Austin Group Defect Tracker

Aardvark Mark III


Viewing Issue Simple Details Jump to Notes ] Issue History ] Print ]
ID Category Severity Type Date Submitted Last Update
0000095 [1003.1(2008)/Issue 7] System Interfaces Comment Omission 2009-06-29 20:07 2009-07-01 18:04
Reporter msbrown View Status public  
Assigned To ajosey
Priority normal Resolution Rejected  
Status Closed  
Name Michael Kerrisk
Organization Linux Foundation
User Reference
Section timer_getoverrun()
Page Number 2114
Line Number 66945
Interp Status ---
Final Accepted Text
Summary 0000095: timer_getoverrun() and notification
Description _____________________________________________________________________________
 COMMENT Enhancement Request Number 25
 mtk.lists:xxxxxxxxx Defect in XSH timer_getoverrun() (rdvk# 2)
 {timer_getoverrun} Wed, 11 Feb 2009 00:32:04 GMT
 _____________________________________________________________________________

The discussion of timer overruns in lines 66945 to 66956
 is phrased entirely in terms of timer notification via
 signals (SIGEV_SIGNAL). However, a timer overrun can also
 occur for for notification via a thread (SIGEV_THREAD),
 since there may be a delay between the timer expiration
 and the invocation of the corresponding thread, and in
 that time, further expirations may occur. Linux, at
 least, supports timer_getoverrun() in the context of
 timer notification via a thread, and perhaps other
  systems also do. I suspect the standard should say
 something about this, an that it may be an accident of
 history that it doesn't.
Desired Action After line 66956, add text something like the following:

 An implementation may also support the use of
 timer_gettoverrun() for retrieving the overrun count
 when timer expirations are notified via invocation of
 a thread (SIGEV_THREAD). In this case, the overrun
 count reflects the number of extra timer expirations
 that have occurred between the first timer expiration
 and the time that the notification thread was invoked.
Tags No tags attached.
Attached Files

- Relationships

-  Notes
(0000136)
msbrown (manager)
2009-06-29 20:07

This aardvark request was rejected since time_getoverrun() is only defined
if the expiration causes signals to be generated.

timer_setoverrun() reports the number of signals not generated because
of overruns. When threads are used instead of signals, a thread will
be created for each event, independent of whether prior threads
handling these events have started or completed.

- Issue History
Date Modified Username Field Change
2009-06-29 20:07 msbrown New Issue
2009-06-29 20:07 msbrown Status New => Under Review
2009-06-29 20:07 msbrown Assigned To => ajosey
2009-06-29 20:07 msbrown Name => Mark Brown
2009-06-29 20:07 msbrown Organization => IBM
2009-06-29 20:07 msbrown Section => timer_getoverrun()
2009-06-29 20:07 msbrown Page Number => 2114
2009-06-29 20:07 msbrown Line Number => 66945
2009-06-29 20:07 msbrown Note Added: 0000136
2009-06-29 20:07 msbrown Status Under Review => Closed
2009-06-29 20:07 msbrown Resolution Open => Rejected
2009-07-01 18:04 Don Cragun Name Mark Brown => Michael Kerrisk
2009-07-01 18:04 Don Cragun Organization IBM => Linux Foundation


Mantis 1.1.6[^]
Copyright © 2000 - 2008 Mantis Group
Powered by Mantis Bugtracker