Just Launched: You can now import projects and releases from Google Code onto SourceForge
We are excited to release new functionality to enable a 1-click import from Google Code onto the Allura platform on SourceForge. You can import tickets, wikis, source, releases, and more with a few simple steps. Read More
From: Brian Mastenbrook <bmastenb@cs...> - 2004-12-14 01:41:07
SBCL does not currently build for me on OS X. Current CVS fails with
the strange error in http://paste.lisp.org/display/4312 - even when
building with a different optimization level and different minor
version of gcc 3.1 - and just for kicks, I tried 0.8.17.23, which
failed with the following much more understandable error:
debugger invoked on a UNDEFINED-FUNCTION in thread 16567:
The function SB-IMPL::DYNAMIC-FOREIGN-SYMBOLS is undefined.
You can type HELP for debugger help, or (SB-EXT:QUIT) to exit from SBCL.
restarts (invokable by number or by possibly-abbreviated name):
0: [ABORT ] Reduce debugger level (leaving debugger, returning to
1: [TOPLEVEL] Restart at toplevel READ/EVAL/PRINT loop.
0: ("undefined function")
1: (SB-KERNEL:FOREIGN-DEINIT 0)[:EXTERNAL]
2: (SAVE-LISP-AND-DIE 3 "output/sbcl.core")[:EXTERNAL]
(SAVE-LISP-AND-DIE "output/sbcl.core" :PURIFY T)
(rest of backtrace elided)
It's my guess that builds without linkage tables are currently broken;
someone who knows more about them would have to take a look and finger
the guilty party. I don't know why the backtrace from HEAD is so
garbled, however. Perhaps one of the arithmetic changes had a subtle
From: Raymond Wiker <Raymond.W<iker@fa...> - 2004-12-14 06:34:06
Brian Mastenbrook writes:
> Hello all,
> SBCL does not currently build for me on OS X. Current CVS fails with
> the strange error in http://paste.lisp.org/display/4312 - even when
> building with a different optimization level and different minor
> version of gcc 3.1 - and just for kicks, I tried 0.8.17.23, which
> failed with the following much more understandable error:
I saw the same problem at the end of last week, and tried to
step through the forms in make-target-2.sh. The build fails in the call to
sb-ext:save-lisp-and-die, but if I choose either of the restarts and
retry sb-ext:save-lisp-and-die, it succeeds. The resulting core can be
loaded, and exhibits the same behaviour with respect to
> debugger invoked on a UNDEFINED-FUNCTION in thread 16567:
> The function SB-IMPL::DYNAMIC-FOREIGN-SYMBOLS is undefined.
This was fixed around 0.8.17.25, I think.