On 04/08/2016 12:06 PM, Keith Busch wrote:
On Fri, Apr 08, 2016 at 01:40:06PM -0400, Matthew Wilcox wrote:
- Inability to use all queues supported by a device. Intel's P3700
supports 31 queues, but block-mq insists on assigning an even multiple
of CPUs to each queue. So if you have
>Hey Willy,
>
>> - Interrupt steering needs to be controlled by block-mq instead of
>> the driver. It's pointless to have each driver implement its own
>> policies on interrupt steering, irqbalanced remains a source of
>> end-user frustration, and block-mq can change the
Hey Willy,
- Interrupt steering needs to be controlled by block-mq instead of
the driver. It's pointless to have each driver implement its own
policies on interrupt steering, irqbalanced remains a source of
end-user frustration, and block-mq can change the queue<->cpu mapping
On Fri, 2016-04-08 at 13:40 -0400, Matthew Wilcox wrote:
> On Fri, Apr 08, 2016 at 01:29:26PM +0200, Hannes Reinecke wrote:
> > - Interrupt steering needs to be controlled by block-mq instead of
> the driver. It's pointless to have each driver implement its own
> policies on interrupt
On Fri, 2016-04-08 at 20:08 +0200, Christoph Hellwig wrote:
> On Fri, Apr 08, 2016 at 11:00:51AM -0700, James Bottomley wrote:
> > > - Inability to use all queues supported by a device. Intel's
> > > P3700
> > >supports 31 queues, but block-mq insists on assigning an even
> > > multiple
> >
On 04/08/2016 10:40 AM, Matthew Wilcox wrote:
- Interrupt steering needs to be controlled by block-mq instead of
the driver. It's pointless to have each driver implement its own
policies on interrupt steering, irqbalanced remains a source of
end-user frustration, and block-mq can
On Fri, Apr 08, 2016 at 11:00:51AM -0700, James Bottomley wrote:
> > - Inability to use all queues supported by a device. Intel's P3700
> >supports 31 queues, but block-mq insists on assigning an even multiple
> >of CPUs to each queue. So if you have 48 CPUs, it will use 24 queues.
> >
On Fri, Apr 08, 2016 at 01:40:06PM -0400, Matthew Wilcox wrote:
> - Inability to use all queues supported by a device. Intel's P3700
>supports 31 queues, but block-mq insists on assigning an even multiple
>of CPUs to each queue. So if you have 48 CPUs, it will use 24 queues.
>If you
On Fri, 2016-04-08 at 13:40 -0400, Matthew Wilcox wrote:
> On Fri, Apr 08, 2016 at 01:29:26PM +0200, Hannes Reinecke wrote:
> > I'd like to propose a topic on block-mq issues with FC.
> > During my performance testing using block/scsi-mq with FC I've hit
> > several issues I'd like to discuss:
>
On Fri, Apr 08, 2016 at 01:29:26PM +0200, Hannes Reinecke wrote:
> I'd like to propose a topic on block-mq issues with FC.
> During my performance testing using block/scsi-mq with FC I've hit
> several issues I'd like to discuss:
If there's a general block-mq bitching session, I have some ideas
On 04/08/2016 04:29 AM, Hannes Reinecke wrote:
I'd like to propose a topic on block-mq issues with FC.
During my performance testing using block/scsi-mq with FC I've hit
several issues I'd like to discuss:
- timeout handling:
Out of necessity the status of any timed out command is undefined.
So
On Fri, 2016-04-08 at 11:51 -0400, Ewan D. Milne wrote:
> On Fri, 2016-04-08 at 08:11 -0700, James Bottomley wrote:
> > On Fri, 2016-04-08 at 13:29 +0200, Hannes Reinecke wrote:
> > > Hi all,
> > >
> > > I'd like to propose a topic on block-mq issues with FC.
> > > During my performance testing
On Fri, 2016-04-08 at 08:11 -0700, James Bottomley wrote:
> On Fri, 2016-04-08 at 13:29 +0200, Hannes Reinecke wrote:
> > Hi all,
> >
> > I'd like to propose a topic on block-mq issues with FC.
> > During my performance testing using block/scsi-mq with FC I've hit
> > several issues I'd like to
13 matches
Mail list logo