othed by commit c9b1d0981fcc
> ("mm: limit growth of 3% hardcoded other user reserve"),
> which limits the impact of this check with 128Mb (tunable via sysctl),
> but it can still be a problem on small machines.
>
> Signed-off-by: Roman Gushchin
> Cc: Andrew Morto
user_reserve_kbytes is required.
>
> Fix this issue by switching to signed arithmetic here.
>
> Signed-off-by: Roman Gushchin
> Cc: Andrew Morton
> Cc: Andrew Shewmaker
> Cc: Rik van Riel
> Cc: Konstantin Khlebnikov
> Cc: sta...@vger.kernel.org
> ---
> mm/mma
Signed-off-by: Andrew Shewmaker
---
Documentation/Changes | 2 +-
net/sched/Kconfig | 7 ---
2 files changed, 5 insertions(+), 4 deletions(-)
diff --git a/Documentation/Changes b/Documentation/Changes
index 1de131b..74bdda9 100644
--- a/Documentation/Changes
+++ b/Documentation/Changes
On Wed, Apr 10, 2013 at 12:11 AM, kbuild test robot
wrote:
> tree: git://git.cmpxchg.org/linux-mmotm.git master
> head: 47ca352cea8ba679f803387d208c739131ecb38a
> commit: 992357a07ee0697a1997c2960c3f88d02b2f2753 [81/499] mm: reinititalise
> user and admin reserves if memory is added or remove
On Mon, Apr 08, 2013 at 01:37:12PM -0700, Andrew Morton wrote:
> On Mon, 8 Apr 2013 15:07:38 -0400 Andrew Shewmaker wrote:
>
> > This patch alters the admin and user reserves of the previous patches
> > in this series when memory is added or removed.
> >
> > If me
On Tue, Apr 9, 2013 at 4:19 PM, Andrew Morton wrote:
> On Mon, 8 Apr 2013 17:00:40 -0400 Andrew Shewmaker wrote:
>
>> Should I add the memory notifier code to mm/nommu.c too?
>> I'm guessing that if a system doesn't have an mmu that it also
>> won't be hot
On Tue, Apr 9, 2013 at 6:05 PM, Simon Jeons wrote:
> Hi Andrew,
>
> On 04/10/2013 07:56 AM, Andrew Shewmaker wrote:
>>
>> On Tue, Apr 9, 2013 at 4:19 PM, Andrew Morton
>> wrote:
>>>
>>> On Mon, 8 Apr 2013 17:00:40 -0400 Andrew Shewmaker
>>>
s the same as the reserve initialization code.
I tested hot addition and removal by triggering it via sysfs. The reserves
shrunk when they were set high and memory was removed. They were reset
higher when memory was added again.
Signed-off-by: Andrew Shewmaker
---
Please see first patch in serie
ted to a need to modify reserves
when memory is hot-added or hot-removed
Signed-off-by: Andrew Shewmaker
---
Documentation/sysctl/vm.txt | 30 ++
include/linux/mm.h | 1 +
kernel/sysctl.c | 7 +++
mm/mmap.c | 30 +
38MB yes
neverno 4* 4998/5424 no 230MB yes 338MB yes
* more memtesters were launched, able to allocate approximately another 100MB
Future Work
- Test larger memory systems.
- Test an embedded image.
- Test other architectures.
- Time malloc microbenchmarks.
On Tue, Apr 9, 2013 at 4:19 PM, Andrew Morton wrote:
> On Mon, 8 Apr 2013 17:00:40 -0400 Andrew Shewmaker wrote:
>
>> Should I add the memory notifier code to mm/nommu.c too?
>> I'm guessing that if a system doesn't have an mmu that it also
>> won't be hot
On Mon, Apr 08, 2013 at 01:37:12PM -0700, Andrew Morton wrote:
> On Mon, 8 Apr 2013 15:07:38 -0400 Andrew Shewmaker wrote:
>
> > This patch alters the admin and user reserves of the previous patches
> > in this series when memory is added or removed.
