Am 05.06.2015 um 18:27 hat Alberto Garcia geschrieben:
This adds a new 'cache-clean-interval' option that cleans all qcow2
cache entries that haven't been used in a certain interval, given in
seconds.
This allows setting a large L2 cache size so it can handle scenarios
with lots of I/O and
Am 26.06.2015 um 11:57 hat Stefan Hajnoczi geschrieben:
On Thu, Jun 25, 2015 at 11:05:20AM -0400, Jeff Cody wrote:
On Thu, Jun 25, 2015 at 03:28:35PM +0100, Stefan Hajnoczi wrote:
On Wed, Jun 24, 2015 at 03:54:27PM -0400, Jeff Cody wrote:
@@ -269,7 +270,9 @@ static int
On Wed, Jul 08, 2015 at 02:01:30PM +0200, Kevin Wolf wrote:
The guest can't cause it, but once the connection is down, I expect
every request to fail. You don't have to have a malicious guest for
filling up the log file, it just needs to be careless enough to continue
trying new requests
On Wed, Jul 08, 2015 at 09:01:27AM +0800, Fam Zheng wrote:
On Tue, 07/07 16:08, Stefan Hajnoczi wrote:
+#define EPOLL_BATCH 128
+static bool aio_poll_epoll(AioContext *ctx, bool blocking)
+{
+AioHandler *node;
+bool was_dispatching;
+int i, ret;
+bool
Am 08.07.2015 um 12:47 hat Laurent Vivier geschrieben:
On 08/07/2015 12:31, Kevin Wolf wrote:
Am 02.07.2015 um 16:18 hat Laurent Vivier geschrieben:
On 02/07/2015 16:03, Paolo Bonzini wrote:
On 02/07/2015 15:58, Laurent Vivier wrote:
Since any /dev entry can be treated as a
On Wed, Jul 08, 2015 at 09:15:57AM +0800, Fam Zheng wrote:
On Tue, 07/07 15:35, Stefan Hajnoczi wrote:
On Tue, Jun 30, 2015 at 09:19:44PM +0800, Fam Zheng wrote:
diff --git a/async.c b/async.c
index 06971f4..1d70cfd 100644
--- a/async.c
+++ b/async.c
@@ -290,12 +290,17 @@
Am 02.07.2015 um 16:18 hat Laurent Vivier geschrieben:
On 02/07/2015 16:03, Paolo Bonzini wrote:
On 02/07/2015 15:58, Laurent Vivier wrote:
Since any /dev entry can be treated as a raw disk image, it is worth
noting which devices can be accessed when and how. /dev/rdisk nodes are
From: Cornelia Huck cornelia.h...@de.ibm.com
This reverts commit f5a5628cf0b65b223fa0c9031714578dfac4cf04.
This was an old patch that had been already superseded by b0e5d90eb
(dataplane: endianness-aware accesses).
Signed-off-by: Cornelia Huck cornelia.h...@de.ibm.com
Reviewed-by: Michael S.
Am 11.06.2015 um 12:01 hat Christoph Hellwig geschrieben:
The first patch ensures that flush actually flushes data so that
data integrity is preserved, the second fixe up WCE reporting.
Thanks, applied to the block branch for 2.4-rc1.
Kevin
Am 08.07.2015 um 03:02 schrieb Fam Zheng:
On Tue, 07/07 16:54, Christian Borntraeger wrote:
Am 30.06.2015 um 15:19 schrieb Fam Zheng:
epoll is more scalable than ppoll. It performs faster than ppoll when the
number of polled fds is high.
See patch 4 for an example of the senario and some
Currently if qemu is connected to a curl source (eg. web server), and
the web server fails / times out / dies, you always see a bogus EIO
Input/output error.
For example, choose a large file located on any local webserver which
you control:
$ qemu-img convert -p http://example.com/large.iso
Currently if qemu is connected to a curl source (eg. web server), and
the web server fails / times out / dies, you always see a bogus EIO
Input/output error.
For example, choose a large file located on any local webserver which
you control:
$ qemu-img convert -p http://example.com/large.iso
You can now write code like this:
FLOOD_COUNTER(errcount, 100);
...
NO_FLOOD(errcount,
error_report(oops, something bad happened),
error_report(further errors suppressed));
which would print the oops, ... error message up to 100 times,
followed by the further errors
Am 08.07.2015 um 14:50 hat Richard W.M. Jones geschrieben:
You can now write code like this:
FLOOD_COUNTER(errcount, 100);
...
NO_FLOOD(errcount,
error_report(oops, something bad happened),
error_report(further errors suppressed));
which would print the oops,
On Jul 8, 2015, at 7:01 AM, Kevin Wolf wrote:
Am 08.07.2015 um 12:47 hat Laurent Vivier geschrieben:
On 08/07/2015 12:31, Kevin Wolf wrote:
Am 02.07.2015 um 16:18 hat Laurent Vivier geschrieben:
On 02/07/2015 16:03, Paolo Bonzini wrote:
On 02/07/2015 15:58, Laurent Vivier wrote:
Add a function to query BlockLimits.max_iov.
