Re: [PATCH v4 1/2] x86/mce: Check for writes ignored in MCA_STATUS register

[Smita Koralahalli]

On 4/20/2022 2:17 AM, Borislav Petkov wrote:
> On Mon, Apr 18, 2022 at 08:24:35PM -0700, Smita Koralahalli wrote:
> > Why are we checking this here? This flag (hw_injection_possible)
> > is set to false inside prepare_msrs() called from
> > smp_call_function_single().
> > Should this check be done after the call to smp_call_function_single()?
> Why would you do that then?
>
> Is any of the code after
>
>          if (inj_type == SW_INJ) {
>                  mce_log(m);
>                  return 0;
>          }
>
> worth running if hardware injection is not possible?

Okay I got the gist of this now. This will be mainly useful for subsequent
hardware error injections.

Also, should we move this slightly before? In inj_bank_set() after we check
for sw injection and before reading IPID value?

> > Also, we already have inj_desc.err which returns error code to userspace
> > when WRIG in status registers. Why is this flag needed?
> To not do unnecessary work when you *know* hardware injection won't
> work.
>
> :-)

Okay.

> Btw, please trim your mails when you reply, just like I did.

I'm sorry. Noted for my next replies!

Thanks,
Smita
> Thx.