On Thu, Oct 3, 2019 at 11:03 AM Thomas Hellström (VMware)
wrote:
>
> >
> > So I think this is the right direction to move into, but I do want
> > people to think about this, and think about that next phase of doing
> > the pmd_trans_huge_lock too.
>
> I think if we take the ptl lock outside the ca
On 10/3/19 6:55 PM, Linus Torvalds wrote:
d) Fix the pte walker to do the right thing, then just use separate
pte walkers in your code
The fix would be those two conceptual changes:
1) don't split if the walker asks for a pmd_entry (the walker itself
can then decide to split, of course, bu
On Thu, Oct 3, 2019 at 12:56 AM Thomas Hellstrom wrote:
>
> Both the cleaning operation and the wp operation operate on shared
> writable mappings, and since they are also both restricted to entries
> that may take part in dirty-tracking (they're ignoring pmds and puds),
> so perhaps a "dirty" may
On 10/2/19 10:28 PM, Linus Torvalds wrote:
> On Wed, Oct 2, 2019 at 12:09 PM Thomas Hellström (VMware)
> wrote:
>> Yes I typically tend towards using a "namespace_object_operation" naming
>> scheme, with "as_dirty" being the namespace here,
> We discourage that kind of mindless namespacing for cor
On Wed, Oct 2, 2019 at 12:09 PM Thomas Hellström (VMware)
wrote:
>
> Yes I typically tend towards using a "namespace_object_operation" naming
> scheme, with "as_dirty" being the namespace here,
We discourage that kind of mindless namespacing for core stuff.
It makes sense in a driver or a filesy
On 10/2/19 8:06 PM, Linus Torvalds wrote:
On Wed, Oct 2, 2019 at 6:48 AM Thomas Hellström (VMware)
wrote:
From: Thomas Hellstrom
Add two utilities to a) write-protect and b) clean all ptes pointing into
a range of an address space.
This one I still don't exactly love.
I'm not entirely sure
On Wed, Oct 02, 2019 at 11:06:43AM -0700, Linus Torvalds wrote:
> On Wed, Oct 2, 2019 at 6:48 AM Thomas Hellström (VMware)
> wrote:
> >
> > From: Thomas Hellstrom
> >
> > Add two utilities to a) write-protect and b) clean all ptes pointing into
> > a range of an address space.
[...]
> Yes, it's a
On Wed, Oct 2, 2019 at 6:48 AM Thomas Hellström (VMware)
wrote:
>
> From: Thomas Hellstrom
>
> Add two utilities to a) write-protect and b) clean all ptes pointing into
> a range of an address space.
This one I still don't exactly love.
I'm not entirely sure what rubs me the wrong way, but part
From: Thomas Hellstrom
Add two utilities to a) write-protect and b) clean all ptes pointing into
a range of an address space.
The utilities are intended to aid in tracking dirty pages (either
driver-allocated system memory or pci device memory).
The write-protect utility should be used in conjunc
9 matches
Mail list logo
