* [PATCH v14 0/7] support dirty restraint on vCPU
@ 2022-02-10 16:17 huangy81
2022-02-10 16:17 ` [PATCH v14 1/7] accel/kvm/kvm-all: Refactor per-vcpu dirty ring reaping huangy81
` (6 more replies)
0 siblings, 7 replies; 22+ messages in thread
From: huangy81 @ 2022-02-10 16:17 UTC (permalink / raw)
To: qemu-devel
Cc: Eduardo Habkost, David Hildenbrand, Hyman, Juan Quintela,
Richard Henderson, Markus ArmBruster, Peter Xu,
Dr. David Alan Gilbert, Paolo Bonzini,
Philippe Mathieu-Daudé
From: Hyman Huang(黄勇) <huangy81@chinatelecom.cn>
v14
- v13 sent by accident, resend patchset.
v13
- rebase on master
- passing NULL to kvm_dirty_ring_reap in commit
"refactor per-vcpu dirty ring reaping" to keep the logic unchanged.
In other word, we still try the best to reap as much PFNs as possible
if dirtylimit not in service.
- move the cpu list gen id changes into a separate patch.
- release the lock before sleep during dirty page rate calculation.
- move the dirty ring size fetch logic into a separate patch.
- drop the DIRTYLIMIT_LINEAR_ADJUSTMENT_WATERMARK MACRO .
- substitute bh with function pointer when implement dirtylimit.
- merge the dirtylimit_start/stop into dirtylimit_change.
- fix "cpu-index" parameter type with "int" to keep consistency.
- fix some syntax error in documents.
Please review. Thanks,
Yong
v12
- rebase on master
- add a new commmit to refactor per-vcpu dirty ring reaping, which can resolve
the "vcpu miss the chances to sleep" problem
- remove the dirtylimit_thread and implemtment throttle in bottom half instead.
- let the dirty ring reaper thread keep sleeping when dirtylimit is in service
- introduce cpu_list_generation_id to identify cpu_list changing.
- keep taking the cpu_list_lock during dirty_stat_wait to prevent vcpu plug/unplug
when calculating the dirty page rate
- move the dirtylimit global initializations out of dirtylimit_set_vcpu and do
some code clean
- add DIRTYLIMIT_LINEAR_ADJUSTMENT_WATERMARK in case of oscillation when throttling
- remove the unmatched count field in dirtylimit_state
- add stub to fix build on non-x86
- refactor the documents
Thanks Peter and Markus for reviewing the previous versions, please review.
Thanks,
Yong
v11
- rebase on master
- add a commit " refactor dirty page rate calculation" so that dirty page rate limit
can reuse the calculation logic.
- handle the cpu hotplug/unplug case in the dirty page rate calculation logic.
- modify the qmp commands according to Markus's advice.
- introduce a standalone file dirtylimit.c to implement dirty page rate limit
- check if dirty limit in service by dirtylimit_state pointer instead of global variable
- introduce dirtylimit_mutex to protect dirtylimit_state
- do some code clean and docs
See the commit for more detail, thanks Markus and Peter very mush for the code
review and give the experienced and insightful advices, most modifications are
based on these advices.
v10:
- rebase on master
- make the following modifications on patch [1/3]:
1. Make "dirtylimit-calc" thread joinable and join it after quitting.
2. Add finalize function to free dirtylimit_calc_state
3. Do some code clean work
- make the following modifications on patch [2/3]:
1. Remove the original implementation of throttle according to
Peter's advice.
2. Introduce a negative feedback system and implement the throttle
on all vcpu in one thread named "dirtylimit".
3. Simplify the algo when calculation the throttle_us_per_full:
increase/decrease linearly when there exists a wide difference
between quota and current dirty page rate, increase/decrease
a fixed time slice when the difference is narrow. This makes
throttle responds faster and reach the quota smoothly.
4. Introduce a unfit_cnt in algo to make sure throttle really
takes effect.
5. Set the max sleep time 99 times more than "ring_full_time_us".
6. Make "dirtylimit" thread joinable and join it after quitting.
- make the following modifications on patch [3/3]:
1. Remove the unplug cpu handling logic.
2. "query-vcpu-dirty-limit" only return dirtylimit information of
vcpus that enable dirtylimit
3. Remove the "dirtylimit_setup" function
4. Trigger the dirtylimit and initialize the global state only
when someone enable dirtylimit, and finalize it after the last
dirtylimit be canceled.
5. Redefine the qmp command vcpu-dirty-limit/query-vcpu-dirty-limit:
enable/disable dirtylimit use a single command "vcpu-dirty-limit",
to enable/disabled dirtylimit on specified vcpu only if "cpu-index"
is specified, otherwise, all vcpu will be affected.
6. Redefine the hmp command vcpu_dirty_limit/info vcpu_dirty_limit
- other points about the code review:
1. "merge the code of calculation dirty page rate"
I think maybe it's not suitable to touch the 'calc-dirty-rate',
because 'calc-dirty-rate' will stop sync log after calculating
the dirtyrate and the 'dirtylimit-cal' will not untill the last
dirtylimit be canceled, if we merge the GLOBAL_DIRTY_LIMIT into
GLOBAL_DIRTY_DIRTYRATE, the two are interacted with each other.
2. The new implementaion of throttle algo enlightened by Peter
responds faster and consume less cpu resource than the older,
we make a impressed progress.
And there is a viewpoint may be discussed, it is that the new
throttle logic is "passive", vcpu sleeps only after dirty ring,
is full, unlike the "auto-converge" which will kick vcpu instead
in a fixed slice time. If the vcpu is memory-write intensive
and the ring size is large, it will produce dirty memory during
the dirty ring full time and the throttle works not so good, it
means the throttle depends on the dirty ring size.
I actually tested the new algo in two case:
case 1: dirty-ring-size: 4096, dirtyrate: 1170MB/s
result: minimum quota dirtyrate is 25MB/s or even less
minimum vcpu util is 6%
case 2: dirty-ring-size: 65536, dirtyrate: 1170MB/s
result: minimum quota dirtyrate is 256MB/s
minimum vcpu util is 24%
I post this just for discussion, i think this is not a big deal
beacase if we set the dirty-ring-size to the maximum value(65536),
we assume the server's bandwidth is capable of handling it.
3. I hard-code the acceptable deviation value to 25MB/s, see the
macro DIRTYLIMIT_TOLERANCE_RANGE. I'm struggling to decide
whether to let it configurable
4. Another point is the unplug cpu handle, current algo affects the
unplugged vcpu, if we set dirty limit on it, we should fork 2
thread "dirtylimit" and "dirtylimit-calc" but do nothing, once the
vcpu is hot-plugged, dirty limit works, i think the logic is ok
but still there can be different advice.
- to let developers play with it easier, i post the hmp usage example:
(qemu) vcpu_dirty_limit -g on -1 500
[Please use 'info vcpu_dirty_limit' to query dirty limit for virtual CPU]
(qemu) info vcpu_dirty_limit
vcpu[0], limit rate 500 (MB/s), current rate 415 (MB/s)
vcpu[1], limit rate 500 (MB/s), current rate 496 (MB/s)
vcpu[2], limit rate 500 (MB/s), current rate 0 (MB/s)
vcpu[3], limit rate 500 (MB/s), current rate 0 (MB/s)
(qemu) vcpu_dirty_limit -g off
[Please use 'info vcpu_dirty_limit' to query dirty limit for virtual CPU]
(qemu) info vcpu_dirty_limit
Dirty page limit not enabled!
