On Fri, Aug 12, 2011 at 14:15:34 +0200, Ana Guerrero wrote: > On Fri, Aug 12, 2011 at 02:05:13PM +0200, Julien Cristau wrote: > > On Fri, Aug 12, 2011 at 13:46:25 +0200, Ana Guerrero wrote: > > > > > --- chrony-1.24.orig/sys_linux.c > > > +++ chrony-1.24/sys_linux.c > > > @@ -819,6 +819,11 @@ > > > LOG_FATAL(LOGF_SysLinux, "Kernel version not supported yet, > > > sorry."); > > > } > > > break; > > > + case 3: > > > + /* These don't need scaling (treat like 2.6.28 and later) */ > > > + freq_scale = 1.0; > > > + have_readonly_adjtime = 2; > > > + break; > > > default: > > > LOG_FATAL(LOGF_SysLinux, "Kernel's major version not supported > > > yet, sorry"); > > > break; > > > > Any particular reason you're not removing the LOG_FATAL thing entirely, > > and making the "new kernel" behaviour be the 'default' case? > > Hi! > > Wouldn't that be very intrusive for a NMU? I don't like the code either but > luckily upstream has redone this in their new release version. Hopefully John > will > do the new release upload himself soon or ack new (co-)maintainers. > Or you were asking for another reason? > I don't know. It just seems wrong for that check to fail if it encounters an unknown kernel version. Not sure why you think it'd be very intrusive, it's at most 2 more lines than your change.
Cheers, Julien -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected]
