|
Home About Us A-Z Index Search * Contact Us Register Login Press ShopThe Open Brand -- Problem Reporting and Interpretations System |
Problem Report 1155 Details
Show help | Quick Search | Submit a Test Suite Support Request | Click here to view your privileges
This page provides all information on Problem Report 1155.
Report 1155 Actions
Problem Report Number 1155 Submitter's Classification Test Suite problem State Resolved Resolution Rejected (REJ) Problem Resolution ID REJ.X.0357 Raised 1970-01-01 08:00 Updated 2003-03-13 08:00 Published null Product Standard Internationalised System Calls and Libraries Extended (UNIX 95) Certification Program The Open Brand certification program Test Suite VSU version 4.1.0 Test Identification Base/sigaction 14,19 Problem Summary PG4U.00113 This request asks for rulings on 2 separate issues at once. It is refuesed since an interpretation may only deal with a single issue. Problem Text
Both of these tests are highly timing dependent and as a result may fail on an
intermittent basis.
In both sigaction #14 and sigaction #19, the tests set up the action for the
SIGCHLD signal. In the case of sigaction #14, the test sets the signal handler
to SIG_DFL and also sets the SA_NOCLDWAIT flag. For sigaction #19, the signal
handler is set to SIG_IGN.
After calling sigaction(), the tests fork a child that exits and then fork a
child that sleeps. In all except for the first test case in sigaction #14
(wait()), this sleep is for 1 second. The tests then call time(&time1), then
one of wait(), waitpid(), wait3(), or waitid() expecting a return of -1 and
errno set to ECHILD. The tests then call time(&time2), expecting that
time2-time1 >= 1. If the value returned is less than one, the tests fail on
the assumption that the wait function under test did not block.
The sleep(1) in the child does not guarantee that the parent will have time to
block for a full second. The calls to the wait() functions are returning the
expected results, however depending on what the scheduler does, the timing can
fall on either side of the tested time interval. Increasing the sleep time
will more likely guarantee that the assertion under test will not be subject
to implementation specific timing/scheduling differences. The real problem
however, is that as the tests are currently written, there is a race condition
between the parent and child and as such there is no guarantee that the
the call time(&time1) will occur before the sleep(). A way to guarantee that
this will occur is to move the first call to time() to before the call to
fork().Test Output
TEST CASE: sigaction
TEST PURPOSE #14
After a call to int sigaction(int sig, const struct
sigaction *act, struct sigaction *oact) when
SA_NOCLDWAIT is set in the sa_flags member of the
structure pointed to by act and sig is SIGCHLD, if the
calling process subsequently waits for its children
and the process has no unwaited for children that were
transformed into zombie processes it shall block until
all its children terminate, and wait(), wait3(),
waitid(), and waitpid() shall fail and set errno to
ECHLD.
PREP: Set up action for SIGCHLD
PREP: fork() a child that exits
PREP: fork() a child that sleeps
TEST: wait() returns -1 and errno = ECHILD
TEST: wait() blocks until child exits
PREP: fork() a child that sleeps
TEST: waitpid() returns -1 and errno = ECHILD
TEST: waitpid() blocks until child exits
PREP: fork() a child that sleeps
TEST: wait3() returns -1 and errno = ECHILD
TEST: wait3() blocks until child exits
PREP: fork() a child that sleeps
TEST: waitid() returns -1 and errno = ECHILD
TEST: waitid() blocks until child exits
ERROR: waitid() did not block
14 FAIL
TEST PURPOSE #19
After a call to int sigaction(int sig, const struct
sigaction *act, struct sigaction *oact) when the
action for SIGCHLD is set to SIG_IGN if the calling
process subsequently waits for its children and the
process has no unwaited for children that were
transformed into zombie processes it shall block until
all its children terminate, and wait(), wait3(),
waitid(), and waitpid() shall fail and set errno to
ECHILD.
PREP: Set SIG_IGN for SIGCHLD
PREP: fork() a child that exits
PREP: fork() a child that sleeps
TEST: wait() returns -1 and errno = ECHILD
TEST: wait() blocks until child exits
PREP: fork() a child that sleeps
TEST: waitpid() returns -1 and errno = ECHILD
TEST: waitpid() blocks until child exits
PREP: fork() a child that sleeps
TEST: wait3() returns -1 and errno = ECHILD
TEST: wait3() blocks until child exits
ERROR: wait3() did not block
PREP: fork() a child that sleeps
TEST: waitid() returns -1 and errno = ECHILD
TEST: waitid() blocks until child exits
19 FAILReview Information
Review Type TSMA Review Start Date null Completed null Status Complete Review Recommendation No Resolution Given Review Response
We recommend this request be refused since the submitter is
asking for rulings on 2 separate issues at once.
We suggest the submitter file 2 new IRs; one for each issue.
At that time we will recommend a TSD for the sleep issue.
But we will recommend that the race condition issue be rejected.
We believe the test should work as long as the time() precedes
the wait*(). Whether the time() precedes the start of the
sleep() or occurs while the child is sleeping should not matter.
Review Type SA Review Start Date null Completed null Status Complete Review Resolution Rejected (REJ) Review Conclusion
This request is refused.
Problem Reporting System Options:
- View Report 1155
- List All PRs
- Search Reports
- Email the System Administrator
- View the The Open Brand Interpretations Database User Manual
Contact the Certification Authority