[Gcl-devel] Re: Axiom sources for Debian release with 2.6.7

[Camm Maguire]

Greetings!

root <address@hidden> writes:

> Camm,
> 
> apply the changes, do a 
>    diff -Naur original changed >changed.patch
> and send me the patch files. I'll merge them into the
> sources and test them before release.
> 
> I've applied Peter's changes as well as a few of my own and
> I'm in the process of testing the --patch-45 release at the
> moment. I expect it to go out real-soon-now.
> 

Just had to make this vs the existing Debian package:
--- ./src/interp/daase.lisp.pamphlet.orig       2005-07-18 20:15:56.000000000 
+0000
+++ ./src/interp/daase.lisp.pamphlet    2005-07-18 20:08:00.000000000 +0000
@@ -840,7 +840,7 @@
  (let (thisdir nrlibs asos asys libs object only dir key 
       (|$forceDatabaseUpdate| t) noexpose)
   (declare (special |$forceDatabaseUpdate|))
-  (setq thisdir (namestring (probe-file ".")))
+  (setq thisdir (namestring (truename ".")))
   (setq noexpose nil)
   (multiple-value-setq (only dir noexpose) (processOptions options))
      ;don't force exposure during database build
@@ -1106,12 +1106,12 @@
   (setq *compressvector* nil)
   (withSpecialConstructors)
   (localdatabase nil
-     (list (list '|dir| (namestring (probe-file "./")) ))
+     (list (list '|dir| (namestring (truename "./")) ))
      'make-database)
   (dolist (dir dirlist)
          (localdatabase nil 
                         (list (list '|dir| 
-                                    (namestring (probe-file 
+                                    (namestring (truename 
                                                  (format nil "./~a" 
                                                          dir)))))
                         'make-database))


No need to abandon probe-file if your just trying to see if a certain
*file* exists.  If you are trying to get a full path, that's what
truename is for.

If you want the other patches I currently have in the Debian package
vs 20050201 (previously posted to the list) again, please let me know.

Take care, and thanks for your work on Axiom!!!

BTW, is there an FC4 fix I can apply to the code, or does FC4 require
root disabling of certain randomization via

echo 0 > /proc/sys/kernel/randomize_va_space

?  

May need a small 2.6.7 errata for ia64 and arm in any case.

Take care,

> t
> 
> 
> 

-- 
Camm Maguire                                            address@hidden
==========================================================================
"The earth is but one country, and mankind its citizens."  --  Baha'u'llah