On Sun, Nov 2, 2014 at 6:15 AM, Johannes Weiner wrote:
> Memory cgroups used to have 5 per-page pointers. To allow users to
> disable that amount of overhead during runtime, those pointers were
> allocated in a separate array, with a translation layer between them
> and struct page.
>
> There is
On Sun, Nov 2, 2014 at 6:15 AM, Johannes Weiner han...@cmpxchg.org wrote:
Memory cgroups used to have 5 per-page pointers. To allow users to
disable that amount of overhead during runtime, those pointers were
allocated in a separate array, with a translation layer between them
and struct
On Tue, Nov 04, 2014 at 04:00:39PM +0100, Michal Hocko wrote:
> On Tue 04-11-14 09:09:37, Johannes Weiner wrote:
> > On Tue, Nov 04, 2014 at 02:41:10PM +0100, Michal Hocko wrote:
> > > On Tue 04-11-14 08:27:01, Johannes Weiner wrote:
> > > > From: Johannes Weiner
> > > > Subject: [patch] mm: move
From: Johannes Weiner
Date: Tue, 4 Nov 2014 09:09:37 -0500
> You either need cgroup memory accounting and limiting or not. There
> is no possible trade-off to be had.
I couldn't have said it better myself, +1.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the
On Tue 04-11-14 09:09:37, Johannes Weiner wrote:
> On Tue, Nov 04, 2014 at 02:41:10PM +0100, Michal Hocko wrote:
> > On Tue 04-11-14 08:27:01, Johannes Weiner wrote:
> > > From: Johannes Weiner
> > > Subject: [patch] mm: move page->mem_cgroup bad page handling into generic
> > > code fix
> > >
On Tue 04-11-14 08:48:41, Johannes Weiner wrote:
> On Tue, Nov 04, 2014 at 02:06:52PM +0100, Michal Hocko wrote:
> > The code size grows (~1.5k) most probably due to struct page pointer
> > arithmetic (but I haven't checked that) but the data section shrinks
> > for SLAB. So we have additional
On Tue, Nov 04, 2014 at 02:41:10PM +0100, Michal Hocko wrote:
> On Tue 04-11-14 08:27:01, Johannes Weiner wrote:
> > From: Johannes Weiner
> > Subject: [patch] mm: move page->mem_cgroup bad page handling into generic
> > code fix
> >
> > Remove obsolete memory saving recommendations from the
On Tue, Nov 04, 2014 at 02:06:52PM +0100, Michal Hocko wrote:
> The code size grows (~1.5k) most probably due to struct page pointer
> arithmetic (but I haven't checked that) but the data section shrinks
> for SLAB. So we have additional 1.6k for SLUB. I guess this is
> acceptable.
>
>text
On Tue 04-11-14 08:27:01, Johannes Weiner wrote:
> From: Johannes Weiner
> Subject: [patch] mm: move page->mem_cgroup bad page handling into generic
> code fix
>
> Remove obsolete memory saving recommendations from the MEMCG Kconfig
> help text.
The memory overhead is still there. So I do not
On Tue, Nov 04, 2014 at 05:36:39PM +0900, Kamezawa Hiroyuki wrote:
> (2014/11/02 12:15), Johannes Weiner wrote:
> > Memory cgroups used to have 5 per-page pointers. To allow users to
> > disable that amount of overhead during runtime, those pointers were
> > allocated in a separate array, with a
On Mon 03-11-14 17:36:26, Johannes Weiner wrote:
[...]
> Also, nobody is using that space currently, and I can save memory by
> moving the pointer in there. Should we later add another pointer to
> struct page we are only back to the status quo - with the difference
> that booting with
(2014/11/02 12:15), Johannes Weiner wrote:
> Memory cgroups used to have 5 per-page pointers. To allow users to
> disable that amount of overhead during runtime, those pointers were
> allocated in a separate array, with a translation layer between them
> and struct page.
>
> There is now only
(2014/11/02 12:15), Johannes Weiner wrote:
Memory cgroups used to have 5 per-page pointers. To allow users to
disable that amount of overhead during runtime, those pointers were
allocated in a separate array, with a translation layer between them
and struct page.
There is now only one page
On Mon 03-11-14 17:36:26, Johannes Weiner wrote:
[...]
