On Tue, 2009-08-18 at 21:11 +0200, Jordi Mallach wrote: > Package: linux-image-2.6.30-1-powerpc > Version: 2.6.30-6 > Severity: important > > When I boot my G4-based laptop, the hw clock can't be accessed and the system > time ends up being the epoch for the platform Jan 1st, 1904. > > Switching back to 2.6.26 fixes the problem. Although CONFIG_RTC_DRV_GENERIC > was compiled in in July, it seems the switch from rtc-ppc to rtc-generic > doesn't work, at least on my system. Another Debian PPC user reports the > same breakage in his G3. > > 2.6.26 from lenny: > [ 7.800992] platform ppc-rtc.0: rtc core: registered ppc_md as rtc0 > [ 7.864990] platform ppc-rtc.0: setting system clock to 2009-08-18 18:37:53 UTC (1250620673) > > Latest 2.6.30 from unstable: > [ 6.473202] rtc-generic rtc-generic: rtc core: registered rtc-generic as rtc0 > [ 6.533547] rtc-generic rtc-generic: hctosys: unable to read the hardware clock > > Simon Raven, from #debianppc, has a working kernel which after config > inspection seems to differ only in the CONFIG_RTC_INTF_DEV_UIE_EMUL=y setting, > which is not enabled in Debian. There's very little difference between rtc-ppc and rtc-generic, and this config option should not be required by hctosys. Could you please apply this patch and report the error number? --- a/drivers/rtc/hctosys.c +++ b/drivers/rtc/hctosys.c @@ -59,7 +59,7 @@ } else dev_err(rtc->dev.parent, - "hctosys: unable to read the hardware clock\n"); + "hctosys: unable to read the hardware clock (%d)\n", err); rtc_class_close(rtc); --- END --- Ben. -- Ben Hutchings If at first you don't succeed, you're doing about average.
Description: This is a digitally signed message part