On Thu, Sep 21, 2017 at 09:29:43PM +0800, Fam Zheng wrote:
On Thu, 09/21 16:17, Manos Pitsidianakis wrote:
BlockDriverState has a bdrv_do_drain() callback but no equivalent for the end
s/bdrv_do_drain/bdrv_co_drain/
of the drain. The throttle driver (block/throttle.c) needs a way to mark
On Thu, Sep 21, 2017 at 09:35:35PM +0800, Fam Zheng wrote:
On Thu, 09/21 16:17, Manos Pitsidianakis wrote:
This patch series renames bdrv_co_drain to bdrv_co_drain_begin and adds a new
bdrv_co_drain_end callback to match bdrv_drained_begin/end and
drained_begin/end of BdrvChild. This is needed
Reviewed-by: Stefan Hajnoczi <stefa...@redhat.com>
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
block/throttle.c | 18 ++
1 file changed, 18 insertions(+)
diff --git a/block/throttle.c b/block/throttle.c
index 5bca76300f..833175ac77 100644
--- a/bloc
g for completion
[Stefan]
Manos Pitsidianakis (3):
block: add bdrv_co_drain_end callback
block: rename bdrv_co_drain to bdrv_co_drain_begin
block/throttle.c: add bdrv_co_drain_begin/end callbacks
include/block/block_int.h | 8 +++-
block/io.c
BlockDriverState has a bdrv_do_drain() callback but no equivalent for the end
of the drain. The throttle driver (block/throttle.c) needs a way to mark the
end of the drain in order to toggle io_limits_disabled correctly, thus
bdrv_co_drain_end is needed.
Signed-off-by: Manos Pitsidianakis <e
Reviewed-by: Stefan Hajnoczi <stefa...@redhat.com>
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
include/block/block_int.h | 2 +-
block/io.c| 4 ++--
block/qed.c | 6 +++---
3 files changed, 6 insertions(+), 6 deletions(-)
diff --git a/i
On Wed, Sep 20, 2017 at 03:26:32PM +0100, Stefan Hajnoczi wrote:
On Wed, Sep 20, 2017 at 01:23:09PM +0300, Manos Pitsidianakis wrote:
@@ -188,7 +194,7 @@ static bool bdrv_drain_recurse(BlockDriverState *bs)
waited = BDRV_POLL_WHILE(bs, atomic_read(>in_flight) > 0);
/* Ensu
On Wed, Sep 20, 2017 at 02:24:21PM +0200, Kevin Wolf wrote:
Am 20.09.2017 um 13:43 hat Manos Pitsidianakis geschrieben:
blk_check_byte_request() is called from the blk_co_pwritev/blk_co_preadv to
check if the request offset and request bytes parameters are valid for the
given Blockbackend
ast since QEMU 2.10.0 with -drive
iops=100:
$ dd if=/dev/zero of=/dev/vdb oflag=direct count=1000
(qemu) stop
(qemu) cont
...I/O is stuck...
Reported-by: Yongxue Hong <yh...@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefa...@redhat.com>
Reviewed-by: Manos Pitsidianakis <el
On Wed, Sep 20, 2017 at 01:08:52PM +0200, Alberto Garcia wrote:
On Wed 20 Sep 2017 12:17:40 PM CEST, Stefan Hajnoczi wrote:
@@ -592,6 +592,17 @@ void throttle_group_attach_aio_context(ThrottleGroupMember
*tgm,
void throttle_group_detach_aio_context(ThrottleGroupMember *tgm)
{
Cleanups for minor stuff I noticed while looking around blk_root_drained_*
Manos Pitsidianakis (2):
block/block-backend.c: add blk_check_byte_request call to
blk_pread/blk_pwrite
block/block-backend.c: remove blk_pread_unthrottled()
include/sysemu/block-backend.h | 2 --
block/block
blk_pread_unthrottled was used to bypass I/O throttling on the BlockBackend in
the case of async I/O. This is not needed anymore and we can just call
blk_pread() directly.
