Re: [PATCH 5/5] x86, AMD: simplify load_microcode_amd() to fix early microcode loading to no longer access uninitialized per-cpu data

2013-08-12 Thread Borislav Petkov
On Mon, Aug 12, 2013 at 03:37:19PM +0200, Johannes Hirte wrote: > Just tested your amd-ucode branch with CONFIG_MICROCODE_AMD_EARLY > re-enabled and it works so far. Ok, that's good news. Thanks a lot! -- Regards/Gruss, Boris. Sent from a fat crate under my desk. Formatting is fine. -- --

Re: [PATCH 5/5] x86, AMD: simplify load_microcode_amd() to fix early microcode loading to no longer access uninitialized per-cpu data

2013-08-12 Thread Johannes Hirte
On Thu, 8 Aug 2013 21:29:43 +0200 Borislav Petkov wrote: > On Wed, Jul 24, 2013 at 09:32:50PM +0200, Borislav Petkov wrote: > > Btw, this patch is the one that fixes the boot issue on your box, > > correct? > > > > If so, please put a minimal version of it in the next patch set > > you're

Re: [PATCH 5/5] x86, AMD: simplify load_microcode_amd() to fix early microcode loading to no longer access uninitialized per-cpu data

2013-08-12 Thread Johannes Hirte
On Thu, 8 Aug 2013 21:29:43 +0200 Borislav Petkov b...@alien8.de wrote: On Wed, Jul 24, 2013 at 09:32:50PM +0200, Borislav Petkov wrote: Btw, this patch is the one that fixes the boot issue on your box, correct? If so, please put a minimal version of it in the next patch set you're

Re: [PATCH 5/5] x86, AMD: simplify load_microcode_amd() to fix early microcode loading to no longer access uninitialized per-cpu data

2013-08-12 Thread Borislav Petkov
On Mon, Aug 12, 2013 at 03:37:19PM +0200, Johannes Hirte wrote: Just tested your amd-ucode branch with CONFIG_MICROCODE_AMD_EARLY re-enabled and it works so far. Ok, that's good news. Thanks a lot! -- Regards/Gruss, Boris. Sent from a fat crate under my desk. Formatting is fine. -- -- To

Re: [PATCH 5/5] x86, AMD: simplify load_microcode_amd() to fix early microcode loading to no longer access uninitialized per-cpu data

2013-08-08 Thread Borislav Petkov
On Wed, Jul 24, 2013 at 09:32:50PM +0200, Borislav Petkov wrote: > Btw, this patch is the one that fixes the boot issue on your box, > correct? > > If so, please put a minimal version of it in the next patch set you're > sending right after Ok, I've wiggled it ontop of the cpu_has_amd_erratum()

Re: [PATCH 5/5] x86, AMD: simplify load_microcode_amd() to fix early microcode loading to no longer access uninitialized per-cpu data

2013-08-08 Thread Borislav Petkov
On Wed, Jul 24, 2013 at 09:32:50PM +0200, Borislav Petkov wrote: Btw, this patch is the one that fixes the boot issue on your box, correct? If so, please put a minimal version of it in the next patch set you're sending right after Ok, I've wiggled it ontop of the cpu_has_amd_erratum() patch,

Re: [PATCH 5/5] x86, AMD: simplify load_microcode_amd() to fix early microcode loading to no longer access uninitialized per-cpu data

2013-07-24 Thread Borislav Petkov
On Tue, Jul 23, 2013 at 11:06:10PM +0200, Torsten Kaiser wrote: > load_microcode_amd() (and the helper it is using) should not have an > cpu parameter. The microcode loading is not depending on the CPU it is > executed and all the loaded patches will end up in a global list for all > CPUs anyway.

Re: [PATCH 5/5] x86, AMD: simplify load_microcode_amd() to fix early microcode loading to no longer access uninitialized per-cpu data

2013-07-24 Thread Borislav Petkov
On Tue, Jul 23, 2013 at 11:06:10PM +0200, Torsten Kaiser wrote: load_microcode_amd() (and the helper it is using) should not have an cpu parameter. The microcode loading is not depending on the CPU it is executed and all the loaded patches will end up in a global list for all CPUs anyway. The

[PATCH 5/5] x86, AMD: simplify load_microcode_amd() to fix early microcode loading to no longer access uninitialized per-cpu data

2013-07-23 Thread Torsten Kaiser
load_microcode_amd() (and the helper it is using) should not have an cpu parameter. The microcode loading is not depending on the CPU it is executed and all the loaded patches will end up in a global list for all CPUs anyway. The change from cpu to x86family in load_microcode_amd() now allows to

[PATCH 5/5] x86, AMD: simplify load_microcode_amd() to fix early microcode loading to no longer access uninitialized per-cpu data

2013-07-23 Thread Torsten Kaiser
load_microcode_amd() (and the helper it is using) should not have an cpu parameter. The microcode loading is not depending on the CPU it is executed and all the loaded patches will end up in a global list for all CPUs anyway. The change from cpu to x86family in load_microcode_amd() now allows to