Ang: Re: recursive lock attempt, SBCL 1.0.9, Linux, x86

It works here (let it ran for two minutes, then breaked):

$ uname -a
Linux asdf 2.6.16.49 #1 Mon May 14 12:31:46 CEST 2007 i686 GNU/Linux

$ sbcl
This is SBCL 1.0.10, 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.
* *features*

(:ANSI-CL :COMMON-LISP :SBCL :SB-DOC :SB-TEST :SB-LDB :SB-THREAD
           :SB-PACKAGE-LOCKS :SB-UNICODE :SB-EVAL :SB-SOURCE-LOCATIONS
            :IEEE-FLOATING-POINT :X86 :UNIX :ELF :LINUX :LARGEFILE :GENCGC
             :STACK-GROWS-DOWNWARD-NOT-UPWARD :C-STACK-IS-CONTROL-STACK
              :COMPARE-AND-SWAP-VOPS :UNWIND-TO-FRAME-AND-CALL-VOP
               :STACK-ALLOCATABLE-CLOSURES :ALIEN-CALLBACKS :LINKAGE-TABLE
                :OS-PROVIDES-DLOPEN :OS-PROVIDES-DLADDR :OS-PROVIDES-PUTWC)
*

$ ls /usr/local/lib/sbcl/site/clx_0.7.3/

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/

Re: #'directory and filename patterns with *

Andreas Franke wrote:
> Hello.
>
> Suppose you have the following three files:
>    touch /tmp/tmp.foo
>    touch /tmp/tmp.bar.bar
>    touch /tmp/tmpbaz
>
> How can I find the same files as
>     ls /tmp/tmp*
> does?
>   
I did (directory "*.*") and it worked..

A quick look at sbcl/src/code/filesys.lisp makes me think that the
directory function could use som help.

> On pretty much every lisp except SBCL,
>     (directory "/tmp/tmp*")
> returns all three of the above files,
> but SBCL only finds "/tmp/tmpbaz".
> (Tested on SBCL 1.0.7 on Linux.)
> Is this a feature or a bug?
>
> Even if I don't need tmpbaz, I cannot use
> "/tmp/tmp.*" because it will miss tmp.bar.bar. 
> On the other hand, "/tmp/tmp*.*" will include
> tmpbaz on most lisps, but not on clisp.
>
> My workaround so far is to use "/tmp/tmp*"

Re: #'directory and filename patterns with *

Andreas Franke <afranke <at> ags.uni-sb.de> writes:

> Hello.
> 
> Suppose you have the following three files:
>    touch /tmp/tmp.foo
>    touch /tmp/tmp.bar.bar
>    touch /tmp/tmpbaz
> 
> How can I find the same files as
>     ls /tmp/tmp*
> does?
> 
> On pretty much every lisp except SBCL,
>     (directory "/tmp/tmp*")
> returns all three of the above files, but SBCL only finds
> "/tmp/tmpbaz".  (Tested on SBCL 1.0.7 on Linux.)  Is this a feature
> or a bug?

It doesn't sound like a bug to me. What you need to understand is that
unlike Unix filenames CL pathnames are structured data. SBCL maps your
test cases (+ an additional relevant one) as follows:

 1.  tmp.foo     -> :name "tmp"     :type "foo"
 2.  tmp.bar.baz -> :name "tmp.bar" :type "baz"
 3.  tmpbaz      -> :name "tmpbaz"  :type nil
 4.  tmpbah.     -> :name "tmpbaz"  :type ""

Now, DIRECTORY is supposed to return the existing files whose
pathnames match the pathname deisgnator given as argument. The

Fwd: SBCL hash test failure

Forwarding a message from a non-subscriber to sbcl-devel.

IIRC this test failure has been intermittently seen by others every
now and then, and it's passing on the buildbot ppc right now, so I'm
guessing that my changes aren't to blame. But it seems really strange
that the error (weak hash-table hanging on to a single key that it
shouldn't be retaining) would happen on ppc, where the gc should be
precise. Does anyone have any bright ideas on what's happening?

