On Mon, Apr 27, 2015 at 10:24:05AM +1000, NeilBrown wrote:
> On Fri, 24 Apr 2015 21:39:04 +0800 Yuanhan Liu
> wrote:
>
> > I noticed heavy spin lock contention at get_active_stripe() with fsmark
> > multiple thread write workloads.
> >
> > Here is how this hot contention comes from. We have
On Fri, 24 Apr 2015 21:39:04 +0800 Yuanhan Liu
wrote:
> I noticed heavy spin lock contention at get_active_stripe() with fsmark
> multiple thread write workloads.
>
> Here is how this hot contention comes from. We have limited stripes, and
> it's a multiple thread write workload. Hence, those
On Fri, 24 Apr 2015 21:39:04 +0800 Yuanhan Liu yuanhan@linux.intel.com
wrote:
I noticed heavy spin lock contention at get_active_stripe() with fsmark
multiple thread write workloads.
Here is how this hot contention comes from. We have limited stripes, and
it's a multiple thread write
On Mon, Apr 27, 2015 at 10:24:05AM +1000, NeilBrown wrote:
On Fri, 24 Apr 2015 21:39:04 +0800 Yuanhan Liu yuanhan@linux.intel.com
wrote:
I noticed heavy spin lock contention at get_active_stripe() with fsmark
multiple thread write workloads.
Here is how this hot contention comes
I noticed heavy spin lock contention at get_active_stripe() with fsmark
multiple thread write workloads.
Here is how this hot contention comes from. We have limited stripes, and
it's a multiple thread write workload. Hence, those stripes will be taken
soon, which puts later processes to sleep for
I noticed heavy spin lock contention at get_active_stripe() with fsmark
multiple thread write workloads.
Here is how this hot contention comes from. We have limited stripes, and
it's a multiple thread write workload. Hence, those stripes will be taken
soon, which puts later processes to sleep for
6 matches
Mail list logo