On Fri, Jan 26, 2024 at 05:10:20PM +0000, Luck, Tony wrote:
> 12 extra bytes divided by (say) 64GB (a very small server these days, may
> laptop has that much)
> = 0.00000001746%
>
> We will need 57000 changes like this one before we get to 0.001% :-)
You're forgetting that those 12 bytes repeat per MCE tracepoint logged.
And there's other code which adds more 0.01% here and there, well,
because we can.
> But the key there is keeping the details of the source machine attached to
> the error record. My first contact with machine check debugging is always
> just the raw error record (from mcelog, rasdaemon, or console log).
Yes, that is somewhat sensible reason to have the PPIN together with the
MCE record.
> Knowing which microcode version was loaded on a core *at the time of
> the error* is critical.
So is the rest of the debug info you're going to need from that machine.
And yet we're not adding that to the tracepoint.
> You've spent enough time with Ashok and Thomas tweaking the Linux
> microcode driver to know that going back to the machine the next day
> to ask about microcode version has a bunch of ways to get a wrong
> answer.
Huh, what does that have to do with this?
IIUC, if someone changes something on the system, whether that is
updating microcode or swapping a harddrive or swapping memory or
whatever, that invalidates the errors reported, pretty much.
You can't put it all in the trace record, you just can't.
--
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquette
