[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Chicken-users] 1069
From: |
felix |
Subject: |
[Chicken-users] 1069 |
Date: |
Sat, 20 Jul 2002 00:00:11 +0200 |
Hello. I successfully survived one week of beach-holidays and
the telecom-bozos have apparently managed to get my phone-
line working...
Everything back to normal (phew!)
Build 1069 of the bird-which-isnt-to-be-named is available here:
http://www-call-with-current-continuation.org/chicken-0.1069.tar.gz
The compiler now outputs comments in FFI-stub headers and
before every compiled function (many small procedures are
generated in the process of translation from Scheme to CPS C,
this is where all those "kNNN" names come from, but the
comments should still be quite useful). Note that the whole
system hasn't yet been completely recompiled with itself, so
only some files contain comments. If you want the comments,
just compile the appropriate library files, like this:
% chicken library.scm -debug-level 0 -optimize-level 2 -no-warnings \
-output-file library.c
And re-making should build the new system.
If this is annoying, please tell me and I will upload a complete
version ASAP.
Moreover, a bug on Windows/DOS has been fixed that resulted
in problems when large amounts of data where allocated on
machines with a stack in low memory: the demand-check could
overflow because of this expression in nursery-size tests:
(assuming a downward growing stack)
(C_stack_pointer - (n)) > C_stack_limit
This can overflow if `n' is larger than `C_stack_pointer'.
Now we use:
(C_stack_pointer - C_stack_limit) > (n)
(I hope I have made the correct changes in `chicken.h'. Everything
*should* work properly on other machines, and machines with an upward
growing stack. Please tell me if there are any problems).
cheers,
felix
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- [Chicken-users] 1069,
felix <=