Re: [Qemu-block] [PATCH v4 8/8] linux-aio: share one LinuxAioState within an AioContext

2016-05-11 Thread Stefan Hajnoczi
On Tue, May 10, 2016 at 11:40:33AM +0200, Kevin Wolf wrote: > Am 10.05.2016 um 11:30 hat Stefan Hajnoczi geschrieben: > > On Mon, May 09, 2016 at 06:31:44PM +0200, Paolo Bonzini wrote: > > > On 19/04/2016 11:09, Stefan Hajnoczi wrote: > > > >> > This has better performance because it executes

Re: [Qemu-block] [PATCH v4 8/8] linux-aio: share one LinuxAioState within an AioContext

2016-05-11 Thread Stefan Hajnoczi
On Tue, May 10, 2016 at 12:32:27PM +0200, Paolo Bonzini wrote: > On 10/05/2016 11:40, Kevin Wolf wrote: > > > Regarding performance, I'm thinking about a guest with 8 disks (queue > > > depth 32). The worst case is when the guest submits 32 requests at once > > > but the Linux AIO event limit has

Re: [Qemu-block] [PATCH v4 8/8] linux-aio: share one LinuxAioState within an AioContext

2016-05-11 Thread Stefan Hajnoczi
On Thu, Apr 07, 2016 at 06:33:36PM +0200, Paolo Bonzini wrote: > This has better performance because it executes fewer system calls > and does not use a bottom half per disk. > > Originally proposed by Ming Lei. > > Signed-off-by: Paolo Bonzini > --- > async.c

Re: [Qemu-block] [PATCH v4 8/8] linux-aio: share one LinuxAioState within an AioContext

2016-05-10 Thread Paolo Bonzini
On 10/05/2016 11:40, Kevin Wolf wrote: > > Regarding performance, I'm thinking about a guest with 8 disks (queue > > depth 32). The worst case is when the guest submits 32 requests at once > > but the Linux AIO event limit has already been reached. Then the disk > > is starved until other

Re: [Qemu-block] [PATCH v4 8/8] linux-aio: share one LinuxAioState within an AioContext

2016-05-10 Thread Kevin Wolf
Am 10.05.2016 um 11:30 hat Stefan Hajnoczi geschrieben: > On Mon, May 09, 2016 at 06:31:44PM +0200, Paolo Bonzini wrote: > > On 19/04/2016 11:09, Stefan Hajnoczi wrote: > > >> > This has better performance because it executes fewer system calls > > >> > and does not use a bottom half per disk. > >

Re: [Qemu-block] [PATCH v4 8/8] linux-aio: share one LinuxAioState within an AioContext

2016-05-10 Thread Stefan Hajnoczi
On Mon, May 09, 2016 at 06:31:44PM +0200, Paolo Bonzini wrote: > On 19/04/2016 11:09, Stefan Hajnoczi wrote: > >> > This has better performance because it executes fewer system calls > >> > and does not use a bottom half per disk. > > Each aio_context_t is initialized for 128 in-flight requests in

Re: [Qemu-block] [PATCH v4 8/8] linux-aio: share one LinuxAioState within an AioContext

2016-05-09 Thread Paolo Bonzini
On 19/04/2016 11:09, Stefan Hajnoczi wrote: >> > This has better performance because it executes fewer system calls >> > and does not use a bottom half per disk. > Each aio_context_t is initialized for 128 in-flight requests in > laio_init(). > > Will it be possible to hit the limit now that

Re: [Qemu-block] [PATCH v4 8/8] linux-aio: share one LinuxAioState within an AioContext

2016-04-19 Thread Stefan Hajnoczi
On Thu, Apr 07, 2016 at 06:33:36PM +0200, Paolo Bonzini wrote: > This has better performance because it executes fewer system calls > and does not use a bottom half per disk. Each aio_context_t is initialized for 128 in-flight requests in laio_init(). Will it be possible to hit the limit now

[Qemu-block] [PATCH v4 8/8] linux-aio: share one LinuxAioState within an AioContext

2016-04-07 Thread Paolo Bonzini
This has better performance because it executes fewer system calls and does not use a bottom half per disk. Originally proposed by Ming Lei. Signed-off-by: Paolo Bonzini --- async.c| 23 +++ block/linux-aio.c | 3 +