Also, nobody is using that space currently, and I can save memory by
moving the pointer in there. Should we later add another pointer to
struct page we are only back to the status quo - with the difference
that booting with
On Tue, Nov 04, 2014 at 05:36:39PM +0900, Kamezawa Hiroyuki wrote:
(2014/11/02 12:15), Johannes Weiner wrote:
Memory cgroups used to have 5 per-page pointers. To allow users to
disable that amount of overhead during runtime, those pointers were
allocated in a separate array, with a
On Tue 04-11-14 08:27:01, Johannes Weiner wrote:
From: Johannes Weiner han...@cmpxchg.org
Subject: [patch] mm: move page-mem_cgroup bad page handling into generic
code fix
Remove obsolete memory saving recommendations from the MEMCG Kconfig
help text.
The memory overhead is still there.
On Tue, Nov 04, 2014 at 02:06:52PM +0100, Michal Hocko wrote:
The code size grows (~1.5k) most probably due to struct page pointer
arithmetic (but I haven't checked that) but the data section shrinks
for SLAB. So we have additional 1.6k for SLUB. I guess this is
acceptable.
textdata
On Tue, Nov 04, 2014 at 02:41:10PM +0100, Michal Hocko wrote:
On Tue 04-11-14 08:27:01, Johannes Weiner wrote:
From: Johannes Weiner han...@cmpxchg.org
Subject: [patch] mm: move page-mem_cgroup bad page handling into generic
code fix
Remove obsolete memory saving recommendations from
On Tue 04-11-14 08:48:41, Johannes Weiner wrote:
On Tue, Nov 04, 2014 at 02:06:52PM +0100, Michal Hocko wrote:
The code size grows (~1.5k) most probably due to struct page pointer
arithmetic (but I haven't checked that) but the data section shrinks
for SLAB. So we have additional 1.6k for
On Tue 04-11-14 09:09:37, Johannes Weiner wrote:
On Tue, Nov 04, 2014 at 02:41:10PM +0100, Michal Hocko wrote:
On Tue 04-11-14 08:27:01, Johannes Weiner wrote:
From: Johannes Weiner han...@cmpxchg.org
Subject: [patch] mm: move page-mem_cgroup bad page handling into generic
code fix
From: Johannes Weiner han...@cmpxchg.org
Date: Tue, 4 Nov 2014 09:09:37 -0500
You either need cgroup memory accounting and limiting or not. There
is no possible trade-off to be had.
I couldn't have said it better myself, +1.
--
To unsubscribe from this list: send the line unsubscribe
On Tue, Nov 04, 2014 at 04:00:39PM +0100, Michal Hocko wrote:
On Tue 04-11-14 09:09:37, Johannes Weiner wrote:
On Tue, Nov 04, 2014 at 02:41:10PM +0100, Michal Hocko wrote:
On Tue 04-11-14 08:27:01, Johannes Weiner wrote:
From: Johannes Weiner han...@cmpxchg.org
Subject: [patch] mm:
On Mon, Nov 03, 2014 at 10:09:42AM -0500, Johannes Weiner wrote:
> Hi Joonsoo,
>
> On Mon, Nov 03, 2014 at 05:02:08PM +0900, Joonsoo Kim wrote:
> > On Sat, Nov 01, 2014 at 11:15:54PM -0400, Johannes Weiner wrote:
> > > Memory cgroups used to have 5 per-page pointers. To allow users to
> > >
On Mon, Nov 03, 2014 at 04:58:07PM -0500, David Miller wrote:
> From: "Kirill A. Shutemov"
> Date: Mon, 3 Nov 2014 23:52:06 +0200
>
> > On Mon, Nov 03, 2014 at 04:36:28PM -0500, Johannes Weiner wrote:
> >> On Mon, Nov 03, 2014 at 11:06:07PM +0200, Kirill A. Shutemov wrote:
> >> > On Sat, Nov 01,
From: "Kirill A. Shutemov"
Date: Mon, 3 Nov 2014 23:52:06 +0200
> On Mon, Nov 03, 2014 at 04:36:28PM -0500, Johannes Weiner wrote:
>> On Mon, Nov 03, 2014 at 11:06:07PM +0200, Kirill A. Shutemov wrote:
>> > On Sat, Nov 01, 2014 at 11:15:54PM -0400, Johannes Weiner wrote:
>> > > Memory cgroups
On Mon, Nov 03, 2014 at 04:36:28PM -0500, Johannes Weiner wrote:
> On Mon, Nov 03, 2014 at 11:06:07PM +0200, Kirill A. Shutemov wrote:
> > On Sat, Nov 01, 2014 at 11:15:54PM -0400, Johannes Weiner wrote:
> > > Memory cgroups used to have 5 per-page pointers. To allow users to
> > > disable that
On Mon, Nov 03, 2014 at 11:06:07PM +0200, Kirill A. Shutemov wrote:
> On Sat, Nov 01, 2014 at 11:15:54PM -0400, Johannes Weiner wrote:
> > Memory cgroups used to have 5 per-page pointers. To allow users to
> > disable that amount of overhead during runtime, those pointers were
> > allocated in a
On Sat, Nov 01, 2014 at 11:15:54PM -0400, Johannes Weiner wrote:
> Memory cgroups used to have 5 per-page pointers. To allow users to
> disable that amount of overhead during runtime, those pointers were
> allocated in a separate array, with a translation layer between them
> and struct page.
