On Saturday 06 February 2010 05:43:21 pm Christoph Pospiech wrote:
> Hi,
> 
 [...]
> 
> The output is as follows.
> 
> <map>
>       <host name="sirius" slots="1" max_slots="0">
>               <process rank="0"/>
>               <process rank="1"/>
>       </host>
> </map>
> <stdout rank="0">main: before ClockTest</stdout>
> <stdout rank="1">Before next</stdout>
> <stdout rank="0">Before next</stdout>
> [...hang...]
> 

To add further information on this case. the two tasks are taking 97% CPU load 
each (I have a two processor system). Attaching a gdb to the tasks, I am 
getting the following.

First task -

(gdb) where
#0  0x009a1422 in __kernel_vsyscall ()
#1  0x00a77e15 in pthread_cond_wait@@GLIBC_2.3.2 ()
    at 
../nptl/sysdeps/unix/sysv/linux/i386/i686/../i486/pthread_cond_wait.S:122
#2  0x08194aec in x10::lang::Thread::park () at x10/lang/Thread.cc:371
#3  0x08145ead in x10::lang::Runtime__Monitor::await (this=0x8e2abc0)
    at gen/x10/lang/Runtime__Monitor.cc:35
#4  0x081c82b8 in x10::lang::Runtime__Latch::await (this=0x8e2abc0)
    at gen/x10/lang/Runtime__Latch.cc:44
#5  0x0815c9b0 in x10::lang::Runtime__RootFinish::waitForFinish (
    this=0x8e2abc0, safe=false) at gen/x10/lang/Runtime__RootFinish.cc:159
#6  0x0812fdcb in x10::lang::Runtime::stopFinish ()
    at gen/x10/lang/Runtime.cc:784
#7  0x0810ed41 in Hello::ClockTest (this=0x8e437d0)
    at /home/cp/xample/MatMul/bummer_X10/bin/x10-generated-dir/Hello.cc:74
#8  0x0810ee2c in Hello::main (args=...)
    at /home/cp/xample/MatMul/bummer_X10/bin/x10-generated-dir/Hello.cc:89
#9  0x08180e7a in x10aux::BootStrapClosure::apply (this=0x8e0bf30)
    at ./x10aux/bootstrap.h:81
#10 0x081419b8 in x10_lang_Runtime__closure__1::apply() ()
#11 0x0815552e in x10::lang::Activity::run (this=0x8e2aee0)
    at gen/x10/lang/Activity.cc:138
#12 0x0815484a in x10_lang_Runtime__Worker__closure__0::apply() ()
---Type <return> to continue, or q <return> to quit---
#13 0x08129508 in x10::lang::Runtime::runAtLocal (id=0, body=...)
    at gen/x10/lang/Runtime.cc:37
#14 0x08154039 in x10::lang::Runtime__Worker::loop (this=0x8e2af60, latch=..., 
    block=true) at gen/x10/lang/Runtime__Worker.cc:247
#15 0x0815419a in x10::lang::Runtime__Worker::apply (this=0x8e2af60)
    at gen/x10/lang/Runtime__Worker.cc:119
#16 0x081c8a8e in x10::lang::Runtime__Pool::apply (this=0x8e25f78)
    at gen/x10/lang/Runtime__Pool.cc:154
#17 0x0812f0ca in x10::lang::Runtime::start (init=..., body=...)
    at gen/x10/lang/Runtime.cc:347
#18 0x0810e08c in x10aux::template_main<x10::lang::Runtime, Hello> (ac=1, 
    av=0xbfdec934)
    at 
/home/cp/.eclipse/org.eclipse.platform_3.5.0_155965261/plugins/x10.dist.linux.x86.fragment_2.0.1.v201002031506/include/x10aux/bootstrap.h:141
#19 0x0810d27f in main (ac=1, av=0xbfdec934)
    at /home/cp/xample/MatMul/bummer_X10/bin/x10-generated-
dir/xxx_main_xxx.cc:4