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
include/sysemu/block-backend.h | 2 --
block/block-bac
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
block/throttle.c | 18 ++
1 file changed, 18 insertions(+)
diff --git a/block/throttle.c b/block/throttle.c
index 5bca76300f..833175ac77 100644
--- a/block/throttle.c
+++ b/block/throttle.c
@@ -197,6 +197,21 @@
io_limits_disabled correctly.
Based-on: <20170918202529.28379-1-el13...@mail.ntua.gr>
"block/throttle-groups.c: allocate RestartData on the heap"
Which fixes a coroutine crash in block/throttle-groups.c
Manos Pitsidianakis (3):
block: add bdrv_co_drain_end callback
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
include/block/block_int.h | 2 +-
block/io.c| 4 ++--
block/qed.c | 6 +++---
3 files changed, 6 insertions(+), 6 deletions(-)
diff --git a/include/block/block_int.h b/include/block/block_int.h
blk_check_byte_request() is called from the blk_co_pwritev/blk_co_preadv to
check if the request offset and request bytes parameters are valid for the
given Blockbackend. Let's do that in blk_pread/blk_pwrite too.
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
block
BlockDriverState has a bdrv_do_drain() callback but no equivalent for the end
of the drain. The throttle driver (block/throttle.c) needs a way to mark the
end of the drain in order to toggle io_limits_disabled correctly, thus
bdrv_co_drain_end is needed.
Signed-off-by: Manos Pitsidianakis <e
RestartData is the opaque data of the throttle_group_restart_queue_entry
coroutine. By being stack allocated, it isn't available anymore if
aio_co_enter schedules the coroutine with a bottom halve and runs after
throttle_group_restart_queue returns.
Signed-off-by: Manos Pitsidianakis <e
On Thu, Sep 14, 2017 at 06:40:06AM -0400, Pradeep Jagadeesh wrote:
This patch factors out code to use the ThrottleLimits
strurcture.
Signed-off-by: Pradeep Jagadeesh
Reviewed-by: Greg Kurz
Reviewed-by: Eric Blake
Reviewed-by:
On Thu, Sep 14, 2017 at 06:40:05AM -0400, Pradeep Jagadeesh wrote:
This patch factors out the duplicate throttle code that was still
present in block and fsdev devices.
Signed-off-by: Pradeep Jagadeesh
Reviewed-by: Alberto Garcia
Reviewed-by:
On Thu, Sep 14, 2017 at 06:40:06AM -0400, Pradeep Jagadeesh wrote:
This patch factors out code to use the ThrottleLimits
strurcture.
Signed-off-by: Pradeep Jagadeesh
Reviewed-by: Greg Kurz
Reviewed-by: Eric Blake
Reviewed-by:
On Fri, Sep 08, 2017 at 06:00:11PM +0200, Kevin Wolf wrote:
Am 08.09.2017 um 17:44 hat Manos Pitsidianakis geschrieben:
On Thu, Sep 07, 2017 at 03:26:11PM +0200, Kevin Wolf wrote:
> We shouldn't really need any throttling code in
> blk_root_drained_begin/end any more now because the th
On Thu, Sep 07, 2017 at 03:26:11PM +0200, Kevin Wolf wrote:
We shouldn't really need any throttling code in
blk_root_drained_begin/end any more now because the throttle node will
be drained. If this code is necessary, a bdrv_drain() on an explicit
throttle node will work differently from one on
On Tue, Sep 05, 2017 at 04:13:39PM -0500, Eric Blake wrote:
On 09/05/2017 02:06 PM, Kevin Wolf wrote:
Am 05.09.2017 um 18:16 hat Kevin Wolf geschrieben:
Am 25.08.2017 um 15:20 hat Manos Pitsidianakis geschrieben:
Reviewed-by: Alberto Garcia <be...@igalia.com>
Signed-off-by:
This is a GSOC project summary required for the project's final
submission. As part of GSOC 2017, I took the project of moving two hard
coded block layer features into filter drivers. I/O Throttling is
implemented in block/throttle.c and before write notifiers are split
into a driver for each
of the
device. The legacy throttle node is managed by the legacy interface
completely. More advanced configurations with the filter drive are
possible using the QMP API, but these will be ignored by the legacy
interface.
