From: Hyman Huang(黄勇)
Introduce migration dirty-limit capability, which can
be turned on before live migration and limit dirty
page rate durty live migration.
Introduce migrate_dirty_limit function to help check
if dirty-limit capability enabled during live migration.
Meanwhile, refactor vcpu_d
From: Hyman Huang(黄勇)
Implement dirty-limit convergence algo for live migration,
which is kind of like auto-converge algo but using dirty-limit
instead of cpu throttle to make migration convergent.
Enable dirty page limit if dirty_rate_high_cnt greater than 2
when dirty-limit capability enabled,
From: Hyman Huang(黄勇)
Add migration dirty-limit capability test if kernel support
dirty ring.
Migration dirty-limit capability introduce dirty limit
capability, two parameters: x-vcpu-dirty-limit-period and
vcpu-dirty-limit are introduced to implement the live
migration with dirty limit.
The te
From: Hyman Huang(黄勇)
Introduce "vcpu-dirty-limit" migration parameter used
to limit dirty page rate during live migration.
"vcpu-dirty-limit" and "x-vcpu-dirty-limit-period" are
two dirty-limit-related migration parameters, which can
be set before and during live migration by qmp
migrate-set-pa
From: Hyman Huang(黄勇)
Extend query-migrate to provide throttle time and estimated
ring full time with dirty-limit capability enabled, through which
we can observe if dirty limit take effect during live migration.
Signed-off-by: Hyman Huang(黄勇)
Signed-off-by: Markus Armbruster
---
include/syse
From: Peter Xu
It's possible that we want to reap a dirty ring on a vcpu that is during
creation, because the vcpu is put onto list (CPU_FOREACH visible) before
initialization of the structures. In this case:
qemu_init_vcpu
x86_cpu_realizefn
cpu_exec_realizefn
cpu_list_a
From: Hyman Huang(黄勇)
v4:
1. Polish the docs and update the release version suggested by Markus
2. Rename the migrate exported info "dirty-limit-throttle-time-per-round"
to "dirty-limit-throttle-time-per-full".
The following 5 commits hasn't been acked or reviewed yet:
kvm: dirty-ring: Fix
From: Hyman Huang(黄勇)
Introduce "x-vcpu-dirty-limit-period" migration experimental
parameter, which is in the range of 1 to 1000ms and used to
make dirtyrate calculation period configurable.
Currently with the "x-vcpu-dirty-limit-period" varies, the
total time of live migration changes, test res
From: Hyman Huang(黄勇)
Check if block migration is running before throttling
guest down in auto-converge way.
Note that this modification is kind of like code clean,
because block migration does not depend on auto-converge
capability, so the order of checks can be adjusted.
Signed-off-by: Hyman
From: Hyman Huang(黄勇)
Coverity points out a overflow problem when computing MB,
dirty_ring_size and TARGET_PAGE_SIZE are both 32 bits,
multiplication will be done as a 32-bit operation, which
could overflow. Simplify the formula.
Meanwhile, fix spelling mistake of variable name.
Reported-by: Pe
From: Hyman Huang(黄勇)
dirty_rate paraemter of hmp command "set_vcpu_dirty_limit" is invalid
if less than 0, so add parameter check for it.
Note that this patch also delete the unsolicited help message and
clean up the code.
Signed-off-by: Hyman Huang(黄勇)
Signed-off-by: Markus Armbruster
Revie
From: Hyman Huang(黄勇)
Use vhost_user_save_acked_features to implemente acked features
saving.
Signed-off-by: Hyman Huang(黄勇)
Signed-off-by: Guoyi Tu
---
net/vhost-user.c | 6 +-
1 file changed, 1 insertion(+), 5 deletions(-)
diff --git a/net/vhost-user.c b/net/vhost-user.c
index f5cb095.
From: Hyman Huang(黄勇)
This patch aims to fix unexpected negotiation features for
vhost-user netdev interface.
When openvswitch reconnect Qemu after an unexpected disconnection
and Qemu therefore start the vhost_dev, acked_features field in
vhost_dev is initialized with value fetched from acked_f
From: Hyman Huang(黄勇)
Abstract vhost acked features saving into
vhost_user_save_acked_features, export it as util function.
Signed-off-by: Hyman Huang(黄勇)
Signed-off-by: Guoyi Tu
---
include/net/vhost-user.h | 1 +
net/vhost-user.c | 21 +++--
2 files changed, 16 inse
From: Hyman Huang(黄勇)
v6:
-rebase on master
-fix the compling issue
vhost_user_save_acked_features symbol only exists when
"vhost-user" be configured. Add MACRO statement before
calling vhost_user_save_acked_features.
Thanks Michael for pointing out that issue in time.
Please review,
Yong
From: Hyman Huang(黄勇)
This patch aims to fix unexpected negotiation features for
vhost-user netdev interface.
When openvswitch reconnect Qemu after an unexpected disconnection
and Qemu therefore start the vhost_dev, acked_features field in
vhost_dev is initialized with value fetched from acked_f
From: Hyman Huang(黄勇)
Use vhost_user_save_acked_features to implemente acked features
saving.
Signed-off-by: Hyman Huang(黄勇)
Signed-off-by: Guoyi Tu
---
net/vhost-user.c | 6 +-
1 file changed, 1 insertion(+), 5 deletions(-)
diff --git a/net/vhost-user.c b/net/vhost-user.c
index f5cb095.
From: Hyman Huang(黄勇)
Abstract vhost acked features saving into
vhost_user_save_acked_features, export it as util function.
Signed-off-by: Hyman Huang(黄勇)
Signed-off-by: Guoyi Tu
---
include/net/vhost-user.h | 1 +
net/vhost-user.c | 21 +++--
2 files changed, 16 inse
From: Hyman Huang(黄勇)
v5(resend):
-rebase on master
v5:
-fix the assert statement in [PATCH v4 3/3], reported by
xiangdong.
v4:
-rebase on master
-add stub function to fix build errors
-code clean on [PATCH v2 1/2]: drop 'cleanup' parameter in
vhost_user_save_acked_features.
-code clean on [
From: Hyman Huang(黄勇)
This patch aims to fix unexpected negotiation features for
vhost-user netdev interface.
When openvswitch reconnect Qemu after an unexpected disconnection
and Qemu therefore start the vhost_dev, acked_features field in
vhost_dev is initialized with value fetched from acked_f
From: Hyman Huang(黄勇)
Abstract vhost acked features saving into
vhost_user_save_acked_features, export it as util function.
Signed-off-by: Hyman Huang(黄勇)
Signed-off-by: Guoyi Tu
---
include/net/vhost-user.h | 1 +
net/vhost-user.c | 21 +++--
2 files changed, 16 inse
From: Hyman Huang(黄勇)
Use vhost_user_save_acked_features to implemente acked features
saving.
Signed-off-by: Hyman Huang(黄勇)
Signed-off-by: Guoyi Tu
---
net/vhost-user.c | 6 +-
1 file changed, 1 insertion(+), 5 deletions(-)
diff --git a/net/vhost-user.c b/net/vhost-user.c
index 40a2fe8.
From: Hyman Huang(黄勇)
v5:
-fix the assert statement in [PATCH v4 3/3], reported by
xiangdong.
v4:
-rebase on master
-add stub function to fix build errors
-code clean on [PATCH v2 1/2]: drop 'cleanup' parameter in
vhost_user_save_acked_features.
-code clean on [PATCH v2 2/2]: make refactor of
From: Hyman Huang(黄勇)
dirty_rate paraemter of hmp command "set_vcpu_dirty_limit" is invalid
if less than 0, so add parameter check for it.
Note that this patch also delete the unsolicited help message and
clean up the code.
Signed-off-by: Hyman Huang(黄勇)
Signed-off-by: Markus Armbruster
Revie
From: Hyman Huang(黄勇)
Introduce migration dirty-limit capability, which can
be turned on before live migration and limit dirty
page rate durty live migration.
Introduce migrate_dirty_limit function to help check
if dirty-limit capability enabled during live migration.
Meanwhile, refactor vcpu_d
From: Hyman Huang(黄勇)
Export dirty limit throttle time and estimated ring full
time, through which we can observe if dirty limit take
effect during live migration.
Signed-off-by: Hyman Huang(黄勇)
---
include/sysemu/dirtylimit.h | 2 ++
migration/migration.c | 10 ++
monitor/hmp-c
From: Hyman Huang(黄勇)
Coverity points out a overflow problem when computing MB,
dirty_ring_size and TARGET_PAGE_SIZE are both 32 bits,
multiplication will be done as a 32-bit operation, which
could overflow. Simplify the formula.
Meanwhile, fix spelling mistake of variable name.
Reported-by: Pe
From: Hyman Huang(黄勇)
v3(resend):
- fix the syntax error of the topic.
v3:
This version make some modifications inspired by Peter and Markus
as following:
1. Do the code clean up in [PATCH v2 02/11] suggested by Markus
2. Replace the [PATCH v2 03/11] with a much simpler patch posted by
Peter
From: Hyman Huang(黄勇)
Add migration dirty-limit capability test if kernel support
dirty ring.
Migration dirty-limit capability introduce dirty limit
capability, two parameters: x-vcpu-dirty-limit-period and
vcpu-dirty-limit are introduced to implement the live
migration with dirty limit.
The te
From: Hyman Huang(黄勇)
Check if block migration is running before throttling
guest down in auto-converge way.
Note that this modification is kind of like code clean,
because block migration does not depend on auto-converge
capability, so the order of checks can be adjusted.
Signed-off-by: Hyman
From: Peter Xu
It's possible that we want to reap a dirty ring on a vcpu that is during
creation, because the vcpu is put onto list (CPU_FOREACH visible) before
initialization of the structures. In this case:
qemu_init_vcpu
x86_cpu_realizefn
cpu_exec_realizefn
cpu_list_a
From: Hyman Huang(黄勇)
Introduce "vcpu-dirty-limit" migration parameter used
to limit dirty page rate during live migration.
"vcpu-dirty-limit" and "x-vcpu-dirty-limit-period" are
two dirty-limit-related migration parameters, which can
be set before and during live migration by qmp
migrate-set-pa
From: Hyman Huang(黄勇)
Implement dirty-limit convergence algo for live migration,
which is kind of like auto-converge algo but using dirty-limit
instead of cpu throttle to make migration convergent.
Enable dirty page limit if dirty_rate_high_cnt greater than 2
when dirty-limit capability enabled,
From: Hyman Huang(黄勇)
Introduce "x-vcpu-dirty-limit-period" migration experimental
parameter, which is in the range of 1 to 1000ms and used to
make dirtyrate calculation period configurable.
Currently with the "x-vcpu-dirty-limit-period" varies, the
total time of live migration changes, test res
From: Peter Xu
It's possible that we want to reap a dirty ring on a vcpu that is during
creation, because the vcpu is put onto list (CPU_FOREACH visible) before
initialization of the structures. In this case:
qemu_init_vcpu
x86_cpu_realizefn
cpu_exec_realizefn
cpu_list_a
From: Hyman Huang(黄勇)
Check if block migration is running before throttling
guest down in auto-converge way.
Note that this modification is kind of like code clean,
because block migration does not depend on auto-converge
capability, so the order of checks can be adjusted.
Signed-off-by: Hyman
From: Hyman Huang(黄勇)
Export dirty limit throttle time and estimated ring full
time, through which we can observe if dirty limit take
effect during live migration.
Signed-off-by: Hyman Huang(黄勇)
---
include/sysemu/dirtylimit.h | 2 ++
migration/migration.c | 10 ++
monitor/hmp-c
From: Hyman Huang(黄勇)
Coverity points out a overflow problem when computing MB,
dirty_ring_size and TARGET_PAGE_SIZE are both 32 bits,
multiplication will be done as a 32-bit operation, which
could overflow. Simplify the formula.
