Am 17.08.2014 um 07:29 hat Paolo Bonzini geschrieben:
> Il 15/08/2014 22:15, Paolo Bonzini ha scritto:
> >> > | Random throughput | Sequential throughput
> >> > +---+---
> >> > master | 442 MB/s | 730 MB/s
> >> >
Il 15/08/2014 22:15, Paolo Bonzini ha scritto:
>> > | Random throughput | Sequential throughput
>> > +---+---
>> > master | 442 MB/s | 730 MB/s
>> > base| 453 MB/s | 757 MB/s
>> > bypass (Ming
On 8/16/14, Paolo Bonzini wrote:
> Il 14/08/2014 12:46, Kevin Wolf ha scritto:
>> So to finally reply with some numbers... I'm running fio tests based on
>> Ming's configuration on a loop-mounted tmpfs image using dataplane.
>
> I'm not sure tmpfs is a particularly useful comparison, since it does
Il 14/08/2014 11:39, Stefan Hajnoczi ha scritto:
> That begs the question whether you should look at PCI passthrough
> instead?
Being able to use logical volumes, or to access multiple remote LUNs
through a single FC card in the host is an obvious reason to avoid PCI
passthrough.
Paolo
Il 14/08/2014 12:46, Kevin Wolf ha scritto:
> So to finally reply with some numbers... I'm running fio tests based on
> Ming's configuration on a loop-mounted tmpfs image using dataplane.
I'm not sure tmpfs is a particularly useful comparison, since it doesn't
support O_DIRECT. O_DIRECT over ramd
On Thu, Aug 14, 2014 at 6:46 PM, Kevin Wolf wrote:
> Am 11.08.2014 um 21:37 hat Paolo Bonzini geschrieben:
>> Il 10/08/2014 05:46, Ming Lei ha scritto:
>> > Hi Kevin, Paolo, Stefan and all,
>> >
>> >
>> > On Wed, 6 Aug 2014 10:48:55 +0200
>> > Kevin Wolf wrote:
>> >
>> >> Am 06.08.2014 um 07:33 h
Am 11.08.2014 um 21:37 hat Paolo Bonzini geschrieben:
> Il 10/08/2014 05:46, Ming Lei ha scritto:
> > Hi Kevin, Paolo, Stefan and all,
> >
> >
> > On Wed, 6 Aug 2014 10:48:55 +0200
> > Kevin Wolf wrote:
> >
> >> Am 06.08.2014 um 07:33 hat Ming Lei geschrieben:
> >
> >>
> >> Anyhow, the corouti
On Thu, Aug 14, 2014 at 5:39 PM, Stefan Hajnoczi wrote:
> On Wed, Aug 13, 2014 at 09:49:23PM +0800, Ming Lei wrote:
>> On Wed, Aug 13, 2014 at 9:16 PM, Paolo Bonzini wrote:
>> > Il 13/08/2014 11:54, Kevin Wolf ha scritto:
>> >> Am 12.08.2014 um 21:08 hat Paolo Bonzini geschrieben:
>> >>> Il 12/08
On Wed, Aug 13, 2014 at 09:49:23PM +0800, Ming Lei wrote:
> On Wed, Aug 13, 2014 at 9:16 PM, Paolo Bonzini wrote:
> > Il 13/08/2014 11:54, Kevin Wolf ha scritto:
> >> Am 12.08.2014 um 21:08 hat Paolo Bonzini geschrieben:
> >>> Il 12/08/2014 10:12, Ming Lei ha scritto:
> >> The below patch is b
On Wed, Aug 13, 2014 at 9:16 PM, Paolo Bonzini wrote:
> Il 13/08/2014 11:54, Kevin Wolf ha scritto:
>> Am 12.08.2014 um 21:08 hat Paolo Bonzini geschrieben:
>>> Il 12/08/2014 10:12, Ming Lei ha scritto:
>> The below patch is basically the minimal change to bypass coroutines.
