Messages in this thread | | | Date | Thu, 15 Aug 2013 12:01:32 +0200 | From | Borislav Petkov <> | Subject | Re: [PATCH 3/3] mce: acpi/apei: trace: Enable ghes memory error trace event |
| |
On Wed, Aug 14, 2013 at 09:15:04PM -0300, Mauro Carvalho Chehab wrote: > > - Two, if ghes_edac is enabled, it prevents other edac drivers > > from being loaded. It looks like the assumption here is that if > > ghes/firmware first is enabled, then *all* memory errors are > > reported through ghes which is not true. We could have (a subset > > of) corrected errors reported through ghes, some through CMCI > > and uncorrected errors through MCE. So, if I'm not mistaken, if > > ghes_edac is enabled, we will only receive ghes error events through > > mc_event and not the others. Mauro, is this accurate? > > Yes, that's the current assumption. It prevents to have both BIOS and a > direct-hardware-access-EDAC-driver to race, as this is known to have > serious issues.
Ok, this is getting confusing so let's shed some more light.
* First of all, Naveen is asking whether other *edac* drivers can be loaded. And no, they cannot once the ghes thing is loaded which potentially is a problem.
For example, if the chipset-specific driver has additional functionality from ghes_edac, then that functionality is gone when ghes_edac loads first. This might be a problem in some cases, we probably need to think about this more in depth.
* Then, there's the trace_mce_record() TP which comes straight from mce.c This one is always enabled unless the mce_disable_bank functionality kicks in which you, Naveen, added :-).
* The CMCI stuff should be synchronized with the MCE TP so that should be ok.
I think those are our current error reporting paths...
-- Regards/Gruss, Boris.
Sent from a fat crate under my desk. Formatting is fine. --
| |