[DPDK/core Bug 1527] Running ring_stress_autotest with > 112 CPUs hangs test
bugzilla at dpdk.org
bugzilla at dpdk.org
Tue Aug 27 00:08:16 CEST 2024
https://bugs.dpdk.org/show_bug.cgi?id=1527
Bug ID: 1527
Summary: Running ring_stress_autotest with > 112 CPUs hangs
test
Product: DPDK
Version: 24.07
Hardware: All
OS: All
Status: UNCONFIRMED
Severity: normal
Priority: Normal
Component: core
Assignee: dev at dpdk.org
Reporter: drc at linux.ibm.com
Target Milestone: ---
System: IBM POWER9 system (PowerNV) with 128 CPUs (2 NUMA nodes), 256GB RAM
** Unverified on x86_64 system with similar size **
OS: RHEL 8.10 (gcc 8.5.0 & gcc 13.2.1)
DPDK: 24.07 (& all prior releases as far back as 21.05)
Running the ring stress autotest causes the test to hang as follows:
~/src/dpdk/build/app/test/dpdk-test --log-level=debug -l 2-127 --no-pci
--no-huge ring_stress_autotest
EAL: Detected CPU lcores: 128
EAL: Detected NUMA nodes: 2
EAL: Static memory layout is selected, amount of reserved memory can be
adjusted with -m or --socket-mem
EAL: Detected static linkage of DPDK
EAL: Multi-process socket /run/user/1000/dpdk/rte/mp_socket
EAL: Selected IOVA mode 'VA'
APP: HPET is not enabled, using TSC as default timer
RTE>>ring_stress_autotest
TEST-CASE MP/MC MT-WRK_ENQ_DEQ-MST_NONE-PRCS START
lcore_stat_dump(AGGREGATE)={
nb_cycle=30720070618(60000137.93 usec),
DEQ+ENQ={
nb_call=24261400,
nb_obj=861278043,
nb_cycle=3701862086301,
obj/call(avg): 35.50
cycles/obj(avg): 4298.10
cycles/call(avg): 152582.38
max cycles/call=16721609(32659.39 usec),
min cycles/call=346(0.68 usec),
},
};
TEST-CASE MP/MC MT-WRK_ENQ_DEQ-MST_NONE-PRCS OK
TEST-CASE MP/MC MT-WRK_ENQ_DEQ-MST_NONE-AVG START
lcore_stat_dump(AGGREGATE)={
nb_cycle=30720078841(60000153.99 usec),
DEQ+ENQ={
nb_call=24328571,
nb_obj=863669930,
nb_cycle=3840005487992,
obj/call(avg): 35.50
cycles/obj(avg): 4446.15
cycles/call(avg): 157839.34
},
};
TEST-CASE MP/MC MT-WRK_ENQ_DEQ-MST_NONE-AVG OK
TEST-CASE MT_RTS MT-WRK_ENQ_DEQ-MST_NONE-PRCS START
<<< Test hangs at this point, have waited up to 24 hours to complete >>>
--
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mails.dpdk.org/archives/dev/attachments/20240826/49f2e509/attachment.htm>
More information about the dev
mailing list