[Mono-bugs] [Bug 74154][Maj] Changed - Mutex class locks on OS X in simple application
Mon, 4 Apr 2005 09:38:36 -0400 (EDT)
Please do not reply to this email- if you want to comment on the bug, go to the
URL shown below and enter your comments there.
Changed by email@example.com.
--- shadow/74154 2005-03-30 18:16:56.000000000 -0500
+++ shadow/74154.tmp.24487 2005-04-04 09:38:36.000000000 -0400
@@ -63,6 +63,18 @@
has been tested and the condition must be tested afterwards to account
for races, spurious wakeups etc.
The backtraces I saw point at a pthread_cod_wait called by
_wapi_handle_wait_signal_handle in io-layer/handles.c: it doesn't look
like the pattern of checking the condition before and after (in a
loop) is followed here and this could be the cause of this bug.
+------- Additional Comments From firstname.lastname@example.org 2005-04-04 09:38 -------
+Theres good news and theres bad news...
+The bad news is that this seems to be due to a particularly stupid
+design mistake, in that the handle mutexes are kept in shared memory
+and multiple processes are locking them. This happened to work in
+linux, but is breaking in macosx.
+The good news is that this mistake has been rectified in the new io
+code, which should be going into widespread testing imminently.