(qemu) vcpu_dirty_limit on 0 300
[Please use 'info vcpu_dirty_limit' to query dirty limit for virtual CPU]
(qemu) vcpu_dirty_limit on 1 500
[Please use 'info vcpu_dirty_limit' to query dirty limit for virtual CPU]
(qemu) info vcpu_dirty_limit
vcpu[0], limit rate 300 (MB/s), current rate 342 (MB/s)
vcpu[1], limit rate 500 (MB/s), current rate 485 (MB/s)
(qemu) vcpu_dirty_limit off 0
[Please use 'info vcpu_dirty_limit' to query dirty limit for virtual CPU]
(qemu) info vcpu_dirty_limit
vcpu[1], limit rate 500 (MB/s), current rate 528 (MB/s)
(qemu) vcpu_dirty_limit off 1
[Please use 'info vcpu_dirty_limit' to query dirty limit for virtual CPU]
(qemu) info vcpu_dirty_limit
Dirty page limit not enabled!
Thanks very much for the instructive algo suggestion given by Peter,
the comment and other code reviews made by Markus.
Please review, thanks!
v9:
- rebase on master
- fix the meson directory change, keep it untouched.
v8:
- rebase on master
- polish the error message and remove the "unlikely" compilation syntax
according to the advice given by Markus.
- keep the dirty tracking enabled during "dirtylimit-calc" lifecycle
so that the overhead can be reduced according to the advice given by
Peter.
- merge the "set/cancel" qmp commands into one named "vcpu-dirty-limit"
and introduce qmp command "query-vcpu-dirty-limit" to query dirty
limit information about virtual CPU, according to the advice given by
Peter.
- check if vcpu index is valid and handle the unplug case before
enabling, disabling dirty limit for virtual CPU.
- introduce hmp commands so developers can play with them easier, use
"vcpu_dirty_limit" to enable dirty limit and "info vcpu_dirty_limit"
to query.
The patch [2/3] has not been touched so far. Any corrections and
suggetions are welcome.
Please review, thanks!
v7:
- rebase on master
- polish the comments and error message according to the
advices given by Markus
- introduce dirtylimit_enabled function to pre-check if dirty
page limit is enabled before canceling.
v6:
- rebase on master
- fix dirtylimit setup crash found by Markus
- polish the comments according to the advice given by Markus
- adjust the qemu qmp command tag to 7.0
v5:
- rebase on master
- adjust the throttle algorithm by removing the tuning in
RESTRAINT_RATIO case so that dirty page rate could reachs the quota
more quickly.
- fix percentage update in throttle iteration.
v4:
- rebase on master
- modify the following points according to the advice given by Markus
1. move the defination into migration.json
2. polish the comments of set-dirty-limit
3. do the syntax check and change dirty rate to dirty page rate
Thanks for the carefule reviews made by Markus.
Please review, thanks!
v3:
- rebase on master
- modify the following points according to the advice given by Markus
1. remove the DirtyRateQuotaVcpu and use its field as option directly
2. add comments to show details of what dirtylimit setup do
3. explain how to use dirtylimit in combination with existing qmp
commands "calc-dirty-rate" and "query-dirty-rate" in documentation.
Thanks for the carefule reviews made by Markus.
Please review, thanks!
Hyman
v2:
- rebase on master
- modify the following points according to the advices given by Juan
1. rename dirtyrestraint to dirtylimit
2. implement the full lifecyle function of dirtylimit_calc, include
dirtylimit_calc and dirtylimit_calc_quit
3. introduce 'quit' field in dirtylimit_calc_state to implement the
dirtylimit_calc_quit
4. remove the ready_cond and ready_mtx since it may not be suitable
5. put the 'record_dirtypage' function code at the beggining of the
file
6. remove the unnecesary return;
- other modifications has been made after code review
1. introduce 'bmap' and 'nr' field in dirtylimit_state to record the
number of running thread forked by dirtylimit
2. stop the dirtyrate calculation thread if all the dirtylimit thread
are stopped
3. do some renaming works
dirtyrate calulation thread -> dirtylimit-calc
dirtylimit thread -> dirtylimit-{cpu_index}
function name do_dirtyrestraint -> dirtylimit_check
qmp command dirty-restraint -> set-drity-limit
qmp command dirty-restraint-cancel -> cancel-dirty-limit
header file dirtyrestraint.h -> dirtylimit.h
Please review, thanks !
thanks for the accurate and timely advices given by Juan. we really
appreciate it if corrections and suggetions about this patchset are
proposed.