>
On Sat, Nov 01, 2014 at 11:15:54PM -0400, Johannes Weiner wrote:
> Memory cgroups used to have 5 per-page pointers. To allow users to
> disable that amount of overhead during runtime, those pointers were
> allocated in a separate array, with a translation layer between them
> and struct page.
>
Documentation/cgroups/memory.txt is outdate even more hopelessly than
before. It deserves a complete rewrite but I guess something like the
following should be added in the meantime to prepare potential readers
about the trap.
---
diff --git a/Documentation/cgroups/memory.txt
On Mon 03-11-14 10:09:42, Johannes Weiner wrote:
> Hi Joonsoo,
>
> On Mon, Nov 03, 2014 at 05:02:08PM +0900, Joonsoo Kim wrote:
> > On Sat, Nov 01, 2014 at 11:15:54PM -0400, Johannes Weiner wrote:
> > > Memory cgroups used to have 5 per-page pointers. To allow users to
> > > disable that amount
On Sat 01-11-14 23:15:54, Johannes Weiner wrote:
> Memory cgroups used to have 5 per-page pointers. To allow users to
> disable that amount of overhead during runtime, those pointers were
> allocated in a separate array, with a translation layer between them
> and struct page.
>
> There is now
From: Johannes Weiner
Date: Mon, 3 Nov 2014 10:09:42 -0500
> Please re-introduce this code when your new usecase is ready to be
> upstreamed. There is little reason to burden an unrelated feature
> with a sizable chunk of dead code for a vague future user.
+1
--
To unsubscribe from this list:
Hi Joonsoo,
On Mon, Nov 03, 2014 at 05:02:08PM +0900, Joonsoo Kim wrote:
> On Sat, Nov 01, 2014 at 11:15:54PM -0400, Johannes Weiner wrote:
> > Memory cgroups used to have 5 per-page pointers. To allow users to
> > disable that amount of overhead during runtime, those pointers were
> > allocated
On Sat, Nov 01, 2014 at 11:15:54PM -0400, Johannes Weiner wrote:
> Memory cgroups used to have 5 per-page pointers. To allow users to
> disable that amount of overhead during runtime, those pointers were
> allocated in a separate array, with a translation layer between them
> and struct page.
On Sat, Nov 01, 2014 at 11:15:54PM -0400, Johannes Weiner wrote:
Memory cgroups used to have 5 per-page pointers. To allow users to
disable that amount of overhead during runtime, those pointers were
allocated in a separate array, with a translation layer between them
and struct page.
Hello,
Hi Joonsoo,
On Mon, Nov 03, 2014 at 05:02:08PM +0900, Joonsoo Kim wrote:
On Sat, Nov 01, 2014 at 11:15:54PM -0400, Johannes Weiner wrote:
Memory cgroups used to have 5 per-page pointers. To allow users to
disable that amount of overhead during runtime, those pointers were
allocated in a
From: Johannes Weiner han...@cmpxchg.org
Date: Mon, 3 Nov 2014 10:09:42 -0500
Please re-introduce this code when your new usecase is ready to be
upstreamed. There is little reason to burden an unrelated feature
with a sizable chunk of dead code for a vague future user.
+1
--
To unsubscribe
On Sat 01-11-14 23:15:54, Johannes Weiner wrote:
Memory cgroups used to have 5 per-page pointers. To allow users to
disable that amount of overhead during runtime, those pointers were
allocated in a separate array, with a translation layer between them
and struct page.
