| commit | 2abefb68404ae6cc38196e4faa766ed0a167cd69 | [log] [tgz] |
|---|---|---|
| author | Hridya Valsaraju <hridya@google.com> | Thu Jan 19 13:06:58 2017 -0800 |
| committer | Hridya Valsaraju <hridya@google.com> | Mon Feb 13 18:41:26 2017 +0000 |
| tree | 20cd441c568651be2647b5da85e21e3fa79487ec | |
| parent | 8f0e8e55c7190db39fd87ee31180dd6a78e50f1d [diff] |
Correct timeOut handling in readBlocking/writeBlocking. Currently, if the EventFlag wait() returns due to a spurious wake or if a notification was received but there is a need to go back to waiting due to insufficient data to read or space to write, the timeOut parameter is not adjusted to account for the time already spent waiting. This CL fixes the same. Bug: 34290175 Test: FMQ unit tests Change-Id: I71399b23d88d5f3fc91ca1a9c0cf9d3ea35e8290