Project

General

Profile

Bug #1510

Hang in tests/charm++/chkpt when using -tracemode perfReport

Added by Sam White 4 months ago. Updated 4 months ago.

Status:
New
Priority:
Normal
Assignee:
Category:
Tracing
Target version:
Start date:
04/17/2017
Due date:
% Done:

0%


Description

Besides this one test, running make test OPTS="-tracemode perfReport" passes all tests.

../../../bin/testrun  ./hello +p4 +restart log  ++local
Charmrun> scalable start enabled. 
Charmrun> started all node programs in 0.005 seconds.
Charm++> Running in non-SMP mode: numPes 4
Converse/Charm++ Commit ID: v6.8.0-beta1-46-g80fa50245
Trace: traceroot: ./hello
Charm++> scheduler running in netpoll mode.
CharmLB> Load balancer assumes all CPUs are same.
Charm++> Running on 1 unique compute nodes (8-way SMP).
Charm++> cpu topology info is gathered in 0.000 seconds.
Received 1 arguments: { |./hello| }
Main's MigCtor. a=987(0x1011404fc), b[0]=654(0x101140500), b[1]=321, old PE number 4
Main's PUPer. a=123(0x1011404fc), b[0]=456(0x101140500), b[1]=789
[3] data on Group 3
[1] data on Group 1
[2] data on Group 2
[3] data on NOdeGroup 3
[1] data on NOdeGroup 1
CHello's PUPer. step=3.
[0] data on Group 0
[2] data on NOdeGroup 2
[0] data on NOdeGroup 0
[0]CkRestartMain done. sending out callback.
myClient. a=123(0x1011404fc), b[0]=456(0x101140500), b[1]=789
step 4 done
myClient. a=123(0x1011404fc), b[0]=456(0x101140500), b[1]=789
step 5 done
myClient. a=123(0x1011404fc), b[0]=456(0x101140500), b[1]=789
step 6 done
myClient. a=123(0x1011404fc), b[0]=456(0x101140500), b[1]=789
step 7 done
myClient. a=123(0x1011404fc), b[0]=456(0x101140500), b[1]=789
step 8 done
myClient. a=123(0x1011404fc), b[0]=456(0x101140500), b[1]=789
step 9 done

Related issues

Related to Charm++ - Bug #1201: SMP builds segfault on NULL lock in tests/charm++/chkpt New 09/07/2016
Related to Charm++ - Bug #1616: Race condition during ckRestart causes hang after merging patch for bug #1576 Merged 06/21/2017

History

#1 Updated by Phil Miller 4 months ago

I think we can reasonably defer this, unless we think it's indicative of some broader deficiency that users will encounter.

#2 Updated by Sam White 4 months ago

  • Target version changed from 6.8.0 to 6.8.1

#3 Updated by Sam White 4 months ago

Trying this on linux now (instead of darwin) it works, so we should be safe to add "-tracemode perfReport" to netlrts-linux or something on autobuild.

#4 Updated by Phil Miller about 1 month ago

  • Related to Bug #1201: SMP builds segfault on NULL lock in tests/charm++/chkpt added

#5 Updated by Phil Miller about 1 month ago

  • Related to Bug #1616: Race condition during ckRestart causes hang after merging patch for bug #1576 added

Also available in: Atom PDF