Second task -

(gdb) where
#0  0x0068a422 in __kernel_vsyscall ()
#1  0x007d5e15 in pthread_cond_wait@@GLIBC_2.3.2 ()
    at 
../nptl/sysdeps/unix/sysv/linux/i386/i686/../i486/pthread_cond_wait.S:122
#2  0x08194aec in x10::lang::Thread::park () at x10/lang/Thread.cc:371
#3  0x08145ead in x10::lang::Runtime__Monitor::await (this=0x91e7ae0)
    at gen/x10/lang/Runtime__Monitor.cc:35
#4  0x081c82b8 in x10::lang::Runtime__Latch::await (this=0x91e7ae0)
    at gen/x10/lang/Runtime__Latch.cc:44
#5  0x081302c9 in x10::lang::Runtime::runAt (place=..., body=...)
    at gen/x10/lang/Runtime.cc:614
#6  0x08123873 in x10::lang::Clock::nextUnsafe (this=0x9201e18)
    at gen/x10/lang/Clock.cc:221
#7  0x0816bd69 in x10::lang::Runtime__ClockPhases::next (this=0x9206e60)
    at gen/x10/lang/Runtime__ClockPhases.cc:90
#8  0x0812fb16 in x10::lang::Runtime::next () at gen/x10/lang/Runtime.cc:709
#9  0x08112245 in Hello__closure__0::apply (this=0x921fff0)
    at /home/cp/xample/MatMul/bummer_X10/bin/x10-generated-dir/Hello.inc:19
#10 0x0815552e in x10::lang::Activity::run (this=0x9206ea0)
    at gen/x10/lang/Activity.cc:138
#11 0x0815484a in x10_lang_Runtime__Worker__closure__0::apply() ()
#12 0x08129508 in x10::lang::Runtime::runAtLocal (id=1, body=...)
    at gen/x10/lang/Runtime.cc:37
---Type <return> to continue, or q <return> to quit---
#13 0x08154039 in x10::lang::Runtime__Worker::loop (this=0x9206f60, latch=..., 
    block=true) at gen/x10/lang/Runtime__Worker.cc:247
#14 0x0815419a in x10::lang::Runtime__Worker::apply (this=0x9206f60)
    at gen/x10/lang/Runtime__Worker.cc:119
#15 0x081c8a8e in x10::lang::Runtime__Pool::apply (this=0x9201f78)
    at gen/x10/lang/Runtime__Pool.cc:154
#16 0x0812f439 in x10::lang::Runtime::start (init=..., body=...)
    at gen/x10/lang/Runtime.cc:383
#17 0x0810e08c in x10aux::template_main<x10::lang::Runtime, Hello> (ac=1, 
    av=0xbf807ab4)
    at 
/home/cp/.eclipse/org.eclipse.platform_3.5.0_155965261/plugins/x10.dist.linux.x86.fragment_2.0.1.v201002031506/include/x10aux/bootstrap.h:141
#18 0x0810d27f in main (ac=1, av=0xbf807ab4)
    at /home/cp/xample/MatMul/bummer_X10/bin/x10-generated-
dir/xxx_main_xxx.cc:4

It indeed appears that one task is waiting on the finish, the other on the 
next. But still, if this is an illegal combination of finish and clocks, - why 
did the compiler let me get away with it ?
-- 

Mit freundlichen Grüßen / Kind regards

Dr. Christoph Pospiech
High Performance & Parallel Computing
Phone:  +49-351 86269826
Mobile: +49-171-765 5871
E-Mail: christoph.pospi...@de.ibm.com
-------------------------------------
IBM Deutschland GmbH
Vorsitzender des Aufsichtsrats: Erich Clementi 
Geschäftsführung: Martin Jetter (Vorsitzender), 
Reinhard Reschke, Christoph Grandpierre, 
Klaus Lintelmann, Michael Diemer, Martina Koederitz 
Sitz der Gesellschaft: Ehningen / Registergericht: Amtsgericht Stuttgart, HRB 
14562 WEEE-Reg.-Nr. DE 99369940


------------------------------------------------------------------------------
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com
_______________________________________________
X10-users mailing list
X10-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/x10-users

Reply via email to