Re: [Starpu-devel] [LU factorisation: gdb debug output]

[Samuel Thibault <samuel.thibault@inria.fr>]

Maxim Abalenkov, on mer. 20 déc. 2017 14:30:11 +0100, wrote:
> Below is the output of “starpu_machine_display”:
> 
> StarPU has found :
> 4 CPU threads
> CPU 0
> CPU 1
> CPU 2
> CPU 3
> 0 CUDA devices
> 0 OpenCL devices
> 
> topology ... (hwloc logical indexes)
> pack 0 core 0 PU 0 CPU 0
> PU 1
> core 1 PU 2 CPU 1
> PU 3
> core 2 PU 4 CPU 2
> PU 5
> core 3 PU 6 CPU 3
> PU 7
> 
> bandwidth and latency ...
> from/to RAM
> RAM 0
> 
> RAM 0

Ok, so there is no reason why node is 1 in your backtrace.
Are you using the nodes or dyn_nodes field in the codelet structure?

Within

Thread 2 (Thread 0x7fffef4df700 (LWP 11305)):
#0  0x00007ffff31ca860 in raise () from /usr/lib/libc.so.6
#1  0x00007ffff31cbec9 in abort () from /usr/lib/libc.so.6
#2  0x00007ffff31c30bc in __assert_fail_base () from /usr/lib/libc.so.6
#3  0x00007ffff31c3133 in __assert_fail () from /usr/lib/libc.so.6
#4  0x00007ffff73bfafd in _starpu_create_data_request (handle=0x555555a2f750, 
src_replicate=0x555555a2f848, dst_replicate=0x555555a2f890, handling_node=1, 
mode=STARPU_W, ndeps=0, is_prefetch=0, 
    is_write_invalidation=0) at datawizard/data_request.c:150
#5  0x00007ffff73bbfff in _starpu_create_request_to_fetch_data 
(handle=0x555555a2f750, dst_replicate=0x555555a2f890, mode=STARPU_W, 
is_prefetch=0, async=0, callback_func=0x0, callback_arg=0x0)
    at datawizard/coherency.c:634
#6  0x00007ffff73bc751 in _starpu_fetch_data_on_node (handle=0x555555a2f750, 
node=1, dst_replicate=0x555555a2f890, mode=STARPU_W, detached=0, 
is_prefetch=0, async=0, callback_func=0x0, callback_arg=0x0)
    at datawizard/coherency.c:754
#7  0x00007ffff73bc8eb in fetch_data (handle=0x555555a2f750, node=1, 
replicate=0x555555a2f890, mode=STARPU_W) at datawizard/coherency.c:782
#8  0x00007ffff73bd75e in _starpu_fetch_task_input (j=0x555555a5e580) at 
datawizard/coherency.c:972
#9  0x00007ffff744197c in execute_job_on_cpu (j=0x555555a5e580, 
worker_task=0x555555a5d900, cpu_args=0x7ffff76d0180 <_starpu_config+1248>, 
rank=0, perf_arch=0x7ffff76d01c8 <_starpu_config+1320>)

it would be useful to run

gdb> thread 2
gdb> frame 8
gdb> print node
gdb> print local_memory_node
gdb> print descrs[index].node

to make sure where this bogus node is coming from.

Samuel