Best Regards !
Hyman
v1:
this patchset introduce a mechanism to impose dirty restraint
on vCPU, aiming to keep the vCPU running in a certain dirtyrate
given by user. dirty restraint on vCPU maybe an alternative
method to implement convergence logic for live migration,
which could improve guest memory performance during migration
compared with traditional method in theory.
For the current live migration implementation, the convergence
logic throttles all vCPUs of the VM, which has some side effects.
-'read processes' on vCPU will be unnecessarily penalized
- throttle increase percentage step by step, which seems
struggling to find the optimal throttle percentage when
dirtyrate is high.
- hard to predict the remaining time of migration if the
throttling percentage reachs 99%
to a certain extent, the dirty restraint machnism can fix these
effects by throttling at vCPU granularity during migration.
the implementation is rather straightforward, we calculate
vCPU dirtyrate via the Dirty Ring mechanism periodically
as the commit 0e21bf246 "implement dirty-ring dirtyrate calculation"
does, for vCPU that be specified to impose dirty restraint,
we throttle it periodically as the auto-converge does, once after
throttling, we compare the quota dirtyrate with current dirtyrate,
if current dirtyrate is not under the quota, increase the throttling
percentage until current dirtyrate is under the quota.
this patchset is the basis of implmenting a new auto-converge method
for live migration, we introduce two qmp commands for impose/cancel
the dirty restraint on specified vCPU, so it also can be an independent
api to supply the upper app such as libvirt, which can use it to
implement the convergence logic during live migration, supplemented
with the qmp 'calc-dirty-rate' command or whatever.
we post this patchset for RFC and any corrections and suggetions about
the implementation, api, throttleing algorithm or whatever are very
appreciated!
Please review, thanks !
Best Regards !
Hyman Huang (7):
accel/kvm/kvm-all: Refactor per-vcpu dirty ring reaping
cpus: Introduce cpu_list_generation_id
migration/dirtyrate: Refactor dirty page rate calculation
softmmu/dirtylimit: Implement vCPU dirtyrate calculation periodically
accel/kvm/kvm-all: Introduce kvm_dirty_ring_size function
softmmu/dirtylimit: Implement virtual CPU throttle
softmmu/dirtylimit: Implement dirty page rate limit
accel/kvm/kvm-all.c | 40 ++-
accel/stubs/kvm-stub.c | 5 +
cpus-common.c | 8 +
hmp-commands-info.hx | 13 +
hmp-commands.hx | 32 +++
include/exec/cpu-common.h | 1 +
include/exec/memory.h | 5 +-
include/hw/core/cpu.h | 6 +
include/monitor/hmp.h | 3 +
include/sysemu/dirtylimit.h | 37 +++
include/sysemu/dirtyrate.h | 29 +++
include/sysemu/kvm.h | 2 +
migration/dirtyrate.c | 222 ++++++++++------
migration/dirtyrate.h | 7 +-
qapi/migration.json | 80 ++++++
softmmu/dirtylimit.c | 620 ++++++++++++++++++++++++++++++++++++++++++++
softmmu/meson.build | 1 +
softmmu/trace-events | 8 +
18 files changed, 1023 insertions(+), 96 deletions(-)
create mode 100644 include/sysemu/dirtylimit.h
create mode 100644 include/sysemu/dirtyrate.h
create mode 100644 softmmu/dirtylimit.c
--
1.8.3.1
^ permalink raw reply [flat|nested] 22+ messages in thread
* [PATCH v14 1/7] accel/kvm/kvm-all: Refactor per-vcpu dirty ring reaping
2022-02-10 16:17 [PATCH v14 0/7] support dirty restraint on vCPU huangy81
@ 2022-02-10 16:17 ` huangy81
2022-02-14 7:57 ` Peter Xu
2022-02-10 16:17 ` [PATCH v14 2/7] cpus: Introduce cpu_list_generation_id huangy81
` (5 subsequent siblings)
6 siblings, 1 reply; 22+ messages in thread
From: huangy81 @ 2022-02-10 16:17 UTC (permalink / raw)
To: qemu-devel
Cc: Eduardo Habkost, David Hildenbrand, Hyman, Juan Quintela,
Richard Henderson, Markus ArmBruster, Peter Xu,
Dr. David Alan Gilbert, Paolo Bonzini,
Philippe Mathieu-Daudé
From: Hyman Huang(黄勇) <huangy81@chinatelecom.cn>
Add a non-required argument 'CPUState' to kvm_dirty_ring_reap so
that it can cover single vcpu dirty-ring-reaping scenario.
