headers
clisp-cvs-request | 4 May 2012 14:05
Picon

clisp-cvs Digest, Vol 70, Issue 1

Send clisp-cvs mailing list submissions to
	clisp-cvs <at> lists.sourceforge.net

To subscribe or unsubscribe via the World Wide Web, visit
	https://lists.sourceforge.net/lists/listinfo/clisp-cvs
or, via email, send a message with subject or body 'help' to
	clisp-cvs-request <at> lists.sourceforge.net

You can reach the person managing the list at
	clisp-cvs-owner <at> lists.sourceforge.net

When replying, please edit your Subject line so it is more specific
than "Re: Contents of clisp-cvs digest..."

CLISP CVS commits for today

Today's Topics:

   1. clisp: update berkeley-db to 5.1
      (clisp-cvs-request <at> lists.sourceforge.net)

----------------------------------------------------------------------

Message: 1
Date: Fri, 04 May 2012 05:04:37 +0000
From: clisp-cvs-request <at> lists.sourceforge.net
Subject: clisp: update berkeley-db to 5.1
To: clisp-cvs <at> lists.sourceforge.net
Message-ID:
	<hg.3d1765c46a62.1336107877.-3352671374714820773 <at> vz-hg-2.sog>
(Continue reading)

Permalink | Reply | 0 comments |
headers
SourceForge.net | 7 May 2012 07:38
Picon

[ clisp-Bugs-3524255 ] ANSI CL forbids mixed-type complex numbers

Bugs item #3524255, was opened at 2012-05-06 22:38
Message generated for change (Tracker Item Submitted) made by johnwcowan
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101355&aid=3524255&group_id=1355

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: clisp
Group: ANSI compliance issue
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: John Cowan (johnwcowan)
Assigned to: Bruno Haible (haible)
Summary: ANSI CL forbids mixed-type complex numbers

Initial Comment:
CLHS (under Function COMPLEX) says that the real and imaginary parts of a complex numbers must either be
both rational or both of the same float type, and if they are not, floating-point coercion must be done.  One
of the given examples is (complex 1 .99) =>  #C(1.0 0.99).

But in CLISP, (integerp (realpart (complex 1 .99)) => #t.  This means that the real part is being stored as a
rational number, since there is no coercion from floats to rationals.  COMPLEX (and #C) should be changed
to make sure both parts are either rational or of the same float type.

This bug is present both in 2.48 on Cygwin and 2.44.1 on Ubuntu Lucid (both from binary packages).  Here are
the clisp --version outputs:
(Continue reading)

Permalink | Reply | 3 comments |
headers
SourceForge.net | 7 May 2012 15:51
Picon

[ clisp-Bugs-3524255 ] ANSI CL forbids mixed-type complex numbers

Bugs item #3524255, was opened at 2012-05-06 22:38
Message generated for change (Comment added) made by sds
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101355&aid=3524255&group_id=1355

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: clisp
Group: ANSI compliance issue
>Status: Pending
>Resolution: Invalid
Priority: 5
Private: No
Submitted By: John Cowan (johnwcowan)
Assigned to: Bruno Haible (haible)
Summary: ANSI CL forbids mixed-type complex numbers

Initial Comment:
CLHS (under Function COMPLEX) says that the real and imaginary parts of a complex numbers must either be
both rational or both of the same float type, and if they are not, floating-point coercion must be done.  One
of the given examples is (complex 1 .99) =>  #C(1.0 0.99).

But in CLISP, (integerp (realpart (complex 1 .99)) => #t.  This means that the real part is being stored as a
rational number, since there is no coercion from floats to rationals.  COMPLEX (and #C) should be changed
to make sure both parts are either rational or of the same float type.

This bug is present both in 2.48 on Cygwin and 2.44.1 on Ubuntu Lucid (both from binary packages).  Here are
the clisp --version outputs:
(Continue reading)

Permalink | Reply | 0 comments |
headers
SourceForge.net | 7 May 2012 16:11
Picon

[ clisp-Bugs-3524255 ] ANSI CL forbids mixed-type complex numbers

Bugs item #3524255, was opened at 2012-05-06 22:38
Message generated for change (Comment added) made by johnwcowan
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101355&aid=3524255&group_id=1355

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: clisp
Group: ANSI compliance issue
>Status: Open
Resolution: Invalid
Priority: 5
Private: No
Submitted By: John Cowan (johnwcowan)
Assigned to: Bruno Haible (haible)
Summary: ANSI CL forbids mixed-type complex numbers

Initial Comment:
CLHS (under Function COMPLEX) says that the real and imaginary parts of a complex numbers must either be
both rational or both of the same float type, and if they are not, floating-point coercion must be done.  One
of the given examples is (complex 1 .99) =>  #C(1.0 0.99).