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
include
ed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
include/sysemu/block-backend.h | 12 +---
block/block-backend.c | 43 +++---
block/qapi.c | 4 ++--
blockdev.c | 4 ++--
4 files changed, 2
.
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
include/block/throttle-groups.h | 1 +
block/block-backend.c | 15 +++--
block/throttle-groups.c | 145 +++-
tests/test-throttle.c | 3 +
4 files chang
graph.
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
include/block/block_int.h | 17 +
block.c | 10 ++
block/qapi.c | 14 +-
blockdev.c| 34 ++
4 files
ctions
add new function to get filter child bs
take ownership of options in bdrv_new_open_driver()
Manos Pitsidianakis (7):
block: skip implicit nodes in snapshots, blockjobs
block: add options parameter to bdrv_new_open_driver()
block: require job-id when device is a node name
block:
"sync": "full",
"target": "backup.img" } }
is not a BlockBackend name, instead of automatically getting it from the
root BS if device is a node name. That information is lost after calling
block_job_create(), so we can do it in its cal
Check that the implicit throttle filter driver node, used for
compatibility with the legacy throttling interface on the BlockBackend
level, works.
Reviewed-by: Alberto Garcia <be...@igalia.com>
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
tests/qemu-iotests/1
Allow passing a QDict *options parameter to bdrv_new_open_driver() so
that it can be used if a driver needs it upon creation. The previous
behaviour (empty bs->options and bs->explicit_options) remains when
options is NULL.
Reviewed-by: Alberto Garcia <be...@igalia.com>
Signed-o
-group=bar
which registers the throttle filter node with the ThrottleGroup 'bar'. The
given group must be created beforehand with object-add or -object.
Reviewed-by: Alberto Garcia <be...@igalia.com>
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
qapi/block-core.json
. This is done by gathering ThrottleGroup membership
details from BlockBackendPublic into ThrottleGroupMember and refactoring
existing code to use the structure.
Reviewed-by: Alberto Garcia <be...@igalia.com>
Reviewed-by: Stefan Hajnoczi <stefa...@redhat.com>
Signed-off-by: Manos Pitsidi
Reviewed-by: Alberto Garcia <be...@igalia.com>
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
tests/qemu-iotests/184 | 205 +++
tests/qemu-iotests/184.out | 300 +
tests/qemu-iotests/group
error in 'add aio_context field in ThrottleGroupMember'
v2:
change QOM throttle group object name
print valid ranges for uint on error
move frees in throttle_group_obj_finalize()
split throttle_group_{set,get}()
add throttle_recurse_is_first_non_filter()
Manos Pitsidianakis (6):
block
o",
"props" : {
"limits": {
"iops-total": 100
}
}
}
}
{ "execute" : "qom-set",
"arguments" : {
"path" : "foo",
"property" : "limits",
"value"
com>
Reviewed-by: Stefan Hajnoczi <stefa...@redhat.com>
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
include/block/throttle-groups.h | 7 -
block/block-backend.c | 15 --
block/throttle-groups.c | 38 -
tes
Move the CoMutex and CoQueue inits inside throttle_group_register_tgm()
which is called whenever a ThrottleGroupMember is initialized. There's
no need for them to be separate.
Reviewed-by: Alberto Garcia <be...@igalia.com>
Reviewed-by: Stefan Hajnoczi <stefa...@redhat.com>
Signed-o
On Thu, Aug 24, 2017 at 07:43:08PM +0100, Stefan Hajnoczi wrote:
On Tue, Aug 22, 2017 at 01:15:35PM +0300, Manos Pitsidianakis wrote:
+_supported_fmt qcow2
What makes this test qcow2-specific? With additional filtering for
IMGFMT it should be possible to run this on any image format.