> >
> > If me
This patch alters the admin and user reserves of the previous patches
in this series when memory is added or removed.
If memory is added and the reserves have been eliminated or increased above
the default max, then we'll trust the admin.
If memory is removed and there isn't enough free memory,
Add an admin_reserve_kbytes knob to allow admins to change the
hardcoded memory reserve to something other than 3%, which
may be multiple gigabytes on large memory systems. Only about
8MB is necessary to enable recovery in the default mode, and
only a few hundred MB are required even when overcommi
338MB yes
neverno 4* 4998/5424 no 230MB yes 338MB yes
* more memtesters were launched, able to allocate approximately another 100MB
Future Work
- Test larger memory systems.
- Test an embedded image.
- Test other architectures.
- Time malloc microbenchm
On Sat, Apr 6, 2013 at 6:48 PM, Simon Jeons wrote:
> Hi Andrew,
>
> On 04/05/2013 11:02 PM, Andrew Shewmaker wrote:
>>
>> On Wed, Apr 3, 2013 at 9:50 PM, Simon Jeons wrote:
>>>>
>>>> FAQ
>>>>
>> ...
>>>>
>>>>
On Wed, Apr 3, 2013 at 9:50 PM, Simon Jeons wrote:
>> FAQ
>>
...
>> * How do you calculate a minimum useful reserve?
>>
>> A user or the admin needs enough memory to login and perform
>> recovery operations, which includes, at a minimum:
>>
>> sshd or login + bash (or some other shel
On Wed, Mar 27, 2013 at 02:28:32PM -0700, Andrew Morton wrote:
> On Mon, 18 Mar 2013 17:44:42 -0400 Andrew Shewmaker wrote:
>
> > Add user_reserve_kbytes knob.
> >
> > Limit the growth of the memory reserved for other user
> > processes to min(3% current pr
5001/5427 no 230MB yes 338MB yes
neverno 4* 4998/5424 no 230MB yes 338MB yes
* more memtesters were launched, able to allocate approximately another 100MB
Future Work
- Test larger memory systems.
- Test an embedded image.
- Time malloc mic
001/5427 no 230MB yes 338MB yes
neverno 4* 4998/5424 no 230MB yes 338MB yes
* more memtesters were launched, able to allocate approximately another 100MB
Future Work
- Test larger memory systems.
- Test an embedded image.
- Time malloc microbe
everno 1 5332/5428 no 0MB no 50MB yes
neverno 1 5293/5429 no 0MB no 90MB yes
neverno 1 5001/5427 no 230MB yes 338MB yes
neverno 4* 4998/5424 no 230MB yes 338MB yes
*
everno 1 5323/5428 no 0MB no 10MB barely
neverno 1 5332/5428 no 0MB no 50MB yes
neverno 1 5293/5429 no 0MB no 90MB yes
neverno 1 5001/5427 no 230MB yes 338M
On Tue, Mar 12, 2013 at 04:01:36PM -0700, Andrew Morton wrote:
> On Wed, 6 Mar 2013 18:52:01 -0500 Andrew Shewmaker wrote:
>
> > Add user_reserve_pages knob.
> >
> > Limit the growth of the memory reserved for other user
> > processes to min(3% curre
-off-by: Andrew Shewmaker
---
v5:
* Custom sysctl handler was unnecessary
now using proc_doulongvec_minmax()
v4:
* Rebased onto v3.8-mmotm-2013-03-01-15-50
* No longer assumes 4kb pages
* Code duplicated for nommu
v3:
* New patch summary because it wasn't unique
New is "m
of each.
This only affects OVERCOMMIT_NEVER mode.