There is now only
On Mon 03-11-14 10:09:42, Johannes Weiner wrote:
Hi Joonsoo,
On Mon, Nov 03, 2014 at 05:02:08PM +0900, Joonsoo Kim wrote:
On Sat, Nov 01, 2014 at 11:15:54PM -0400, Johannes Weiner wrote:
Memory cgroups used to have 5 per-page pointers. To allow users to
disable that amount of overhead
Documentation/cgroups/memory.txt is outdate even more hopelessly than
before. It deserves a complete rewrite but I guess something like the
following should be added in the meantime to prepare potential readers
about the trap.
---
diff --git a/Documentation/cgroups/memory.txt
On Sat, Nov 01, 2014 at 11:15:54PM -0400, Johannes Weiner wrote:
Memory cgroups used to have 5 per-page pointers. To allow users to
disable that amount of overhead during runtime, those pointers were
allocated in a separate array, with a translation layer between them
and struct page.
On Sat, Nov 01, 2014 at 11:15:54PM -0400, Johannes Weiner wrote:
Memory cgroups used to have 5 per-page pointers. To allow users to
disable that amount of overhead during runtime, those pointers were
allocated in a separate array, with a translation layer between them
and struct page.
On Mon, Nov 03, 2014 at 11:06:07PM +0200, Kirill A. Shutemov wrote:
On Sat, Nov 01, 2014 at 11:15:54PM -0400, Johannes Weiner wrote:
Memory cgroups used to have 5 per-page pointers. To allow users to
disable that amount of overhead during runtime, those pointers were
allocated in a
On Mon, Nov 03, 2014 at 04:36:28PM -0500, Johannes Weiner wrote:
On Mon, Nov 03, 2014 at 11:06:07PM +0200, Kirill A. Shutemov wrote:
On Sat, Nov 01, 2014 at 11:15:54PM -0400, Johannes Weiner wrote:
Memory cgroups used to have 5 per-page pointers. To allow users to
disable that amount of
From: Kirill A. Shutemov kir...@shutemov.name
Date: Mon, 3 Nov 2014 23:52:06 +0200
On Mon, Nov 03, 2014 at 04:36:28PM -0500, Johannes Weiner wrote:
On Mon, Nov 03, 2014 at 11:06:07PM +0200, Kirill A. Shutemov wrote:
On Sat, Nov 01, 2014 at 11:15:54PM -0400, Johannes Weiner wrote:
Memory
On Mon, Nov 03, 2014 at 04:58:07PM -0500, David Miller wrote:
From: Kirill A. Shutemov kir...@shutemov.name
Date: Mon, 3 Nov 2014 23:52:06 +0200
On Mon, Nov 03, 2014 at 04:36:28PM -0500, Johannes Weiner wrote:
On Mon, Nov 03, 2014 at 11:06:07PM +0200, Kirill A. Shutemov wrote:
On Sat,
On Mon, Nov 03, 2014 at 10:09:42AM -0500, Johannes Weiner wrote:
Hi Joonsoo,
On Mon, Nov 03, 2014 at 05:02:08PM +0900, Joonsoo Kim wrote:
On Sat, Nov 01, 2014 at 11:15:54PM -0400, Johannes Weiner wrote:
Memory cgroups used to have 5 per-page pointers. To allow users to
disable that
From: Johannes Weiner
Date: Sat, 1 Nov 2014 23:15:54 -0400
> Memory cgroups used to have 5 per-page pointers. To allow users to
> disable that amount of overhead during runtime, those pointers were
> allocated in a separate array, with a translation layer between them
> and struct page.
>
>
From: Johannes Weiner han...@cmpxchg.org
Date: Sat, 1 Nov 2014 23:15:54 -0400
Memory cgroups used to have 5 per-page pointers. To allow users to
disable that amount of overhead during runtime, those pointers were
allocated in a separate array, with a translation layer between them
and
Memory cgroups used to have 5 per-page pointers. To allow users to
disable that amount of overhead during runtime, those pointers were
allocated in a separate array, with a translation layer between them
and struct page.
There is now only one page pointer remaining: the memcg pointer, that
Memory cgroups used to have 5 per-page pointers. To allow users to
disable that amount of overhead during runtime, those pointers were
allocated in a separate array, with a translation layer between them
and struct page.
There is now only one page pointer remaining: the memcg pointer, that
52 matches
Mail list logo