Signed-off-by: Hyman Huang(黄勇) <huangy81@chinatelecom.cn>
---
accel/kvm/kvm-all.c | 23 +++++++++++++----------
1 file changed, 13 insertions(+), 10 deletions(-)
diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c
index 0e66ebb..7b06b8a 100644
--- a/accel/kvm/kvm-all.c
+++ b/accel/kvm/kvm-all.c
@@ -756,17 +756,20 @@ static uint32_t kvm_dirty_ring_reap_one(KVMState *s, CPUState *cpu)
}
/* Must be with slots_lock held */
-static uint64_t kvm_dirty_ring_reap_locked(KVMState *s)
+static uint64_t kvm_dirty_ring_reap_locked(KVMState *s, CPUState* cpu)
{
int ret;
- CPUState *cpu;
uint64_t total = 0;
int64_t stamp;
stamp = get_clock();
- CPU_FOREACH(cpu) {
- total += kvm_dirty_ring_reap_one(s, cpu);
+ if (cpu) {
+ total = kvm_dirty_ring_reap_one(s, cpu);
+ } else {
+ CPU_FOREACH(cpu) {
+ total += kvm_dirty_ring_reap_one(s, cpu);
+ }
}
if (total) {
@@ -787,7 +790,7 @@ static uint64_t kvm_dirty_ring_reap_locked(KVMState *s)
* Currently for simplicity, we must hold BQL before calling this. We can
* consider to drop the BQL if we're clear with all the race conditions.
*/
-static uint64_t kvm_dirty_ring_reap(KVMState *s)
+static uint64_t kvm_dirty_ring_reap(KVMState *s, CPUState *cpu)
{
uint64_t total;
@@ -807,7 +810,7 @@ static uint64_t kvm_dirty_ring_reap(KVMState *s)
* reset below.
*/
kvm_slots_lock();
- total = kvm_dirty_ring_reap_locked(s);
+ total = kvm_dirty_ring_reap_locked(s, cpu);
kvm_slots_unlock();
return total;
@@ -854,7 +857,7 @@ static void kvm_dirty_ring_flush(void)
* vcpus out in a synchronous way.
*/
kvm_cpu_synchronize_kick_all();
- kvm_dirty_ring_reap(kvm_state);
+ kvm_dirty_ring_reap(kvm_state, NULL);
trace_kvm_dirty_ring_flush(1);
}
@@ -1398,7 +1401,7 @@ static void kvm_set_phys_mem(KVMMemoryListener *kml,
* Not easy. Let's cross the fingers until it's fixed.
