svn+ssh://pythondev@svn.python.org/python/branches/py3k
........
r83921 | antoine.pitrou | 2010-08-10 01:39:31 +0200 (mar., 10 août 2010) | 4 lines
Issue #6915: Under Windows, os.listdir() didn't release the Global
Interpreter Lock around all system calls. Original patch by Ryan Kelly.
........
svn+ssh://pythondev@svn.python.org/python/branches/py3k
........
r83088 | ronald.oussoren | 2010-07-23 14:53:51 +0100 (Fri, 23 Jul 2010) | 8 lines
This fixes issue7900 by adding code that deals
with the fact that getgroups(2) might return
more that MAX_GROUPS on OSX.
See the issue (and python-dev archives) for the
gory details. Summarized: OSX behaves rather oddly
and Apple says this is intentional.
........
os.kill takes one of two newly added signals, CTRL_C_EVENT and
CTRL_BREAK_EVENT, or any integer value. The events are a special case
which work with subprocess console applications which implement a
special console control handler. Any other value but those two will
cause os.kill to use TerminateProcess, outright killing the process.
This change adds win_console_handler.py, which is a script to implement
SetConsoleCtrlHandler and applicable handler function, using ctypes.
subprocess also gets another attribute which is a necessary flag to
creationflags in Popen in order to send the CTRL events.
- Issue #7714: Use ``gcc -dumpversion`` to detect the version of GCC on
MacOSX.
- Make configure look for util.h as well as libutil.h. The former
is the header file that on OSX contains the defition of openpty.
(Needed to compile for OSX 10.4 on OSX 10.6)
- Use the correct definition of CC to compile the pythonw executable
acquiring the import lock around fork() calls. This prevents other threads
from having that lock while the fork happens, and is the recommended way of
dealing with such issues. There are two other locks we care about, the GIL
and the Thread Local Storage lock. The GIL is obviously held when calling
Python functions like os.fork(), and the TLS lock is explicitly reallocated
instead, while also deleting now-orphaned TLS data.
This only fixes calls to os.fork(), not extension modules or embedding
programs calling C's fork() directly. Solving that requires a new set of API
functions, and possibly a rewrite of the Python/thread_*.c mess. Add a
warning explaining the problem to the documentation in the mean time.
This also changes behaviour a little on AIX. Before, AIX (but only AIX) was
getting the import lock reallocated, seemingly to avoid this very same
problem. This is not the right approach, because the import lock is a
re-entrant one, and reallocating would do the wrong thing when forking while
holding the import lock.
Will backport to 2.6, minus the tiny AIX behaviour change.
Dynamically discoverd the size of the ioinfo struct used by the crt for its file descriptors. This should work across all flavors of the CRT. Thanks to Amaury Forgeot d'Arc
Needs porting to 3.1
process rather than both parent and child.
Does anyone actually use fork1()? It appears to be a Solaris thing
but if Python is built with pthreads on Solaris, fork1() and fork()
should be the same.
uid and gid input to accept values >=2**31 as valid while still accepting
negative numbers to pass -1 to chown for "no change".
Fixes issue1747858.
This should be backported to release25-maint.
Antoine Pitrou
Ronald Oussoren
Stefan Krah
Senthil Kumaran
Amaury Forgeot d'Arc
Victor Stinner
Brian Curtin
Matthias Klose
Gregory P. Smith
Martin v. Löwis
Thomas Wouters
Hirokazu Yamamoto
Kristján Valur Jónsson
Mark Dickinson
Benjamin Peterson
Georg Brandl
Facundo Batista
Christian Heimes
Brett Cannon