Signed-off-by: Stefan Hajnoczi stefa...@redhat.com
---
block/block-backend.c | 5 +
include/sysemu/block-backend.h | 1 +
2 files changed, 6 insertions(+)
diff --git a/block/block-backend.c b/block/block-backend.c
index aee8a12..5108aec
The maximum number of struct iovec elements depends on the
BlockDriverState. The raw-posix protocol has a maximum of IOV_MAX but
others could have different values.
Instead of assuming raw-posix and hardcoding IOV_MAX in several places,
put the limit into BlockLimits.
Cc: Peter Lieven
Use blk_get_max_iov() instead of hardcoding IOV_MAX, which may not apply
to all BlockDrivers.
Signed-off-by: Stefan Hajnoczi stefa...@redhat.com
---
block/mirror.c | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/block/mirror.c b/block/mirror.c
index 985ad00..0f4445c
Request merging must not result in a huge request that exceeds the
maximum number of iovec elements. Use BlockLimits.max_iov instead of
hardcoding IOV_MAX.
Signed-off-by: Stefan Hajnoczi stefa...@redhat.com
---
block/io.c| 3 ++-
hw/block/virtio-blk.c | 2 +-
2 files changed, 3
qcow2_cache_flush() writes dirty cache to the disk and invokes bdrv_flush()
to make the data durable. But even if there is no dirty cache,
qcow2_cache_flush() would invoke bdrv_flush(). In fact, bdrv_flush() will
invoke fdatasync(), and it is an expensive operation. The patch will not
invoke
Block drivers may still want to access their child nodes in their
.bdrv_close handler. If they unref and/or detach a child by themselves,
this should not result in a double free.
There is additional code for backing files, which are just a special
case of child nodes. The same applies for them.
This patch moves bdrv_attach_child() from the individual places that add
a backing file to a BDS to bdrv_set_backing_hd(), which is called by all
of them. It also adds bdrv_detach_child() there.
For normal operation (starting with one backing file chain and not
changing it until the topmost image
Limiting to sectors_per_chunk for each bdrv_is_allocated_above is slow,
because the underlying protocol driver would issue much more queries
than necessary. We should coalesce the query.
Signed-off-by: Fam Zheng f...@redhat.com
---
block/mirror.c | 15 ++-
1 file changed, 6
Sleeping for 0 second may not be as effective as we want, use
block_job_relax_cpu.
Signed-off-by: Fam Zheng f...@redhat.com
---
block/mirror.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/block/mirror.c b/block/mirror.c
index 62db031..ca55578 100644
--- a/block/mirror.c
* Wen Congyang (we...@cn.fujitsu.com) wrote:
Signed-off-by: Wen Congyang we...@cn.fujitsu.com
Signed-off-by: zhanghailiang zhang.zhanghaili...@huawei.com
Signed-off-by: Gonglei arei.gong...@huawei.com
Cc: Jeff Cody jc...@redhat.com
---
block/backup.c | 13 +
* Wen Congyang (we...@cn.fujitsu.com) wrote:
On 07/08/2015 11:49 PM, Michael R. Hines wrote:
On 07/07/2015 08:38 PM, Wen Congyang wrote:
On 07/08/2015 12:56 AM, Michael R. Hines wrote:
On 07/07/2015 04:23 AM, Paolo Bonzini wrote:
On 07/07/2015 11:13, Dr. David Alan Gilbert wrote:
This
This supersedes:
http://patchwork.ozlabs.org/patch/491415/
and [1] which is currently in Jeff's tree.
Although [1] fixed the QMP responsiveness, Alexandre DERUMIER reported that
guest responsiveness still suffers when we are busy in the initial dirty bitmap
scanning loop of mirror job. That is
On 07/09/2015 09:55 AM, Dr. David Alan Gilbert wrote:
* Wen Congyang (we...@cn.fujitsu.com) wrote:
On 07/08/2015 11:49 PM, Michael R. Hines wrote:
On 07/07/2015 08:38 PM, Wen Congyang wrote:
On 07/08/2015 12:56 AM, Michael R. Hines wrote:
On 07/07/2015 04:23 AM, Paolo Bonzini wrote:
On
On 07/08/2015 11:49 PM, Michael R. Hines wrote:
On 07/07/2015 08:38 PM, Wen Congyang wrote:
On 07/08/2015 12:56 AM, Michael R. Hines wrote:
On 07/07/2015 04:23 AM, Paolo Bonzini wrote:
On 07/07/2015 11:13, Dr. David Alan Gilbert wrote:
This log is very stange. The NBD client connects to NBD
On 07/09/2015 09:28 AM, Dr. David Alan Gilbert wrote:
* Wen Congyang (we...@cn.fujitsu.com) wrote:
Signed-off-by: Wen Congyang we...@cn.fujitsu.com
Signed-off-by: zhanghailiang zhang.zhanghaili...@huawei.com
Signed-off-by: Gonglei arei.gong...@huawei.com
Cc: Jeff Cody jc...@redhat.com
---
This series is extracted from my work towards removing bdrv_swap(),
which is targeted for 2.5. It contains a fix for dangling pointers when
modifying the BDS graph and its dependencies.
I didn't bother to split patches of which only a part is required, nor
did I remove references to the future
It is the same as bdrv_open_image(), except that it doesn't only return
success or failure, but the newly created BdrvChild object for the new
child node.
As the BdrvChild object already contains a BlockDriverState pointer (and
this is supposed to become the only pointer so that bdrv_append() and
32 matches
Mail list logo