Re: [PATCH] x86/mce/amd: Fix inverted interrupt enablement during storm handling

On Fri, Aug 07, 2026 at 04:26:59PM -0600, Jasjeet Rangi wrote:
> mce_amd_handle_storm() currently does the opposite of what storm
> handling needs: it enables threshold interrupts when a storm is detected
> and disables them when the storm subsides.

the subside path:

printk_deferred(KERN_NOTICE "CPU%d BANK%d CMCI storm subsided\n", smp_processor_id(), mce->bank);
mce_handle_storm(mce->bank, false);
|->mce_amd_handle_storm(bank, on)
|-> threshold_restart_bank(bank, on);
|-> tr.b->interrupt_enable = false;

while that commit Fixes: 5c4663ed1eac ("x86/mce: Handle AMD threshold
interrupt storms") says:

    Rather, disable the interrupt on the corresponding CPU and bank.
    Re-enable back the interrupts if enough consecutive polls of the bank show no
    corrected errors (30, as programmed by Intel).

Yap, I see it.

I'm not sure about your approach to change the bool meaning - "on" is fine at
a first glance and this fix should be minimal so that it is more easily
backportable to stable - but yap, looks like we're doing it wrong.

Thx.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette