View Issue Details
| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 0000388 | 1003.1(2008)/Issue 7 | System Interfaces | public | 2011-02-28 18:09 | 2011-05-05 16:10 |
| Reporter | dvhart | Assigned To | ajosey | ||
| Priority | normal | Severity | Objection | Type | Enhancement Request |
| Status | Closed | Resolution | Rejected | ||
| Name | Darren Hart | ||||
| Organization | Intel | ||||
| User Reference | |||||
| Section | pthread_cond_init | ||||
| Page Number | Of which document? | ||||
| Line Number | Of which document? | ||||
| Interp Status | --- | ||||
| Final Accepted Text | |||||
| Summary | 0000388: pthread condvars are not priority inheritance aware | ||||
| Description | When using a PTHREAD_PRIO_INHERIT mutex with a condvar, the pthread_cond* calls can still cause an unbounded priority inversion via the internal condvar lock. The POSIX specification doesn't provide a mechanism to specify the protocol of the condvar. We would like to do this at runtime, but unfortunately it is legal to call pthread_cond_signal() or pthread_cond_broadcast() without first waiting on the lock, so the mutex type may not be known the first time the condvar is used. A new API, pthread_condattr_setprotocol() and pthread_condattr_getprotocol(), would allow the user to create a PTHREAD_PRIO_INHERIT condvar. This would use a PTHREAD_PRIO_INHERIT mutex for the internal condvar lock, eliminating the potential for hitting an unbounded priority inversion on that lock. This topic was initially discussed here: http://sources.redhat.com/ml/libc-alpha/2010-01/msg00011.html And again here: http://sources.redhat.com/ml/libc-alpha/2010-02/msg00089.html Test cases and patches for evaluation have been provided in the bugzilla record for glibc: http://sources.redhat.com/bugzilla/show_bug.cgi?id=11588 | ||||
| Desired Action | Add pthread_condattr_setprotocol() and pthread_condattr_getprotocol() to the specification in order to specify PTHREAD_PRIOR_INHERIT in the condattr at pthread_cond_init() time. | ||||
| Tags | No tags attached. | ||||
|
|
Given that there is no existing practice for this, we are closing until there is an existing implementation, or until a complete specification for one is received. At such time this issue can be re-evaluated. |
| Date Modified | Username | Field | Change |
|---|---|---|---|
| 2011-02-28 18:09 | dvhart | New Issue | |
| 2011-02-28 18:09 | dvhart | Status | New => Under Review |
| 2011-02-28 18:09 | dvhart | Assigned To | => ajosey |
| 2011-02-28 18:09 | dvhart | Name | => Darren Hart |
| 2011-02-28 18:09 | dvhart | Organization | => Intel |
| 2011-02-28 18:09 | dvhart | Section | => pthread_cond_init |
| 2011-02-28 18:09 | dvhart | Page Number | => Of which document? |
| 2011-02-28 18:09 | dvhart | Line Number | => Of which document? |
| 2011-05-05 16:10 | msbrown | Interp Status | => --- |
| 2011-05-05 16:10 | msbrown | Note Added: 0000778 | |
| 2011-05-05 16:10 | msbrown | Status | Under Review => Closed |
| 2011-05-05 16:10 | msbrown | Resolution | Open => Rejected |
