Re: dropping d=1 w=0

[Maxime Villard <max%m00nbsd.net@localhost>]

Le 28/02/2020 à 19:51, is%netbsd.org@localhost a écrit :
> On Fri, Feb 28, 2020 at 07:14:47PM +0100, Maxime Villard wrote:
>> We need to eliminate the pages that have the "dirty" bit set without the
>> "writable" bit set.
>>
>> The biggest cause of these pages is pmap_write_protect(), which drops
>> PTE_W but leaves PTE_D.
> 
> Is this a bug or intentional? - as  in
> "no more changes here, but need still to write what was changed earlier"

Actually, I'm not so sure now. Maybe UVM will have to be fixed too.

The reason is that the combination of d=1 and w=0 will have a new meaning
on future x86 CPUs, which will have nothing to do with actual dirtiness
and readonlyness.