On Fri, Nov 30, 2012 at 08:58:31PM +0100, Ingo Molnar wrote:
> I'm pleased to announce the latest, -v18 numa/core release.
>
I collected the results for the following kernels
stats-v8r6TLB flush optimisations, stats from balancenuma tree
numacore-20121130 numacore v17 (tip/master as of N
On Mon, 3 Dec 2012, Mel Gorman wrote:
> On Fri, Nov 30, 2012 at 12:37:49PM -0800, Linus Torvalds wrote:
> > So if this is a migration-specific scalability issue, then it might be
> > possible to solve by making the mutex be a rwsem instead, and have
> > migration only take it for reading.
> >
> >
On Fri, Nov 30, 2012 at 12:37:49PM -0800, Linus Torvalds wrote:
> On Fri, Nov 30, 2012 at 11:58 AM, Ingo Molnar wrote:
> >
> > When pushed hard enough via threaded workloads (for example via the
> > numa02 test) then the upstream page migration code in mm/migration.c
> > becomes unscalable, result
I was away for the weekend so did not see this until Sunday night. I
queued up tip/master as it looked at that time and ran it overnight. In
general, I have not looked closely at any of the patches.
As a heads-up, I'm also flying very early tomorrow morning and will be
travelling for the week. I'l
On Fri, Nov 30, 2012 at 11:58 AM, Ingo Molnar wrote:
>
> When pushed hard enough via threaded workloads (for example via the
> numa02 test) then the upstream page migration code in mm/migration.c
> becomes unscalable, resulting in lot of scheduling on the anon vma
> mutex and a subsequent drop in
I'm pleased to announce the latest, -v18 numa/core release.
This release fixes regressions and improves NUMA performance.
It has the following main changes:
- Introduce directed NUMA convergence, which is based on
the 'task buddy' relation introduced in -v17, and make
use of the new "ta
6 matches
Mail list logo