But in CLISP, (integerp (realpart (complex 1 .99)) => #t.  This means that the real part is being stored as a
rational number, since there is no coercion from floats to rationals.  COMPLEX (and #C) should be changed
to make sure both parts are either rational or of the same float type.

This bug is present both in 2.48 on Cygwin and 2.44.1 on Ubuntu Lucid (both from binary packages).  Here are
the clisp --version outputs:
(Continue reading)

Permalink | Reply | 0 comments |
headers
SourceForge.net | 7 May 2012 16:40
Picon

[ clisp-Bugs-3524255 ] ANSI CL forbids mixed-type complex numbers

Bugs item #3524255, was opened at 2012-05-06 22:38
Message generated for change (Comment added) made by sds
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101355&aid=3524255&group_id=1355

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: clisp
Group: ANSI compliance issue
>Status: Pending
Resolution: Invalid
Priority: 5
Private: No
Submitted By: John Cowan (johnwcowan)
Assigned to: Bruno Haible (haible)
Summary: ANSI CL forbids mixed-type complex numbers

Initial Comment:
CLHS (under Function COMPLEX) says that the real and imaginary parts of a complex numbers must either be
both rational or both of the same float type, and if they are not, floating-point coercion must be done.  One
of the given examples is (complex 1 .99) =>  #C(1.0 0.99).

But in CLISP, (integerp (realpart (complex 1 .99)) => #t.  This means that the real part is being stored as a
rational number, since there is no coercion from floats to rationals.  COMPLEX (and #C) should be changed
to make sure both parts are either rational or of the same float type.

This bug is present both in 2.48 on Cygwin and 2.44.1 on Ubuntu Lucid (both from binary packages).  Here are
the clisp --version outputs:
(Continue reading)

Permalink | Reply | 0 comments |
headers
SourceForge.net | 10 May 2012 14:48
Picon

[ clisp-Bugs-3525509 ] CLISP treats CL:REQUIRE specially at compile time

Bugs item #3525509, was opened at 2012-05-10 05:48
Message generated for change (Tracker Item Submitted) made by xach
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101355&aid=3525509&group_id=1355

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: clisp
Group: lisp error
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Zach Beane (xach)
Assigned to: Bruno Haible (haible)
Summary: CLISP treats CL:REQUIRE specially at compile time

Initial Comment:
I have a file "test.lisp" that contains only this:

  (require 'bogus)

When I start "clisp -ansi -norc" and issue (compile-file "test.lisp"), I get this:

    [1]> (compile-file "test.lisp")
    ;; Compiling file /home/xach/test.lisp ...
    *** - LOAD: A file with name BOGUS does not exist
    The following restarts are available:
    ABORT          :R1      Abort main loop
(Continue reading)

Permalink | Reply | 1 comment |
headers
SourceForge.net | 10 May 2012 14:51
Picon

[ clisp-Bugs-3525509 ] CLISP treats CL:REQUIRE specially at compile time

Bugs item #3525509, was opened at 2012-05-10 05:48
Message generated for change (Comment added) made by xach
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=101355&aid=3525509&group_id=1355

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: clisp
Group: lisp error
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Zach Beane (xach)
Assigned to: Bruno Haible (haible)
Summary: CLISP treats CL:REQUIRE specially at compile time

Initial Comment:
I have a file "test.lisp" that contains only this:

  (require 'bogus)

When I start "clisp -ansi -norc" and issue (compile-file "test.lisp"), I get this:

    [1]> (compile-file "test.lisp")
    ;; Compiling file /home/xach/test.lisp ...
    *** - LOAD: A file with name BOGUS does not exist
    The following restarts are available:
    ABORT          :R1      Abort main loop
(Continue reading)

Permalink | Reply | 0 comments |
headers
Jerry James | 18 May 2012 19:39
Picon

Re: Test failure on ARM

On Mon, Apr 23, 2012 at 12:49 PM, Sam Steingold <sds <at> gnu.org> wrote:
>> * Jerry James <ybtnawreel <at> tznvy.pbz> [2012-04-23 10:56:12 -0600]:
>> I was afraid that would be the answer. :-)  I'll see if I can get
>> access to some ARM hardware so I can get more details on the problem.
>
> Please file the bug report on
> https://savannah.gnu.org/bugs/?func=additem&group=libffcall
> and put the patches there too!
>
>> In the meantime, I'll turn FFI off for the ARM platform to get the
>> builds going again.  Wish me luck.
>
> Good luck!

