Bruce Momjian a4c40f140d Here's an attempt at new socket and signal code for win32. ... It works on the principle of turning sockets into non-blocking, and then emulate blocking behaviour on top of that, while allowing signals to run. Signals are now implemented using an event instead of APCs, thus getting rid of the issue of APCs not being compatible with "old style" sockets functions. It also moves the win32 specific code away from pqsignal.h/c into port/win32, and also removes the "thread style workaround" of the APC issue previously in place. In order to make things work, a few things are also changed in pgstat.c: 1) There is now a separate pipe to the collector and the bufferer. This is required because the pipe will otherwise only be signalled in one of the processes when the postmaster goes down. The MS winsock code for select() must have some kind of workaround for this behaviour, but I have found no stable way of doing that. You really are not supposed to use the same socket from more than one process (unless you use WSADuplicateSocket(), in which case the docs specifically say that only one will be flagged). 2) The check for "postmaster death" is moved into a separate select() call after the main loop. The previous behaviour select():ed on the postmaster pipe, while later explicitly saying "we do NOT check for postmaster exit inside the loop". The issue was that the code relies on the same select() call seeing both the postmaster pipe *and* the pgstat pipe go away. This does not always happen, and it appears that useing WSAEventSelect() makes it even more common that it does not. Since it's only called when the process exits, I don't think using a separate select() call will have any significant impact on how the stats collector works. Magnus Hagander		2004-04-12 16:19:18 +00:00
..
access	This is a cleanup patch for access/transam/xact.c. It only removes some	2004-04-05 03:11:39 +00:00
bootstrap	Replace TupleTableSlot convention for whole-row variables and function	2004-04-01 21:28:47 +00:00
catalog	check_sql_fn_retval has always thought that we supported doing	2004-04-02 23:14:08 +00:00
commands	ALTER SEQUENCE RESTART did the wrong thing if sequence last_value was	2004-04-06 16:39:30 +00:00
executor	Still another place to make the world safe for zero-column tables.	2004-04-07 18:46:12 +00:00
lib	make sure the $Id tags are converted to $PostgreSQL as well ...	2003-11-29 22:41:33 +00:00
libpq	Here's an attempt at new socket and signal code for win32.	2004-04-12 16:19:18 +00:00
main	Windows uses codepages rather than the environment, so we work around	2004-03-15 16:14:26 +00:00
nodes	Reimplement CASE val WHEN compval1 THEN ... WHEN compval2 THEN ... END	2004-03-17 20:48:43 +00:00
optimizer	Extend set-operation planning to keep track of the sort ordering induced	2004-04-07 18:17:25 +00:00
parser	Allow LIKE/ILIKE to appear in more places in a query.	2004-04-05 03:07:26 +00:00
po	Translation updates	2004-04-09 06:49:35 +00:00
port	Here's an attempt at new socket and signal code for win32.	2004-04-12 16:19:18 +00:00
postmaster	Here's an attempt at new socket and signal code for win32.	2004-04-12 16:19:18 +00:00
regex	$Header: -> $PostgreSQL Changes ...	2003-11-29 19:52:15 +00:00
rewrite	Replace TupleTableSlot convention for whole-row variables and function	2004-04-01 21:28:47 +00:00
storage	When changing select() calls for delays into pg_usleep(), two comments	2004-03-23 21:39:46 +00:00
tcop	Exit backend from SIGTERM or FATAL by simulating client EOF, rather than	2004-04-11 00:54:45 +00:00
utils	Exit backend from SIGTERM or FATAL by simulating client EOF, rather than	2004-04-11 00:54:45 +00:00
Makefile	Briefly,	2004-02-02 00:11:31 +00:00
nls.mk	$Header: -> $PostgreSQL Changes ...	2003-11-29 19:52:15 +00:00