>> Of course
Il 13/08/2014 11:54, Kevin Wolf ha scritto:
> Am 12.08.2014 um 21:08 hat Paolo Bonzini geschrieben:
>> Il 12/08/2014 10:12, Ming Lei ha scritto:
> The below patch is basically the minimal change to bypass coroutines. Of
> course
> the block.c part is not acceptable as is (the change t
On Wed, Aug 13, 2014 at 8:35 PM, Paolo Bonzini wrote:
> Il 13/08/2014 13:43, Ming Lei ha scritto:
>>> > The below patch is basically the minimal change to bypass coroutines. Of
>>> > course
>>> > the block.c part is not acceptable as is (the change to
>>> > refresh_total_sectors
>>> > is broken
Il 13/08/2014 13:43, Ming Lei ha scritto:
>> > The below patch is basically the minimal change to bypass coroutines. Of
>> > course
>> > the block.c part is not acceptable as is (the change to
>> > refresh_total_sectors
>> > is broken, the others are just ugly), but it is a start. Please run it
Il 13/08/2014 12:19, Ming Lei ha scritto:
>> > The problem is that your patches to do touch too much code and subtly
>> > break too much stuff. The one I wrote does have a little breakage
> Could you give a hint about which stuff are broken? Last time, you mention
> virtio-scsi need to keep AIOCB
Hi Paolo,
On Tue, Aug 12, 2014 at 3:37 AM, Paolo Bonzini wrote:
> Il 10/08/2014 05:46, Ming Lei ha scritto:
>> Hi Kevin, Paolo, Stefan and all,
>>
>>
>> On Wed, 6 Aug 2014 10:48:55 +0200
>> Kevin Wolf wrote:
>>
>>> Am 06.08.2014 um 07:33 hat Ming Lei geschrieben:
>>
>>>
>>> Anyhow, the coroutine
On Wed, Aug 13, 2014 at 3:08 AM, Paolo Bonzini wrote:
> Il 12/08/2014 10:12, Ming Lei ha scritto:
>>> > The below patch is basically the minimal change to bypass coroutines. Of
>>> > course
>>> > the block.c part is not acceptable as is (the change to
>>> > refresh_total_sectors
>>> > is broken
Am 12.08.2014 um 21:08 hat Paolo Bonzini geschrieben:
> Il 12/08/2014 10:12, Ming Lei ha scritto:
> >> > The below patch is basically the minimal change to bypass coroutines.
> >> > Of course
> >> > the block.c part is not acceptable as is (the change to
> >> > refresh_total_sectors
> >> > is br
On Mon, Aug 11, 2014 at 09:37:01PM +0200, Paolo Bonzini wrote:
> Il 10/08/2014 05:46, Ming Lei ha scritto:
> @@ -4356,6 +4353,20 @@ BlockDriverAIOCB *bdrv_aio_readv(BlockDriverState *bs,
> int64_t sector_num,
> {
> trace_bdrv_aio_readv(bs, sector_num, nb_sectors, opaque);
>
> +if (bs->
Il 12/08/2014 10:12, Ming Lei ha scritto:
>> > The below patch is basically the minimal change to bypass coroutines. Of
>> > course
>> > the block.c part is not acceptable as is (the change to
>> > refresh_total_sectors
>> > is broken, the others are just ugly), but it is a start. Please run it
On Tue, Aug 12, 2014 at 7:40 PM, Kevin Wolf wrote:
> Am 12.08.2014 um 09:53 hat Ming Lei geschrieben:
>> On Mon, Aug 11, 2014 at 10:03 PM, Kevin Wolf wrote:
>> > Am 10.08.2014 um 05:46 hat Ming Lei geschrieben:
>> >> Hi Kevin, Paolo, Stefan and all,
>> >>
>> >>
>> >> On Wed, 6 Aug 2014 10:48:55 +
Am 12.08.2014 um 09:53 hat Ming Lei geschrieben:
> On Mon, Aug 11, 2014 at 10:03 PM, Kevin Wolf wrote:
> > Am 10.08.2014 um 05:46 hat Ming Lei geschrieben:
> >> Hi Kevin, Paolo, Stefan and all,
> >>
