login
Header Space

 
 

Re: [PATCH 0/5] make slab gfp fair

Score:
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
To: Peter Zijlstra <a.p.zijlstra@...>
Cc: <clameter@...>, <mpm@...>, <linux-kernel@...>, <linux-mm@...>, <tgraf@...>, <davem@...>, <akpm@...>, <phillips@...>, <penberg@...>
Date: Thursday, May 17, 2007 - 5:44 pm

> The way I read the cpuset page allocator, it will only respect the

Wrong.  Well, only a little right.

For allocations that can't fail (the kernel could die if it failed)
then yes, the kernel will eventually take any damn page it can find,
regardless of cpusets.

Allocations for user space are hardwall enforced to be in the current
tasks cpuset.

Allocations off interrupts ignore the current tasks cpuset (such allocations
don't have a valid current contect.)

Allocations for most kernel space allocations will try to fit in the
current tasks cpuset, but may come from the possibly larger context of
the closest ancestor cpuset that is marked memory_exclusive.

-- 
                  I won't rest till it's the best ...
                  Programmer, Linux Scalability
                  Paul Jackson <pj@sgi.com> 1.925.600.0401
-
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:
[PATCH 0/5] make slab gfp fair, Peter Zijlstra, (Mon May 14, 9:19 am)
Re: [PATCH 0/5] make slab gfp fair, Christoph Lameter, (Wed May 16, 11:02 pm)
Re: [PATCH 0/5] make slab gfp fair, Peter Zijlstra, (Thu May 17, 3:08 am)
Re: [PATCH 0/5] make slab gfp fair, Christoph Lameter, (Thu May 17, 1:29 pm)
Re: [PATCH 0/5] make slab gfp fair, Matt Mackall, (Thu May 17, 1:53 pm)
Re: [PATCH 0/5] make slab gfp fair, Christoph Lameter, (Thu May 17, 2:02 pm)
Re: [PATCH 0/5] make slab gfp fair, Peter Zijlstra, (Thu May 17, 3:18 pm)
Re: [PATCH 0/5] make slab gfp fair, Christoph Lameter, (Thu May 17, 3:24 pm)
Re: [PATCH 0/5] make slab gfp fair, Peter Zijlstra, (Thu May 17, 5:26 pm)
Re: [PATCH 0/5] make slab gfp fair, Christoph Lameter, (Thu May 17, 6:27 pm)
Re: [PATCH 0/5] make slab gfp fair, Peter Zijlstra, (Fri May 18, 5:54 am)
Re: [PATCH 0/5] make slab gfp fair, Christoph Lameter, (Fri May 18, 1:11 pm)
Re: [PATCH 0/5] make slab gfp fair, Peter Zijlstra, (Sun May 20, 4:39 am)
Re: [PATCH 0/5] make slab gfp fair, Christoph Lameter, (Mon May 21, 12:45 pm)
Re: [PATCH 0/5] make slab gfp fair, Peter Zijlstra, (Mon May 21, 3:33 pm)
Re: [PATCH 0/5] make slab gfp fair, Christoph Lameter, (Mon May 21, 3:43 pm)
Re: [PATCH 0/5] make slab gfp fair, Peter Zijlstra, (Mon May 21, 4:08 pm)
Re: [PATCH 0/5] make slab gfp fair, Christoph Lameter, (Mon May 21, 4:32 pm)
Re: [PATCH 0/5] make slab gfp fair, Peter Zijlstra, (Mon May 21, 4:54 pm)
Re: [PATCH 0/5] make slab gfp fair, Christoph Lameter, (Mon May 21, 5:04 pm)
Re: [PATCH 0/5] make slab gfp fair, Paul Jackson, (Fri May 18, 1:11 pm)
Re: [PATCH 0/5] make slab gfp fair, Paul Jackson, (Thu May 17, 5:44 pm)
