"184.108.40.206": threads: Add memory-barrier framework.
* New file, src/code/barrier.lisp, containing the baseline
* Added the barrier functions to the compiler function database.
* Export the interface macro, BARRIER, from SB!THREAD and the
* Document a new architecture-dependent build-time feature,
[6e9a41] by Christophe Rhodes
220.127.116.11: much faster MAKE-INITIAL-DFUN
Don't attempt to precompute any kind of effective methods in
This allows us to delete FINALIZE-SPECIALIZERS and
18.104.22.168: socket-connect was not thread safe
Committing diffs as supplied in the bug report after review and
Credits to Andrew Golding for the diffs and Jaap de Heer for the
[cea294] by Christophe Rhodes
22.214.171.124: call an appropriate existing ctor from MAKE-INSTANCE (CLASS)
The intuition here is that searching for a ctor is much cheaper than
Only call optimized ctors so that we don't risk endless recursion
[ab7aed] by Christophe Rhodes
126.96.36.199: better non-ctor make-instance
Actually, the really expensive bit is the initarg validity checking.
With this change, asdf.fasl loads about 10% faster.
[57eae6] by Christophe Rhodes
188.8.131.52: ctor machinery bugfixes
One bug was a simple harmless thinko, limiting the cache for validity
The other is actually a correctness problem: addition or removal of
184.108.40.206: fix return value of WRITE
Used to return NIL when the compiler macro for the constant keyword
220.127.116.11: win32: Fix foreign function name conversion.
* In 18.104.22.168, the win32 case for foreign function names was
22.214.171.124: Fix run-sbcl.sh to use absolute paths when necessary.
* The new ASDF 2 requires absolute paths in places where SBCL
126.96.36.199: "upgrade" to ASDF 2.003
188.8.131.52: more ASDF2 cleanup
* update the asdf README and texinfo documentation
* update asdf license
* fix pulling documentation from upstream
184.108.40.206: more asdf logical pathname cleanup
* use #p"..." for logical pathnames instead of 1) just "" or
220.127.116.11: asdf2 merge from upstream
* replaced asdf.lisp with asdf.lisp 2.102
* wrap logical pathname in sb-rotate-byte with #.(logical-pathname ...)
18.104.22.168: Use SB-VM:WORD in (DEFSTRUCT HASH-TABLE).
22.214.171.124: Reduce stress level of sb-concurrency tests.
Reduce the stress level of the MAILBOX tests by a factor of 10.
a) As contrib tests are run as part of a normal SBCL build, the
b) On BSD platforms, the tests were reported several times in past
126.96.36.199: tls-slot allocation not thread-safe
Committing James Knight's patch from
Tested on amd64 and x86 with same failures before and after.
It seems as though the ALLOC-TLS-INDEX-IN-* code is not thread-safe. I
But this test case shows that it's not.
sbcl --eval '(load (compile-file "thread-test.lisp"))' --eval "(run-test)" --eval "(quit)"
It will fail with something like:
SBCL 188.8.131.52. Debian Linux 2.6.32-amd64, x86-64.
Okay, turns out to be an utterly trivial problem. This is the
0x20000f38: mov %rbp,0xb8(%r12)
[03833f] by Andreas Fuchs
GIT GATEWAY: Remove files that were deleted in 2006.
These file removals were never picked up, probably due to a git bug.
* b66385e2031fc2cac17dd129df0af400beb48a22 (0.9.16.38)
Thanks to Cyrus Harmon for alerting me to this issue.
184.108.40.206: delete all traces of the darwin dlshim
* following up on previous commits which still used the shim with the
220.127.116.11: prepend an underscore to ldso stubs
* a leading underscore keeps the cross-compiling linker happy
18.104.22.168: restore darwin/ppc building
* fix ldso stubs to work without the dlshim
* hack to avoid calling dlclose on the main executable