> >>
> >> On Wed, 6 Aug 2014 10:48:55 +0200
> >> Kevin Wolf wrote:
> >>
> >> > Am 06.08.2014 um 07
On Tue, Aug 12, 2014 at 3:37 AM, Paolo Bonzini wrote:
> Il 10/08/2014 05:46, Ming Lei ha scritto:
>> Hi Kevin, Paolo, Stefan and all,
>>
>>
>> On Wed, 6 Aug 2014 10:48:55 +0200
>> Kevin Wolf wrote:
>>
>>> Am 06.08.2014 um 07:33 hat Ming Lei geschrieben:
>>
>>>
>>> Anyhow, the coroutine version of
On Mon, Aug 11, 2014 at 10:03 PM, Kevin Wolf wrote:
> Am 10.08.2014 um 05:46 hat Ming Lei geschrieben:
>> Hi Kevin, Paolo, Stefan and all,
>>
>>
>> On Wed, 6 Aug 2014 10:48:55 +0200
>> Kevin Wolf wrote:
>>
>> > Am 06.08.2014 um 07:33 hat Ming Lei geschrieben:
>>
>> >
>> > Anyhow, the coroutine ve
Il 10/08/2014 05:46, Ming Lei ha scritto:
> Hi Kevin, Paolo, Stefan and all,
>
>
> On Wed, 6 Aug 2014 10:48:55 +0200
> Kevin Wolf wrote:
>
>> Am 06.08.2014 um 07:33 hat Ming Lei geschrieben:
>
>>
>> Anyhow, the coroutine version of your benchmark is buggy, it leaks all
>> coroutines instead of
Am 10.08.2014 um 05:46 hat Ming Lei geschrieben:
> Hi Kevin, Paolo, Stefan and all,
>
>
> On Wed, 6 Aug 2014 10:48:55 +0200
> Kevin Wolf wrote:
>
> > Am 06.08.2014 um 07:33 hat Ming Lei geschrieben:
>
> >
> > Anyhow, the coroutine version of your benchmark is buggy, it leaks all
> > coroutine
Hi Kevin, Paolo, Stefan and all,
On Wed, 6 Aug 2014 10:48:55 +0200
Kevin Wolf wrote:
> Am 06.08.2014 um 07:33 hat Ming Lei geschrieben:
>
> Anyhow, the coroutine version of your benchmark is buggy, it leaks all
> coroutines instead of exiting them, so it can't make any use of the
> coroutine
On Fri, Aug 8, 2014 at 6:32 PM, Ming Lei wrote:
> On Thu, Aug 7, 2014 at 9:51 PM, Kevin Wolf wrote:
>> Am 07.08.2014 um 12:27 hat Ming Lei geschrieben:
>>> On Wed, Aug 6, 2014 at 11:40 PM, Kevin Wolf wrote:
>>> > Am 06.08.2014 um 13:28 hat Ming Lei geschrieben:
>>> >> On Wed, Aug 6, 2014 at 6:09
On Thu, Aug 7, 2014 at 9:51 PM, Kevin Wolf wrote:
> Am 07.08.2014 um 12:27 hat Ming Lei geschrieben:
>> On Wed, Aug 6, 2014 at 11:40 PM, Kevin Wolf wrote:
>> > Am 06.08.2014 um 13:28 hat Ming Lei geschrieben:
>> >> On Wed, Aug 6, 2014 at 6:09 PM, Kevin Wolf wrote:
>> >> > Am 06.08.2014 um 11:37
Am 07.08.2014 um 12:27 hat Ming Lei geschrieben:
> On Wed, Aug 6, 2014 at 11:40 PM, Kevin Wolf wrote:
> > Am 06.08.2014 um 13:28 hat Ming Lei geschrieben:
> >> On Wed, Aug 6, 2014 at 6:09 PM, Kevin Wolf wrote:
> >> > Am 06.08.2014 um 11:37 hat Ming Lei geschrieben:
> >> >> On Wed, Aug 6, 2014 at
On Thu, Aug 7, 2014 at 7:06 PM, Kevin Wolf wrote:
> Am 07.08.2014 um 12:52 hat Ming Lei geschrieben:
>> On Thu, Aug 7, 2014 at 6:27 PM, Ming Lei wrote:
>> > On Wed, Aug 6, 2014 at 11:40 PM, Kevin Wolf wrote:
>>
>> > Also there are some problems with your patches which can't boot a
>> > VM in my
Am 07.08.2014 um 12:52 hat Ming Lei geschrieben:
> On Thu, Aug 7, 2014 at 6:27 PM, Ming Lei wrote:
> > On Wed, Aug 6, 2014 at 11:40 PM, Kevin Wolf wrote:
>
> > Also there are some problems with your patches which can't boot a
> > VM in my environment:
> >
> > - __thread patch: looks there is no
On Thu, Aug 7, 2014 at 6:27 PM, Ming Lei wrote:
> On Wed, Aug 6, 2014 at 11:40 PM, Kevin Wolf wrote:
> Also there are some problems with your patches which can't boot a
> VM in my environment:
>
> - __thread patch: looks there is no '__thread' used, and the patch
> basically makes bypass not wor
On Wed, Aug 6, 2014 at 11:40 PM, Kevin Wolf wrote:
> Am 06.08.2014 um 13:28 hat Ming Lei geschrieben:
>> On Wed, Aug 6, 2014 at 6:09 PM, Kevin Wolf wrote:
>> > Am 06.08.2014 um 11:37 hat Ming Lei geschrieben:
>> >> On Wed, Aug 6, 2014 at 4:48 PM, Kevin Wolf wrote:
>> >> > Am 06.08.2014 um 07:33
Am 06.08.2014 um 13:28 hat Ming Lei geschrieben:
> On Wed, Aug 6, 2014 at 6:09 PM, Kevin Wolf wrote:
> > Am 06.08.2014 um 11:37 hat Ming Lei geschrieben:
> >> On Wed, Aug 6, 2014 at 4:48 PM, Kevin Wolf wrote:
> >> > Am 06.08.2014 um 07:33 hat Ming Lei geschrieben:
> >> >> Hi Kevin,
> >> >>
> >> >
On Wed, Aug 6, 2014 at 4:50 PM, Paolo Bonzini wrote:
> Il 06/08/2014 10:38, Ming Lei ha scritto:
>> On Wed, Aug 6, 2014 at 3:45 PM, Paolo Bonzini wrote:
>>> Il 06/08/2014 07:33, Ming Lei ha scritto:
>> I played a bit with the following, I hope it's not too naive. I couldn't
>> see a diffe
On Wed, Aug 6, 2014 at 7:28 PM, Ming Lei wrote:
> On Wed, Aug 6, 2014 at 6:09 PM, Kevin Wolf wrote:
>
> I use the /dev/nullb0 block device to test, which is available in linux kernel
> 3.13+, and follows the difference, which looks not very big(< 10%):
>
> And I added two parameter to your img-b
On Wed, Aug 6, 2014 at 6:09 PM, Kevin Wolf wrote:
> Am 06.08.2014 um 11:37 hat Ming Lei geschrieben:
>> On Wed, Aug 6, 2014 at 4:48 PM, Kevin Wolf wrote:
>> > Am 06.08.2014 um 07:33 hat Ming Lei geschrieben:
>> >> Hi Kevin,
>> >>
>> >> On Tue, Aug 5, 2014 at 10:47 PM, Kevin Wolf wrote:
>> >> > A
Am 06.08.2014 um 11:37 hat Ming Lei geschrieben:
> On Wed, Aug 6, 2014 at 4:48 PM, Kevin Wolf wrote:
> > Am 06.08.2014 um 07:33 hat Ming Lei geschrieben:
> >> Hi Kevin,
> >>
> >> On Tue, Aug 5, 2014 at 10:47 PM, Kevin Wolf wrote:
> >> > Am 05.08.2014 um 15:48 hat Stefan Hajnoczi geschrieben:
> >>
On Wed, Aug 06, 2014 at 01:33:36PM +0800, Ming Lei wrote:
> With coroutine, the running time for same loading is increased
> ~50%(1.325s vs. 0.903s), and dcache load events is increased
I agree with Paolo about microbenchmarks. We need to do I/O to get a
realistic picture of performance, since th
On Wed, Aug 6, 2014 at 4:48 PM, Kevin Wolf wrote:
> Am 06.08.2014 um 07:33 hat Ming Lei geschrieben:
>> Hi Kevin,
>>
>> On Tue, Aug 5, 2014 at 10:47 PM, Kevin Wolf wrote:
>> > Am 05.08.2014 um 15:48 hat Stefan Hajnoczi geschrieben:
>> >> I have been wondering how to prove that the root cause is t
Il 06/08/2014 10:38, Ming Lei ha scritto:
> On Wed, Aug 6, 2014 at 3:45 PM, Paolo Bonzini wrote:
>> Il 06/08/2014 07:33, Ming Lei ha scritto:
> I played a bit with the following, I hope it's not too naive. I couldn't
> see a difference with your patches, but at least one reason for this is
Am 06.08.2014 um 07:33 hat Ming Lei geschrieben:
> Hi Kevin,
>
> On Tue, Aug 5, 2014 at 10:47 PM, Kevin Wolf wrote:
> > Am 05.08.2014 um 15:48 hat Stefan Hajnoczi geschrieben:
> >> I have been wondering how to prove that the root cause is the ucontext
> >> coroutine mechanism (stack switching).
On Wed, Aug 6, 2014 at 3:45 PM, Paolo Bonzini wrote:
> Il 06/08/2014 07:33, Ming Lei ha scritto:
>>> > I played a bit with the following, I hope it's not too naive. I couldn't
>>> > see a difference with your patches, but at least one reason for this is
>>> > probably that my laptop SSD isn't fast
Il 06/08/2014 07:33, Ming Lei ha scritto:
>> > I played a bit with the following, I hope it's not too naive. I couldn't
>> > see a difference with your patches, but at least one reason for this is
>> > probably that my laptop SSD isn't fast enough to make the CPU the
>> > bottleneck. Haven't tried
Hi Kevin,
On Tue, Aug 5, 2014 at 10:47 PM, Kevin Wolf wrote:
> Am 05.08.2014 um 15:48 hat Stefan Hajnoczi geschrieben:
>> On Tue, Aug 05, 2014 at 06:00:22PM +0800, Ming Lei wrote:
>> > On Tue, Aug 5, 2014 at 5:48 PM, Kevin Wolf wrote:
>> > > Am 05.08.2014 um 05:33 hat Ming Lei geschrieben:
>> >
Am 05.08.2014 um 15:48 hat Stefan Hajnoczi geschrieben:
> On Tue, Aug 05, 2014 at 06:00:22PM +0800, Ming Lei wrote:
> > On Tue, Aug 5, 2014 at 5:48 PM, Kevin Wolf wrote:
> > > Am 05.08.2014 um 05:33 hat Ming Lei geschrieben:
> > >> Hi,
> > >>
> > >> These patches bring up below 4 changes:
> > >>
On Tue, Aug 05, 2014 at 05:50:42PM +0800, Ming Lei wrote:
> On Tue, Aug 5, 2014 at 5:38 PM, Stefan Hajnoczi wrote:
> > On Tue, Aug 05, 2014 at 11:33:01AM +0800, Ming Lei wrote:
> >> These patches bring up below 4 changes:
> >> - introduce object allocation pool and apply it to
> >>
On Tue, Aug 05, 2014 at 06:00:22PM +0800, Ming Lei wrote:
> On Tue, Aug 5, 2014 at 5:48 PM, Kevin Wolf wrote:
> > Am 05.08.2014 um 05:33 hat Ming Lei geschrieben:
> >> Hi,
> >>
> >> These patches bring up below 4 changes:
> >> - introduce object allocation pool and apply it to
> >>
Il 05/08/2014 12:00, Ming Lei ha scritto:
>> >
>> > I also think that the device emulation has no business in deciding
>> > whether the bypass is used (it depends solely on conditions outside of
>> > the device) and that leaking the fd number out of raw-posix is wrong.
>> > Both of them are layerin
On Tue, Aug 5, 2014 at 5:56 PM, Kevin Wolf wrote:
> Am 05.08.2014 um 11:50 hat Ming Lei geschrieben:
>> On Tue, Aug 5, 2014 at 5:38 PM, Stefan Hajnoczi wrote:
>> > On Tue, Aug 05, 2014 at 11:33:01AM +0800, Ming Lei wrote:
>> >> These patches bring up below 4 changes:
>> >> - introduce obj
On Tue, Aug 5, 2014 at 5:48 PM, Kevin Wolf wrote:
> Am 05.08.2014 um 05:33 hat Ming Lei geschrieben:
>> Hi,
>>
>> These patches bring up below 4 changes:
>> - introduce object allocation pool and apply it to
>> virtio-blk dataplane for improving its performance
>>
>> - intr
Am 05.08.2014 um 11:50 hat Ming Lei geschrieben:
> On Tue, Aug 5, 2014 at 5:38 PM, Stefan Hajnoczi wrote:
> > On Tue, Aug 05, 2014 at 11:33:01AM +0800, Ming Lei wrote:
> >> These patches bring up below 4 changes:
> >> - introduce object allocation pool and apply it to
> >> virtio-b
On Tue, Aug 5, 2014 at 5:38 PM, Stefan Hajnoczi wrote:
> On Tue, Aug 05, 2014 at 11:33:01AM +0800, Ming Lei wrote:
>> These patches bring up below 4 changes:
>> - introduce object allocation pool and apply it to
>> virtio-blk dataplane for improving its performance
>>
>> -
Am 05.08.2014 um 05:33 hat Ming Lei geschrieben:
> Hi,
>
> These patches bring up below 4 changes:
> - introduce object allocation pool and apply it to
> virtio-blk dataplane for improving its performance
>
> - introduce selective coroutine bypass mechanism
> for i
On Tue, Aug 05, 2014 at 11:33:01AM +0800, Ming Lei wrote:
> These patches bring up below 4 changes:
> - introduce object allocation pool and apply it to
> virtio-blk dataplane for improving its performance
>
> - introduce selective coroutine bypass mechanism
> for i
Hi,
These patches bring up below 4 changes:
- introduce object allocation pool and apply it to
virtio-blk dataplane for improving its performance
- introduce selective coroutine bypass mechanism
for improving performance of virtio-blk dataplane with
raw for
56 matches
Mail list logo
