<div dir="ltr"><div><div><div>Hi Stephan Oepen, </div>This is the result I've got when I execute this command (print *features*) (:ALLEGRO-CL-ENTERPRISE :CHUNGA :FLEXI-STREAMS :ASDF2 :ASDF :ACLDNS NET.ASERVE::QUEUE-DOES-NOT-TIMEOUT :IO-TIMEOUT :ZLIB-DEFLATE :INFLATE-BUG20472 :RETROACTIVITY :SMAF :LSP :LM :PSQL :LUI :PREPROCESSOR :POOLING :PXML :XML :LKB-V5.4 :CL-PPCRE :PPCRE :REGISTRY :CL-FAD :FAD :|YACC Release 0.98| :HAS-LOCKF :HAS-MKDTEMP :PVM :DRAKMA :ITSDB :DISCO-LOADUP :ANSI-FUNCTIONP :ANSI-EVAL-WHEN :CASE-INSENSITIVE :MK-DEFSYSTEM :CLTL2 :SLAVE :BDB :LOGON :MT :TSDB :MRS :LKB :IPV6 :ACL-SOCKET :HIPER-SOCKET :RUNTIME-STANDARD :STANDARD-RUNTIME :CLIM-MOTIF :ANSI-90 :SILICA :CLIM-2.1 :CLIM-2 :CLIM :ALLEGRO-CL-RUNTIME :CLOS :ATOMIC-SUBWORD-SETF :ATOMIC-SETF :DYNLOAD :DLFCN :UNIX :LINUX :AMD64 :LINUX86-64 :X86-64 :LITTLE-ENDIAN :ENCAPSULATING-EFS :DOUBLE-FLOAT-RANDOM-FIX :RELATIVE-PACKAGE-NAMES :MODULE-VERSIONS :IEEE :IEEE-FLOATING-POINT :CONFORMING-IEEE :GNU-SOURCE :ICS :COMMON-LISP :ANSI-CL :DRAFT-ANSI-CL-2 :X3J13 :ALLEGRO :EXCL :FRANZ-INC :ALLEGRO-VERSION>= :ALLEGRO-VERSION= :NEW-ENVIRONMENTS :GSGC :FLAVORS :MULTIPROCESSING :USE-THREAD-LIBS :DYNLOAD-ACL-LIBRARY :ALLEGRO-V8.2 :64BIT :SMP-MACROS :SSL-SUPPORT) (:ALLEGRO-CL-ENTERPRISE :CHUNGA :FLEXI-STREAMS :ASDF2 :ASDF :ACLDNS NET.ASERVE::QUEUE-DOES-NOT-TIMEOUT :IO-TIMEOUT :ZLIB-DEFLATE :INFLATE-BUG20472 ...) </div>The LKB version I'm using is the default version in the trunk of Logon repo. I have double checked that I have all three listed libraries installed (glibc.i686, libXext.i686, and fontconfig.i686) </div>Sincerely, <div> </div><div>Le Tuan Anh </div><div class="gmail_extra"> <div class="gmail_quote">On 14 October 2014 03:19, Stephan Oepen <<a href="mailto:oe@ifi.uio.no" target="_blank">oe@ifi.uio.no</a>> wrote: <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">hi again, i am afraid, i have no good theory why it should make a difference whether the grammar loading is initiated from the GUI or the command prompt. i assume in all cases you have loaded the same ‘script’ file, right? from the trace you sent, it appears the system was in the process of reporting an error from loading the grammar. this should not cause the behavior you get, of course, but i also wonder why there should occur an error during the grammar loading? are you actually running the 64-bit version of the LKB? could you (at the Lisp prompt) evaluate the following and email me the result: (pprint *features*) from what we see so far, i have no reason to suspect a lack of 32-bit compatibility libraries. but, just in case, here is a report i received earlier this year about required 32-bit libraries on Fedora 20: > [...] installing glibc.i686, libXext.i686, and fontconfig.i686 gave me an error free startup. best wishes, oe <div class="HOEnZb"><div class="h5"> On Fri, Oct 10, 2014 at 4:12 AM, Tuấn Anh Lê <<a href="mailto:tuananh.ke@gmail.com">tuananh.ke@gmail.com</a>> wrote: > Hi Stephan Oepen, > > Thank you so much for your fast reply. I'm working on a grammar with Dan > Flickinger and I hope I can fix this before Francis try to convert my laptop > to Ubuntu, lol. Strangely, It's pretty simple to make Logon works on Ubuntu > 64 bit. > > I understand that (rsa :eng) will try to load a grammar at > ~/logon/uio/tm/eng/lkb/script. I tried that and surprisingly it's working > fine. I also tried to load a grammar with (lkb::read-script-file-aux > "/home/tuananh/erg/lkb/script") and I can parse sentence with > (lkb::do-parse-tty "Logon works."). The tree can be shown too. > > I tried to load grammar from the menu of LKB and it's failed as before. So I > guess that the bug was raised by the LKB menu. I have attached the > *common-lisp* buffer for the three tries with this email for your reference. > Thanks again for your efforts to help. > > Sincerely yours, > -- > Tuan Anh, Le > > > On 9 October 2014 18:29, Stephan Oepen <<a href="mailto:oe@ifi.uio.no">oe@ifi.uio.no</a>> wrote: >> >> hi tuấn anh >> >> > [...] every time I load a grammar, this error show up on Emacs >> > >> > Error: Received signal number 11 (Segmentation fault) >> > [condition type: SYNCHRONOUS-OPERATING-SYSTEM-SIGNAL] >> >> i develop on 64-bit Fedora 20 and RHEL6 platforms, so your setup >> should certainly be supported. >> >> but to understand better what is going on, we will need more information. >> >> in a stock (i.e. unmodified) LOGON tree, i assume you run M-x logon >> RET in emacs, and both the LKB and [incr tsdb()] windows pop up? if >> so, could you run the following from the Lisp prompt in the emacs >> *common-list* buffer, please: >> >> (rsa :erg) >> >> from your earlier report, i expect this to lead to the same error you >> reported above. as you look at the error message, could you please >> evaluate the following commands at the Lisp prompt: >> >> :bt >> :zoom >> :local >> >> then, please save the entire *common-lisp* buffer, i.e. the log of >> your session, into a file and email it to this list. >> >> with thanks in advance, oe > </div></div></blockquote></div> </div></div>