Revision 0x2006a08 cause cpu stall or kernel panic on 06-55-04

Bug #1905047 reported by Alessandro Ratti
10
This bug affects 1 person
Affects Status Importance Assigned to Milestone
intel-microcode (Ubuntu)
Invalid
Undecided
Unassigned

Bug Description

Hello,

I'm running Ubuntu Focal with latest kernel 5.4.0-54-generic on this cpu:

Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Byte Order: Little Endian
Address sizes: 46 bits physical, 48 bits virtual
CPU(s): 56
On-line CPU(s) list: 0-55
Thread(s) per core: 2
Core(s) per socket: 14
Socket(s): 2
NUMA node(s): 2
Vendor ID: GenuineIntel
CPU family: 6
Model: 85
Model name: Intel(R) Xeon(R) Gold 5120 CPU @ 2.20GHz
Stepping: 4
CPU MHz: 1000.066
CPU max MHz: 3200.0000
CPU min MHz: 1000.0000
BogoMIPS: 4400.00
Virtualization: VT-x
L1d cache: 896 KiB
L1i cache: 896 KiB
L2 cache: 28 MiB
L3 cache: 38.5 MiB
NUMA node0 CPU(s): 0-13,28-41
NUMA node1 CPU(s): 14-27,42-55
Vulnerability Itlb multihit: KVM: Mitigation: Split huge pages
Vulnerability L1tf: Mitigation; PTE Inversion; VMX conditional cache flushes, SMT vulnerable
Vulnerability Mds: Mitigation; Clear CPU buffers; SMT vulnerable
Vulnerability Meltdown: Mitigation; PTI
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl and seccomp
Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2: Mitigation; Full generic retpoline, IBPB conditional, IBRS_FW, STIBP conditional, RSB filling
Vulnerability Srbds: Not affected
Vulnerability Tsx async abort: Mitigation; Clear CPU buffers; SMT vulnerable
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 p
                                 be syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aper
                                 fmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2api
                                 c movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb cat_l3 cdp_l3 i
                                 nvpcid_single pti intel_ppin ssbd mba ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjus
                                 t bmi1 hle avx2 smep bmi2 erms invpcid rtm cqm mpx rdt_a avx512f avx512dq rdseed adx smap clflushopt clwb intel_pt av
                                 x512cd avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local dtherm ida
                                 arat pln pts pku ospke md_clear flush_l1d

Using the microcode rev. 0x2006a08 - Ubuntu version intel-microcode-3.20201110.0ubuntu0.20.04.2 - I'm not able to boot on this cpu anymore.
Doing an early-load update the RCU's CPU stall detector seems to be entering in an infinite loop (see the strace in the attachment).
This keeps repeting forever and it's neccessary a power reset to regain the machine control.

Performing a late-load update, with the Ubuntu kernel 5.4.0-52, the machine crashes after a while (see the strace in the attachment).
The Ubuntu kernel 5.4.0-54, disabling microcode loading - adding cmdline option `dis_ucode_ldr` - boots without any issue.

Please note: I posted this very same issue upstream: https://github.com/intel/Intel-Linux-Processor-Microcode-Data-Files/issues/45

Revision history for this message
Alessandro Ratti (lord2y) wrote :
description: updated
description: updated
description: updated
Revision history for this message
lotuspsychje (lotuspsychje) wrote :

Thank you for taking the time to report this bug and helping to make Ubuntu better. Please execute the following command only once, as it will automatically gather debugging information, in a terminal:
apport-collect 1905047

When reporting bugs in the future please use apport by using 'ubuntu-bug' and the name of the package affected. You can learn more about this functionality at https://wiki.ubuntu.com/ReportingBugs.

Revision history for this message
Alessandro Ratti (lord2y) wrote :
Download full text (4.0 KiB)

Here it is the apport-collect output for intel-microcode before the machine crashes:

