Please don't reply to lustre-devel. Instead, comment in Bugzilla by using the following link: https://bugzilla.lustre.org/show_bug.cgi?id=11636
It seems our non-SMP kernel configs are actually set to enable SMP, and we don't use them anyway. I propose we remove them. >From #cfs: 13:13 < nathan> scjody, green - why are the kernel_patches/kernel_configs/*smp t he same as the non-smp ones? 13:16 < scjody> nathan: hmm, interesting. 13:16 < scjody> I think the question is really "why are the non-SMP ones the sam e as the SMP ones?" 13:16 < nathan> right. 13:17 < scjody> Looks like a bug to me. 13:17 < scjody> I'd actually be in favour of dropping the non-SMP ones entirely - we don't use them ourselves, and I question their usefulness in general. 13:17 < scjody> (especially given that they've likely been broken for a long tim e without any complaints.) 13:18 < nathan> I agree. Except kernel-2.4.21-rhel-2.4-i686 with smp doesn't bu ild with our current patches. 13:19 < nathan> but I suppose that's another issue. 13:19 < nathan> are these config files used by ltest? 13:20 < scjody> The -smp ones are. 13:20 < scjody> Does kernel-2.4.21-rhel-2.4-i686 without SMP build? 13:20 < scjody> I would guess not... 13:22 < nathan> seems to get farther 13:23 < nathan> scjody, if only -smp ones are used, and non-smp ones are smp any how, that seems a good argument to drop the non-smp. 13:24 < scjody> nathan: Yes.. I didn't realize the non-smp ones were really smp. 13:24 < nathan> looks like patchless is different, but the others I checked are not. 13:26 < scjody> OK. I'll deal with this sometime soon. _______________________________________________ Lustre-devel mailing list [email protected] https://mail.clusterfs.com/mailman/listinfo/lustre-devel
