Coding Error?


Advanced search

Message boards : SZTAKI Desktop Grid : Coding Error?

AuthorMessage
gwg
Avatar
Send message
Joined: Aug 1 06
Posts: 58
Credit: 213,886
RAC: 0
Message 4034 - Posted 18 Sep 2006 5:13:01 UTC

    I am running Search 2.00 under Boinc 5.4.9 on a dual 1GHz PowerPC G4 Mac with OS X 10.4.7. Two WUs are being run, and I was looking into them to see what the depth of recursion was, and I noticed a pair of spin lock in the trace. The problem is that the spin locks and unlocks were not nested! Unless this is a thread transfer, then the code may be in error. Not sure what WU it was now, because they have been rescheduled, but think it was

    21b342bd-9be3-49d2-8c12-48e176df7b34_4d826580-7fc9-4fe1-94a8-09040e396cb7_581_1.

    Trace below:

    Analysis of sampling pid 2293 every 10.000000 milliseconds
    Call graph:
    273 Thread_0f0f
    273 _dyld_start
    273 _start
    273 main
    273 high_coeff_iterate
    273 low_coeff_iterate
    273 high_coeff_iterate
    273 low_coeff_iterate
    273 high_coeff_iterate
    273 low_coeff_iterate
    272 high_coeff_iterate
    271 coeff_iterate
    237 DC_checkClientEvent
    120 szone_free
    118 vm_deallocate
    112 mach_msg
    104 mach_msg_trap
    104 mach_msg_trap
    8 mach_msg
    3 mig_get_reply_port
    3 mig_get_reply_port
    3 vm_deallocate
    2 szone_free
    110 malloc
    109 szone_malloc
    109 large_and_huge_malloc
    106 vm_allocate
    106 mach_msg
    97 mach_msg_trap
    97 mach_msg_trap
    9 mach_msg
    1 __spin_lock
    1 __spin_lock
    1 __spin_unlock
    1 __spin_unlock
    1 large_and_huge_malloc
    1 large_and_huge_malloc
    1 large_and_huge_malloc
    4 free
    2 free
    2 szone_size
    2 szone_size
    1 DC_checkClientEvent
    1 boinc_receive_trickle_down
    1 boinc_receive_trickle_down
    1 szone_malloc
    1 szone_malloc
    28 is_exp_polpointer
    28 is_exp_polpointer
    2 coeff_iterate
    2 szone_free
    2 szone_free
    1 coeff_iterate
    1 coeff_iterate
    1 passed_cnt_finish
    1 passed_cnt_finish
    1 high_coeff_iterate
    1 coeff_iterate
    1 DC_checkClientEvent
    1 malloc
    1 szone_malloc
    1 large_and_huge_malloc
    1 vm_allocate
    1 mach_msg
    1 mach_msg_trap
    1 mach_msg_trap
    273 Thread_1003
    273 _pthread_body
    273 timer_thread(void*)
    273 boinc_sleep(double)
    273 sleep
    273 nanosleep
    273 mach_wait_until
    273 mach_wait_until

    Total number in stack (recursive counted multiple, when >=5):

    Sort by top of stack, same collapsed (when >= 5):
    mach_wait_until 273
    mach_msg_trap 202
    is_exp_polpointer 28
    mach_msg 17
    Sample analysis of process 2293 written to file /dev/stdout
    Sampling process 2293 each 10 msecs 300 times

    ____________
    Dr George W Gerrity
    4 Coral Place
    Campbell, ACT 2612
    AUSTRALIA

    Ph: +61 2 6156 0286
    Time: +10 hours (ref GMT)
    PGP RSA Public Key Fingerprint:
    73EF 318A DFF5 EB8A 6810 49AC 0763 AF07

    gwg
    Avatar
    Send message
    Joined: Aug 1 06
    Posts: 58
    Credit: 213,886
    RAC: 0
    Message 4035 - Posted 18 Sep 2006 5:16:19 UTC - in response to Message 4034.

      I am running Search 2.00 under Boinc 5.4.9 on a dual 1GHz PowerPC G4 Mac with OS X 10.4.7. Two WUs are being run, and I was looking into them to see what the depth of recursion was, and I noticed a pair of spin lock in the trace. The problem is that the spin locks and unlocks were not nested! Unless this is a thread transfer, then the code may be in error. Not sure what WU it was now, because they have been rescheduled, but think it was

      21b342bd-9be3-49d2-8c12-48e176df7b34_4d826580-7fc9-4fe1-94a8-09040e396cb7_581_1.

      Trace below: ...



      Sorry, should have used \\[pre\\] instead of \\[code\\] to preserve stack indents. Trying again:

      Analysis of sampling pid 2293 every 10.000000 milliseconds Call graph: 273 Thread_0f0f 273 _dyld_start 273 _start 273 main 273 high_coeff_iterate 273 low_coeff_iterate 273 high_coeff_iterate 273 low_coeff_iterate 273 high_coeff_iterate 273 low_coeff_iterate 272 high_coeff_iterate 271 coeff_iterate 237 DC_checkClientEvent 120 szone_free 118 vm_deallocate 112 mach_msg 104 mach_msg_trap 104 mach_msg_trap 8 mach_msg 3 mig_get_reply_port 3 mig_get_reply_port 3 vm_deallocate 2 szone_free 110 malloc 109 szone_malloc 109 large_and_huge_malloc 106 vm_allocate 106 mach_msg 97 mach_msg_trap 97 mach_msg_trap 9 mach_msg 1 __spin_lock 1 __spin_lock 1 __spin_unlock 1 __spin_unlock 1 large_and_huge_malloc 1 large_and_huge_malloc 1 large_and_huge_malloc 4 free 2 free 2 szone_size 2 szone_size 1 DC_checkClientEvent 1 boinc_receive_trickle_down 1 boinc_receive_trickle_down 1 szone_malloc 1 szone_malloc 28 is_exp_polpointer 28 is_exp_polpointer 2 coeff_iterate 2 szone_free 2 szone_free 1 coeff_iterate 1 coeff_iterate 1 passed_cnt_finish 1 passed_cnt_finish 1 high_coeff_iterate 1 coeff_iterate 1 DC_checkClientEvent 1 malloc 1 szone_malloc 1 large_and_huge_malloc 1 vm_allocate 1 mach_msg 1 mach_msg_trap 1 mach_msg_trap 273 Thread_1003 273 _pthread_body 273 timer_thread(void*) 273 boinc_sleep(double) 273 sleep 273 nanosleep 273 mach_wait_until 273 mach_wait_until Total number in stack (recursive counted multiple, when >=5): Sort by top of stack, same collapsed (when >= 5): mach_wait_until 273 mach_msg_trap 202 is_exp_polpointer 28 mach_msg 17 Sample analysis of process 2293 written to file /dev/stdout Sampling process 2293 each 10 msecs 300 times

      ____________
      Dr George W Gerrity
      4 Coral Place
      Campbell, ACT 2612
      AUSTRALIA

      Ph: +61 2 6156 0286
      Time: +10 hours (ref GMT)
      PGP RSA Public Key Fingerprint:
      73EF 318A DFF5 EB8A 6810 49AC 0763 AF07

      Post to thread

      Message boards : SZTAKI Desktop Grid : Coding Error?


      Home | My Account | Message Boards


      Copyright © 2017 SZTAKI Desktop Grid