From: Sam S. <sd...@gn...> - 2003-04-29 13:42:36
|
> * In message <9F8582E37B2EE5498E76392AEDDCD3FE03ECEC33@G8PQD.blf01.telekom.de> > * On the subject of "apology: CVS CLISP is not broken beyond belief" > * Sent on Mon, 28 Apr 2003 11:14:20 +0200 > * Honorable "Hoehle, Joerg-Cyril" <Joe...@t-...> writes: > > Interpreted.mem was rebuilt by loading from the *old* .fas files > (except for init.lsp). That causes old information (different > signatures of functions) to mix badly with new definitions. No crash > occured during generation of the various *.mem files. It's only > afterwards, in interactive sessions, or running the testsuite when I > found the new behaviour so suspicious, that I experienced the spurious > crashes and prepared several bug reports which I later had to reject > when I found this single cause of all but a few crashes. > > Maybe the Makefile should handle this situation? this is a bootstrapping issue, not Makefile issue. (see two definitions of search-file in init.lisp) "make clean" should help. > BTW, another area where Makefile did not do what I expected: > > I also called "make full". But neither full/lisp.run nor > full/lispinit.mem were regenerated. Thus, ./lisp.run is newer than > full/lisp.run "this is a feature, not a bug". you have to do $ rm -rf full && make full I am not sure why Bruno made it this way. -- Sam Steingold (http://www.podval.org/~sds) running RedHat9 GNU/Linux <http://www.camera.org> <http://www.iris.org.il> <http://www.memri.org/> <http://www.mideasttruth.com/> <http://www.palestine-central.com/links.html> I don't want to be young again, I just don't want to get any older. |