Hello,
qcow2-rs[1] is one pure Rust library for reading/writing qcow2 image, it is
based on rsd's[2] internal qcow2 implementation, but with lots of change, so
far:
- supports read/write on data file, backing file and compressed image
- block device like interface, minimized read/write unit is
On Wed, Oct 12, 2022 at 10:15:28AM -0400, Stefan Hajnoczi wrote:
> On Thu, 6 Oct 2022 at 06:14, Richard W.M. Jones wrote:
> >
> > On Tue, Oct 04, 2022 at 09:53:32AM -0400, Stefan Hajnoczi wrote:
> > > qemu-nbd doesn't use io_uring to handle the backend IO,
> >
> > Would this be fixed by your (not
On Fri, Oct 07, 2022 at 07:21:51PM +0800, Yongji Xie wrote:
> On Fri, Oct 7, 2022 at 6:51 PM Ming Lei wrote:
> >
> > On Fri, Oct 07, 2022 at 06:04:29PM +0800, Yongji Xie wrote:
> > > On Thu, Oct 6, 2022 at 7:24 PM Ming Lei wrote:
> > > >
> > > >
On Thu, Oct 06, 2022 at 02:29:55PM -0400, Stefan Hajnoczi wrote:
> On Thu, Oct 06, 2022 at 11:09:48PM +0800, Ming Lei wrote:
> > On Thu, Oct 06, 2022 at 09:59:40AM -0400, Stefan Hajnoczi wrote:
> > > On Thu, Oct 06, 2022 at 06:26:15PM +0800, Ming Lei wrote:
> > > >
On Fri, Oct 07, 2022 at 06:04:29PM +0800, Yongji Xie wrote:
> On Thu, Oct 6, 2022 at 7:24 PM Ming Lei wrote:
> >
> > On Wed, Oct 05, 2022 at 08:21:45AM -0400, Stefan Hajnoczi wrote:
> > > On Wed, 5 Oct 2022 at 00:19, Ming Lei wrote:
> > > >
> > &g
On Thu, Oct 06, 2022 at 09:59:40AM -0400, Stefan Hajnoczi wrote:
> On Thu, Oct 06, 2022 at 06:26:15PM +0800, Ming Lei wrote:
> > On Wed, Oct 05, 2022 at 11:11:32AM -0400, Stefan Hajnoczi wrote:
> > > On Tue, Oct 04, 2022 at 01:57:50AM +0200, Denis V. Lunev wrote:
> > >
On Wed, Oct 05, 2022 at 08:21:45AM -0400, Stefan Hajnoczi wrote:
> On Wed, 5 Oct 2022 at 00:19, Ming Lei wrote:
> >
> > On Tue, Oct 04, 2022 at 09:53:32AM -0400, Stefan Hajnoczi wrote:
> > > On Tue, 4 Oct 2022 at 05:44, Ming Lei wrote:
> > > >
> > &g
On Wed, Oct 05, 2022 at 11:11:32AM -0400, Stefan Hajnoczi wrote:
> On Tue, Oct 04, 2022 at 01:57:50AM +0200, Denis V. Lunev wrote:
> > On 10/3/22 21:53, Stefan Hajnoczi wrote:
> > > On Fri, Sep 30, 2022 at 05:24:11PM +0800, Ming Lei wrote:
> > > > ublk-qcow2 is ava
On Tue, Oct 04, 2022 at 09:53:32AM -0400, Stefan Hajnoczi wrote:
> On Tue, 4 Oct 2022 at 05:44, Ming Lei wrote:
> >
> > On Mon, Oct 03, 2022 at 03:53:41PM -0400, Stefan Hajnoczi wrote:
> > > On Fri, Sep 30, 2022 at 05:24:11PM +0800, Ming Lei wrote:
> > &
On Mon, Oct 03, 2022 at 03:53:41PM -0400, Stefan Hajnoczi wrote:
> On Fri, Sep 30, 2022 at 05:24:11PM +0800, Ming Lei wrote:
> > ublk-qcow2 is available now.
>
> Cool, thanks for sharing!
>
> >
> > So far it provides basic read/write function, and compression and
+
> > +# This test only runs on Linux hosts with raw image files.
> > +_supported_fmt raw
> > +_supported_proto file
> > +_supported_os Linux
> > +
> > +QEMU_IO="build/qemu-io"
> > +IMG="--image-opts driver=zoned_host_device,filename=/dev/nullb0"
> > +
e synchronize_rcu() from blk_mq_del_queue_tag_set()
5815839b3ca1 blk-mq: introduce new lock for protecting hctx->dispatch_wait
2278d69f030f blk-mq: don't pass **hctx to blk_mq_mark_tag_wait()
8ab6bb9ee8d0 blk-mq: cleanup blk_mq_get_driver_tag()
Thanks,
Ming Lei
On Wed, Dec 10, 2014 at 11:02 PM, Paolo Bonzini wrote:
>
>
> On 10/12/2014 15:35, Ming Lei wrote:
>>>> It is _not_ never happen at all, and easy to be triggered when using
>>>> mkfs.
>>>
>>> mkfs is not something to optimize for, it's just s
On Wed, Dec 10, 2014 at 8:55 PM, Paolo Bonzini wrote:
>
>
> On 10/12/2014 13:23, Ming Lei wrote:
>>> >
>>> > Again, we're talking of 2GB and this is something that should never
>>> > happen in practice.
>> Again, the 2GB limit can be av
On Wed, Dec 10, 2014 at 5:56 PM, Paolo Bonzini wrote:
>
>
> On 10/12/2014 02:41, Ming Lei wrote:
>> On Wed, Dec 10, 2014 at 1:45 AM, Paolo Bonzini wrote:
>>>
>>>
>>> On 08/12/2014 08:19, Ming Lei wrote:
>>>>>>
>>>>>
On Wed, Dec 10, 2014 at 1:45 AM, Paolo Bonzini wrote:
>
>
> On 08/12/2014 08:19, Ming Lei wrote:
>>> >
>>> > Alternatively, I'd accept a SCSI patch setting max_ws_blocks and friends
>>> > to 2GB - 1 block.
>> It should be better to not intr
Currently block can not handle big write well when write
size is bigger than INT_MAX, so split the write zero into
smaller size of chunks to meet block's requirement.
This patch fixes one WRITE SAME 16 failure in linux VM side.
Cc: Max Reitz
Signed-off-by: Ming Lei
---
hw/scsi/scsi-d
On Sat, Dec 6, 2014 at 12:33 AM, Paolo Bonzini wrote:
>
>
> On 05/12/2014 17:15, Ming Lei wrote:
>> From: Ming Lei
>>
>> QEMU block should have supported to read/write at most
>> 0x7f * 512 bytes, unfortunately INT_MAX is used to check
>>
From: Ming Lei
QEMU block should have supported to read/write at most
0x7f * 512 bytes, unfortunately INT_MAX is used to check
bytes in both bdrv_co_do_writev() and bdrv_check_byte_request(),
so cause write failure if nr_sectors is equal or more
than 0x40.
There are still other INT_MAX
- host driver may prefer WRITE/WRITE 16, then try to parse
"max write same length"
Signed-off-by: Ming Lei
---
hw/scsi/scsi-disk.c | 14 ++
1 file changed, 14 insertions(+)
diff --git a/hw/scsi/scsi-disk.c b/hw/scsi/scsi-disk.c
index 2f75d7d..b15bf4f 100644
--- a/
On Thu, Dec 4, 2014 at 11:39 PM, Kevin Wolf wrote:
> Am 04.12.2014 um 16:22 hat Ming Lei geschrieben:
>> On Thu, Dec 4, 2014 at 10:37 PM, Kevin Wolf wrote:
>> > Am 26.11.2014 um 15:46 hat Kevin Wolf geschrieben:
>> >> When getting an error while submitting r
On Thu, Dec 4, 2014 at 10:37 PM, Kevin Wolf wrote:
> Am 26.11.2014 um 15:46 hat Kevin Wolf geschrieben:
>> When getting an error while submitting requests, we must be careful to
>> wake up only inactive coroutines. Therefore we must special-case the
>> currently active coroutine and communicate an
On Mon, Dec 1, 2014 at 8:41 PM, Paolo Bonzini wrote:
>
>
> On 01/12/2014 02:28, Ming Lei wrote:
>>> > - (unsigned long)(10 * duration) / maxcycles);
>>> > + (unsigned long)(10.0 * duration / maxcycles));
>&g
viewed-by: Stefan Hajnoczi
Reviewed-by: Paolo Bonzini
Suggested-by: Paolo Bonzini
Signed-off-by: Ming Lei
---
block/linux-aio.c | 32
1 file changed, 24 insertions(+), 8 deletions(-)
diff --git a/block/linux-aio.c b/block/linux-aio.c
index 53c5616..9403b17 1006
No one uses the 'node' field any more, so remove it
from 'struct qemu_laiocb', and this can save 16byte
for the struct on 64bit arch.
