[LU-10086] LNET_MINOR conflicts with USERIO_MINOR

• Type: Bug
• Resolution: Fixed
• Priority: Major
• Fix Version/s: Lustre 2.11.0, Lustre 2.10.2
• Affects Version/s: Lustre 2.10.1
• Environment: MOFED4
• Severity: 3
• Assignee: Dmitry Eremin (Inactive)
• Reporter: Mahmoud Hanafi

Description

When loading lustre module we sometimes get this error.

r147i0n0 ~ # modprobe libcfs
modprobe: ERROR: could not insert 'libcfs': Device or resource busy

 

dmesg

[ 3067.259489] LNet: HW NUMA nodes: 2, HW CPU cores: 80, npartitions: 2
[ 3067.259497] LNetError: 4183:0:(module.c:742:libcfs_init()) misc_register: error -16

module configs

options lnet networks=o2ib314(ib1) routes="o2ib 10.149.26.[60,140,141,142,143,144,145]@o2ib314 10.149.25.[195,196,197,198,205]@o2ib314" dead_router_check_interval=60 live_router_check_interval=31
options lnet avoid_asym_router_failure=1 check_routers_before_use=1 small_router_buffers=65536 large_router_buffers=8192

lscpu

147i0n0 ~ # lscpu
Architecture:          x86_64
CPU op-mode(s):        32-bit, 64-bit
Byte Order:            Little Endian
CPU(s):                80
On-line CPU(s) list:   0-79
Thread(s) per core:    2
Core(s) per socket:    20
Socket(s):             2
NUMA node(s):          2
Vendor ID:             GenuineIntel
CPU family:            6
Model:                 85
Model name:            Intel(R) Xeon(R) Gold 6148 CPU @ 2.40GHz
Stepping:              4
CPU MHz:               1000.000
CPU max MHz:           2401.0000
CPU min MHz:           1000.0000
BogoMIPS:              4799.99
Virtualization:        VT-x
L1d cache:             32K
L1i cache:             32K
L2 cache:              1024K
L3 cache:              28160K
NUMA node0 CPU(s):     0-19,40-59
NUMA node1 CPU(s):     20-39,60-79
Flags:                 fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc aperfmperf eagerfpu pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch ida arat epb pln pts dtherm hwp hwp_act_window hwp_epp hwp_pkg_req intel_pt tpr_shadow vnmi flexpriority ept vpid fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm cqm mpx avx512f avx512dq rdseed adx smap clflushopt clwb avx512cd avx512bw avx512vl xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc

Issue Links

• is related to: LU-9679 Prepare lustre for adoption into the linux kernel

Comments

[Gerrit Updater]

John L. Hammond ([e-mail]) merged in patch https://review.whamcloud.com/29945/
Subject: LU-10086 libcfs: use dynamic minors for /dev/

{lnet,obd}

Project: fs/lustre-release
Branch: b2_10
Current Patch Set:
Commit: 705863d6e6d03c47b67c179f7b07e54ea36817e5

[Gerrit Updater]

Minh Diep ([e-mail]) uploaded a new patch: https://review.whamcloud.com/29945
Subject: LU-10086 libcfs: use dynamic minors for /dev/

{lnet,obd}

Project: fs/lustre-release
Branch: b2_10
Current Patch Set: 1
Commit: 01b68a07a2024742c8e12c13d51b0f64dde7ad05

[Peter Jones]

Landed for 2.11

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch https://review.whamcloud.com/29741/
Subject: LU-10086 libcfs: use dynamic minors for /dev/

{lnet,obd}

Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: e446c1667dc9136e3a1cb3603c3851975b68e0ac

[Gerrit Updater]

John L. Hammond ([e-mail]) uploaded a new patch: https://review.whamcloud.com/29741
Subject: LU-10086 libcfs: use dynamic minors for /dev/

{lnet,obd}

Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 7cee7dc7e102847f3b089a83b68ad1ac0d233212

[John Hammond]

It would be good to remove LNET_MINOR entirely and switch to using MISC_DYNAMIC_MINOR when we register libcfs_dev. At the same time we should remove the (fairly questionable) code in lctl the creates /dev/lnet if it doesn't exist.

[Mahmoud Hanafi]

Not critical for us. But it cause some confusion when trying to figure it all out. 

 

[Dmitry Eremin (Inactive)]

How critical to have both dirvers loaded simultaneously?

[Dmitry Eremin (Inactive)]

LNet driver and UserIO have the same driver MINOR. Therefore they cannot be loaded simultaneously.

#define USERIO_MINOR 240

#define LNET_MINOR 240

[Joseph Gmitter (Inactive)]

Hi Dmitry,

Can you please advise here?

Thanks.
Joe