Re: [PATCH 3/4] blk-mq: establish new mapping before cpu starts handling requests

2015-06-30 Thread Ming Lei
On Mon, Jun 29, 2015 at 10:25 PM, Akinobu Mita wrote: > 2015-06-28 1:08 GMT+09:00 Ming Lei : >> On Sat, Jun 27, 2015 at 1:14 AM, Akinobu Mita wrote: >>> Akinobu Mita wrote: 2015-06-26 0:40 GMT+09:00 Ming Lei : > On Thu, 25 Jun 2015 21:49:43 +0900 > Akinobu Mita wrote: >> For

Re: [PATCH 3/4] blk-mq: establish new mapping before cpu starts handling requests

2015-06-29 Thread Akinobu Mita
2015-06-28 1:08 GMT+09:00 Ming Lei : > On Sat, Jun 27, 2015 at 1:14 AM, Akinobu Mita wrote: >> Akinobu Mita wrote: >>> 2015-06-26 0:40 GMT+09:00 Ming Lei : >>> > On Thu, 25 Jun 2015 21:49:43 +0900 >>> > Akinobu Mita wrote: >>> >> For example, there is a single hw queue (hctx) and two CPU queues

Re: [PATCH 3/4] blk-mq: establish new mapping before cpu starts handling requests

2015-06-27 Thread Ming Lei
On Sat, Jun 27, 2015 at 1:14 AM, Akinobu Mita wrote: > Akinobu Mita wrote: >> 2015-06-26 0:40 GMT+09:00 Ming Lei : >> > On Thu, 25 Jun 2015 21:49:43 +0900 >> > Akinobu Mita wrote: >> >> For example, there is a single hw queue (hctx) and two CPU queues >> >> (ctx0 for CPU0, and ctx1 for CPU1). N

Re: [PATCH 3/4] blk-mq: establish new mapping before cpu starts handling requests

2015-06-26 Thread Akinobu Mita
Akinobu Mita wrote: > 2015-06-26 0:40 GMT+09:00 Ming Lei : > > On Thu, 25 Jun 2015 21:49:43 +0900 > > Akinobu Mita wrote: > >> For example, there is a single hw queue (hctx) and two CPU queues > >> (ctx0 for CPU0, and ctx1 for CPU1). Now CPU1 is just onlined and > >> a request is inserted into c

Re: [PATCH 3/4] blk-mq: establish new mapping before cpu starts handling requests

2015-06-25 Thread Akinobu Mita
2015-06-26 0:40 GMT+09:00 Ming Lei : > On Thu, 25 Jun 2015 21:49:43 +0900 > Akinobu Mita wrote: >> For example, there is a single hw queue (hctx) and two CPU queues >> (ctx0 for CPU0, and ctx1 for CPU1). Now CPU1 is just onlined and >> a request is inserted into ctx1->rq_list and set bit0 in pend

Re: [PATCH 3/4] blk-mq: establish new mapping before cpu starts handling requests

2015-06-25 Thread Ming Lei
On Thu, 25 Jun 2015 21:49:43 +0900 Akinobu Mita wrote: > 2015-06-25 17:07 GMT+09:00 Ming Lei : > > On Thu, Jun 25, 2015 at 10:56 AM, Akinobu Mita > > wrote: > >> 2015-06-25 1:24 GMT+09:00 Ming Lei : > >>> On Wed, Jun 24, 2015 at 10:34 PM, Akinobu Mita > >>> wrote: > Hi Ming, > > >>>

Re: [PATCH 3/4] blk-mq: establish new mapping before cpu starts handling requests

2015-06-25 Thread Akinobu Mita
2015-06-25 17:07 GMT+09:00 Ming Lei : > On Thu, Jun 25, 2015 at 10:56 AM, Akinobu Mita wrote: >> 2015-06-25 1:24 GMT+09:00 Ming Lei : >>> On Wed, Jun 24, 2015 at 10:34 PM, Akinobu Mita >>> wrote: Hi Ming, 2015-06-24 18:46 GMT+09:00 Ming Lei : > On Sun, Jun 21, 2015 at 9:52 PM,

Re: [PATCH 3/4] blk-mq: establish new mapping before cpu starts handling requests

2015-06-25 Thread Ming Lei
On Thu, Jun 25, 2015 at 10:56 AM, Akinobu Mita wrote: > 2015-06-25 1:24 GMT+09:00 Ming Lei : >> On Wed, Jun 24, 2015 at 10:34 PM, Akinobu Mita >> wrote: >>> Hi Ming, >>> >>> 2015-06-24 18:46 GMT+09:00 Ming Lei : On Sun, Jun 21, 2015 at 9:52 PM, Akinobu Mita wrote: > ctx->index_hw

Re: [PATCH 3/4] blk-mq: establish new mapping before cpu starts handling requests

2015-06-24 Thread Akinobu Mita
2015-06-25 1:24 GMT+09:00 Ming Lei : > On Wed, Jun 24, 2015 at 10:34 PM, Akinobu Mita wrote: >> Hi Ming, >> >> 2015-06-24 18:46 GMT+09:00 Ming Lei : >>> On Sun, Jun 21, 2015 at 9:52 PM, Akinobu Mita >>> wrote: ctx->index_hw is zero for the CPUs which have never been onlined since the b

Re: [PATCH 3/4] blk-mq: establish new mapping before cpu starts handling requests

2015-06-24 Thread Ming Lei
On Wed, Jun 24, 2015 at 10:34 PM, Akinobu Mita wrote: > Hi Ming, > > 2015-06-24 18:46 GMT+09:00 Ming Lei : >> On Sun, Jun 21, 2015 at 9:52 PM, Akinobu Mita wrote: >>> ctx->index_hw is zero for the CPUs which have never been onlined since >>> the block queue was initialized. If one of those CPUs

Re: [PATCH 3/4] blk-mq: establish new mapping before cpu starts handling requests

2015-06-24 Thread Akinobu Mita
Hi Ming, 2015-06-24 18:46 GMT+09:00 Ming Lei : > On Sun, Jun 21, 2015 at 9:52 PM, Akinobu Mita wrote: >> ctx->index_hw is zero for the CPUs which have never been onlined since >> the block queue was initialized. If one of those CPUs is hotadded and >> starts handling request before new mappings

Re: [PATCH 3/4] blk-mq: establish new mapping before cpu starts handling requests

2015-06-24 Thread Ming Lei
On Sun, Jun 21, 2015 at 9:52 PM, Akinobu Mita wrote: > ctx->index_hw is zero for the CPUs which have never been onlined since > the block queue was initialized. If one of those CPUs is hotadded and > starts handling request before new mappings are established, pending Could you explain a bit wha