Reviewed-by: Kevin Wolf
Reviewed-by: Paolo Bonzini
Signed-off-by: Ming Lei
---
block/linux-aio.c |1 -
1 file changed, 1 deletion(-)
ntil
aborting is handled
Reviewed-by: Paolo Bonzini
Signed-off-by: Ming Lei
---
block/linux-aio.c | 116 -
1 file changed, 97 insertions(+), 19 deletions(-)
diff --git a/block/linux-aio.c b/block/linux-aio.c
index d92513b..53c5616 100644
---
:
- code style fix and commit log fix as suggested by Benoît Canet
V1:
- rebase on latest QEMU master
block/linux-aio.c | 139 -
1 file changed, 116 insertions(+), 23 deletions(-)
Thanks,
Ming Lei
On Mon, 01 Dec 2014 08:05:17 +0100
Peter Lieven wrote:
> On 01.12.2014 06:55, Ming Lei wrote:
> > On Fri, Nov 28, 2014 at 10:12 PM, Paolo Bonzini wrote:
> >> As discussed in the other thread, this brings speedups from
> >> dropping the coroutine mutex (which se
cost just measures
this part of cost), but also some implicit/invisible part. I have some
test cases which can show the problem. If someone is interested,
I can post them in list.
Thanks,
Ming Lei
per coroutine",
> maxcycles,
> duration, ops,
> - (unsigned long)(10 * duration) / maxcycles);
> + (unsigned long)(10.0 * duration / maxcycles));
One more single bracket.
thanks,
Ming Lei
On 11/28/14, Markus Armbruster wrote:
> Ming Lei writes:
>
>> On 11/28/14, Markus Armbruster wrote:
>>> Ming Lei writes:
>>>
>>>> Hi Kevin,
>>>>
>>>> On Wed, Nov 26, 2014 at 10:46 PM, Kevin Wolf wrote:
>>>>&g
On 11/28/14, Markus Armbruster wrote:
> Ming Lei writes:
>
>> Hi Kevin,
>>
>> On Wed, Nov 26, 2014 at 10:46 PM, Kevin Wolf wrote:
>>> This improves the performance of requests because an ACB doesn't need to
>>> be allocated on the heap any mor
o debug in the future.
>> It's so easy to add that I'm against merging the patch without this
>> protection.
>
> Yeah, either that or an assertion must be there.
OK.
Thanks,
Ming Lei
34.968s
> run 5 | 30.041s | 26.263s| 35.224s | 35.000s
>
> TODO: Do some more serious benchmarking in VMs with less variance.
> Results of a quick fio run are vaguely positive.
I will do the test with Paolo's fast path approach under
VM I/O situation.
Thanks,
Ming Lei
Hi Kevin,
On Wed, Nov 26, 2014 at 7:27 PM, Kevin Wolf wrote:
> Am 25.11.2014 um 08:23 hat Ming Lei geschrieben:
>> Previously -EAGAIN is simply ignored for !s->io_q.plugged case,
>> and sometimes it is easy to cause -EIO to VM, such as NVME device.
>>
>> This pat
On Wed, Nov 26, 2014 at 7:18 PM, Kevin Wolf wrote:
> Am 25.11.2014 um 08:23 hat Ming Lei geschrieben:
>> In the submit path, we can't complete request directly,
>> otherwise "Co-routine re-entered recursively" may be caused,
>> so this patch fixes the issue
re hasn't the potential problem in current tree, could you
agree on merging it first?
BTW, there isn't sort of handling for 'completion_bh' of linux aio too, :-)
Thanks,
Ming Lei
On Tue, Nov 25, 2014 at 9:08 PM, Stefan Hajnoczi wrote:
> On Tue, Nov 25, 2014 at 03:23:11PM +0800, Ming Lei wrote:
>> @@ -296,12 +370,14 @@ void laio_detach_aio_context(void *s_, AioContext
>> *old_context)
>>
>> aio_set_event_notifier(old_context, &s->
ted-by: Paolo Bonzini
Signed-off-by: Ming Lei
---
block/linux-aio.c | 24
1 file changed, 16 insertions(+), 8 deletions(-)
diff --git a/block/linux-aio.c b/block/linux-aio.c
index 11ac828..ac25722 100644
--- a/block/linux-aio.c
+++ b/block/linux-aio.c
@@ -282,8 +282,13 @@
tyle fix and commit log fix as suggested by Benoît Canet
V1:
- rebase on latest QEMU master
block/linux-aio.c | 129 +++--
1 file changed, 106 insertions(+), 23 deletions(-)
Thanks,
Ming Lei
ntil
aborting is handled
Reviewed-by: Paolo Bonzini
Signed-off-by: Ming Lei
---
block/linux-aio.c | 114 -
1 file changed, 95 insertions(+), 19 deletions(-)
diff --git a/block/linux-aio.c b/block/linux-aio.c
index d92513b..11ac828 100644
---
No one uses the 'node' field any more, so remove it
from 'struct qemu_laiocb', and this can save 16byte
for the struct on 64bit arch.
Reviewed-by: Paolo Bonzini
Signed-off-by: Ming Lei
---
block/linux-aio.c |1 -
1 file changed, 1 deletion(-)
diff --git a/block/linux
On Mon, Nov 24, 2014 at 10:31 PM, Paolo Bonzini wrote:
>
>
> On 24/11/2014 15:29, Ming Lei wrote:
>> The 1st patch fixes batch submission.
>>
>> The 2nd one fixes -EAGAIN for non-batch case.
>>
>> The 3rd one is a cleanup.
>>
>> This pa
No one uses the 'node' field any more, so remove it
from 'struct qemu_laiocb', and this can save 16byte
for the struct on 64bit arch.
Reviewed-by: Paolo Bonzini
Signed-off-by: Ming Lei
---
block/linux-aio.c |1 -
1 file changed, 1 deletion(-)
diff --git a/block/linux
off-by: Ming Lei
---
block/linux-aio.c | 28 ++--
1 file changed, 18 insertions(+), 10 deletions(-)
diff --git a/block/linux-aio.c b/block/linux-aio.c
index 70312a4..0cb098d 100644
--- a/block/linux-aio.c
+++ b/block/linux-aio.c
@@ -282,12 +282,17 @@ static int ioq_e
e changed, 107 insertions(+), 24 deletions(-)
Thanks,
Ming Lei
ntil
aborting is handled
Signed-off-by: Ming Lei
---
block/linux-aio.c | 116 -
1 file changed, 96 insertions(+), 20 deletions(-)
diff --git a/block/linux-aio.c b/block/linux-aio.c
index d92513b..70312a4 100644
--- a/block/linux-aio.c
+++
On Mon, Nov 24, 2014 at 7:39 PM, Paolo Bonzini wrote:
>
>
> On 24/11/2014 12:31, Ming Lei wrote:
>> +static int ioq_enqueue(struct qemu_laio_state *s, struct iocb *iocb)
>> {
>> unsigned int idx = s->io_q.idx;
>>
>> +if (unlikely(idx ==
On Mon, Nov 24, 2014 at 7:38 PM, Paolo Bonzini wrote:
>
>
> On 24/11/2014 12:31, Ming Lei wrote:
>> +/* don't submit until next completion for -EAGAIN of non plug case */
>> +if (unlikely(!s->io_q.plugged)) {
>> +return 0;
>> +}
&g
No one uses the 'node' field any more, so remove it
from 'struct qemu_laiocb', and this can save 16byte
for the struct on 64bit arch.
Reviewed-by: Paolo Bonzini
Signed-off-by: Ming Lei
---
block/linux-aio.c |1 -
1 file changed, 1 deletion(-)
diff --git a/block/linux
- rebase on latest QEMU master
block/linux-aio.c | 108 +
1 file changed, 84 insertions(+), 24 deletions(-)
Thanks
Ming Lei
off-by: Ming Lei
---
block/linux-aio.c | 20 ++--
1 file changed, 14 insertions(+), 6 deletions(-)
diff --git a/block/linux-aio.c b/block/linux-aio.c
index 11fcedb..974e4f9 100644
--- a/block/linux-aio.c
+++ b/block/linux-aio.c
@@ -259,6 +259,11 @@ static int ioq_enqueue(
n in BH context
- for part of completion, update the io queue too
- for case of io queue full, submit queued requests
immediatelly and return failure to caller
- for other failure, abort all queued requests in BH context
Signed-off-by: Ming Lei
---
block/lin
ting some I/O if the queue is getting too full. In the
> * former case, instead, wait until an I/O operation is completed.
> */
> if (likely(idx <= s->io_q.size * 2 / 3) || unlikely(!s->io_q.plugged) {
> return 0;
> }
Yes, the above change need the corresponding comment, but this
patch needn't since the 1st case won't reach here, :-)
Thanks,
Ming Lei
On Mon, Nov 24, 2014 at 5:47 PM, Paolo Bonzini wrote:
>
>
> On 22/11/2014 13:16, Ming Lei wrote:
>> > > +static int ioq_enqueue(struct qemu_laio_state *s, struct iocb *iocb)
>> > > {
>> > > unsigned int idx = s->io_q.idx;
>
On Tue, Nov 18, 2014 at 9:57 PM, Paolo Bonzini wrote:
>
>
> On 09/11/2014 08:42, Ming Lei wrote:
>> This patch implements AioContext wide IO submission as batch, and
>> the idea behind is very simple:
>>
>> - linux native aio(io_submit) supports to enqueu
On Tue, Nov 18, 2014 at 10:06 PM, Paolo Bonzini wrote:
>
>
> On 06/11/2014 16:10, Ming Lei wrote:
>> +/* don't submit until next completion for -EAGAIN of non plug case */
>> +if (unlikely(!s->io_q.plugged)) {
>> +return 0;
>> +}
>
n ret;
> }
>
> abort_queue(s);
> ret = len;
> }
>
> if (ret > 0) {
> memmove(...)
> s->io_q.idx -= ret;
> }
> return ret;
The above is better.
>> + * update io queue, for partial completion, retry will be
>> + * started automatically in following completion cb.
>> + */
>> +s->io_q.idx -= ret;
>> +
>> return ret;
>> }
>>
>> -static void ioq_enqueue(struct qemu_laio_state *s, struct iocb *iocb)
>> +static void ioq_submit_retry(void *opaque)
>> +{
>> +struct qemu_laio_state *s = opaque;
>> +ioq_submit(s, false);
>> +}
>> +
>> +static int ioq_enqueue(struct qemu_laio_state *s, struct iocb *iocb)
>> {
>> unsigned int idx = s->io_q.idx;
>>
>> +if (unlikely(idx == s->io_q.size)) {
>> +return -1;
>
> return -EAGAIN?
It means the io queue is full, so the code has to fail the current
request.
Thanks,
Ming Lei
This patch uses vring_should_notify() to suppress
guest notification, and looks notification frequency
can be decreased from ~33K/sec to ~2K/sec in my test
environment.
Suggested-by: Stefan Hajnoczi
Signed-off-by: Ming Lei
---
hw/scsi/virtio-scsi-dataplane.c |7 ++-
1 file changed, 6
On Wed, Nov 12, 2014 at 1:28 AM, Stefan Hajnoczi wrote:
> On Tue, Nov 11, 2014 at 09:17:10AM +0800, Ming Lei wrote:
>> +static void notify_guest_bh(void *opaque)
>> +{
>> +VirtIOSCSI *s = opaque;
>> +unsigned int qid;
>> +uint64_t pending = s->
On Tue, Nov 11, 2014 at 11:24 AM, Ming Lei wrote:
> On Mon, Nov 10, 2014 at 6:00 PM, alvise rigo
> wrote:
>> Hi Claudio,
>>
>> On Fri, Nov 7, 2014 at 4:40 PM, Claudio Fontana
>> wrote:
>>>
>>> Hi Alvise,
>>>
>>> I now got to te
tches, and it works fine
on ARMv7 VM, but ARMv8 VM can't boot. 'git bisect' told
me the 1st bad patch is below one:
generic_pci: generate dt node after devices init
Thanks,
--
Ming Lei
On Tue, Nov 11, 2014 at 10:52 AM, Fam Zheng wrote:
> On Tue, 11/11 10:29, Ming Lei wrote:
>> On Tue, Nov 11, 2014 at 9:52 AM, Fam Zheng wrote:
>> > On Tue, 11/11 09:17, Ming Lei wrote:
>> >> It isn't necessery to notify guest each time when one request
>>
On Tue, Nov 11, 2014 at 9:52 AM, Fam Zheng wrote:
> On Tue, 11/11 09:17, Ming Lei wrote:
>> It isn't necessery to notify guest each time when one request
>> is completed, and it should be enough to just notify one time
>> for each running of virtio_scsi_iothread_hand
It isn't necessery to notify guest each time when one request
is completed, and it should be enough to just notify one time
for each running of virtio_scsi_iothread_handle_cmd().
This patch supresses about 30K/sec write on eventfd.
Signed-off-by: Ming Lei
---
hw/scsi/virtio-scsi-datapl
Hi,
The 1st patch fixes an allocation problem.
The 2nd one supresses writing eventfd a lot(~30K/sec in my test).
V1:
- use g_new() in 1/2
- avoid to dereference VIRTIO_SCSI() directly 2/2
- add build check on queue depth
Thanks,
Ming Lei
The size of each element should be sizeof(VirtIOSCSIVring *).
Signed-off-by: Ming Lei
---
hw/scsi/virtio-scsi-dataplane.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/hw/scsi/virtio-scsi-dataplane.c b/hw/scsi/virtio-scsi-dataplane.c
index 855439e..df17229 100644
--- a
On Mon, Nov 10, 2014 at 5:21 PM, Kevin Wolf wrote:
> Am 10.11.2014 um 10:14 hat Ming Lei geschrieben:
>> On Mon, Nov 10, 2014 at 4:24 PM, Markus Armbruster wrote:
>> > Ming Lei writes:
>> >
>> >> The size of each element should be sizeof(VirtIOSCSIVring
On Mon, Nov 10, 2014 at 4:24 PM, Markus Armbruster wrote:
> Ming Lei writes:
>
>> The size of each element should be sizeof(VirtIOSCSIVring *).
>>
>> Signed-off-by: Ming Lei
>> ---
>> hw/scsi/virtio-scsi-dataplane.c |2 +-
>> 1 file changed, 1 inse
It isn't necessery to notify guest each time when one request
is completed, and it should be enough to just notify one time
for each running of virtio_scsi_iothread_handle_cmd().
This patch supresses about 30K/sec write on eventfd.
Signed-off-by: Ming Lei
---
hw/scsi/virtio-scsi-datapl
The size of each element should be sizeof(VirtIOSCSIVring *).
Signed-off-by: Ming Lei
---
hw/scsi/virtio-scsi-dataplane.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/hw/scsi/virtio-scsi-dataplane.c b/hw/scsi/virtio-scsi-dataplane.c
index 855439e..8a7cd9f 100644
--- a
The 1st patch fixes an allocation problem.
The 2nd one supresses writing eventfd a lot(~30K/sec in my test).
Thanks,
Ming Lei
Replace previous usage with AioContext Wide approach.
Signed-off-by: Ming Lei
---
hw/scsi/virtio-scsi-dataplane.c |8
hw/scsi/virtio-scsi.c |2 --
2 files changed, 8 insertions(+), 2 deletions(-)
diff --git a/hw/scsi/virtio-scsi-dataplane.c b/hw/scsi/virtio-scsi
Because we will support AioContext wide IO submission as batch,
so IO resources should be allocated according to how many there
are backends in the same AioContext.
Signed-off-by: Ming Lei
---
block/linux-aio.c | 38 +++---
1 file changed, 31 insertions(+), 7
y does the reallocation if there aren't
any pending I/O.
Signed-off-by: Ming Lei
---
block/linux-aio.c | 30 ++
1 file changed, 30 insertions(+)
diff --git a/block/linux-aio.c b/block/linux-aio.c
index e219b80..c5a88e8 100644
--- a/block/linux-aio.c
+++ b/block/linux
These two APIs are introduced for using AioContext wide
IO submission as batch.
Signed-off-by: Ming Lei
---
block.c | 16
include/block/block.h |3 +++
2 files changed, 19 insertions(+)
diff --git a/block.c b/block.c
index dacd881..0200af0 100644
--- a
This patch introduces laio_alloc_resource() for allocating
resources for linux aio, then in the following patchs we
can allocate IO resources just in demand.
Signed-off-by: Ming Lei
---
block/linux-aio.c | 55 ++---
1 file changed, 35 insertions
This patch allocates events array of 'struct qemu_laio_state'
dynamically so that in the following patch we can allocate
resource elasticly in case of AioContext wide IO submission
as batch.
Signed-off-by: Ming Lei
---
block/linux-aio.c |4 +++-
1 file changed, 3 insertions(+),
ill borrow the io queue from this 'bs',
so that we can maximum IO submission as batch.
aio_detach_aio_bs() will detach the 'bs' when all 'bs'
in the AioContext is detached.
Signed-off-by: Ming Lei
---
async.c |1 +
include/block/aio.h | 27
This patch introduces parameter of 'BlockDriverState' to
laio_attach_aio_context() and its pair, so that it will be
easier to support IO submission as batch in AioContext wide.
Signed-off-by: Ming Lei
---
block/linux-aio.c |6 --
block/raw-aio.h |6 --
block/raw-posi
So that it can be distinguished from the 'ctx' in qemu_laio_state.
Signed-off-by: Ming Lei
---
block/linux-aio.c |8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/block/linux-aio.c b/block/linux-aio.c
index b9db28e..cf8691e 100644
--- a/block/linux-aio.c
+
This patch supports IO submission as batch in AioContext wide
by sharing 'struct qemu_laio_state' instance among all linux-aio
backend in same AioContext.
Signed-off-by: Ming Lei
---
block/linux-aio.c | 36 ++--
1 file changed, 34 insertions(+), 2
Now lifetime of 'completion_bh', io queue and io context is same,
so move their allocation into laio_state_alloc() and their
releasing into laio_state_free().
Signed-off-by: Ming Lei
---
block/linux-aio.c | 26 --
1 file changed, 12 insertions(+), 14 deletion
This patch allocates 'struct qemu_laio_state' in aio attach,
and frees it in aio detach, so that in the following patch
we can share one same instance of the structure among multiple
linux-aio backend in same AioContext.
Signed-off-by: Ming Lei
---
block/linux-ai
This patch allocates io queue dynamically so that we
can support aio_context wide io queue in the following
patch.
Signed-off-by: Ming Lei
---
block/linux-aio.c | 66 +
1 file changed, 41 insertions(+), 25 deletions(-)
diff --git a/block
| 27 +
include/block/block.h |3 +
9 files changed, 259 insertions(+), 59 deletions(-)
Thanks,
Ming Lei
- for part of completion, also update the io queue
- for other failure, return the failure if in enqueue path,
otherwise, abort all queued I/O
Signed-off-by: Ming Lei
---
block/linux-aio.c | 101 +
1 file changed, 79 in
off-by: Ming Lei
---
block/linux-aio.c | 22 +-
1 file changed, 21 insertions(+), 1 deletion(-)
diff --git a/block/linux-aio.c b/block/linux-aio.c
index f66e8ad..f5ca41d 100644
--- a/block/linux-aio.c
+++ b/block/linux-aio.c
@@ -263,6 +263,11 @@ static int ioq_enqueue(
The 1st patch fixes batch submission.
The 2nd one fixes -EAGAIN for non-batch case.
The 3rd one is a cleanup.
This patchset is splitted from previous patchset(dataplane: optimization
and multi virtqueue support), as suggested by Stefan.
v3:
- rebase on QEMU master
v2:
- code sty
No one uses the 'node' field any more, so remove it
from 'struct qemu_laiocb', and this can save 16byte
for the struct on 64bit arch.
Signed-off-by: Ming Lei
---
block/linux-aio.c |1 -
1 file changed, 1 deletion(-)
diff --git a/block/linux-aio.c b/block/linux-aio.c
in
Hi Stefan and Guys,
On Thu, Sep 4, 2014 at 6:27 PM, Ming Lei wrote:
> The 1st patch fixes batch submission.
>
> The 2nd one fixes -EAGAIN for non-batch case.
>
> The 3rd one is a cleanup.
>
> The 4th one increase max event to 256 for supporting the comming
> multi virt-
Hi Paolo, Stefan and Kevin,
On Fri, Sep 5, 2014 at 12:27 AM, Ming Lei wrote:
> The 1st patch fixes batch submission.
>
> The 2nd one fixes -EAGAIN for non-batch case.
>
> The 3rd one is a cleanup.
>
> The 4th one increase max event to 256 for supporting the comming
> mu
This patch increases max event to 256 for the coming
virtio-blk multi virtqueue support.
Signed-off-by: Ming Lei
---
block/linux-aio.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/block/linux-aio.c b/block/linux-aio.c
index f45a142..5d565ad 100644
--- a/block/linux
g completion cb
- for part of completion, also update the io queue
- for other failure, return the failure if in enqueue path,
otherwise, abort all queued I/O
Signed-off-by: Ming Lei
---
block/linux-aio.c | 106 -
1 file changed
No one uses the 'node' field any more, so remove it
from 'struct qemu_laiocb', and this can save 16byte
for the struct on 64bit arch.
Signed-off-by: Ming Lei
---
block/linux-aio.c |1 -
1 file changed, 1 deletion(-)
diff --git a/block/linux-aio.c b/block/linux-aio.c
in
off-by: Ming Lei
---
block/linux-aio.c | 22 +-
1 file changed, 21 insertions(+), 1 deletion(-)
diff --git a/block/linux-aio.c b/block/linux-aio.c
index a06576d..20a87ec 100644
--- a/block/linux-aio.c
+++ b/block/linux-aio.c
@@ -276,6 +276,11 @@ static int ioq_enqueue(
-aio.c | 131 +
1 file changed, 103 insertions(+), 28 deletions(-)
Thanks,
--
Ming Lei
On Thu, Sep 4, 2014 at 10:59 PM, Benoît Canet wrote:
> The Thursday 14 Aug 2014 à 17:41:41 (+0800), Ming Lei wrote :
>> In the enqueue path, we can't complete request, otherwise
>> "Co-routine re-entered recursively" may be caused, so this
>> patch fixes th
off-by: Ming Lei
---
block/linux-aio.c | 22 +-
1 file changed, 21 insertions(+), 1 deletion(-)
diff --git a/block/linux-aio.c b/block/linux-aio.c
index ff66d1e..a979331 100644
--- a/block/linux-aio.c
+++ b/block/linux-aio.c
@@ -275,6 +275,11 @@ static int ioq_enqueue(
This patch increases max event to 256 for the comming
virtio-blk multi virtqueue support.
Signed-off-by: Ming Lei
---
block/linux-aio.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/block/linux-aio.c b/block/linux-aio.c
index ee125bc..4536106 100644
--- a/block/linux
1 - 100 of 339 matches
Mail list logo