I
Reviewed-by: Alberto Garcia <be...@igalia.com>
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
tests/qemu-iotests/184 | 205 +++
tests/qemu-iotests/184.out | 300 +
tests/qemu-iotests/group
print valid ranges for uint on error
move frees in throttle_group_obj_finalize()
split throttle_group_{set,get}()
add throttle_recurse_is_first_non_filter()
Manos Pitsidianakis (6):
block: move ThrottleGroup membership to ThrottleGroupMember
block: add aio_context field
o",
"props" : {
"limits": {
"iops-total": 100
}
}
}
}
{ "execute" : "qom-set",
"arguments" : {
"path" : "foo",
"property" : "limits",
"value"
-group=bar
which registers the throttle filter node with the ThrottleGroup 'bar'. The
given group must be created beforehand with object-add or -object.
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
qapi/block-core.json| 18 ++-
include/block/throttle-groups.h
com>
Reviewed-by: Stefan Hajnoczi <stefa...@redhat.com>
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
include/block/throttle-groups.h | 7 -
block/block-backend.c | 15 --
block/throttle-groups.c | 38 -
tes
. This is done by gathering ThrottleGroup membership
details from BlockBackendPublic into ThrottleGroupMember and refactoring
existing code to use the structure.
Reviewed-by: Alberto Garcia <be...@igalia.com>
Reviewed-by: Stefan Hajnoczi <stefa...@redhat.com>
Signed-off-by: Manos Pitsidi
Move the CoMutex and CoQueue inits inside throttle_group_register_tgm()
which is called whenever a ThrottleGroupMember is initialized. There's
no need for them to be separate.
Reviewed-by: Alberto Garcia <be...@igalia.com>
Reviewed-by: Stefan Hajnoczi <stefa...@redhat.com>
Signed-o
On Tue, Aug 22, 2017 at 04:16:26PM +0200, Alberto Garcia wrote:
On Tue 22 Aug 2017 12:15:34 PM CEST, Manos Pitsidianakis wrote:
@@ -548,6 +548,11 @@ void throttle_group_unregister_tgm(ThrottleGroupMember
*tgm)
ThrottleGroupMember *token;
int i;
+if (!ts) {
+/* Discard
On Tue, Aug 22, 2017 at 03:25:41PM +0200, Alberto Garcia wrote:
+/* This function edits throttle_groups and must be called under the global
+ * mutex */
+static void throttle_group_obj_complete(UserCreatable *obj, Error **errp)
+{
+ThrottleGroup *tg = THROTTLE_GROUP(obj);
+ThrottleConfig
On Tue, Aug 22, 2017 at 11:57:28AM +0200, Alberto Garcia wrote:
On Mon 21 Aug 2017 05:39:48 PM CEST, Manos Pitsidianakis wrote:
-if (job_id == NULL && !(flags & BLOCK_JOB_INTERNAL)) {
-job_id = bdrv_get_device_name(bs);
-if (!*job_id) {
-err
o",
"props" : {
"limits": {
"iops-total": 100
}
}
}
}
{ "execute" : "qom-set",
"arguments" : {
"path" : "foo",
"property" : "limits",
"value"
-group=bar
which registers the throttle filter node with the ThrottleGroup 'bar'. The
given group must be created beforehand with object-add or -object.
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
qapi/block-core.json| 18 ++-
include/block/throttle-groups.h
com>
Reviewed-by: Stefan Hajnoczi <stefa...@redhat.com>
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
include/block/throttle-groups.h | 7 -
block/block-backend.c | 15 --
block/throttle-groups.c | 38 -
tes
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
tests/qemu-iotests/184 | 205 +++
tests/qemu-iotests/184.out | 300 +
tests/qemu-iotests/group | 1 +
3 files changed, 506 insertions(+)
creat
Move the CoMutex and CoQueue inits inside throttle_group_register_tgm()
which is called whenever a ThrottleGroupMember is initialized. There's
no need for them to be separate.
Reviewed-by: Alberto Garcia <be...@igalia.com>
Reviewed-by: Stefan Hajnoczi <stefa...@redhat.com>
Signed-o
. This is done by gathering ThrottleGroup membership
details from BlockBackendPublic into ThrottleGroupMember and refactoring
existing code to use the structure.
