Re: [PATCH v2 13/16] x86/mce: Unify AMD DFR handler with MCA Polling

[Yazen Ghannam <yazen.ghannam@amd.com>]

On Tue, Feb 18, 2025 at 07:37:18AM +0000, Zhuo, Qiuxu wrote:
> > From: Yazen Ghannam <yazen.ghannam@amd.com>
> > [...]
> > +static bool smca_should_log_poll_error(enum mcp_flags flags, struct
> > +mce_hw_err *err) {
> > +	struct mce *m = &err->m;
> > +
> > +	/*
> > +	 * If this is a deferred error found in MCA_STATUS, then clear
> > +	 * the redundant data from the MCA_DESTAT register.
> > +	 */
> > +	if (m->status & MCI_STATUS_VAL) {
> > +		if (m->status & MCI_STATUS_DEFERRED)
> > +			mce_wrmsrl(MSR_AMD64_SMCA_MCx_DESTAT(m-
> > >bank), 0);
> > +
> > +		return true;
> > +	}
> > +
> > +	/*
> > +	 * If the MCA_DESTAT register has valid data, then use
> > +	 * it as the status register.
> > +	 */
> > +	m->status = mce_rdmsrl(MSR_AMD64_SMCA_MCx_DESTAT(m-
> > >bank));
> > +
> > +	if (!(m->status & MCI_STATUS_VAL))
> > +		return false;
> > +
> > +	/*
> > +	 * Gather all relevant data now and log the record before clearing
> > +	 * the deferred status register. This avoids needing to go back to
> > +	 * the polling function for these actions.
> > +	 */
> > +	mce_read_aux(err, m->bank);
> > +
> > +	if (m->status & MCI_STATUS_ADDRV)
> > +		m->addr =
> > mce_rdmsrl(MSR_AMD64_SMCA_MCx_DEADDR(m->bank));
> > +
> > +	smca_extract_err_addr(m);
> > +	m->severity = mce_severity(m, NULL, NULL, false);
> > +
> 
> Is the following check in machine_check_poll() needed before 
> queuing/logging AMD's deferred error?
> 
>        if (mca_cfg.dont_log_ce && !mce_usable_address(m))
>              //Just clear MCA_STATUS, but not queue/log errors.
> 

Good question. Deferred errors are uncorrectable errors that don't need
immediate action. They are not correctable errors, so the 'dont_log_ce'
flag shouldn't apply.

Thanks,
Yazen