Meanwhile, fix spelling mistake of variable name.
Reported-by: Pe
From: Hyman Huang(黄勇)
v3:
This version make some modifications inspired by Peter and Markus
as following:
1. Do the code clean up in [PATCH v2 02/11] suggested by Markus
2. Replace the [PATCH v2 03/11] with a much simpler patch posted by
Peter to fix the following bug:
https://bugzilla.red
From: Hyman Huang(黄勇)
Add migration dirty-limit capability test if kernel support
dirty ring.
Migration dirty-limit capability introduce dirty limit
capability, two parameters: x-vcpu-dirty-limit-period and
vcpu-dirty-limit are introduced to implement the live
migration with dirty limit.
The te
From: Hyman Huang(黄勇)
Introduce "x-vcpu-dirty-limit-period" migration experimental
parameter, which is in the range of 1 to 1000ms and used to
make dirtyrate calculation period configurable.
Currently with the "x-vcpu-dirty-limit-period" varies, the
total time of live migration changes, test res
From: Hyman Huang(黄勇)
Introduce migration dirty-limit capability, which can
be turned on before live migration and limit dirty
page rate durty live migration.
Introduce migrate_dirty_limit function to help check
if dirty-limit capability enabled during live migration.
Meanwhile, refactor vcpu_d
From: Hyman Huang(黄勇)
Introduce "vcpu-dirty-limit" migration parameter used
to limit dirty page rate during live migration.
"vcpu-dirty-limit" and "x-vcpu-dirty-limit-period" are
two dirty-limit-related migration parameters, which can
be set before and during live migration by qmp
migrate-set-pa
From: Hyman Huang(黄勇)
Implement dirty-limit convergence algo for live migration,
which is kind of like auto-converge algo but using dirty-limit
instead of cpu throttle to make migration convergent.
Enable dirty page limit if dirty_rate_high_cnt greater than 2
when dirty-limit capability enabled,
From: Hyman Huang(黄勇)
dirty_rate paraemter of hmp command "set_vcpu_dirty_limit" is invalid
if less than 0, so add parameter check for it.
Note that this patch also delete the unsolicited help message and
clean up the code.
Signed-off-by: Hyman Huang(黄勇)
Signed-off-by: Markus Armbruster
Revie
From: Hyman Huang(黄勇)
Export dirty limit throttle time and estimated ring full
time, through which we can observe the process of dirty
limit during live migration.
Signed-off-by: Hyman Huang(黄勇)
---
include/sysemu/dirtylimit.h | 2 ++
migration/migration.c | 10 ++
monitor/hmp-c
From: Hyman Huang(黄勇)
Implement dirty-limit convergence algo for live migration,
which is kind of like auto-converge algo but using dirty-limit
instead of cpu throttle to make migration convergent.
Enable dirty page limit if dirty_rate_high_cnt greater than 2
when dirty-limit capability enabled,
From: Hyman Huang(黄勇)
Guestperf tool does not enable diry ring feature when test
migration by default.
To support dirty ring migration performance test, introduce
dirty-ring-size option into guestperf tools, which ranges in
[1024, 65536].
To set dirty ring size with 4096 during migration test:
From: Hyman Huang(黄勇)
When tested large vcpu size vm with dirtylimit feature, Qemu crashed
due to the assertion in kvm_dirty_ring_reap_one, which assert that
vcpu's kvm_dirty_gfns has been allocated and not NULL.
Because dirty ring reaper thread races with Qemu main thread, reaper
may reap vcpu'
From: Hyman Huang(黄勇)
Introduce "vcpu-dirty-limit" migration parameter used
to limit dirty page rate during live migration.
"vcpu-dirty-limit" and "x-vcpu-dirty-limit-period" are
two dirty-limit-related migration parameters, which can
be set before and during live migration by qmp
migrate-set-pa
From: Hyman Huang(黄勇)
Introduce "x-vcpu-dirty-limit-period" migration experimental
parameter, which is in the range of 1 to 1000ms and used to
make dirtyrate calculation period configurable.
Signed-off-by: Hyman Huang(黄勇)
---
migration/migration.c | 26 ++
monitor/hmp-c
From: Hyman Huang(黄勇)
Add migration dirty-limit capability test if kernel support
dirty ring.
Migration dirty-limit capability introduce dirty limit
capability, two parameters: x-vcpu-dirty-limit-period and
x-vcpu-dirty-limit are introduced to implement the live
migration with dirty limit.
The
From: Hyman Huang(黄勇)
Guestperf tool does not cover the dirty-limit migration
currently, support this feature.
To enable dirty-limit, setting x-vcpu-dirty-limit-period
as 500ms and x-vcpu-dirty-limit as 10MB/s:
$ ./tests/migration/guestperf.py \
--dirty-limit --x-vcpu-dirty-limit-period 500
From: Hyman Huang(黄勇)
Introduce migration dirty-limit capability, which can
be turned on before live migration and limit dirty
page rate durty live migration.
Introduce migrate_dirty_limit function to help check
if dirty-limit capability enabled during live migration.
Meanwhile, refactor vcpu_d
From: Hyman Huang(黄勇)
v2:
This version make a little bit modifications comparing with
version 1 as following:
1. fix the overflow issue reported by Peter Maydell
2. add parameter check for hmp "set_vcpu_dirty_limit" command
3. fix the racing issue between dirty ring reaper thread and
Qemu mai
From: Hyman Huang(黄勇)
overity points out a overflow problem when computing MB,
dirty_ring_size and TARGET_PAGE_SIZE are both 32 bits,
multiplication will be done as a 32-bit operation, which
could overflow. Simplify the formula.
Meanwhile, fix spelling mistake of variable name.
Reported-by: Pet
From: Hyman Huang(黄勇)
dirty_rate paraemter of hmp command "set_vcpu_dirty_limit" is invalid
if less than 0, so add parameter check for it.
Signed-off-by: Hyman Huang(黄勇)
---
softmmu/dirtylimit.c | 5 +
1 file changed, 5 insertions(+)
diff --git a/softmmu/dirtylimit.c b/softmmu/dirtylimit.
From: Hyman Huang(黄勇)
Use vhost_user_save_acked_features to implemente acked features
saving.
Signed-off-by: Hyman Huang(黄勇)
Signed-off-by: Guoyi Tu
---
net/vhost-user.c | 6 +-
1 file changed, 1 insertion(+), 5 deletions(-)
diff --git a/net/vhost-user.c b/net/vhost-user.c
index f5cb095.
From: Hyman Huang(黄勇)
v4:
-rebase on master
-add stub function to fix build errors
-code clean on [PATCH v2 1/2]: drop 'cleanup' parameter in
vhost_user_save_acked_features.
-code clean on [PATCH v2 2/2]: make refactor of chr_closed_bh
a standalone patch.
Above changes are suggested by Michael
From: Hyman Huang(黄勇)
Abstract vhost acked features saving into
vhost_user_save_acked_features, export it as util function.
Signed-off-by: Hyman Huang(黄勇)
Signed-off-by: Guoyi Tu
---
include/net/vhost-user.h | 1 +
net/vhost-user.c | 21 +++--
2 files changed, 16 inse
From: Hyman Huang(黄勇)
This patch aims to fix unexpected negotiation features for
vhost-user netdev interface.
When openvswitch reconnect Qemu after an unexpected disconnection
and Qemu therefore start the vhost_dev, acked_features field in
vhost_dev is initialized with value fetched from acked_f
From: Hyman Huang(黄勇)
Introduce "info netdev" command so developers can play with
it easier.
Signed-off-by: Hyman Huang(黄勇)
---
hmp-commands-info.hx | 14 ++
include/monitor/hmp.h | 1 +
net/net.c | 31 +++
3 files changed, 46 insertions(+)
From: Hyman Huang(黄勇)
For netdev device that can offload virtio-net dataplane to slave,
such as vhost-net, vhost-user and vhost-vdpa, exporting it's
capability information and acked features would be more friendly for
developers. These infomation can be analyzed and compare to slave
capability pr
From: Hyman Huang(黄勇)
Add netdev information into output of hmp command hmp_info_network
so developers can analyze interface capability more easily.
Signed-off-by: Hyman Huang(黄勇)
---
net/net.c | 15 +++
1 file changed, 15 insertions(+)
diff --git a/net/net.c b/net/net.c
index c27
From: Hyman Huang(黄勇)
For vhost-user network device, Qemu backup the final features as
acked_features internally after guest acknowledging features during
virtio-net driver initialization, so the acked_features could be
used as input of VHOST_USER_SET_FEATURES command when slave device
restore fr
From: Hyman Huang(黄勇)
This series is enlightened by Michael when we fixed a virtio features
negotiation flaw, see the details here:
https://lore.kernel.org/qemu-devel/cover.1667136717.git.huang...@chinatelecom.cn/
And a test is suggested to be added to test behavior of virtio-net features
negoti
From: Hyman Huang(黄勇)
v3:
-rebase on master
-code clean on [PATCH v2 1/2]: keep the commit self-consistent and
do not modify the logic of saving acked_features. Just abstract the
util function.
-modify the [PATCH v2 2/2] logic: change the behavior of saving
acked_features in chr_closed_bh: sav
From: Hyman Huang(黄勇)
Save the acked_features once it be configured by guest
virtio driver so it can't miss any features.
Note that this patch also change the features saving logic
in chr_closed_bh, which originally backup features no matter
whether the features are 0 or not, but now do it only
From: Hyman Huang(黄勇)
Abstract vhost acked features saving into
vhost_user_save_acked_features, export it as util function.
Signed-off-by: Hyman Huang(黄勇)
Signed-off-by: Guoyi Tu
---
include/net/vhost-user.h | 2 ++
net/vhost-user.c | 29 ++---
2 files changed
From: Hyman Huang(黄勇)
Abstract vhost acked features saving into
vhost_user_save_acked_features, export it as util function.
Signed-off-by: Hyman Huang(黄勇)
Signed-off-by: Guoyi Tu
---
include/net/vhost-user.h | 2 ++
net/vhost-user.c | 35 +++
2 files c
From: Hyman Huang(黄勇)
Save the acked_features once it be configured by guest
virtio driver so it can't miss any features.
Signed-off-by: Hyman Huang(黄勇)
Signed-off-by: Guoyi Tu
---
hw/net/vhost_net.c | 9 +
hw/net/virtio-net.c | 5 +
include/net/vhost_net.h | 2 ++
3 file
From: Hyman Huang(黄勇)
v2:
Fix the typo in subject of [PATCH v2 2/2]
v1:
This is the version 1 of the series and it is exactly the same as
RFC version, but fixing a typo in subject, which is reported by Michael.
As for test for the behavior suggested by Michael, IMHO, it could be
post in anoth
From: Hyman Huang(黄勇)
Save the acked_features once it be configured by guest
virtio driver so it can't miss any features.
Signed-off-by: Hyman Huang(黄勇)
Signed-off-by: Guoyi Tu
---
hw/net/vhost_net.c | 9 +
hw/net/virtio-net.c | 5 +
include/net/vhost_net.h | 2 ++
3 file
From: Hyman Huang(黄勇)
Abstract vhost acked features saving into
vhost_user_save_acked_features, export it as util function.
Signed-off-by: Hyman Huang(黄勇)
Signed-off-by: Guoyi Tu
---
include/net/vhost-user.h | 2 ++
net/vhost-user.c | 35 +++
2 files c
From: Hyman Huang(黄勇)
v1:
This is the version 1 of the series and it is exactly the same as
RFC version, but fixing a typo in subject, which is reported by Michael.
As for test for the behavior suggested by Michael, IMHO, it could be
post in another series, since i found that testing the negoti
From: "Hyman Huang(黄勇)"
This patchset aim to fix the unexpected negotiation features for
vhost-user netdev interface.
Steps to reproduce the issue:
Prepare a vm (CentOS 8 in my work scenario) with vhost-user
backend interface and configure qemu as server mode. So dpdk
would connect qemu's unix
From: Hyman Huang(黄勇)
Abstract vhost acked features saving into
vhost_user_save_acked_features, export it as util function.
Signed-off-by: Hyman Huang(黄勇)
Signed-off-by: Guoyi Tu
---
include/net/vhost-user.h | 2 ++
net/vhost-user.c | 35 +++
2 files c
From: Hyman Huang(黄勇)
Save the acked_features once it be configured by guest
virtio driver so it can't miss any features.
Signed-off-by: Hyman Huang(黄勇)
Signed-off-by: Guoyi Tu
---
hw/net/vhost_net.c | 9 +
hw/net/virtio-net.c | 5 +
include/net/vhost_net.h | 2 ++
3 file
From: Hyman Huang(黄勇)
v1:
- make parameter vcpu-dirty-limit experimental
- switch dirty limit off when cancel migrate
- add cancel logic in migration test
Please review, thanks,
Yong
Abstract
This series added a new migration capability called "dirtylimit". It can
be enabled when
From: Hyman Huang(黄勇)
Introduce migration dirty-limit capability, which can
be turned on before live migration and limit dirty
page rate durty live migration.
Introduce migrate_dirty_limit function to help check
if dirty-limit capability enabled during live migration.
Meanwhile, refactor vcpu_d
From: Hyman Huang(黄勇)
Guestperf tool does not cover the dirty-limit migration
currently, support this feature.
To enable dirty-limit, setting x-vcpu-dirty-limit-period
as 500ms and x-vcpu-dirty-limit as 10MB/s:
$ ./tests/migration/guestperf.py \
--dirty-limit --x-vcpu-dirty-limit-period 500
From: Hyman Huang(黄勇)
Implement dirty-limit convergence algo for live migration,
which is kind of like auto-converge algo but using dirty-limit
instead of cpu throttle to make migration convergent.
Signed-off-by: Hyman Huang(黄勇)
---
migration/migration.c | 1 +
migration/ram.c| 53 ++
From: Hyman Huang(黄勇)
Export dirty limit throttle time and estimated ring full
time, through which we can observe the process of dirty
limit during live migration.
Signed-off-by: Hyman Huang(黄勇)
---
include/sysemu/dirtylimit.h | 2 ++
migration/migration.c | 10 ++
monitor/hmp-c
From: Hyman Huang(黄勇)
Guestperf tool does not enable diry ring feature when test
migration by default.
To support dirty ring migration performance test, introduce
dirty-ring-size option into guestperf tools, which ranges in
[1024, 65536].
To set dirty ring size with 4096 during migration test:
From: Hyman Huang(黄勇)
Introduce "x-vcpu-dirty-limit" migration parameter used
to limit dirty page rate during live migration.
"x-vcpu-dirty-limit" and "x-vcpu-dirty-limit-period" are
two dirty-limit-related migration parameters, which can
be set before and during live migration by qmp
migrate-se
From: Hyman Huang(黄勇)
Add migration dirty-limit capability test if kernel support
dirty ring.
Migration dirty-limit capability introduce dirty limit
capability, two parameters: x-vcpu-dirty-limit-period and
x-vcpu-dirty-limit are introduced to implement the live
migration with dirty limit.
The
From: Hyman Huang(黄勇)
Introduce "x-vcpu-dirty-limit-period" migration experimental
parameter, which is used to make dirtyrate calculation period
configurable.
Signed-off-by: Hyman Huang(黄勇)
---
migration/migration.c | 16
monitor/hmp-cmds.c| 8
qapi/migration.jso
From: Hyman Huang(黄勇)
Coverity points out a overflow problem when computing MB,
dirty_ring_size and TARGET_PAGE_SIZE are both 32 bits,
multiplication will be done as a 32-bit operation, which
could overflow. Simplify the formula.
Meanwhile, fix spelling mistake of variable name.
Reported-by: Pe
From: Hyman Huang(黄勇)
Coverity points out a overflow problem when computing MB,
dirty_ring_size and TARGET_PAGE_SIZE are both 32 bits,
multiplication will be done as a 32-bit operation, which
could overflow. Simplify the formula.
Meanwhile, fix spelling mistake of variable name.
Reported-by: Pe
From: Hyman Huang(黄勇)
Guestperf tool does not cover the dirty-limit migration
currently, support this feature.
To enable dirty-limit, setting x-vcpu-dirty-limit-period
as 500ms and vcpu-dirty-limit as 10MB/s:
$ ./tests/migration/guestperf.py \
--dirty-limit --x-vcpu-dirty-limit-period 500 \
From: Hyman Huang(黄勇)
Add migration dirty-limit capability test if kernel support
dirty ring.
Migration dirty-limit capability introduce dirty limit
capability, two parameters: x-vcpu-dirty-limit-period and
vcpu-dirty-limit are introduced to implement the live
migration with dirty limit.
The te
From: Hyman Huang(黄勇)
Export dirty limit throttle time and estimated ring full
time, through which we can observe the process of dirty
limit during live migration.
Signed-off-by: Hyman Huang(黄勇)
---
include/sysemu/dirtylimit.h | 2 ++
migration/migration.c | 10 ++
monitor/hmp-c
From: Hyman Huang(黄勇)
Introduce migration dirty-limit capability, which can
be turned on before live migration and limit dirty
page rate durty live migration.
Introduce migrate_dirty_limit function to help check
if dirty-limit capability enabled during live migration.
Meanwhile, refactor vcpu_d
From: Hyman Huang(黄勇)
Introduce "x-vcpu-dirty-limit-period" migration experimental
parameter, which is used to make dirtyrate calculation period
configurable.
Signed-off-by: Hyman Huang(黄勇)
---
migration/migration.c | 16
monitor/hmp-cmds.c| 8
qapi/migration.jso
From: Hyman Huang(黄勇)
Implement dirty-limit convergence algo for live migration,
which is kind of like auto-converge algo but using dirty-limit
instead of cpu throttle to make migration convergent.
Signed-off-by: Hyman Huang(黄勇)
---
migration/ram.c| 53 +
From: Hyman Huang(黄勇)
Guestperf tool does not enable diry ring feature when test
migration by default.
To support dirty ring migration performance test, introduce
dirty-ring-size option into guestperf tools, which ranges in
[1024, 65536].
To set dirty ring size with 4096 during migration test:
From: Hyman Huang(黄勇)
Introduce "vcpu-dirty-limit" migration parameter used
to limit dirty page rate during live migration.
"vcpu-dirty-limit" and "x-vcpu-dirty-limit-period" are
two dirty-limit-related migration parameters, which can
be set before and during live migration by qmp
migrate-set-pa
From: Hyman Huang(黄勇)
Abstract
This series added a new migration capability called "dirtylimit". It can
be enabled when dirty ring is enabled, and it'll improve the vCPU performance
during the process of migration. It is based on the previous patchset:
https://lore.kernel.org/qemu-deve
From: Hyman Huang(黄勇)
Setup a negative feedback system when vCPU thread
handling KVM_EXIT_DIRTY_RING_FULL exit by introducing
throttle_us_per_full field in struct CPUState. Sleep
throttle_us_per_full microseconds to throttle vCPU
if dirtylimit is in service.
Signed-off-by: Hyman Huang(黄勇)
Revie
From: Hyman Huang(黄勇)
Add dirty page rate limit test if kernel support dirty ring,
The following qmp commands are covered by this test case:
"calc-dirty-rate", "query-dirty-rate", "set-vcpu-dirty-limit",
"cancel-vcpu-dirty-limit" and "query-vcpu-dirty-limit".
Signed-off-by: Hyman Huang(黄勇)
Ack
1 - 100 of 413 matches
Mail list logo