*/
if (kvm_state->kvm_dirty_ring_size) {
- kvm_dirty_ring_reap_locked(kvm_state);
+ kvm_dirty_ring_reap_locked(kvm_state, NULL);
} else {
kvm_slot_get_dirty_log(kvm_state, mem);
}
@@ -1470,7 +1473,7 @@ static void *kvm_dirty_ring_reaper_thread(void *data)
r->reaper_state = KVM_DIRTY_RING_REAPER_REAPING;
qemu_mutex_lock_iothread();
- kvm_dirty_ring_reap(s);
+ kvm_dirty_ring_reap(s, NULL);
qemu_mutex_unlock_iothread();
r->reaper_iteration++;
@@ -2956,7 +2959,7 @@ int kvm_cpu_exec(CPUState *cpu)
*/
trace_kvm_dirty_ring_full(cpu->cpu_index);
qemu_mutex_lock_iothread();
- kvm_dirty_ring_reap(kvm_state);
+ kvm_dirty_ring_reap(kvm_state, NULL);
qemu_mutex_unlock_iothread();
ret = 0;
break;
--
1.8.3.1
^ permalink raw reply related [flat|nested] 22+ messages in thread
* [PATCH v14 2/7] cpus: Introduce cpu_list_generation_id
2022-02-10 16:17 [PATCH v14 0/7] support dirty restraint on vCPU huangy81
2022-02-10 16:17 ` [PATCH v14 1/7] accel/kvm/kvm-all: Refactor per-vcpu dirty ring reaping huangy81
@ 2022-02-10 16:17 ` huangy81
2022-02-14 7:59 ` Peter Xu
2022-02-10 16:17 ` [PATCH v14 3/7] migration/dirtyrate: Refactor dirty page rate calculation huangy81
` (4 subsequent siblings)
6 siblings, 1 reply; 22+ messages in thread
From: huangy81 @ 2022-02-10 16:17 UTC (permalink / raw)
To: qemu-devel
Cc: Eduardo Habkost, David Hildenbrand, Hyman, Juan Quintela,
Richard Henderson, Markus ArmBruster, Peter Xu,
Dr. David Alan Gilbert, Paolo Bonzini,
Philippe Mathieu-Daudé
From: Hyman Huang(黄勇) <huangy81@chinatelecom.cn>
Introduce cpu_list_generation_id to track cpu list generation so
that cpu hotplug/unplug can be detected during measurement of
dirty page rate.
Signed-off-by: Hyman Huang(黄勇) <huangy81@chinatelecom.cn>
---
cpus-common.c | 8 ++++++++
include/exec/cpu-common.h | 1 +
2 files changed, 9 insertions(+)
diff --git a/cpus-common.c b/cpus-common.c
index 6e73d3e..31c6415 100644
--- a/cpus-common.c
+++ b/cpus-common.c
@@ -73,6 +73,12 @@ static int cpu_get_free_index(void)
}
CPUTailQ cpus = QTAILQ_HEAD_INITIALIZER(cpus);
+static unsigned int cpu_list_generation_id;
+
+unsigned int cpu_list_generation_id_get(void)
+{
+ return cpu_list_generation_id;
+}
void cpu_list_add(CPUState *cpu)
{
@@ -84,6 +90,7 @@ void cpu_list_add(CPUState *cpu)
assert(!cpu_index_auto_assigned);
}
QTAILQ_INSERT_TAIL_RCU(&cpus, cpu, node);
+ cpu_list_generation_id++;
}
void cpu_list_remove(CPUState *cpu)
@@ -96,6 +103,7 @@ void cpu_list_remove(CPUState *cpu)
QTAILQ_REMOVE_RCU(&cpus, cpu, node);
cpu->cpu_index = UNASSIGNED_CPU_INDEX;
+ cpu_list_generation_id++;
}
CPUState *qemu_get_cpu(int index)
diff --git a/include/exec/cpu-common.h b/include/exec/cpu-common.h
index de5f444..eb33642 100644
--- a/include/exec/cpu-common.h
+++ b/include/exec/cpu-common.h
@@ -20,6 +20,7 @@ extern intptr_t qemu_host_page_mask;
void qemu_init_cpu_list(void);
void cpu_list_lock(void);
void cpu_list_unlock(void);
+unsigned int cpu_list_generation_id_get(void);
void tcg_flush_softmmu_tlb(CPUState *cs);
--
1.8.3.1
^ permalink raw reply related [flat|nested] 22+ messages in thread
* [PATCH v14 3/7] migration/dirtyrate: Refactor dirty page rate calculation
2022-02-10 16:17 [PATCH v14 0/7] support dirty restraint on vCPU huangy81
2022-02-10 16:17 ` [PATCH v14 1/7] accel/kvm/kvm-all: Refactor per-vcpu dirty ring reaping huangy81
2022-02-10 16:17 ` [PATCH v14 2/7] cpus: Introduce cpu_list_generation_id huangy81
@ 2022-02-10 16:17 ` huangy81
2022-02-14 7:57 ` Peter Xu
2022-02-10 16:17 ` [PATCH v14 4/7] softmmu/dirtylimit: Implement vCPU dirtyrate calculation periodically huangy81
` (3 subsequent siblings)
6 siblings, 1 reply; 22+ messages in thread
From: huangy81 @ 2022-02-10 16:17 UTC (permalink / raw)
To: qemu-devel
Cc: Eduardo Habkost, David Hildenbrand, Hyman, Juan Quintela,
Richard Henderson, Markus ArmBruster, Peter Xu,
Dr. David Alan Gilbert, Paolo Bonzini,
Philippe Mathieu-Daudé
From: Hyman Huang(黄勇) <huangy81@chinatelecom.cn>
abstract out dirty log change logic into function
global_dirty_log_change.