From: Robert Brown <brown <at> google.com>
Subject: SBCL hash test failure
Date: 2007-10-02 01:56:52 GMT

I'm sorry for mailing this to you directly.  I should send it to the SBCL
developer mailing list, but I'd probably have to join the mailing list first
in order to report a bug.  Also, I know you've been working on the hash
table code recently, so the test failure I noticed may be related to your
recent changes.

Anyway, I built the latest SBCL today and noticed that one of the weak hash
table tests fails on my PowerPC Macintosh -- output below.

Re: #'directory and filename patterns with *

Thanks for your hints and explanation, Christophe, Knut Olav and Juho!
Now I feel a lot more comfortable with this formerly pretty scary topic...

Best,
Andreas

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/

Addition for the 'save core' documentation

Hi all,

It would be nice to have documented in '2.2.3 Saving a Core Image'
how the command line arguments can be accessed. It took me quite
some time to discover it.

Perhaps just appending the following:

- Variable: *posix-argv*

      The command line arugments given the sbcl core.

Kind regards,
Daniel

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/

sb-posix docs

Greetings,

1.0.10.16 contains some sb-posix documentation, basically a mutilated  
version of sb-posix/README.  This was just a quick hack, I'd be  
grateful if someone currently using sb-posix could skim that chapter  
and point out glaring errors.

Cheers,

Rudi

Re: sb-posix docs

Rudi Schlatte <rudi <at> constantly.at> writes:

> 1.0.10.16 contains some sb-posix documentation, basically a mutilated  
> version of sb-posix/README.  This was just a quick hack, I'd be  
> grateful if someone currently using sb-posix could skim that chapter  
> and point out glaring errors.

Some suggestions (most implemented in the attached patch):

(1) AFAICT, the README was written as a specification proposal for a
    low level POSIX API, and so a bit of the language in it isn't
    needed in a manual.  For example, I don't think it's worthwhile to
    include the motivations for the various design decisions.  (I've
    preserved them as marked comments, however.)

(2) There are a handful of functions in sb-posix that don't exactly
    work like their POSIX analogues: the bindings for readlink, getcwd
    and syslog come to mind.  Since users can't guess what the
    sb-posix bindings do without looking at the source, I propose that
    such functions have docstrings and that the docstring-to-texinfo
    infrastructure be used for sb-posix.  (I've done this in the
    patch.)

(3) Several points about objects corresponding to C structures:

    (a) I don't see anything in the Texinfo about the names of objects
        and slots corresponding to C structures and structure members.
        I've copied the stuff about these kinds of objects from the
        README into the document.

posix group patch

The following patch adds support for the C struct group found in  
grp.h. I've submitted this here rather than committing this because  
I'm doing something a bit hinky, which is ignoring the gr_mem struct  
member as I don't remember the proper way to deal with char **'s.  
I've ignored it for the moment and it seems to work as is, but if  
there's a good, clean, easy solution for adding gr_mem, that would  
probably be a good thing to do.

The motivation for this is that hunchentoot wants to use getgrnam,  
but this wasn't previously supported by sb-posix.

Thanks,

Cyrus

cvs diff: Diffing contrib/sb-posix
Index: contrib/sb-posix/constants.lisp
===================================================================
RCS file: /cvsroot/sbcl/sbcl/contrib/sb-posix/constants.lisp,v
retrieving revision 1.38
diff -u -r1.38 constants.lisp
--- contrib/sb-posix/constants.lisp	2 Oct 2007 07:17:26 -0000	1.38
+++ contrib/sb-posix/constants.lisp	5 Oct 2007 06:56:04 -0000
 <at>  <at>  -21,6 +21,7  <at>  <at> 
   "errno.h"
   "dirent.h" "signal.h"
   #-win32 "pwd.h"
+ #-win32 "grp.h"
   "unistd.h"
   #-win32 "termios.h"

Windows download link not working

Hello,
   The windows SBCL 1.0.9 download link does not work from any mirror I have tried coming from
http://www.sbcl.org/platform-table.html:  http://sourceforge.net/project/downloading.php?groupname=sbcl&filename=sbcl-1.0.9-x86-win32-binary.msi&use_mirror=internap

Is the file non-existent?

Thanks,
Bryan

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/