On 7/19/2017 5:20 AM, Sagi Grimberg wrote:
>> Code is moving the completion queue doorbell after processing all completed
>> events and sending callbacks to the block layer on each iteration.
>>
>> This is causing a performance drop when a lot of jobs are queued towards
>> the HW. Move the
On 7/19/2017 5:20 AM, Sagi Grimberg wrote:
>> Code is moving the completion queue doorbell after processing all completed
>> events and sending callbacks to the block layer on each iteration.
>>
>> This is causing a performance drop when a lot of jobs are queued towards
>> the HW. Move the
Code is moving the completion queue doorbell after processing all completed
events and sending callbacks to the block layer on each iteration.
This is causing a performance drop when a lot of jobs are queued towards
the HW. Move the completion queue doorbell on each loop instead and allow new
Code is moving the completion queue doorbell after processing all completed
events and sending callbacks to the block layer on each iteration.
This is causing a performance drop when a lot of jobs are queued towards
the HW. Move the completion queue doorbell on each loop instead and allow new
On Tue, Jul 18, 2017 at 02:52:26PM -0400, Sinan Kaya wrote:
> On 7/18/2017 10:36 AM, Keith Busch wrote:
>
> I do see that the NVMe driver is creating a completion interrupt on
> each CPU core for the completions. No problems with that.
>
> However, I don't think you can guarantee that there
On Tue, Jul 18, 2017 at 02:52:26PM -0400, Sinan Kaya wrote:
> On 7/18/2017 10:36 AM, Keith Busch wrote:
>
> I do see that the NVMe driver is creating a completion interrupt on
> each CPU core for the completions. No problems with that.
>
> However, I don't think you can guarantee that there
On 7/18/2017 10:36 AM, Keith Busch wrote:
> On Mon, Jul 17, 2017 at 07:07:00PM -0400, ok...@codeaurora.org wrote:
>> Maybe, I need to understand the design better. I was curious why completion
>> and submission queues were protected by a single lock causing lock
>> contention.
> Ideally the queues
On 7/18/2017 10:36 AM, Keith Busch wrote:
> On Mon, Jul 17, 2017 at 07:07:00PM -0400, ok...@codeaurora.org wrote:
>> Maybe, I need to understand the design better. I was curious why completion
>> and submission queues were protected by a single lock causing lock
>> contention.
> Ideally the queues
On Mon, Jul 17, 2017 at 07:07:00PM -0400, ok...@codeaurora.org wrote:
> Maybe, I need to understand the design better. I was curious why completion
> and submission queues were protected by a single lock causing lock
> contention.
Ideally the queues are tied to CPUs, so you couldn't have one
On Mon, Jul 17, 2017 at 07:07:00PM -0400, ok...@codeaurora.org wrote:
> Maybe, I need to understand the design better. I was curious why completion
> and submission queues were protected by a single lock causing lock
> contention.
Ideally the queues are tied to CPUs, so you couldn't have one
On 2017-07-17 18:56, Keith Busch wrote:
On Mon, Jul 17, 2017 at 06:46:11PM -0400, Sinan Kaya wrote:
Hi Keith,
On 7/17/2017 6:45 PM, Keith Busch wrote:
> On Mon, Jul 17, 2017 at 06:36:23PM -0400, Sinan Kaya wrote:
>> Code is moving the completion queue doorbell after processing all completed
>>
On 2017-07-17 18:56, Keith Busch wrote:
On Mon, Jul 17, 2017 at 06:46:11PM -0400, Sinan Kaya wrote:
Hi Keith,
On 7/17/2017 6:45 PM, Keith Busch wrote:
> On Mon, Jul 17, 2017 at 06:36:23PM -0400, Sinan Kaya wrote:
>> Code is moving the completion queue doorbell after processing all completed
>>
On Mon, Jul 17, 2017 at 06:46:11PM -0400, Sinan Kaya wrote:
> Hi Keith,
>
> On 7/17/2017 6:45 PM, Keith Busch wrote:
> > On Mon, Jul 17, 2017 at 06:36:23PM -0400, Sinan Kaya wrote:
> >> Code is moving the completion queue doorbell after processing all completed
> >> events and sending callbacks
On Mon, Jul 17, 2017 at 06:46:11PM -0400, Sinan Kaya wrote:
> Hi Keith,
>
> On 7/17/2017 6:45 PM, Keith Busch wrote:
> > On Mon, Jul 17, 2017 at 06:36:23PM -0400, Sinan Kaya wrote:
> >> Code is moving the completion queue doorbell after processing all completed
> >> events and sending callbacks
Hi Keith,
On 7/17/2017 6:45 PM, Keith Busch wrote:
> On Mon, Jul 17, 2017 at 06:36:23PM -0400, Sinan Kaya wrote:
>> Code is moving the completion queue doorbell after processing all completed
>> events and sending callbacks to the block layer on each iteration.
>>
>> This is causing a performance
Hi Keith,
On 7/17/2017 6:45 PM, Keith Busch wrote:
> On Mon, Jul 17, 2017 at 06:36:23PM -0400, Sinan Kaya wrote:
>> Code is moving the completion queue doorbell after processing all completed
>> events and sending callbacks to the block layer on each iteration.
>>
>> This is causing a performance
On Mon, Jul 17, 2017 at 06:36:23PM -0400, Sinan Kaya wrote:
> Code is moving the completion queue doorbell after processing all completed
> events and sending callbacks to the block layer on each iteration.
>
> This is causing a performance drop when a lot of jobs are queued towards
> the HW.
On Mon, Jul 17, 2017 at 06:36:23PM -0400, Sinan Kaya wrote:
> Code is moving the completion queue doorbell after processing all completed
> events and sending callbacks to the block layer on each iteration.
>
> This is causing a performance drop when a lot of jobs are queued towards
> the HW.
Code is moving the completion queue doorbell after processing all completed
events and sending callbacks to the block layer on each iteration.
This is causing a performance drop when a lot of jobs are queued towards
the HW. Move the completion queue doorbell on each loop instead and allow new
Code is moving the completion queue doorbell after processing all completed
events and sending callbacks to the block layer on each iteration.
This is causing a performance drop when a lot of jobs are queued towards
the HW. Move the completion queue doorbell on each loop instead and allow new
20 matches
Mail list logo