ProblemType: Bug
ApportVersion: 2.20.11-0ubuntu27.12
Architecture: amd64
CasperMD5CheckResult: skip
Date: Mon Nov 23 12:39:18 2020
Dependencies:
 adduser 3.118ubuntu2
 busybox-initramfs 1:1.30.1-4ubuntu6.2
 coreutils 8.30-3ubuntu2
 cpio 2.13+dfsg-2
 debconf 1.5.73
 dpkg 1.19.7ubuntu3
 gcc-10-base 10.2.0-5ubuntu1~20.04
 initramfs-tools 0.136ubuntu6.3
 initramfs-tools-bin 0.136ubuntu6.3
 initramfs-tools-core 0.136ubuntu6.3
 intel-microcode 3.20201110.0ubuntu0.20.04.2
 iucode-tool 2.3.1-1
 klibc-utils 2.0.7-1ubuntu5
 kmod 27-1ubuntu2
 libacl1 2.2.53-6
 libattr1 1:2.4.48-5
 libaudit-common 1:2.8.5-2ubuntu6
 libaudit1 1:2.8.5-2ubuntu6
 libblkid1 2.34-0.1ubuntu9.1
 libbz2-1.0 1.0.8-2
 libc6 2.31-0ubuntu9.1
 libcap-ng0 0.7.9-2.1build1
 libcrypt1 1:4.4.10-10ubuntu4
 libdb5.3 5.3.28+dfsg1-0.6ubuntu2
 libgcc-s1 10.2.0-5ubuntu1~20.04
 libgcrypt20 1.8.5-5ubuntu1
 libgpg-error0 1.37-1
 libidn2-0 2.2.0-2
 libklibc 2.0.7-1ubuntu5
 libkmod2 27-1ubuntu2
 liblz4-1 1.9.2-2
 liblzma5 5.2.4-1ubuntu1
 libmount1 2.34-0.1ubuntu9.1
 libpam-modules 1.3.1-5ubuntu4.1
 libpam-modules-bin 1.3.1-5ubuntu4.1
 libpam-runtime 1.3.1-5ubuntu4.1
 libpam0g 1.3.1-5ubuntu4.1
 libpcre2-8-0 10.34-7
 libselinux1 3.0-1build2
 libsemanage-common 3.0-1build2
 libsemanage1 3.0-1build2
 libsepol1 3.0-1
 libsmartcols1 2.34-0.1ubuntu9.1
 libssl1.1 1.1.1f-1ubuntu2
 libsystemd0 245.4-4ubuntu3.3
 libtinfo6 6.2-0ubuntu2
 libudev1 245.4-4ubuntu3.3
 libunistring2 0.9.10-2
 libuuid1 2.34-0.1ubuntu9.1
 libzstd1 1.4.4+dfsg-3
 linux-base 4.5ubuntu3.1
 login 1:4.8.1-1ubuntu5.20.04
 logsave 1.45.5-2ubuntu1
 lsb-base 11.1.0ubuntu2
 lz4 1.9.2-2
 passwd 1:4.8.1-1ubuntu5.20.04
 perl-base 5.30.0-9ubuntu0.2
 tar 1.30+dfsg-7
 udev 245.4-4ubuntu3.3
 util-linux 2.34-0.1ubuntu9.1
 uuid-runtime 2.34-0.1ubuntu9.1
 zlib1g 1:1.2.11.dfsg-2ubuntu1.2
DistroRelease: Ubuntu 20.04
Package: intel-microcode 3.20201110.0ubuntu0.20.04.2
PackageArchitecture: amd64
ProcCpuinfoMinimal:
 processor : 55
 vendor_id : GenuineIntel
 cpu family : 6
 model : 85
 model name : Intel(R) Xeon(R) Gold 5120 CPU @ 2.20GHz
 stepping : 4
 microcode : 0x2006a08
 cpu MHz : 1000.564
 cache size : 19712 KB
 physical id : 1
 siblings : 28
 core id : 14
 cpu cores : 14
 apicid : 61
 initial apicid : 61
 fpu : yes
 fpu_exception : yes
 cpuid level : 22
 wp : yes
 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 pti intel_ppin ssbd mba ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm cqm mpx rdt_a avx512f avx512dq rdseed adx smap clflusho...

Read more...

Revision history for this message
Alessandro Ratti (lord2y) wrote :

Updating the BIOS fixed those issue.
Sorry for the noise.

Changed in intel-microcode (Ubuntu):
status: New → Invalid
Revision history for this message
Colin Ian King (colin-king) wrote :

No problem, thanks for reporting back, Can you add the version of the BIOS to the bug report so that other folk will know which version addresses the problem if they have the same issue.

Revision history for this message
Alex Murray (alexmurray) wrote :

Updating the BIOS may have the effect of also updating the microcode in the BIOS to the latest one and hence early loading of microcode won't happen in the initramfs (since you are already running the same microcode version). As such this just papers over this bug - so I suspect there is perhaps still a bug in either the microcode from Intel or the kernel such that early loading of it causes issues for this particular CPU etc.

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.