-
Notifications
You must be signed in to change notification settings - Fork 117
Description
Cray's gdb4hpc will break if our slurm config varies too much ===> it would be nice to have a reframe check for daint-gpu/daint-mc * gnu/intel/pgi/cce.
While gdb4hpc is normally used inside a salloc session, it will be hard to do that with reframe (?).
It's also possible to launch it in batch mode (-b) but must wait for Cray Case #224617 to be fixed.
I use this issue as a placeholder.
- srun is replaced with:
/opt/cray/pe/gdb4hpc/3.0.10/libexec/guard_proxy -h daint104 -p 36081
daint-gpu
cc
-dynamic-g ~/reframe.git/cscs-checks/prgenv/src/hello_world_mpi.c
srun -n2 -Cgpu -Ausup -t1 -pdebug ./a.out
Hello World from thread 0 out of 1 from process 0 out of 2
# MPI-3.1 = MPI VERSION : CRAY MPICH version 7.7.2 (ANL base 3.2)
MPI BUILD INFO : Built Thu Jul 19 20:48:27 2018 (git hash 5d705bad2) MT-G
Hello World from thread 0 out of 1 from process 1 out of 2
# MPI-3.1 = MPI VERSION : CRAY MPICH version 7.7.2 (ANL base 3.2)
MPI BUILD INFO : Built Thu Jul 19 20:48:27 2018 (git hash 5d705bad2) MT-G
salloc -N1 -t2 -Cgpu -pdebug -Ausup
module load gdb4hpc
gdb4hpc
gdb4hpc 3.0 - Cray Line Mode Parallel Debugger
...
dbg all> launch --debug $tst{8} ./a.out
Starting application, please wait...
Creating MRNet communication network...
Waiting for debug servers to attach to MRNet communications network...
Timeout in 400 seconds. Please wait for the attach to complete.
Number of dbgsrvs connected: [1]; Timeout Counter: [0]
Number of dbgsrvs connected: [1]; Timeout Counter: [1]
Number of dbgsrvs connected: [8]; Timeout Counter: [0]
Finalizing setup...
Launch complete.
tst{0..7}: Initial breakpoint, main at /
apps/common/UES/sandbox/jgp/reframe.git.users/cscs-checks/prgenv/src/
hello_world_mpi.c:8
dbg all> cont
<$tst>: Hello World from thread 0 out of 1 from process 0 out of 8
<$tst>: # MPI-3.1 = MPI VERSION : CRAY MPICH version 7.7.2 (ANL base 3.2)
<$tst>: MPI BUILD INFO : Built Thu Jul 19 20:48:27 2018 (git hash 5d705bad2) MT-G
<$tst>:
tst{0..7}: The application has reached an exit breakpoint.
<$tst>: Hello World from thread 0 out of 1 from process 1 out of 8
<$tst>: # MPI-3.1 = MPI VERSION : CRAY MPICH version 7.7.2 (ANL base 3.2)
<$tst>: MPI BUILD INFO : Built Thu Jul 19 20:48:27 2018 (git hash 5d705bad2) MT-G
<$tst>:
<$tst>: Hello World from thread 0 out of 1 from process 2 out of 8
<$tst>: # MPI-3.1 = MPI VERSION : CRAY MPICH version 7.7.2 (ANL base 3.2)
<$tst>: MPI BUILD INFO : Built Thu Jul 19 20:48:27 2018 (git hash 5d705bad2) MT-G
<$tst>:
<$tst>: Hello World from thread 0 out of 1 from process 3 out of 8
<$tst>: # MPI-3.1 = MPI VERSION : CRAY MPICH version 7.7.2 (ANL base 3.2)
<$tst>: MPI BUILD INFO : Built Thu Jul 19 20:48:27 2018 (git hash 5d705bad2) MT-G
<$tst>:
<$tst>: Hello World from thread 0 out of 1 from process 4 out of 8
<$tst>: # MPI-3.1 = MPI VERSION : CRAY MPICH version 7.7.2 (ANL base 3.2)
<$tst>: MPI BUILD INFO : Built Thu Jul 19 20:48:27 2018 (git hash 5d705bad2) MT-G
<$tst>:
<$tst>: Hello World from thread 0 out of 1 from process 5 out of 8
<$tst>: # MPI-3.1 = MPI VERSION : CRAY MPICH version 7.7.2 (ANL base 3.2)
<$tst>: MPI BUILD INFO : Built Thu Jul 19 20:48:27 2018 (git hash 5d705bad2) MT-G
<$tst>:
<$tst>: Hello World from thread 0 out of 1 from process 6 out of 8
<$tst>: # MPI-3.1 = MPI VERSION : CRAY MPICH version 7.7.2 (ANL base 3.2)
<$tst>: MPI BUILD INFO : Built Thu Jul 19 20:48:27 2018 (git hash 5d705bad2) MT-G
<$tst>:
<$tst>: Hello World from thread 0 out of 1 from process 7 out of 8
<$tst>: # MPI-3.1 = MPI VERSION : CRAY MPICH version 7.7.2 (ANL base 3.2)
<$tst>: MPI BUILD INFO : Built Thu Jul 19 20:48:27 2018 (git hash 5d705bad2) MT-G
<$tst>:
dbg all> quit
Shutting down debugger and killing application for 'tst'.
exit
salloc: Relinquishing job allocation 10097647
daint-mc
- idem