Re: [PATCH] x86: Optimize load_mm_cr4 to load_mm_cr4_irqsoff

[Dave Hansen]

On 6/18/19 12:32 AM, Jan Kiszka wrote:
> Thus, we can avoid disabling interrupts again in cr4_set/clear_bits.

Seems reasonable.

Your *_irqsoff() variants need lockdep_assert_irqs_disabled(), at least,
though.

Can you talk a bit about the motivation here, though?  Did you encounter
some performance issue that led you to make this patch, or was it simply
an improvement you realized you could make from code inspection?