Messages in this thread | | | Date | Thu, 1 Dec 2022 09:24:05 -0800 | From | "Paul E. McKenney" <> | Subject | Re: [PATCH clocksource 1/3] clocksource: Reject bogus watchdog clocksource measurements |
| |
On Wed, Nov 30, 2022 at 02:00:04PM +0800, Feng Tang wrote: > On Tue, Nov 29, 2022 at 09:50:24PM -0800, Paul E. McKenney wrote: > [...] > > > > > > Great! As both HPET and PM_TIMER get the same calibration 1975.000 MHz, > > > and it matches the 40ms drift per second you mentioned earlier, this > > > seems like the CPUID(0x15) gave the wrong frequence number. > > > > > > Or unlikely, HPET and PM_TIMER are driven by the same circuit source, > > > which has deviation. > > > > > > Either way, I think the HW/FW could have some problem. > > > > And this time with your printk()s: > > > > [ 0.000000] tsc: using CPUID[0x15] crystal_khz= 24000 kHz ebx=158 eax=2 > > [ 0.000000] tsc: Detected 1900.000 MHz processor > > [ 0.000000] tsc: Detected 1896.000 MHz TSC > > [ 5.268858] clocksource: refined-jiffies: mask: 0xffffffff max_cycles: 0xffffffff, max_idle_ns: 1910969940391419 ns > > [ 25.706231] clocksource: tsc-early: mask: 0xffffffffffffffff max_cycles: 0x36a8d32ce31, max_idle_ns: 881590731004 ns > > [ 32.223011] clocksource: jiffies: mask: 0xffffffff max_cycles: 0xffffffff, max_idle_ns: 1911260446275000 ns > > [ 57.823933] clocksource: Switched to clocksource tsc-early > > [ 58.144840] clocksource: acpi_pm: mask: 0xffffff max_cycles: 0xffffff, max_idle_ns: 2085701024 ns > > [ 63.613713] clocksource: tsc: mask: 0xffffffffffffffff max_cycles: 0x36a8d32ce31, max_idle_ns: 881590731004 ns > > [ 63.637323] clocksource: Switched to clocksource tsc > > [ 64.673579] tsc: Warning: TSC freq calibrated by CPUID/MSR differs from what is calibrated by HW timer, please check with vendor!! > > [ 64.703719] tsc: Previous calibrated TSC freq: 1896.000 MHz > > [ 64.716816] tsc: TSC freq recalibrated by [PM_TIMER]: 1974.999 MHz > > This confirms the tsc frequency is calculated from CPUID(0x15). > > > What would be good next steps to check up on the hardware and firmware? > > Maybe raise it to vendor? I have no idea how to check thos black boxes :)
Done, but no high hopes here. (What, me cynical? Better believe it!!!)
> > (My next step involves a pillow, but will follow up tomorrow morning > > Pacific Time.) > > Really thanks for checking this through late night!
No problem, and I guess it is instead the day after tomorrow, but I thought you might be interested in chronyd's opinion:
[root@rtptest1029.snc8 ~]# cat /var/lib/chrony/drift 40001.074911 0.002098
In contrast, on my Fedora laptop:
$ sudo cat /var/lib/chrony/drift 2.074313 0.186606
I am (perhaps incorrectly) taking this to indicate that TSC is in fact drifting with respect to standard time.
Thoughts?
Thanx, Paul
| |