(I absent-mindedly sent this reply to sbcl-devel instead of sbcl-help.
Trying again...)
----- Forwarded message from William Harold Newman <william.newman@...> -----
On Tue, May 21, 2002 at 11:14:43PM +0200, Antonio Martínez wrote:
> As part of an unhealthy Lisp obsession, I'm trying to build SBCL 0.7.3
> with CMU CL 2.4.19 under Debian 2.2 (on an AMD K7) and I'd appreciate
> it a lot if anyone has time to send along any help with the following
> problem.
>
> I do the first step as recommended in INSTALL:
>
> sh make.sh 'lisp -noinit -batch'
>
> which prints lots of compilation info, ending with the lines
>
> [...]
>
> obj/from-host/src/code/late-type.lisp-obj-tmp written.
> deleted "obj/from-host/src/code/late-type.lisp-obj-tmp"
> Error in batch processing:
>
> Error in function COMPILE-STEM:
>
> The sbcl and sbcl.core files aren't written. I've searched backwards
> through the compiler output, but COMPILE-STEM isn't referenced
> anywhere. I don't know where to look next. Any suggestions as to
> which parts of my local setup could be causing this behaviour? What
> other info could help me clarify the problem? Is this a valid build
> target for SBCL?
SBCL is supposed to build on that system, so if it doesn't work
I'd like to know about it.
COMPILE-STEM is part of the machinery for cross-compiling,
defined in src/cold/shared.lisp.
COMPILE-STEM will signal an error, using CERROR, if COMPILE-FILE
returns FAILURE-P. I haven't used CMU CL for a while, but I'd
guess that when you use the '-batch' option, it's handling the
CERROR in a way which is hard to understand.
To go about finding the problem you might try these things:
* Read backwards from the COMPILE-STEM failure looking for
any compilation warnings or compilation errors. Any compilation
error should cause FAILURE-P, causing COMPILE-STEM to fail.
And as far as I know, CMU CL still doesn't make a systematic
distinction between STYLE-WARNINGs and full WARNINGs, so any
compilation warning might or might not cause FAILURE-P,
causing COMPILE-STEM to fail.
* Run the failed compilation step by hand, with CMU CL
running interactively instead of with "-batch". E.g.
if the failure is happening in make-host-1.sh, basically
you start CMU CL in an inferior buffer in Emacs, paste
the Lisp code from make-host-1.sh into the buffer as
command input, let it run, and then use the debugger
to investigate where it stops.
--
William Harold Newman <william.newman@...>
"When I am picking problems to work on, ones that stumped John von
Neumann go at the bottom of the stack."
-- <http://www.daviddfriedman.com/laws_order/index.shtml>, ch. 8
PGP key fingerprint 85 CE 1C BA 79 8D 51 8C B9 25 FB EE E0 C3 E5 7C
_______________________________________________________________
Don't miss the 2002 Sprint PCS Application Developer's Conference
August 25-28 in Las Vegas -- http://devcon.sprintpcs.com/adp/index.cfm
_______________________________________________
Sbcl-devel mailing list
Sbcl-devel@...
https://lists.sourceforge.net/lists/listinfo/sbcl-devel
----- End forwarded message -----
--
William Harold Newman <william.newman@...>
"When I am picking problems to work on, ones that stumped John von
Neumann go at the bottom of the stack."
-- <http://www.daviddfriedman.com/laws_order/index.shtml>, ch. 8
PGP key fingerprint 85 CE 1C BA 79 8D 51 8C B9 25 FB EE E0 C3 E5 7C
|