[RTEMS Project] #2274: Enable libgomp build in GCC

RTEMS trac trac at rtems.org
Thu Jul 23 07:59:54 UTC 2015


#2274: Enable libgomp build in GCC
-----------------------------+------------------------------
 Reporter:  sebastian.huber  |       Owner:  sebastian.huber
     Type:  enhancement      |      Status:  accepted
 Priority:  normal           |   Milestone:  4.11.1
Component:  GCC              |     Version:  4.11
 Severity:  normal           |  Resolution:
 Keywords:                   |
-----------------------------+------------------------------

Comment (by sebastian.huber):

 Performance with self-contained objects defined in Newlib <sys/lock.h>.
 The barrier implementation is virtually identical to the Linux futex
 barrier present in libgomp.

 Microbench (2 processor T4240):

 barrier bench 0.47163 seconds
 parallel bench 0.355602 seconds
 static bench 0.0223162 seconds
 dynamic bench 0.181986 seconds
 guided bench 0.00362983 seconds
 runtime bench 0.180831 seconds
 single bench 0.106319 seconds

 [[Image(libgomp-parallel-bench-sys-lock.png, center, 75%)]]

 Microbench (24 processor T4240):

 barrier bench 5.74687 seconds
 parallel bench 2.38893 seconds
 static bench 0.118236 seconds
 dynamic bench 0.2516 seconds
 guided bench 0.00146854 seconds
 runtime bench 0.250789 seconds
 single bench 0.543456 seconds

 This is a major improvement compared to the previous versions. In the
 parallel bench profile, the only operating system function is
 _Futex_Wake() with 13% processor utilization. This is all right, since
 barrier operations are heavily used in this test case.

--
Ticket URL: <http://devel.rtems.org/ticket/2274#comment:6>
RTEMS Project <http://www.rtems.org/>
RTEMS Project


More information about the bugs mailing list