Hi,Jeff
On Friday, October 23, 2015 03:04 AM, Jeff Moyer wrote:
Jens Axboe writes:
On 10/22/2015 09:53 AM, Jeff Moyer wrote:
I think that percolating BLK_MQ_F_TAG_SHARED up to the tag set would
allow newly created hctxs to simply inherit the shared state (in
blk_mq_init_hctx), and you won't
Hi,Jeff
On Friday, October 23, 2015 03:04 AM, Jeff Moyer wrote:
Jens Axboe writes:
On 10/22/2015 09:53 AM, Jeff Moyer wrote:
I think that percolating BLK_MQ_F_TAG_SHARED up to the tag set would
allow newly created hctxs to simply inherit the shared state (in
On Thu, Oct 22, 2015 at 5:15 PM, jason wrote:
>
>
> On Thursday, October 22, 2015 04:47 PM, Tejun Heo wrote:
>>
>> Hello,
>>
>> On Mon, Oct 19, 2015 at 07:40:13AM -0700, Zhangqing Luo wrote:
>>
>> > So every time blk_mq_freeze_queue_start, it runs in this way
>> >
>> >
Jens Axboe writes:
> On 10/22/2015 09:53 AM, Jeff Moyer wrote:
>> I think that percolating BLK_MQ_F_TAG_SHARED up to the tag set would
>> allow newly created hctxs to simply inherit the shared state (in
>> blk_mq_init_hctx), and you won't need to freeze every queue in order to
>> guarantee that.
On 10/22/2015 09:53 AM, Jeff Moyer wrote:
Jens Axboe writes:
I agree with the optimizing hot paths by cheaper percpu operation,
but how much does it affect the performance?
A lot, since the queue referencing happens twice per IO. The switch to
percpu was done to use shared/common code for
Jens Axboe writes:
>> I agree with the optimizing hot paths by cheaper percpu operation,
>> but how much does it affect the performance?
>
> A lot, since the queue referencing happens twice per IO. The switch to
> percpu was done to use shared/common code for this, the previous
> version was a
On 10/22/2015 03:15 AM, jason wrote:
On Thursday, October 22, 2015 04:47 PM, Tejun Heo wrote:
Hello,
On Mon, Oct 19, 2015 at 07:40:13AM -0700, Zhangqing Luo wrote:
> So every time blk_mq_freeze_queue_start, it runs in this way
>
> blk_mq_freeze_queue_start
>
On Thursday, October 22, 2015 04:47 PM, Tejun Heo wrote:
Hello,
On Mon, Oct 19, 2015 at 07:40:13AM -0700, Zhangqing Luo wrote:
> So every time blk_mq_freeze_queue_start, it runs in this way
>
> blk_mq_freeze_queue_start
> ->percpu_ref_kill->percpu_ref_kill_and_confirm
>
Hello,
On Mon, Oct 19, 2015 at 07:40:13AM -0700, Zhangqing Luo wrote:
> So every time blk_mq_freeze_queue_start, it runs in this way
>
> blk_mq_freeze_queue_start
> ->percpu_ref_kill->percpu_ref_kill_and_confirm
> ->__percpu_ref_switch_to_atomic
>
Hello,
On Mon, Oct 19, 2015 at 07:40:13AM -0700, Zhangqing Luo wrote:
> So every time blk_mq_freeze_queue_start, it runs in this way
>
> blk_mq_freeze_queue_start
> ->percpu_ref_kill->percpu_ref_kill_and_confirm
> ->__percpu_ref_switch_to_atomic
>
Jens Axboe writes:
> On 10/22/2015 09:53 AM, Jeff Moyer wrote:
>> I think that percolating BLK_MQ_F_TAG_SHARED up to the tag set would
>> allow newly created hctxs to simply inherit the shared state (in
>> blk_mq_init_hctx), and you won't need to freeze every queue in order to
On Thu, Oct 22, 2015 at 5:15 PM, jason wrote:
>
>
> On Thursday, October 22, 2015 04:47 PM, Tejun Heo wrote:
>>
>> Hello,
>>
>> On Mon, Oct 19, 2015 at 07:40:13AM -0700, Zhangqing Luo wrote:
>>
>> > So every time blk_mq_freeze_queue_start, it runs in this way
>> >
On Thursday, October 22, 2015 04:47 PM, Tejun Heo wrote:
Hello,
On Mon, Oct 19, 2015 at 07:40:13AM -0700, Zhangqing Luo wrote:
> So every time blk_mq_freeze_queue_start, it runs in this way
>
> blk_mq_freeze_queue_start
> ->percpu_ref_kill->percpu_ref_kill_and_confirm
>
On 10/22/2015 03:15 AM, jason wrote:
On Thursday, October 22, 2015 04:47 PM, Tejun Heo wrote:
Hello,
On Mon, Oct 19, 2015 at 07:40:13AM -0700, Zhangqing Luo wrote:
> So every time blk_mq_freeze_queue_start, it runs in this way
>
> blk_mq_freeze_queue_start
>
Jens Axboe writes:
>> I agree with the optimizing hot paths by cheaper percpu operation,
>> but how much does it affect the performance?
>
> A lot, since the queue referencing happens twice per IO. The switch to
> percpu was done to use shared/common code for this, the previous
On 10/22/2015 09:53 AM, Jeff Moyer wrote:
Jens Axboe writes:
I agree with the optimizing hot paths by cheaper percpu operation,
but how much does it affect the performance?
A lot, since the queue referencing happens twice per IO. The switch to
percpu was done to use
Hi, Jens,Tejun
Current problem we meet:
When Multiple Queue is used for scsi, the period between each LUN probe
is increasing as the number of block request queue goes up, eventually
it takes hours for Initiator to finish scanning thousands of LUNs.
Kernel version we're using: 4.1.6-14
I
Hi, Jens,Tejun
Current problem we meet:
When Multiple Queue is used for scsi, the period between each LUN probe
is increasing as the number of block request queue goes up, eventually
it takes hours for Initiator to finish scanning thousands of LUNs.
Kernel version we're using: 4.1.6-14
I
18 matches
Mail list logo