abstract out dirty page rate calculation logic via
dirty-ring into function vcpu_calculate_dirtyrate.
abstract out mathematical dirty page rate calculation
into do_calculate_dirtyrate, decouple it from DirtyStat.
rename set_sample_page_period to dirty_stat_wait, which
is well-understood and will be reused in dirtylimit.
handle cpu hotplug/unplug scenario during measurement of
dirty page rate.
export util functions outside migration.
Signed-off-by: Hyman Huang(黄勇) <huangy81@chinatelecom.cn>
---
include/sysemu/dirtyrate.h | 29 ++++++
migration/dirtyrate.c | 222 +++++++++++++++++++++++++++++----------------
migration/dirtyrate.h | 7 +-
3 files changed, 173 insertions(+), 85 deletions(-)
create mode 100644 include/sysemu/dirtyrate.h
diff --git a/include/sysemu/dirtyrate.h b/include/sysemu/dirtyrate.h
new file mode 100644
index 0000000..cb6f02b
--- /dev/null
+++ b/include/sysemu/dirtyrate.h
@@ -0,0 +1,29 @@
+/*
+ * dirty page rate helper functions
+ *
+ * Copyright (c) 2022 CHINA TELECOM CO.,LTD.
+ *
+ * Authors:
+ * Hyman Huang(黄勇) <huangy81@chinatelecom.cn>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+ * See the COPYING file in the top-level directory.
+ */
+
+#ifndef QEMU_DIRTYRATE_H
+#define QEMU_DIRTYRATE_H
+
+typedef struct VcpuStat {
+ int nvcpu; /* number of vcpu */
+ DirtyRateVcpu *rates; /* array of dirty rate for each vcpu */
+} VcpuStat;
+
+int64_t vcpu_calculate_dirtyrate(int64_t calc_time_ms,
+ int64_t init_time_ms,
+ VcpuStat *stat,
+ unsigned int flag,
+ bool one_shot);
+
+void global_dirty_log_change(unsigned int flag,
+ bool start);
+#endif
diff --git a/migration/dirtyrate.c b/migration/dirtyrate.c
index d65e744..6be6664 100644
--- a/migration/dirtyrate.c
+++ b/migration/dirtyrate.c
@@ -46,7 +46,7 @@ static struct DirtyRateStat DirtyStat;
static DirtyRateMeasureMode dirtyrate_mode =
DIRTY_RATE_MEASURE_MODE_PAGE_SAMPLING;
-static int64_t set_sample_page_period(int64_t msec, int64_t initial_time)
+static int64_t dirty_stat_wait(int64_t msec, int64_t initial_time)
{
int64_t current_time;
@@ -60,6 +60,130 @@ static int64_t set_sample_page_period(int64_t msec, int64_t initial_time)
return msec;
}
+static inline void record_dirtypages(DirtyPageRecord *dirty_pages,
+ CPUState *cpu, bool start)
+{
+ if (start) {
+ dirty_pages[cpu->cpu_index].start_pages = cpu->dirty_pages;
+ } else {
+ dirty_pages[cpu->cpu_index].end_pages = cpu->dirty_pages;
+ }
+}
+
+static int64_t do_calculate_dirtyrate(DirtyPageRecord dirty_pages,
