Re: [BUG,2.6.28,s390] Fails to boot in Hercules S/390 emulator

[Frans Pop <elendil@planet.nl>]

On Wednesday 11 March 2009, Frans Pop wrote:
> From this I conclude that clock->error is getting buggered somewhere
> else: we get a completely different value back from what is calculated
> here.

Some extra printks at least close the loop:

0: scale/shift: 32/8, xtime_ns old: 168555264000, new: 168555264256
     tv_ns: 658419001, rem: -256, old_err: 0, error: -4294867296
bigadjust: adj: 2097152, int: 260046848, offs: 4194304
CS adjust: -4294867296 -> -4292487689804800

1: scale/shift: 32/8, xtime_ns old: 168551069952, new: 168551070208
     tv_ns: 658402618, rem: -256, old_err: -4292487689804800, error: -4292501984672096
bigadjust: adj: 4194304, int: 520093696, offs: 12582912
CS adjust: -4292501984672096 -> -12807120030269440

2: scale/shift: 32/8, xtime_ns old: 168538487296, new: 168538487552
    tv_ns: 658353467, rem: -256, old_err: -12807120030269440, error: -12807124325236736
bigadjust: adj: 1048576, int: 130023424, offs: 4194304
CS adjust: -12807124325236736 -> -14918186650466656

"bigadjust" is from just before the return in clocksource_bigadjust.
"CS adjust" is from after clock->error gets calculated in clocksource_adjust.