Hi all,
I have compiled openmp benchmarks using m5threads. command lines are as follow:
"./build/X86/gem5.opt configs/example/se.py -n 4 -c ../m5threads/tests/test_omp
-o '2 2' --caches --l2cache --l1d_size=128kB --l1i_size=128kB --l2_size=1MB
--l1d_assoc=2 --l1i_assoc=2 --l2_assoc=1 --cacheline_size=64"
I got this message on the terminal:
info: Entering event queue @ 0. Starting simulation...
warn: readlink() called on '/proc/self/exe' may yield unexpected results in
various settings.
Returning '/home/wangxiang/gem5_my/m5threads/tests/omp_true'
info: Increasing stack size by one page.
warn: ignoring syscall access(...)
Setting OMP threads to 2
Starting with row/col size=2
A initialized
B initialized
Computing A*B with 2 threads
warn: ClockedObject: Already in the requested power state, request ignored
Done
Exiting @ tick 18446744073709551615 because simulate() limit reached
stats.txt are as follow:
sim_seconds 18446744.073710
# Number of seconds simulated
sim_ticks18446744073709551616
# Number of ticks simulated
final_tick 18446744073709551616
# Number of ticks from beginning of simulation (restored from checkpoints
and never reset
The problem is that I want to get correct sim_seconds of test_omp, but
"sim_seconds 18446744.073710" is impossible. It seems like openmp program has
stalled somewhere. But when I test it on my own computer, it works well and
exits correctly. What should I do to get the true stats of my bench?
Any help will be appreciated!
___
gem5-users mailing list
gem5-users@gem5.org
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users