https://bugzilla.novell.com/show_bug.cgi?id=852007 https://bugzilla.novell.com/show_bug.cgi?id=852007#c1 Thomas Renninger changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |ASSIGNED CC| |kim.naru@amd.com, | |trenn@suse.com AssignedTo|kernel-maintainers@forge.pr |trenn@suse.com |ovo.novell.com | --- Comment #1 from Thomas Renninger 2013-11-25 00:06:34 UTC --- Not sure we have such a CPU, I'll search tomorrow. Ideal would be a platform with serial console. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=852007 https://bugzilla.novell.com/show_bug.cgi?id=852007#c3 --- Comment #3 from Borislav Petkov 2013-11-25 16:57:25 UTC --- Ok, I can reproduce it exactly as described. @Kim: you might want to take a look at this. In the meantime, I'll try to reproduce it with the upstream kernel. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=852007 https://bugzilla.novell.com/show_bug.cgi?id=852007#c5 --- Comment #5 from Kim Naru 2013-11-25 18:30:08 UTC --- (In reply to comment #4)

As you may know this week AMD is shutdown so more than likely more attention will be given to this issue.

Next week. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=852007 https://bugzilla.novell.com/show_bug.cgi?id=852007#c7 --- Comment #7 from Borislav Petkov 2013-11-27 17:30:31 UTC --- Ok, I think I've rootcaused it: we're doing vmalloc() too early. Now I need to think about how to fix it, stay tuned. :-) -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=852007 https://bugzilla.novell.com/show_bug.cgi?id=852007#c9 --- Comment #9 from Borislav Petkov 2013-12-01 16:55:12 UTC --- Created an attachment (id=569744) --> (http://bugzilla.novell.com/attachment.cgi?id=569744) test patch -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=852007 https://bugzilla.novell.com/show_bug.cgi?id=852007#c10 Borislav Petkov changed: What |Removed |Added ---------------------------------------------------------------------------- Status|ASSIGNED |NEEDINFO InfoProvider| |trenn@suse.com --- Comment #10 from Borislav Petkov 2013-12-06 09:15:03 UTC --- Ok, I sent out a couple of fixes upstream. Due to their intrusive nature we're probably going to not backport them to oS13.1 so that microcode loading will not be done early there. Thomas knows this and we won't be adding the ucode to the initrd. Btw, Thomas, would it be possible to add a oneliner to some init script which does: echo 1 > /sys/devices/system/cpu/microcode/reload unconditionally on each boot? It doesn't matter whether it is Intel or AMD - it won't hurt anything because if there is no new ucode, we do exit early anyway. Thanks. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=852007 https://bugzilla.novell.com/show_bug.cgi?id=852007#c12 Thomas Renninger changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |ASSIGNED InfoProvider|trenn@suse.com | --- Comment #12 from Thomas Renninger 2013-12-09 15:27:26 UTC ---

... to include ucodes and the early microcode loading fixes into factory ... Makes sense. First Borislav should submit the patches mainline for a first review, not sure whether this happened already.

If it comes out that things are too intrusive for stable kernels and won't make it into 3.12, then the patches should be added to our openSUSE Factory kernel for now. Removing myself from needinfo. Boris is already assigned and should set the bug resolved fixed as soon as fixes hit factory, so that people can give it a try easily. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=852007 https://bugzilla.novell.com/show_bug.cgi?id=852007#c14 Borislav Petkov changed: What |Removed |Added ---------------------------------------------------------------------------- Status|ASSIGNED |CLOSED Resolution| |UPSTREAM --- Comment #14 from Borislav Petkov 2014-03-05 21:25:20 UTC --- Ok, here's the deal: Changes are too complex for 13.1 so we won't be supporting early microcode loading on AMD on 13.1. Here, we're loading the microcode when the module loads. Refreshing it without system reboot -> comment #10. The working solution will come down from upstream probably in the next oS release - something which picks kernel 3.14 or later. SLE12 has it, of course. Closing, thank you all! -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.