Re: [PATCH 0/5] make slab gfp fair, Peter Zijlstra, (Thu May 17, 1:52 pm)
Re: [PATCH 0/5] make slab gfp fair, Christoph Lameter, (Thu May 17, 1:59 pm)
Re: [PATCH 0/5] make slab gfp fair, Christoph Lameter, (Mon May 14, 11:53 am)
Re: [PATCH 0/5] make slab gfp fair, Matt Mackall, (Mon May 14, 12:12 pm)
Re: [PATCH 0/5] make slab gfp fair, Christoph Lameter, (Mon May 14, 12:29 pm)
Re: [PATCH 0/5] make slab gfp fair, Peter Zijlstra, (Mon May 14, 1:40 pm)
Re: [PATCH 0/5] make slab gfp fair, Christoph Lameter, (Mon May 14, 1:57 pm)
Re: [PATCH 0/5] make slab gfp fair, Peter Zijlstra, (Mon May 14, 3:28 pm)
Re: [PATCH 0/5] make slab gfp fair, Christoph Lameter, (Mon May 14, 3:56 pm)
Re: [PATCH 0/5] make slab gfp fair, Peter Zijlstra, (Mon May 14, 4:03 pm)
Re: [PATCH 0/5] make slab gfp fair, Christoph Lameter, (Mon May 14, 4:25 pm)
Re: [PATCH 0/5] make slab gfp fair, Christoph Lameter, (Mon May 14, 4:06 pm)
Re: [PATCH 0/5] make slab gfp fair, Peter Zijlstra, (Mon May 14, 4:12 pm)
Re: [PATCH 0/5] make slab gfp fair, Peter Zijlstra, (Tue May 15, 1:27 pm)
Re: [PATCH 0/5] make slab gfp fair, Christoph Lameter, (Tue May 15, 6:02 pm)
Re: [PATCH 0/5] make slab gfp fair, Peter Zijlstra, (Wed May 16, 2:59 am)
Re: [PATCH 0/5] make slab gfp fair, Christoph Lameter, (Wed May 16, 2:43 pm)
Re: [PATCH 0/5] make slab gfp fair, Peter Zijlstra, (Wed May 16, 3:25 pm)
Re: [PATCH 0/5] make slab gfp fair, Christoph Lameter, (Wed May 16, 3:53 pm)
Re: [PATCH 0/5] make slab gfp fair, Peter Zijlstra, (Wed May 16, 4:18 pm)
Re: [PATCH 0/5] make slab gfp fair, Christoph Lameter, (Wed May 16, 4:27 pm)
Re: [PATCH 0/5] make slab gfp fair, Peter Zijlstra, (Wed May 16, 4:40 pm)
Re: [PATCH 0/5] make slab gfp fair, Christoph Lameter, (Wed May 16, 4:44 pm)
Re: [PATCH 0/5] make slab gfp fair, Peter Zijlstra, (Wed May 16, 4:54 pm)
Re: [PATCH 0/5] make slab gfp fair, Christoph Lameter, (Wed May 16, 4:59 pm)
Re: [PATCH 0/5] make slab gfp fair, Peter Zijlstra, (Wed May 16, 5:04 pm)
Re: [PATCH 0/5] make slab gfp fair, Christoph Lameter, (Wed May 16, 5:13 pm)
Re: [PATCH 0/5] make slab gfp fair, Peter Zijlstra, (Wed May 16, 5:20 pm)
Re: [PATCH 0/5] make slab gfp fair, Christoph Lameter, (Wed May 16, 5:42 pm)
Re: [PATCH 0/5] make slab gfp fair, Peter Zijlstra, (Thu May 17, 3:28 am)
Re: [PATCH 0/5] make slab gfp fair, Christoph Lameter, (Thu May 17, 1:30 pm)
Re: [PATCH 0/5] make slab gfp fair, Peter Zijlstra, (Thu May 17, 1:53 pm)
Re: [PATCH 0/5] make slab gfp fair, Christoph Lameter, (Thu May 17, 2:01 pm)
Re: [PATCH 0/5] make slab gfp fair, Andrew Morton, (Mon May 14, 3:44 pm)
Re: [PATCH 0/5] make slab gfp fair, Matt Mackall, (Mon May 14, 4:01 pm)
Re: [PATCH 0/5] make slab gfp fair, Peter Zijlstra, (Mon May 14, 4:05 pm)
Re: [PATCH 0/5] make slab gfp fair, Peter Zijlstra, (Mon May 14, 12:10 pm)
Re: [PATCH 0/5] make slab gfp fair, Christoph Lameter, (Mon May 14, 12:37 pm)
speck-geostationary