accel-sim / accel-sim-framework

This is the top-level repository for the Accel-Sim framework.
https://accel-sim.github.io
Other
294 stars 114 forks source link

Getting both Pass and Fail from the PTX execution of lud (only sometimes) #275

Open tgrogers opened 8 months ago

tgrogers commented 8 months ago

Jenkins build failed: https://tgrogers-pc01.ecn.purdue.edu/job/Accel-Sim/job/accel-sim-framework/job/dev/39/

Looks like something transient in the detection of the FAILED string in lud on one particular card. It seems that it needs further investigation. Here is the part of the output where LUD fails. It looks like it both passes and fails, not sure why...:

Using logfiles ['/home/tgrogers-raid/a/jenkin99/workspace/ccel-Sim_accel-sim-framework_dev/util/job_launching/../job_launching/logfiles/sim_log.short-ptx-39.24.01.28-Sunday.txt'] 16732 squeue.id » Node » App » AppArgs » Version » Config » RunningTime »Mem » JobStatus » Basic GPGPU-Sim Stats 16733 ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- ... 16764 177721 » tgrogers-bigram-02 » lud-rodinia-2.0-ft » _v__b__i___data_64_d» gpgpu-sim_git-commit» RTX2060-PT» UNKNOWN »2 G » FUNC_TEST_PASSED, FUNC_TEST_FA» SIMRATE_IPS=7 K»SIM_TIME=1 min, 37 sec (97 sec)»TOT_ IPC=3» TOT_INSN=685 K» TOT_CYCLE=265 K ....