[RTEMS Project] #2129: sync() implementation is dangerous

[RTEMS trac]

#2129: sync() implementation is dangerous
-----------------------------+------------------------------
 Reporter:  sebastian.huber  |       Owner:  sebastian.huber
     Type:  defect           |      Status:  assigned
 Priority:  normal           |   Milestone:  4.11
Component:  cpukit           |     Version:  HEAD
 Severity:  normal           |  Resolution:
 Keywords:                   |
-----------------------------+------------------------------
Changes (by joel.sherrill):

 * owner:  joel.sherrill => sebastian.huber
 * status:  new => assigned


Old description:

> The sync() implementation has some weaknesses.
>
> 1. No protection of concurrent sync() operations.
> 2. Since we have no Newlib locks, other threads must not use file
> operations during the sync() period.
> 3. Deletion of threads during the sync() may lead to all sorts of things.

New description:

 The sync() implementation has some weaknesses.

 1. No protection of concurrent sync() operations.
 2. Since we have no Newlib locks, other threads must not use file
 operations during the sync() period.
 3. Deletion of threads during the sync() may lead to all sorts of things.

--

--
Ticket URL: <http://devel.rtems.org/ticket/2129#comment:2>
RTEMS Project <http://www.rtems.org/>
RTEMS Project