I was checking dmesg of good/bad cases, but there was no major
difference.

The time init looks pretty much the same - one outlier is the processor
frequency on time init which differs slightly:

Failing guests (diamond)
[    0.000000] time_init: decrementer frequency = 512.000000 MHz
[    0.000000] time_init: processor frequency   = 3458.000000 MHz
[    0.000001] clocksource: timebase: mask: 0xffffffffffffffff max_cycles 
0x761537d007, max_idle_ns: 440795202126 ns
[    0.000003] clocksource: timebase mult[1f40000] shift[24] registered
[...]
[    0.023316] clocksource: jiffies: mask: 0xffffffff max_cycles: 0xffffffff, 
max_idle_ns: 7645041785100000 ns
[...]
[    0.054877] clocksource: Switched to clocksource timebase

Working guests (wichita): TODO

[    0.000000] time_init: decrementer frequency = 512.000000 MHz
[    0.000000] time_init: processor frequency   = 3425.000000 MHz
[    0.000002] clocksource: timebase: mask: 0xffffffffffffffff max_cycles: 
0x761537d007, max_idle_ns: 440795202126 ns
[    0.000004] clocksource: timebase mult[1f40000] shift[24] registered
[...]
[    0.024374] clocksource: jiffies: mask: 0xffffffff max_cycles: 0xffffffff, 
max_idle_ns: 7645041785100000 ns
[...]
[    0.061523] clocksource: Switched to clocksource timebase

As soon as systemd starts things get a bit noisier, but no major
difference can be seen

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1668241

Title:
  Guests trash time on start, crashing the guest

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/qemu/+bug/1668241/+subscriptions

-- 
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Reply via email to