Signed-off-by: Andrew Shewmaker
---
v5:
* Change k in min(3% process size, k) into user_reserve_pages knob
* user_reserve_pages defaults to min(3% free pages, 128MB)
previous k=8MB wasn't enough for OVERCOMMIT_NEVER mode
and 128MB w
On Wed, Mar 06, 2013 at 07:46:29AM +0800, Simon Jeons wrote:
> On 03/06/2013 07:38 AM, Andrew Shewmaker wrote:
> >Limit the growth of the memory reserved for other processes
> >to the smaller of 3% or 8MB.
> >
> >This affects only OVERCOMMIT_NEVER.
> >
&
Add an admin_reserve_pages knob to allow admins of large memory systems
to change the hardcoded memory reserve to something other than 3%.
This affects OVERCOMMIT_GUESS and OVERCOMMIT_NEVER.
admin_reserve_pages is initialized to min(3% free pages, 8MB)
Signed-off-by: Andrew Shewmaker
Limit the growth of the memory reserved for other processes
to the smaller of 3% or 8MB.
This affects only OVERCOMMIT_NEVER.
Signed-off-by: Andrew Shewmaker
---
Rebased onto v3.8-mmotm-2013-03-01-15-50
No longer assumes 4kb pages.
Code duplicated for nommu.
diff --git a/mm/mmap.c b/mm
, bash, and top
to recover if necessary.
This affects OVERCOMMIT_GUESS and OVERCOMMIT_NEVER modes.
Signed-off-by: Andrew Shewmaker
---
I changed the name of the knob from "rootuser" to "admin"
because it better matches the cap_sys_admin conditional.
Documentatio
Limit the growth of the memory reserved for other processes
to the smaller of 3% or 2000 pages.
This affects OVERCOMMIT_NEVER mode.
Signed-off-by: Andrew Shewmaker
---
I had simply removed the reserve previously, but that caused forks
to fail easily. This allows a user to recover similar to
On Fri, Mar 01, 2013 at 10:40:43AM +0800, Ric Mason wrote:
> On 02/28/2013 11:48 AM, Andrew Shewmaker wrote:
> >On Thu, Feb 28, 2013 at 02:12:00PM -0800, Andrew Morton wrote:
> >>On Wed, 27 Feb 2013 15:56:30 -0500
> >>Andrew Shewmaker wrote:
> >>
> >>
On Thu, Feb 28, 2013 at 02:12:00PM -0800, Andrew Morton wrote:
> On Wed, 27 Feb 2013 15:56:30 -0500
> Andrew Shewmaker wrote:
>
> > The following patches are against the mmtom git tree as of February 27th.
> >
> > The first patch only affects OVERCOMMIT_NEVER mode, e
On Thu, Feb 28, 2013 at 02:14:41PM -0800, Andrew Morton wrote:
> On Wed, 27 Feb 2013 16:09:25 -0500
> Andrew Shewmaker wrote:
>
> > Add a rootuser_reserve_pages knob to allow admins of large memory
> > systems running with overcommit disabled to change the hardcoded
Add a rootuser_reserve_pages knob to allow admins of large memory
systems running with overcommit disabled to change the hardcoded
memory reserve to something other than 3%.
Signed-off-by: Andrew Shewmaker
---
Patch based off of mmotm git tree as of February 27th.
I set rootuser_reserve
for the root user with a
tunable knob.
Signed-off-by: Andrew Shewmaker
---
__vm_enough_memory reserves 3% of free pages with the default
overcommit mode and 6% when overcommit is disabled. These hardcoded
values have become less reasonable as memory sizes have grown.
On scientific clusters
and when it is disabled. I've made it tunable in private patches, and I plan on
submitting some version of them, but I can't decide whether a ratio or a byte
count would be more acceptable. What would people prefer see?
Signed-off-by: Andrew Shewmaker
diff --git a/mm/mmap.c b/mm/mmap
uff.
http://www.linuxbios.org/pipermail/linuxbios/2006-December/017861.html
same thread, but continued
http://www.linuxbios.org/pipermail/linuxbios/2007-January/017905.html
--
Andrew Shewmaker
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a mes
37 matches
Mail list logo