Reviewed-by: Alberto Garcia <be...@igalia.com>
Reviewed-by: Stefan Hajnoczi <stefa...@redhat.com>
Signed-off-by: Manos Pitsidi
on error
move frees in throttle_group_obj_finalize()
split throttle_group_{set,get}()
add throttle_recurse_is_first_non_filter()
Manos Pitsidianakis (6):
block: move ThrottleGroup membership to ThrottleGroupMember
block: add aio_context field in ThrottleGroupMember
block: tidy
On Mon, Aug 21, 2017 at 05:05:30PM +0200, Alberto Garcia wrote:
On Wed 09 Aug 2017 04:02:53 PM CEST, Manos Pitsidianakis wrote:
@@ -622,20 +622,14 @@ void *block_job_create(const char *job_id, const
BlockJobDriver *driver,
return NULL;
}
-if (job_id == NULL &&
On Mon, Aug 21, 2017 at 02:10:59PM +0200, Alberto Garcia wrote:
On Sat 19 Aug 2017 05:23:12 PM CEST, Manos Pitsidianakis wrote:
-/* If the bucket is full then we have to wait */
-extra = bkt->level - bkt->max * bkt->burst_length;
+if (!bkt->max) {
+/* If bkt-
Move the CoMutex and CoQueue inits inside throttle_group_register_tgm()
which is called whenever a ThrottleGroupMember is initialized. There's
no need for them to be separate.
Reviewed-by: Alberto Garcia <be...@igalia.com>
Reviewed-by: Stefan Hajnoczi <stefa...@redhat.com>
Signed-o
. This is done by gathering ThrottleGroup membership
details from BlockBackendPublic into ThrottleGroupMember and refactoring
existing code to use the structure.
Reviewed-by: Alberto Garcia <be...@igalia.com>
Reviewed-by: Stefan Hajnoczi <stefa...@redhat.com>
Signed-off-by: Manos Pitsidi
{
"iops-total" : 99
}
}
}
This also means a group's configuration can be fetched with qom-get.
ThrottleGroups can be anonymous which means they can't get accessed by
other users ie they will always be units instead of group (Because they
have one ThrottleGroupMember
parameters and their semantics are identical to the
hardcoded throttling ones. limits.* cannot be specified at the same time
as throttle-group.
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
qapi/block-core.json| 22 ++-
include/qapi/qmp/qdict.h| 1 +
includ
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
tests/qemu-iotests/184 | 244
tests/qemu-iotests/184.out | 339 +
tests/qemu-iotests/group | 1 +
3 files changed, 584 insertions(+)
creat
com>
Reviewed-by: Stefan Hajnoczi <stefa...@redhat.com>
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
include/block/throttle-groups.h | 7 -
block/block-backend.c | 15 --
block/throttle-groups.c | 38 -
tes
:
fix style error in 'add aio_context field in ThrottleGroupMember'
v2:
change QOM throttle group object name
print valid ranges for uint on error
move frees in throttle_group_obj_finalize()
split throttle_group_{set,get}()
add throttle_recurse_is_first_non_filter()
Manos Pitsidianakis (6
On Thu, Aug 17, 2017 at 05:28:14PM +0300, Alberto Garcia wrote:
The throttling code can change internally the value of bkt->max if it
hasn't been set by the user. The problem with this is that if we want
to retrieve the original value we have to undo this change first. This
is ugly and
their value changes depending on the fraction of time elapsed
since the previous I/O operation.
Signed-off-by: Alberto Garcia <be...@igalia.com>
Reviewed-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
include/qemu/throttle.h | 4 ++--
util/throttle.c | 4 ++--
2 files changed, 4 inse
On Thu, Aug 17, 2017 at 05:28:12PM +0300, Alberto Garcia wrote:
The level of the burst bucket is stored in bkt.burst_level, not
bkt.burst_lenght.
s/lenght/length, otherwise:
Reviewed-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
Signed-off-by: Alberto Garcia <be...@igalia.com>
On Fri, Aug 18, 2017 at 03:05:31PM +0200, Alberto Garcia wrote:
On Fri 18 Aug 2017 05:10:17 AM CEST, Manos Pitsidianakis wrote:
* If no ThrottleGroup is found with the given name a new one is
* created.
*
- * @name: the name of the ThrottleGroup
+ * This function edits throttle_groups
On Fri, Aug 18, 2017 at 10:23:09AM +0200, Alberto Garcia wrote:
On Fri 18 Aug 2017 05:10:18 AM CEST, Manos Pitsidianakis wrote:
block/throttle.c uses existing I/O throttle infrastructure inside a
block filter driver. I/O operations are intercepted in the filter's
read/write coroutines
{
"iops-total" : 99
}
}
}
This also means a group's configuration can be fetched with qom-get.
ThrottleGroups can be anonymous which means they can't get accessed by
other users ie they will always be units instead of group (Because they
have one ThrottleGroupMember
. This is done by gathering ThrottleGroup membership
details from BlockBackendPublic into ThrottleGroupMember and refactoring
existing code to use the structure.
Reviewed-by: Alberto Garcia <be...@igalia.com>
Reviewed-by: Stefan Hajnoczi <stefa...@redhat.com>
Signed-off-by: Manos Pitsidi
com>
Reviewed-by: Stefan Hajnoczi <stefa...@redhat.com>
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
include/block/throttle-groups.h | 7 -
block/block-backend.c | 15 --
block/throttle-groups.c | 38 -
tes
hrottling filter driver infrastructure
+ *
+ * Copyright (c) 2017 Manos Pitsidianakis
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either ve
Move the CoMutex and CoQueue inits inside throttle_group_register_tgm()
which is called whenever a ThrottleGroupMember is initialized. There's
no need for them to be separate.
Reviewed-by: Alberto Garcia <be...@igalia.com>
Reviewed-by: Stefan Hajnoczi <stefa...@redhat.com>
Signed-o
in BlockDevOptionsThrottle
v3:
fix style error in 'add aio_context field in ThrottleGroupMember'
v2:
change QOM throttle group object name
print valid ranges for uint on error
move frees in throttle_group_obj_finalize()
split throttle_group_{set,get}()
add throttle_recurse_is_first_non_filter()
Manos
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
tests/qemu-iotests/184 | 310 +
tests/qemu-iotests/184.out | 422 +
tests/qemu-iotests/group | 1 +
3 files changed, 733 insertions(+)
creat
On Wed, Aug 16, 2017 at 02:25:44PM +0100, Stefan Hajnoczi wrote:
On Tue, Aug 15, 2017 at 11:18:53AM +0300, Manos Pitsidianakis wrote:
block/backup.c currently uses before write notifiers on the targeted
node. We can create a filter node instead to intercept write requests
for the backup job
On Wed, Aug 16, 2017 at 06:59:25AM -0500, Eric Blake wrote:
On 08/16/2017 04:41 AM, Manos Pitsidianakis wrote:
+##
+# @block-insert-node:
+#
+# Insert a filter node between a specific edge in the block driver
state graph.
+# @parent: the name of the parent node or device
+# @node:the name
On Tue, Aug 15, 2017 at 05:12:42PM -0500, Eric Blake wrote:
On 08/15/2017 02:45 AM, Manos Pitsidianakis wrote:
block-insert-node and its pair command block-remove-node provide runtime
insertion and removal of filter nodes.
block-insert-node takes a (parent, child) and (node, child) pair
On Tue, Aug 15, 2017 at 03:24:38PM +0200, Alberto Garcia wrote:
On Wed 09 Aug 2017 04:02:51 PM CEST, Manos Pitsidianakis wrote:
@@ -2988,6 +3004,9 @@ void qmp_block_stream(bool has_job_id, const char
*job_id, const char *device,
return;
}
+/* Skip implicit filter nodes
into the block layer. Block filter drivers are inserted into the graph
only when a feature is needed. This makes the block layer more modular
and reuses the block driver abstraction that is already present.
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
block.c
present.
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
block/qapi.c| 2 +-
block/write-threshold.c | 264 +++-
include/block/write-threshold.h | 22 ++--
qapi/block-core.json| 19 ++-
tests/test
) to using the filter driver.
v2:
add motivation in commit messages
dropped hunk that removed before-write notifiers
Manos Pitsidianakis (2):
block: use internal filter node in backup
block: add filter driver to block/write-threshold.c
block.c | 89
On Tue, Aug 15, 2017 at 10:44:15AM +0300, Vladimir Sementsov-Ogievskiy wrote:
15.08.2017 09:19, Manos Pitsidianakis wrote:
block/backup.c currently uses before write notifiers on the targeted
node. We can create a filter node instead to intercept write requests
for the backup job on the BDS
BdrvChildRole.
This is a different approach than x-blockdev-change which uses the driver
methods bdrv_add_child() and bdrv_del_child(),
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
block.c| 192
blockdev.c | 44 ++
include/block/b
driver, so it is not affected.
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
block/qapi.c| 2 +-
block/write-threshold.c | 264 +++-
include/block/write-threshold.h | 22 ++--
qapi/block-core.json
block/backup.c currently uses before write notifiers on the targeted
node. We can create a filter node instead to intercept write requests
for the backup job on the BDS level, instead of the BlockBackend level.
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
b
-threshold) to using the filter driver.
Based-on: <20170809140256.25584-1-el13...@mail.ntua.gr>
Manos Pitsidianakis (2):
block: use internal filter node in backup
block: add filter driver to block/write-threshold.c
block.c | 89 --
block/ba
On Wed, Aug 09, 2017 at 05:39:42PM +0200, Kevin Wolf wrote:
Am 09.08.2017 um 16:45 hat Alberto Garcia geschrieben:
On Wed 09 Aug 2017 03:42:07 PM CEST, Manos Pitsidianakis wrote:
> On Wed, Aug 09, 2017 at 02:36:20PM +0200, Alberto Garcia wrote:
>>On Wed 09 Aug 2017 11:36:12 AM CE
On Wed, Aug 09, 2017 at 01:07:32PM +0300, Manos Pitsidianakis wrote:
+static int coroutine_fn throttle_co_preadv(BlockDriverState *bs,
+ uint64_t offset, uint64_t bytes,
+ QEMUIOVector *qiov, int flags
On Thu, Aug 10, 2017 at 03:54:02PM +0200, Alberto Garcia wrote:
On Wed 09 Aug 2017 12:07:32 PM CEST, Manos Pitsidianakis wrote:
+/* Extract ThrottleConfig options. Assumes cfg is initialized and will be
+ * checked for validity.
+ *
+ * Returns -1 and sets errp if a burst_length value is over
of the
device. The legacy throttle node is managed by the legacy interface
completely. More advanced configurations with the filter drive are
possible using the QMP API, but these will be ignored by the legacy
interface.
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
block
-id when device is a node name
new commit: remove BlockBackendPublic
new commit: add dedicated iotest
cleanup reference counting in block/block-backend.c functions
add new function to get filter child bs
take ownership of options in bdrv_new_open_driver()
Manos Pitsidianakis (6):
block
"sync": "full",
"target": "backup.img" } }
is not a BlockBackend name, instead of automatically getting it from the
root BS if device is a node name. That information is lost after calling
block_job_create(), so we can do it in its
Check that the implicit throttle filter driver node, used for
compatibility with the legacy throttling interface on the BlockBackend
level, works.
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
tests/qemu-iotests/191 | 138 +
BlockBackend, which we want to be skipped during automatic
operations on the graph since the user does not necessarily know about
their existence.
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
---
block.c | 10 ++
block/qapi.c
Allow passing a QDict *options parameter to bdrv_new_open_driver() so
that it can be used if a driver needs it upon creation. The previous
behaviour (empty bs->options and bs->explicit_options) remains when
options is NULL.
Signed-off-by: Manos Pitsidianakis <el13...@mail.ntua.gr>
401 - 500 of 659 matches
Mail list logo