Uploaded image for project: 'OpenShift Bugs'
  1. OpenShift Bugs
  2. OCPBUGS-13330

OCP 4.13 | NTO profile's default parameter workloadHints.perPodPowerManagement = false seems to be hitting kernel taint bit 2 in nodes using x86_energy_perf function

XMLWordPrintable

    • Icon: Bug Bug
    • Resolution: Done-Errata
    • Icon: Critical Critical
    • None
    • 4.13
    • Node Tuning Operator
    • None
    • No
    • Rejected
    • False
    • Hide

      None

      Show
      None

      Description of problem:

      NTO profile's default parameter workloadHints.perPodPowerManagement = false seems to be hitting kernel taint bit 2 in nodes using x86_energy_perf function
      
      This feature activates the x86_energy_perf function in the node(s) where the profile is applied, and if this function is enabled and available in the CPUs of the node(s), then these nodes are tainted with taint bit 2 (TAINT_CPU_OUT_OF_SPEC) (i.e., the kernel is running on a processor or system that is out of specification).
      
      Since the default NTO profiles deployed by NTO are not changing the value of this parameter, this means all nodes running default NTO profiles are tainted with taint bit 2.

      Version-Release number of selected component (if applicable):

      OCP 4.13

      How reproducible:

      Always, starting 4.13

      Steps to Reproduce:

      1. Deploy a OCP cluster using IPI
      2. Check NTO profiles created by default (e.g. the ones for master nodes). They are hitting the kernel taint
      3. Create a PerformanceProfile setting workloadHints.perPodPowerManagement = true, e.g. in worker nodes. The kernel taint disappear (you need to wait until the MCP are updated and the nodes are rebooted with the new config).
      
      ---
      kind: PerformanceProfile
      apiVersion: "performance.openshift.io/v2"
      metadata:
        name: blueprint-profile
      spec:
        cpu:
          isolated: "1-19,21-39,41-59,61-79"
          reserved: "0,40,20,60"
        hugepages:
          pages:
            - size: "1G"
              count: 32
              node: 0
            - size: "1G"
              count: 32
              node: 1
            - size: "2M"
              count: 12000
              node: 0
            - size: "2M"
              count: 12000
              node: 1
        realTimeKernel:
          enabled: false
        workloadHints:
          realTime: false
          highPowerConsumption: false
          perPodPowerManagement: true
        net:
          userLevelNetworking: false
        numa:
          topologyPolicy: "single-numa-node"
        nodeSelector:
          node-role.kubernetes.io/worker: ""
      ...

      Actual results:

      $ cat /proc/sys/kernel/tainted 
      4
      

      Expected results:

      Kernel should not be tainted

      Additional info:

      $ uname -a
      Linux worker-1 5.14.0-284.13.1.el9_2.x86_64 #1 SMP PREEMPT_DYNAMIC Thu Apr 27 13:35:10 EDT 2023 x86_64 x86_64 x86_64 GNU/Linux
      $ lscpu 
      Architecture:            x86_64
        CPU op-mode(s):        32-bit, 64-bit
        Address sizes:         46 bits physical, 48 bits virtual
        Byte Order:            Little Endian
      CPU(s):                  80
        On-line CPU(s) list:   0-79
      Vendor ID:               GenuineIntel
        Model name:            Intel(R) Xeon(R) Gold 6248 CPU @ 2.50GHz
          CPU family:          6
          Model:               85
          Thread(s) per core:  2
          Core(s) per socket:  20
          Socket(s):           2
          Stepping:            7
          CPU max MHz:         3900.0000
          CPU min MHz:         1000.0000
          BogoMIPS:            5000.00
          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 cpuid aperfmperf 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 cpuid_fault epb cat_l3 cdp_l3 invpcid_single i
                               ntel_ppin ssbd mba ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid cqm mpx rdt_a 
                               avx512f avx512dq rdseed adx smap clflushopt clwb intel_pt avx512cd avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_l
                               ocal dtherm ida arat pln pts hwp hwp_act_window hwp_pkg_req pku ospke avx512_vnni md_clear flush_l1d arch_capabilities
      Virtualization features: 
        Virtualization:        VT-x
      Caches (sum of all):     
        L1d:                   1.3 MiB (40 instances)
        L1i:                   1.3 MiB (40 instances)
        L2:                    40 MiB (40 instances)
        L3:                    55 MiB (2 instances)
      NUMA:                    
        NUMA node(s):          2
        NUMA node0 CPU(s):     0-19,40-59
        NUMA node1 CPU(s):     20-39,60-79
      Vulnerabilities:         
        Itlb multihit:         KVM: Mitigation: VMX disabled
        L1tf:                  Not affected
        Mds:                   Not affected
        Meltdown:              Not affected
        Mmio stale data:       Mitigation; Clear CPU buffers; SMT vulnerable
        Retbleed:              Mitigation; Enhanced IBRS
        Spec store bypass:     Mitigation; Speculative Store Bypass disabled via prctl
        Spectre v1:            Mitigation; usercopy/swapgs barriers and __user pointer sanitization
        Spectre v2:            Mitigation; Enhanced IBRS, IBPB conditional, RSB
       filling, PBRSB-eIBRS SW sequence
        Srbds:                 Not affected
        Tsx async abort:       Mitigation; TSX disabled
      
      $ sudo  dmidecode -t system
      # dmidecode 3.3
      Getting SMBIOS data from sysfs.
      SMBIOS 3.2.1 present.Handle 0x008E, DMI type 1, 27 bytes
      System Information
          Manufacturer: HPE
          Product Name: ProLiant DL360 Gen10
          Version: Not Specified
          Serial Number: MXQ01901HD
          UUID: 39373638-3935-584d-5130-313930314844
          Wake-up Type: Power Switch
          SKU Number: 867959-B21
          Family: ProLiantHandle 0x00C0, DMI type 32, 11 bytes
      System Boot Information
          Status: No errors detected
      

      This issue was originally called "OCP 4.13 | Deprecated Driver nft_compat creates a Kernel taint", but the discussion has been moved to review NTO profiles. You can follow the whole history in the comments section of this Jira card.

            jmencak Jiri Mencak
            josearod@redhat.com Jose Alberto Rodriguez
            Liquan Cui Liquan Cui
            Jose Alberto Rodriguez
            Votes:
            0 Vote for this issue
            Watchers:
            19 Start watching this issue

              Created:
              Updated:
              Resolved: