Increase bmds->cur_dirty after submit io, so reduce the frequency
involve into blk_drain, and improve the performance obviously
when block migration.
The performance test result of this patch:
During the block dirty save phase, this patch improve guest os IOPS
from 4.0K to 9.5K. and improve the m
On Wed, Mar 15, 2017 at 10:57 AM, Fam Zheng wrote:
> On Wed, 03/15 10:28, 858585 jemmy wrote:
>> On Tue, Mar 14, 2017 at 11:12 PM, Eric Blake wrote:
>> > On 03/14/2017 02:57 AM, jemmy858...@gmail.com wrote:
>> >> From: Lidong Chen
>> >>
>> >> Increase bmds->cur_dirty after submit io, so reduce t
On Tue, 03/14 08:28, Eric Blake wrote:
> On 03/13/2017 09:30 PM, Fam Zheng wrote:
> > bdrv_child_set_perm alone is not very usable because the caller must
> > call bdrv_child_check_perm first. This is already encapsulated
> > conveniently in bdrv_child_try_set_perm, so remove the other prototypes
>
On Wed, 03/15 10:28, 858585 jemmy wrote:
> On Tue, Mar 14, 2017 at 11:12 PM, Eric Blake wrote:
> > On 03/14/2017 02:57 AM, jemmy858...@gmail.com wrote:
> >> From: Lidong Chen
> >>
> >> Increase bmds->cur_dirty after submit io, so reduce the frequency involve
> >> into blk_drain, and improve the
On Tue, Mar 14, 2017 at 11:15 PM, Eric Blake wrote:
>
> On 03/14/2017 10:12 AM, Eric Blake wrote:
> > On 03/14/2017 02:57 AM, jemmy858...@gmail.com wrote:
> >> From: Lidong Chen
> >>
> >> Increase bmds->cur_dirty after submit io, so reduce the frequency involve
> >> into blk_drain, and improve t
On Tue, Mar 14, 2017 at 11:12 PM, Eric Blake wrote:
> On 03/14/2017 02:57 AM, jemmy858...@gmail.com wrote:
>> From: Lidong Chen
>>
>> Increase bmds->cur_dirty after submit io, so reduce the frequency involve
>> into blk_drain, and improve the performance obviously when block migration.
>
> Long
在 2017/3/14 22:13, Dr. David Alan Gilbert 写道:
* QingFeng Hao (ha...@linux.vnet.ibm.com) wrote:
This problem affects s390x only if we are running without KVM.
Basically, S390CPU.irqstate is unused if we do not use KVM,
and thus no buffer is allocated.
This causes size=0, first_elem=NULL and n_e
On 03/14/2017 11:03 AM, Kevin Wolf wrote:
> Signed-off-by: Kevin Wolf
> ---
> block/file-posix.c | 2 --
> 1 file changed, 2 deletions(-)
Reviewed-by: Eric Blake
>
> diff --git a/block/file-posix.c b/block/file-posix.c
> index 2cb2f64..a03a3e1 100644
> --- a/block/file-posix.c
> +++ b/block/f
On 03/14/2017 11:04 AM, Kevin Wolf wrote:
> Signed-off-by: Kevin Wolf
> ---
> block/file-win32.c | 1 -
> 1 file changed, 1 deletion(-)
Reviewed-by: Eric Blake
>
> diff --git a/block/file-win32.c b/block/file-win32.c
> index 800fabd..e132ba1 100644
> --- a/block/file-win32.c
> +++ b/block/fil
On 03/14/2017 11:28 AM, Kevin Wolf wrote:
> Am 14.03.2017 um 17:20 hat Eric Blake geschrieben:
>> On 03/14/2017 11:12 AM, Fam Zheng wrote:
>>> Signed-off-by: Fam Zheng
>>> ---
>>> block/file-posix.c | 3 +++
>>> 1 file changed, 3 insertions(+)
>>>
>>> diff --git a/block/file-posix.c b/block/file-
If we have separate iothread for cdrom, we lose connection to it on
qmp_blockdev_change_medium, as aio_context is on bds which is dropped
and switched with new one.
As an example result, after such media change we have crash on
virtio_scsi_ctx_check: Assertion `blk_get_aio_context(d->conf.blk) ==
Am 14.03.2017 um 17:20 hat Eric Blake geschrieben:
> On 03/14/2017 11:12 AM, Fam Zheng wrote:
> > Signed-off-by: Fam Zheng
> > ---
> > block/file-posix.c | 3 +++
> > 1 file changed, 3 insertions(+)
> >
> > diff --git a/block/file-posix.c b/block/file-posix.c
> > index c4c0663..d670be3 100644
>
On 03/14/2017 11:12 AM, Fam Zheng wrote:
> Signed-off-by: Fam Zheng
> ---
> block/file-posix.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/block/file-posix.c b/block/file-posix.c
> index c4c0663..d670be3 100644
> --- a/block/file-posix.c
> +++ b/block/file-posix.c
> @@ -703,6 +70
Am 14.03.2017 um 17:12 hat Fam Zheng geschrieben:
> Signed-off-by: Fam Zheng
Thanks, applied to the block branch.
Kevin
On 03/14/2017 11:14 AM, Eric Blake wrote:
> On 03/14/2017 11:11 AM, Eric Blake wrote:
>> On 03/14/2017 10:43 AM, Fam Zheng wrote:
>>> Signed-off-by: Fam Zheng
>>> ---
>>> block/file-posix.c | 1 +
>>> 1 file changed, 1 insertion(+)
>>>
>>> diff --git a/block/file-posix.c b/block/file-posix.c
>>>
Am 14.03.2017 um 17:11 hat Eric Blake geschrieben:
> On 03/14/2017 10:43 AM, Fam Zheng wrote:
> > Signed-off-by: Fam Zheng
> > ---
> > block/file-posix.c | 1 +
> > 1 file changed, 1 insertion(+)
> >
> > diff --git a/block/file-posix.c b/block/file-posix.c
> > index c4c0663..e6170f4 100644
> > -
Signed-off-by: Fam Zheng
---
block/file-posix.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/block/file-posix.c b/block/file-posix.c
index c4c0663..d670be3 100644
--- a/block/file-posix.c
+++ b/block/file-posix.c
@@ -703,6 +703,9 @@ static int hdev_get_max_segments(const struct stat *st
On 03/14/2017 11:11 AM, Eric Blake wrote:
> On 03/14/2017 10:43 AM, Fam Zheng wrote:
>> Signed-off-by: Fam Zheng
>> ---
>> block/file-posix.c | 1 +
>> 1 file changed, 1 insertion(+)
>>
>> diff --git a/block/file-posix.c b/block/file-posix.c
>> index c4c0663..e6170f4 100644
>> --- a/block/file-po
On 03/14/2017 10:43 AM, Fam Zheng wrote:
> Signed-off-by: Fam Zheng
> ---
> block/file-posix.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/block/file-posix.c b/block/file-posix.c
> index c4c0663..e6170f4 100644
> --- a/block/file-posix.c
> +++ b/block/file-posix.c
> @@ -703,6 +703,7
On Tue, 03/14 16:50, Kevin Wolf wrote:
> Am 14.03.2017 um 16:43 hat Fam Zheng geschrieben:
> > Signed-off-by: Fam Zheng
> > ---
> > block/file-posix.c | 1 +
> > 1 file changed, 1 insertion(+)
> >
> > diff --git a/block/file-posix.c b/block/file-posix.c
> > index c4c0663..e6170f4 100644
> > ---
Signed-off-by: Kevin Wolf
---
block/file-win32.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/block/file-win32.c b/block/file-win32.c
index 800fabd..e132ba1 100644
--- a/block/file-win32.c
+++ b/block/file-win32.c
@@ -24,7 +24,6 @@
#include "qemu/osdep.h"
#include "qapi/error.h"
#include
Signed-off-by: Kevin Wolf
---
block/file-posix.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/block/file-posix.c b/block/file-posix.c
index 2cb2f64..a03a3e1 100644
--- a/block/file-posix.c
+++ b/block/file-posix.c
@@ -25,8 +25,6 @@
#include "qapi/error.h"
#include "qemu/cutils.h"
#incl
Am 14.03.2017 um 16:43 hat Fam Zheng geschrieben:
> Signed-off-by: Fam Zheng
> ---
> block/file-posix.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/block/file-posix.c b/block/file-posix.c
> index c4c0663..e6170f4 100644
> --- a/block/file-posix.c
> +++ b/block/file-posix.c
> @@ -703
Signed-off-by: Fam Zheng
---
block/file-posix.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/block/file-posix.c b/block/file-posix.c
index c4c0663..e6170f4 100644
--- a/block/file-posix.c
+++ b/block/file-posix.c
@@ -703,6 +703,7 @@ static int hdev_get_max_segments(const struct stat *st)
On 03/14/2017 10:12 AM, Eric Blake wrote:
> On 03/14/2017 02:57 AM, jemmy858...@gmail.com wrote:
>> From: Lidong Chen
>>
>> Increase bmds->cur_dirty after submit io, so reduce the frequency involve
>> into blk_drain, and improve the performance obviously when block migration.
>
> Long line; plea
On 03/14/2017 02:57 AM, jemmy858...@gmail.com wrote:
> From: Lidong Chen
>
> Increase bmds->cur_dirty after submit io, so reduce the frequency involve
> into blk_drain, and improve the performance obviously when block migration.
