• Constant log of tpmcrb0: Device has Error bit set on 15.2

    From =?UTF-8?Q?Ulrich_Sp=C3=B6rlein?=@uqs@freebsd.org to muc.lists.freebsd.stable on Tue Nov 10 12:43:58 2026
    From Newsgroup: muc.lists.freebsd.stable

    Hi there, some time this year, one of my machines started logging this constantly:

    tpmcrb0: Device has Error bit set

    I don't even wanna use TPM or anything, so this is kinda annoying.

    % sudo grep -i tpm /var/run/dmesg.boot
    tpmcrb0: <Trusted Platform Module 2.0, CRB mode> iomem
    0xfed40000-0xfed44fff on acpi0
    random: registering fast source TPM

    That's not much, it's an Intel mini PC using this:

    CPU: Intel(R) Core(TM) i7-8550U CPU @ 1.80GHz (2000.00-MHz K8-class CPU)
    Origin="GenuineIntel" Id=0x806ea Family=0x6 Model=0x8e Stepping=10
    Features=0xbfebfbff<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,HTT,TM,PBE>
    Features2=0x7ffafbbf<SSE3,PCLMULQDQ,DTES64,MON,DS_CPL,VMX,EST,TM2,SSSE3,SDBG,FMA,CX16,xTPR,PDCM,PCID,SSE4.1,SSE4.2,x2APIC,MOVBE,POPCNT,TSCDLT,AESNI,XSAVE,OSXSAVE,AVX,F16C,RDRAND>
    AMD Features=0x2c100800<SYSCALL,NX,Page1GB,RDTSCP,LM>
    AMD Features2=0x121<LAHF,ABM,Prefetch>
    Structured Extended Features=0x29c67af<FSGSBASE,TSCADJ,SGX,BMI1,AVX2,SMEP,BMI2,ERMS,INVPCID,NFPUSG,MPX,RDSEED,ADX,SMAP,CLFLUSHOPT,PROCTRACE>
    Structured Extended Features3=0x9c002600<MCUOPT,MD_CLEAR,TSXFA,IBPB,STIBP,L1DFL,SSBD>
    XSAVE Features=0xf<XSAVEOPT,XSAVEC,XINUSE,XSAVES>
    VT-x: PAT,HLT,MTF,PAUSE,EPT,UG,VPID
    TSC: P-state invariant, performance statistics

    Anything I should try here for debugging?

    Cheers
    Uli


    --
    Posted automagically by a mail2news gateway at muc.de e.V.
    Please direct questions, flames, donations, etc. to news-admin@muc.de
    --- Synchronet 3.22a-Linux NewsLink 1.2
  • From Kevin Bowling@kevin.bowling@kev009.com to muc.lists.freebsd.stable on Tue Nov 10 21:32:11 2026
    From Newsgroup: muc.lists.freebsd.stable

    On Tue, Nov 10, 2026 at 4:44rC>AM Ulrich Sp||rlein <uqs@freebsd.org> wrote:

    Hi there, some time this year, one of my machines started logging this constantly:
    Sometime this year is a wide window, can you supply any constraints
    like a particular date range or commits?
    tpmcrb0: Device has Error bit set

    I don't even wanna use TPM or anything, so this is kinda annoying.
    You could build a kernel without it using a small include
    GENERIC-NODEBUG and an override configuration. tpm's recent absence
    was due to fatal suspend/resume bugs, not some design decision.
    % sudo grep -i tpm /var/run/dmesg.boot
    tpmcrb0: <Trusted Platform Module 2.0, CRB mode> iomem
    0xfed40000-0xfed44fff on acpi0
    random: registering fast source TPM

    That's not much, it's an Intel mini PC using this:

    CPU: Intel(R) Core(TM) i7-8550U CPU @ 1.80GHz (2000.00-MHz K8-class CPU)
    Origin="GenuineIntel" Id=0x806ea Family=0x6 Model=0x8e Stepping=10
    Features=0xbfebfbff<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,HTT,TM,PBE>
    Features2=0x7ffafbbf<SSE3,PCLMULQDQ,DTES64,MON,DS_CPL,VMX,EST,TM2,SSSE3,SDBG,FMA,CX16,xTPR,PDCM,PCID,SSE4.1,SSE4.2,x2APIC,MOVBE,POPCNT,TSCDLT,AESNI,XSAVE,OSXSAVE,AVX,F16C,RDRAND>
    AMD Features=0x2c100800<SYSCALL,NX,Page1GB,RDTSCP,LM>
    AMD Features2=0x121<LAHF,ABM,Prefetch>
    Structured Extended Features=0x29c67af<FSGSBASE,TSCADJ,SGX,BMI1,AVX2,SMEP,BMI2,ERMS,INVPCID,NFPUSG,MPX,RDSEED,ADX,SMAP,CLFLUSHOPT,PROCTRACE>
    Structured Extended Features3=0x9c002600<MCUOPT,MD_CLEAR,TSXFA,IBPB,STIBP,L1DFL,SSBD>
    XSAVE Features=0xf<XSAVEOPT,XSAVEC,XINUSE,XSAVES>
    VT-x: PAT,HLT,MTF,PAUSE,EPT,UG,VPID
    TSC: P-state invariant, performance statistics

    Anything I should try here for debugging?
    Booting a -CURRENT snapshot and telling us any change and which
    version would be an interesting datapoint, some changes to https://cgit.freebsd.org/src/log/sys/dev/tpm/tpm_crb.c have not MFCed
    yet.
    Cheers
    Uli
    --
    Posted automagically by a mail2news gateway at muc.de e.V.
    Please direct questions, flames, donations, etc. to news-admin@muc.de
    --- Synchronet 3.22a-Linux NewsLink 1.2
  • From Alexander Burke@alex@alexburke.ca to muc.lists.freebsd.stable on Wed Nov 11 07:44:54 2026
    From Newsgroup: muc.lists.freebsd.stable

    I don't even wanna use TPM or anything

    Then feel free to disable it in BIOS, or clear its state (also in BIOS).

    Sometimes the easiest solution really is the best...


    --
    Posted automagically by a mail2news gateway at muc.de e.V.
    Please direct questions, flames, donations, etc. to news-admin@muc.de
    --- Synchronet 3.22a-Linux NewsLink 1.2
  • From Kevin Bowling@kevin.bowling@kev009.com to muc.lists.freebsd.stable on Wed Nov 11 12:37:34 2026
    From Newsgroup: muc.lists.freebsd.stable

    On Wed, Nov 11, 2026 at 5:05rC>AM Ulrich Sp||rlein <uqs@freebsd.org> wrote:

    On Tue, 2026-11-10 at 21:32:11 -0700, Kevin Bowling wrote:
    On Tue, Nov 10, 2026 at 4:44rC>AM Ulrich Sp||rlein <uqs@freebsd.org> wrote: >>
    Hi there, some time this year, one of my machines started logging this
    constantly:

    Sometime this year is a wide window, can you supply any constraints
    like a particular date range or commits?

    It's hard to reconstruct this, I'm 80% sure that I actually went from 14-STABLE to 15-STABLE in July, so it could be any of them from 2026.

    tpmcrb0: Device has Error bit set

    I don't even wanna use TPM or anything, so this is kinda annoying.

    You could build a kernel without it using a small include
    GENERIC-NODEBUG and an override configuration. tpm's recent absence
    was due to fatal suspend/resume bugs, not some design decision.

    Thanks, I think I'll just suffer the low-rate spam and wait for
    16-STABLE to roll around. This is a Protectli FW6 device with coreboot
    4.12 and I don't recall this having a TPM setting in its BIOS, I
    would've likely turned it off back then.
    There's a nonzero chance your bug will never be confirmed or fixed
    since you are the only reporter so far. I would recommend booting the
    -CURRENT installer at some point.
    I can't easily reach its BIOS (doesn't do serial) so I'll have to leave
    it as is for now.

    Cheers
    Uli
    --
    Posted automagically by a mail2news gateway at muc.de e.V.
    Please direct questions, flames, donations, etc. to news-admin@muc.de
    --- Synchronet 3.22a-Linux NewsLink 1.2
  • From =?UTF-8?Q?Olivier_Cochard=2DLabb=C3=A9?=@olivier@freebsd.org to muc.lists.freebsd.stable on Thu Nov 12 22:21:22 2026
    From Newsgroup: muc.lists.freebsd.stable

    On Wed, Nov 11, 2026 at 1:07rC>PM Ulrich Sp||rlein <uqs@freebsd.org> wrote:

    On Tue, 2026-11-10 at 21:32:11 -0700, Kevin Bowling wrote:
    On Tue, Nov 10, 2026 at 4:44rC>AM Ulrich Sp||rlein <uqs@freebsd.org> wrote: >>
    Hi there, some time this year, one of my machines started logging this
    constantly:

    Sometime this year is a wide window, can you supply any constraints
    like a particular date range or commits?

    It's hard to reconstruct this, I'm 80% sure that I actually went from 14-STABLE to 15-STABLE in July, so it could be any of them from 2026.
    Hi,
    15.2 is the first release with tpm(4) and TPM entropy harvesting in
    GENERIC (every 10 seconds).
    So we need to get more data about what is the status of your TPM:
    coreboot wrongly enabling it or a FreeBSD bug.
    I haven't seen a report on Intel hardware before, so could you send:
    - the output of "uname -v";
    - whether the message repeats about every 10 seconds;
    - the output of "(sudo) tpm2_getcap properties-fixed" (from pkg tpm2-tools).
    - the output of "(sudo) acpidump -t | grep -A8 TPM2"
    To silence it in the meantime, try to add into /boot/loader.conf this
    line and reboot:
    hint.tpmcrb.0.disabled="1"
    Regards,
    Olivier
    --
    Posted automagically by a mail2news gateway at muc.de e.V.
    Please direct questions, flames, donations, etc. to news-admin@muc.de
    --- Synchronet 3.22a-Linux NewsLink 1.2
  • From =?UTF-8?Q?Olivier_Cochard=2DLabb=C3=A9?=@olivier@freebsd.org to muc.lists.freebsd.stable on Fri Nov 13 23:57:33 2026
    From Newsgroup: muc.lists.freebsd.stable

    On Fri, Nov 13, 2026 at 9:04rC>AM Ulrich Sp||rlein <uqs@freebsd.org> wrote:

    Thanks, here's the info:

    % uname -v
    FreeBSD 15.2-PRERELEASE stable/15-n286530-1d57423b4c0f GENERIC
    % tail -30 /var/log/messages
    ...
    Nov 13 08:19:24 coyote kernel: tpmcrb0: Device has Error bit set
    Nov 13 08:20:04 coyote syslogd: last message repeated 4 times
    Nov 13 08:22:14 coyote syslogd: last message repeated 13 times
    Nov 13 08:32:14 coyote syslogd: last message repeated 60 times
    [so indeed every 10s]
    Thanks, that helps.
    The start method is the standard one, and every command fails, not
    just the entropy harvester.
    My current guess, not confirmed: there may be no working TPM behind
    that address.
    Firmware can still advertise one in ACPI, for example when Intel ME or
    PTT is disabled.
    Could you send the output of these 2 commands? (both only read):
    sudo dd if=/dev/mem bs=4 count=20 skip=$((0xfed40000/4)) 2>/dev/null |
    od -A x -t x4
    pciconf -lv | grep -i -B3 -A2 'heci\|management engine'
    The first value and the one at offset 0x30 tell whether anything
    answers at that address.
    The pciconf line shows whether the Intel ME interface is visible. It
    is normally hidden when ME is disabled,
    and the firmware TPM runs inside ME.
    If the registers read ffffffff, nothing answers there, and the driver
    should reject the device at attach. I'll write that fix.
    If they look sane, a device is there, and it will be a little more
    complex to troubleshoot.
    The hint.tpmcrb.0.disabled="1" workaround in /boot/loader.conf still
    applies in the meantime.
    Regards,
    Olivier
    --
    Posted automagically by a mail2news gateway at muc.de e.V.
    Please direct questions, flames, donations, etc. to news-admin@muc.de
    --- Synchronet 3.22a-Linux NewsLink 1.2