I finally got around to figuring out how to create an ARM VM with
qemu.  I've got the upcoming Fedora 17 release running on that VM.
The problem is not in ffcall after all.  The illegal instruction error
is occurring on the very last instruction of copy_loop_up_l2() in
src/ariarm.d.  A disassembly with GDB shows:

0x00106b3c <+0>:     ldm     r0!, {r2, r4, r12, lr}
0x00106b40 <+4>:     stmia   r1!, {r2, r4, r12, lr}
0x00106b44 <+8>:     subs    r3, r3, #8
0x00106b48 <+12>:    ldmge   r0!, {r2, r4, r12, lr}
0x00106b4c <+16>:    stmiage r1!, {r2, r4, r12, lr}
0x00106b50 <+20>:    bgt     0x106b3c <copy_loop_up_l2>
0x00106b54 <+24>:    mov     r0, r1
0x00106b58 <+28>:    ldm     sp!, {r4, pc}^

It's that last ldm instruction, at +28, that is the problem.  Also, I
(Continue reading)

Permalink | Reply | 4 comments |
headers
Sam Steingold | 18 May 2012 22:05
Picon

Re: Test failure on ARM

> * Jerry James <ybtnawreel <at> tznvy.pbz> [2012-05-18 11:39:39 -0600]:
>
> I finally got around to figuring out how to create an ARM VM with
> qemu.  I've got the upcoming Fedora 17 release running on that VM.
> The problem is not in ffcall after all.  The illegal instruction error
> is occurring on the very last instruction of copy_loop_up_l2() in
> src/ariarm.d.  A disassembly with GDB shows:
>
> 0x00106b3c <+0>:     ldm     r0!, {r2, r4, r12, lr}
> 0x00106b40 <+4>:     stmia   r1!, {r2, r4, r12, lr}
> 0x00106b44 <+8>:     subs    r3, r3, #8
> 0x00106b48 <+12>:    ldmge   r0!, {r2, r4, r12, lr}
> 0x00106b4c <+16>:    stmiage r1!, {r2, r4, r12, lr}
> 0x00106b50 <+20>:    bgt     0x106b3c <copy_loop_up_l2>
> 0x00106b54 <+24>:    mov     r0, r1
> 0x00106b58 <+28>:    ldm     sp!, {r4, pc}^
>
> It's that last ldm instruction, at +28, that is the problem.  Also, I
> see this from compilation of src/ariarm.s:
> ariarm.s: Assembler messages:
> ariarm.s:1: Warning: ignoring attempt to redefine built-in register 'a1'
> ariarm.s:2: Warning: ignoring attempt to redefine built-in register 'a2'
> ariarm.s:3: Warning: ignoring attempt to redefine built-in register 'a3'
> ariarm.s:4: Warning: ignoring attempt to redefine built-in register 'a4'
> ariarm.s:5: Warning: ignoring attempt to redefine built-in register 'v1'
> ariarm.s:6: Warning: ignoring attempt to redefine built-in register 'v2'
> ariarm.s:7: Warning: ignoring attempt to redefine built-in register 'v3'
> ariarm.s:8: Warning: ignoring attempt to redefine built-in register 'v4'
> ariarm.s:9: Warning: ignoring attempt to redefine built-in register 'v5'
> ariarm.s:10: Warning: ignoring attempt to redefine built-in register 'v6'
(Continue reading)

Permalink | Reply | 3 comments |
headers
Jerry James | 18 May 2012 23:28
Picon

Re: Test failure on ARM

On Fri, May 18, 2012 at 2:05 PM, Sam Steingold <sds <at> gnu.org> wrote:
> I cannot help with the specific problem, but I will ask you to file a
> CLISP bug report on SF.
>
> You can disable all assembly by adding -DNO_ASM to CFLAGS (either at
> configure time as "./configure CFLAGS=-DNO_ASM build-no-asm" or by editing
> the Makefile and doing "make clean"). Please see clisp/unix/PLATFORMS
> for more information.

Thanks, Sam.  I'm going to make one more attempt before I file a bug.
I'm going to try building the latest mercurial sources and see if the
same problem occurs.  We're stilling distributing vanilla 2.49 in
Fedora.  Should we be?

Anyway, compiling on the ARM VM takes a very long time, so I probably
won't have time to get to it until Monday or Tuesday.  If it suffers
the same failure, I'll file a bug.  If not, I'll see if I can figure
out what commit fixes the problem.

Regards,
--

-- 
Jerry James
http://www.jamezone.org/

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
(Continue reading)

Permalink | Reply | 2 comments |

Gmane