On Thu, 4 Feb 2016, Johannes Weiner wrote:
> On Wed, Feb 03, 2016 at 05:39:08PM -0800, Hugh Dickins wrote:
>
> > And (even more off-topic), I'm slightly sad to see that the lrucare
> > arg which mem_cgroup_migrate() used to have (before I renamed it and
> > you renamed it back!) has gone, so
On Thu, 4 Feb 2016, Johannes Weiner wrote:
> On Wed, Feb 03, 2016 at 05:39:08PM -0800, Hugh Dickins wrote:
>
> > And (even more off-topic), I'm slightly sad to see that the lrucare
> > arg which mem_cgroup_migrate() used to have (before I renamed it and
> > you renamed it back!) has gone, so
On Wed, Feb 03, 2016 at 05:39:08PM -0800, Hugh Dickins wrote:
> On Wed, 3 Feb 2016, Johannes Weiner wrote:
>
> > CCing Hugh and Greg, they have worked on the memcg migration code most
> > recently. AFAIK the only reason newpage->mem_cgroup had to be set up
> > that early in migration was because
On Wed, Feb 03, 2016 at 05:39:08PM -0800, Hugh Dickins wrote:
> On Wed, 3 Feb 2016, Johannes Weiner wrote:
>
> > CCing Hugh and Greg, they have worked on the memcg migration code most
> > recently. AFAIK the only reason newpage->mem_cgroup had to be set up
> > that early in migration was because
On Wed, 3 Feb 2016, Johannes Weiner wrote:
> CCing Hugh and Greg, they have worked on the memcg migration code most
> recently. AFAIK the only reason newpage->mem_cgroup had to be set up
> that early in migration was because of the way dirty accounting used
> to work. But Hugh took memcg out of
CCing Hugh and Greg, they have worked on the memcg migration code most
recently. AFAIK the only reason newpage->mem_cgroup had to be set up
that early in migration was because of the way dirty accounting used
to work. But Hugh took memcg out of the equation there, so moving
mem_cgroup_migrate() to
On Wed, Feb 03, 2016 at 02:17:49PM +0100, Mateusz Guzik wrote:
> On Fri, Jan 29, 2016 at 06:19:31PM -0500, Johannes Weiner wrote:
> > Changing a page's memcg association complicates dealing with the page,
> > so we want to limit this as much as possible. Page migration e.g. does
> > not have to do
On Fri, Jan 29, 2016 at 06:19:31PM -0500, Johannes Weiner wrote:
> Changing a page's memcg association complicates dealing with the page,
> so we want to limit this as much as possible. Page migration e.g. does
> not have to do that. Just like page cache replacement, it can forcibly
> charge a
On Fri, Jan 29, 2016 at 06:19:31PM -0500, Johannes Weiner wrote:
> Changing a page's memcg association complicates dealing with the page,
> so we want to limit this as much as possible. Page migration e.g. does
> not have to do that. Just like page cache replacement, it can forcibly
> charge a
On Fri, Jan 29, 2016 at 06:19:31PM -0500, Johannes Weiner wrote:
> Changing a page's memcg association complicates dealing with the page,
> so we want to limit this as much as possible. Page migration e.g. does
> not have to do that. Just like page cache replacement, it can forcibly
> charge a
On Fri, Jan 29, 2016 at 06:19:31PM -0500, Johannes Weiner wrote:
> Changing a page's memcg association complicates dealing with the page,
> so we want to limit this as much as possible. Page migration e.g. does
> not have to do that. Just like page cache replacement, it can forcibly
> charge a
On Wed, Feb 03, 2016 at 02:17:49PM +0100, Mateusz Guzik wrote:
> On Fri, Jan 29, 2016 at 06:19:31PM -0500, Johannes Weiner wrote:
> > Changing a page's memcg association complicates dealing with the page,
> > so we want to limit this as much as possible. Page migration e.g. does
> > not have to do
On Wed, 3 Feb 2016, Johannes Weiner wrote:
> CCing Hugh and Greg, they have worked on the memcg migration code most
> recently. AFAIK the only reason newpage->mem_cgroup had to be set up
> that early in migration was because of the way dirty accounting used
> to work. But Hugh took memcg out of
CCing Hugh and Greg, they have worked on the memcg migration code most
recently. AFAIK the only reason newpage->mem_cgroup had to be set up
that early in migration was because of the way dirty accounting used
to work. But Hugh took memcg out of the equation there, so moving
mem_cgroup_migrate() to
Changing a page's memcg association complicates dealing with the page,
so we want to limit this as much as possible. Page migration e.g. does
not have to do that. Just like page cache replacement, it can forcibly
charge a replacement page, and then uncharge the old page when it gets
freed.
Changing a page's memcg association complicates dealing with the page,
so we want to limit this as much as possible. Page migration e.g. does
not have to do that. Just like page cache replacement, it can forcibly
charge a replacement page, and then uncharge the old page when it gets
freed.
16 matches
Mail list logo