Long line; please wrap your commit messages, preferably around 70
From: Lidong Chen
Increase bmds->cur_dirty after submit io, so reduce the frequency involve into
blk_drain, and improve the performance obviously when block migration.
Signed-off-by: Lidong Chen
---
migration/block.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/migration/block.c b/mi
On 03/14/2017 06:11 AM, Paolo Bonzini wrote:
> After the switch to reading replies in a coroutine, nothing is
> reentering pending receive coroutines if the connection hangs.
> Move nbd_recv_coroutines_enter_all to the reply read coroutine,
> which is the place where hangups are detected. nbd_tear
On 03/14/2017 06:11 AM, Paolo Bonzini wrote:
> While it is true that bdrv_set_aio_context only works on a single
> BlockDriverState subtree (see commit message for 53ec73e, "block: Use
> bdrv_drain to replace uncessary bdrv_drain_all", 2015-07-07), it works
I was about to correct the typo, but see
* QingFeng Hao (ha...@linux.vnet.ibm.com) wrote:
> This problem affects s390x only if we are running without KVM.
> Basically, S390CPU.irqstate is unused if we do not use KVM,
> and thus no buffer is allocated.
> This causes size=0, first_elem=NULL and n_elems=1 in
> vmstate_load_state and vmstate_
On 03/13/2017 09:30 PM, Fam Zheng wrote:
> bdrv_child_set_perm alone is not very usable because the caller must
> call bdrv_child_check_perm first. This is already encapsulated
> conveniently in bdrv_child_try_set_perm, so remove the other prototypes
> from the header and fix the one wrong caller,
hi everybody,
how could I (easily) follow file modifications made by guest os (Windows)
with qemu ?
could I exploit the growing overlay image based on an original Windows
image ?
regards, lacsaP.
Am 14.03.2017 um 12:46 hat Changlong Xie geschrieben:
> Even if hidden_disk, secondary_disk are backing files, they all need
> write permissions in replication scenario. Otherwise we will encouter
> below exceptions on secondary side during adding nbd server:
>
> {'execute': 'nbd-server-add', 'arg
Even if hidden_disk, secondary_disk are backing files, they all need
write permissions in replication scenario. Otherwise we will encouter
below exceptions on secondary side during adding nbd server:
{'execute': 'nbd-server-add', 'arguments': {'device': 'colo-disk', 'writable':
true } }
{"error":
While it is true that bdrv_set_aio_context only works on a single
BlockDriverState subtree (see commit message for 53ec73e, "block: Use
bdrv_drain to replace uncessary bdrv_drain_all", 2015-07-07), it works
at the AioContext level rather than the BlockDriverState level.
Therefore, it is also neces
After the switch to reading replies in a coroutine, nothing is
reentering pending receive coroutines if the connection hangs.
Move nbd_recv_coroutines_enter_all to the reply read coroutine,
which is the place where hangups are detected. nbd_teardown_connection
can simply wait for the reply read co
On 14 March 2017 at 10:29, Kevin Wolf wrote:
> Am 13.03.2017 um 19:05 hat Peter Maydell geschrieben:
>> On 13 March 2017 at 15:54, Kevin Wolf wrote:
>> > The following changes since commit
>> > dd4d2578215cd380f40a38028a9904e15b135ef3:
>> >
>> > Merge remote-tracking branch 'remotes/kraxel/tag
On Tue, 03/14 17:09, Stefan Hajnoczi wrote:
> The following pattern is unsafe:
>
> char buf[32];
> ret = read(fd, buf, sizeof(buf));
> ...
> buf[ret] = 0;
>
> If read(2) returns 32 then a byte beyond the end of the buffer is
> zeroed.
>
> In practice this buffer overflow does not occur b
Am 13.03.2017 um 19:05 hat Peter Maydell geschrieben:
> On 13 March 2017 at 15:54, Kevin Wolf wrote:
> > The following changes since commit dd4d2578215cd380f40a38028a9904e15b135ef3:
> >
> > Merge remote-tracking branch 'remotes/kraxel/tags/pull-fixes-20170309-1'
> > into staging (2017-03-09 13:
Am 14.03.2017 um 10:09 hat Stefan Hajnoczi geschrieben:
> The following pattern is unsafe:
>
> char buf[32];
> ret = read(fd, buf, sizeof(buf));
> ...
> buf[ret] = 0;
>
> If read(2) returns 32 then a byte beyond the end of the buffer is
> zeroed.
>
> In practice this buffer overflow does
The following pattern is unsafe:
char buf[32];
ret = read(fd, buf, sizeof(buf));
...
buf[ret] = 0;
If read(2) returns 32 then a byte beyond the end of the buffer is
zeroed.
In practice this buffer overflow does not occur because the sysfs
max_segments file only contains an unsigned short
41 matches
Mail list logo
