Bug 154471

Summary: pthread_cancel() cannot kill threads waiting on the same condition
Product: Red Hat Enterprise Linux 3 Reporter: Steven Lee <shl1>
Component: glibcAssignee: Jakub Jelinek <jakub>
Status: CLOSED NOTABUG QA Contact: Brian Brock <bbrock>
Severity: medium Docs Contact:
Priority: medium    
Version: 3.0CC: vivi
Target Milestone: ---   
Target Release: ---   
Hardware: i686   
OS: Linux   
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2005-04-12 09:22:24 EDT Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---
Description Flags
C code to reproduce the problem none

Description Steven Lee 2005-04-11 21:12:08 EDT
Description of problem:

Create 2 threads that wait on the same condition.  Then try to call
pthread_cancel() to kill them.  The first thread is terminated correctly; the
second doesn't.

In the attached test code, t2 never gets terminated and pthread_join(t2, NULL)
never returns.

Version-Release number of selected component (if applicable):

kernel: kernel-smp-2.4.21-27.0.2.EL
gcc: gcc-3.2.3-49
glibc: glibc-2.3.2-95.30

kernel: kernel-2.6.9-5.0.3.EL
gcc: gcc-3.4.3-9.EL4
glibc: glibc-2.3.4-2

How reproducible: Always

Steps to Reproduce:

Compile the test code with:

1. gcc pthr_cancel_test.c -pthread
2. ./a.out
Actual results:

pthread_join(t2) never returns

Expected results:

pthread_join(t2) should return

Additional info:
Comment 1 Steven Lee 2005-04-11 21:12:08 EDT
Created attachment 113005 [details]
C code to reproduce the problem
Comment 2 Steven Lee 2005-04-11 21:14:42 EDT
BTW, if t1 and t2 waits on 2 different conditions, pthread_cancel() works correctly.
Comment 3 Jakub Jelinek 2005-04-12 09:22:24 EDT
That's the correct behaviour.
A side effect of acting on a cancellation request while a thread is blocked on a
condition variable is to re-acquire the
mutex before calling any of the cancellation cleanup handlers. This is done in
order to ensure that the cancellation
cleanup handler is executed in the same state as the critical code that lies both
before and after the call to the
condition wait function. This rule is also required when interfacing to POSIX
threads from languages, such as Ada or
C++, which may choose to map cancellation onto a language exception; this rule
ensures that each exception handler
guarding a critical section can always safely depend upon the fact that the
associated mutex has already been locked
regardless of exactly where within the critical section the exception was raised.
Without this rule, there would not be
a uniform rule that exception handlers could follow regarding the lock, and so
coding would become very cumbersome.

Because there is no pthread_cleanup_push/pop around the pthread_cond_wait call
that would unlock the mutex, the second thread is blocking on the mutex while
trying to re-acquire it.
Comment 4 Steven Lee 2005-04-12 10:12:02 EDT

Aha!  Thank you for the quick response to our sloppy coding :-)  Adding
pthread_cleanup_push() and pthread_cleanup_pop() to unblock the mutex worked