi-devel] *parse-bare-structs-as-pointers*
by Stelian Ionescu 01 May '12

01 May '12

Dow we really need *parse-bare-structs-as-pointers* ? It's only used in one place(parse-deprecated-struct-type) and IMO it's useless: we should parse bare structs as pointers unconditionally and issue a warning that it's deprecated and will be removed on 2013-05-01 -- Stelian Ionescu a.k.a. fe[nl]ix Quidquid latine dictum sit, altum videtur. http://common-lisp.net/project/iolib

2 1

[cffi-devel] Fixed regression
by Stelian Ionescu 16 Apr '12

16 Apr '12

https://github.com/cffi/cffi/commit/6eb2778eb We should do a little more testing and re-tag later -- Stelian Ionescu a.k.a. fe[nl]ix Quidquid latine dictum sit, altum videtur. http://common-lisp.net/project/iolib

2 2

[cffi-devel] Git repositories for CFFI
by Liam Healy 01 Apr '12

01 Apr '12

Hi, The CFFI web page http://common-lisp.net/project/cffi/ points to the git repository http://common-lisp.net/gitweb?p=projects/cffi/cffi.git, but that is no longer the official git repository for CFFI, it is on github now. The cl.net repo is severely out of date (master looks OK, but the fsbv branch is ancient and libffi doesn't even show). The web page should point to github, and the cl.net repo should either be deleted or synced up with github. Thanks, Liam

4 4