[PATCH] um: implement arch_sync_kernel_mappings

[Johannes Berg]

On Tue, 2021-03-16 at 10:02 +0000, Anton Ivanov wrote:
> On 16/03/2021 09:59, Johannes Berg wrote:
> > On Tue, 2021-03-16 at 09:58 +0000, anton.ivanov at cambridgegreys.com
> > wrote:
> > > From: Anton Ivanov <anton.ivanov at cambridgegreys.com>
> > > 
> > > This improves the tlb flushing behavior by syncing modified
> > > ranges out of map_kernel_range_noflush in a way which is
> > > similar to x86.
> > So I take it this replaces my patch?
> 
> I get slightly better performance on userspace,
> etc compared to your patch. It is marginal -
> sub 1% close to the experimental error.

Why does that even affect userspace at all, btw? We're talking about
kernel mappings?

Do you see this speed up userspace relative to my patch, or relative to
unpatched?

johannes