On Thu, Jan 02, 2025 at 10:46:54AM +0800, Huang, Ying wrote:
> IIUC, the hardware error will be written out on non-volatile storage at
> the same time with or without ghes_panic_timeout overriding. The
> difference is that after rebooting, the error information in
> non-volatile storage can be extracted and reported via UI, SNMP, or
> MCTP earlier.
... which basically makes this forceful rebooting even worse. The user can
reboot whenever she sees fit - not when by some questionable decision done by
kernel devs.
One more reason to whack this thing.
Thx.
--
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquette
