Skip to Content.
Sympa Menu

charm - Re: [charm] Load balancing with Charm++ >= 6.6.1

charm AT lists.cs.illinois.edu

Subject: Charm++ parallel programming system

List archive

Re: [charm] Load balancing with Charm++ >= 6.6.1


Chronological Thread 
  • From: Phil Miller <mille121 AT illinois.edu>
  • To: James Bordner <jobordner AT gmail.com>
  • Cc: charm <charm AT lists.cs.illinois.edu>
  • Subject: Re: [charm] Load balancing with Charm++ >= 6.6.1
  • Date: Wed, 6 Jul 2016 13:33:58 -0500

Hi James,

Could you specifically confirm that this crash is still observed with
6.7.0 and 6.7.1? I feel like this is somewhat familiar, but would have
to do more digging than the brief search I've done to figure out
where/why/when.

Phil

On Wed, Jul 6, 2016 at 1:28 PM, James Bordner
<jobordner AT gmail.com>
wrote:
> Hello,
>
> I have a load balancing test that works with Charm++ versions <= 6.6.0, but
> with Charm++ >= 6.6.1 I get the following error:
>
> [0] Assertion "n<len" failed in file cklists.h line 221.
> ------------- Processor 0 Exiting: Called CmiAbort ------------
> Reason:
> [0] Stack Traceback:
> [0:0] CmiAbort+0x5b [0x6a51ea]
> [0:1] __cmi_assert+0x42 [0x6af4c3]
> [0:2] _ZN5CkVecI9LDObjDataEixEm+0x32 [0x58d928]
> [0:3] _ZN6BaseLB7LDStats7getHashERK8_LDObjidRK7_LDOMid+0xd5 [0x636deb]
> [0:4] _ZN6BaseLB7LDStats7getHashERK9_LDObjKey+0x47 [0x636ee1]
> [0:5] _ZN6BaseLB7LDStats11getSendHashER10LDCommData+0x33 [0x636f17]
> [0:6] _ZN9CentralLB27removeCommDataOfDeletedObjsEPN6BaseLB7LDStatsE+0x99
> [0x63ede3]
> [0:7] _ZN9CentralLB11LoadBalanceEv+0x1ef [0x63cc39]
> [0:8] _ZN17CkIndex_CentralLB22_call_LoadBalance_voidEPvS0_+0x30
> [0x643dae]
> [0:9] CkDeliverMessageFree+0x4e [0x5b05c8]
> [0:10] [0x5b070e]
> [0:11] [0x5b082a]
> [0:12] [0x5b1f0d]
> [0:13] [0x5b1fb5]
> [0:14] _Z15_processHandlerPvP11CkCoreState+0x126 [0x5b24cc]
> [0:15] CmiHandleMessage+0x4d [0x6ac169]
> [0:16] CsdScheduleForever+0xad [0x6ac3ea]
> [0:17] CsdScheduler+0x16 [0x6ac31b]
> [0:18] [0x6aa334]
> [0:19] ConverseInit+0x32e [0x6aa7f1]
> [0:20] main+0x3f [0x5a0fb2]
> [0:21] __libc_start_main+0xf5 [0x7fb33dda8f45]
> [0:22] [0x5764ff]
> Fatal error on PE 0>
>
> I can provide more details, but thought I'd start by asking if this looks
> familiar to anyone?
>
> Thanks!
> James



Archive powered by MHonArc 2.6.16.

Top of Page