+ int64_t calc_time_ms)
+{
+ uint64_t memory_size_MB;
+ uint64_t increased_dirty_pages =
+ dirty_pages.end_pages - dirty_pages.start_pages;
+
+ memory_size_MB = (increased_dirty_pages * TARGET_PAGE_SIZE) >> 20;
+
+ return memory_size_MB * 1000 / calc_time_ms;
+}
+
+void global_dirty_log_change(unsigned int flag, bool start)
+{
+ qemu_mutex_lock_iothread();
+ if (start) {
+ memory_global_dirty_log_start(flag);
+ } else {
+ memory_global_dirty_log_stop(flag);
+ }
+ qemu_mutex_unlock_iothread();
+}
+
+/*
+ * global_dirty_log_sync
+ * 1. sync dirty log from kvm
+ * 2. stop dirty tracking if needed.
+ */
+static void global_dirty_log_sync(unsigned int flag, bool one_shot)
+{
+ qemu_mutex_lock_iothread();
+ memory_global_dirty_log_sync();
+ if (one_shot) {
+ memory_global_dirty_log_stop(flag);
+ }
+ qemu_mutex_unlock_iothread();
+}
+
+static DirtyPageRecord *vcpu_dirty_stat_alloc(VcpuStat *stat)
+{
+ CPUState *cpu;
+ DirtyPageRecord *records;
+ int nvcpu = 0;
+
+ CPU_FOREACH(cpu) {
+ nvcpu++;
+ }
+
+ stat->nvcpu = nvcpu;
+ stat->rates = g_malloc0(sizeof(DirtyRateVcpu) * nvcpu);
+
+ records = g_malloc0(sizeof(DirtyPageRecord) * nvcpu);
+
+ return records;
+}
+
+static void vcpu_dirty_stat_collect(VcpuStat *stat,
+ DirtyPageRecord *records,
+ bool start)
+{
+ CPUState *cpu;
+
+ CPU_FOREACH(cpu) {
+ record_dirtypages(records, cpu, start);
+ }
+}
+
+int64_t vcpu_calculate_dirtyrate(int64_t calc_time_ms,
+ int64_t init_time_ms,
+ VcpuStat *stat,
+ unsigned int flag,
+ bool one_shot)
+{
+ DirtyPageRecord *records;
+ int64_t duration;
+ int64_t dirtyrate;
+ int i = 0;
+ unsigned int gen_id;
+
+retry:
+ cpu_list_lock();
+ gen_id = cpu_list_generation_id_get();
+ records = vcpu_dirty_stat_alloc(stat);
+ vcpu_dirty_stat_collect(stat, records, true);
+ cpu_list_unlock();
+
+ duration = dirty_stat_wait(calc_time_ms, init_time_ms);
+
+ global_dirty_log_sync(flag, one_shot);
+
+ cpu_list_lock();
+ if (gen_id != cpu_list_generation_id_get()) {
+ g_free(records);
+ g_free(stat->rates);
+ cpu_list_unlock();
+ goto retry;
+ }
+ vcpu_dirty_stat_collect(stat, records, false);
+ cpu_list_unlock();
+
+ for (i = 0; i < stat->nvcpu; i++) {
+ dirtyrate = do_calculate_dirtyrate(records[i], duration);
+
+ stat->rates[i].id = i;
+ stat->rates[i].dirty_rate = dirtyrate;
+
+ trace_dirtyrate_do_calculate_vcpu(i, dirtyrate);
+ }
+
+ g_free(records);
+
+ return duration;
+}
+
static bool is_sample_period_valid(int64_t sec)
{
if (sec < MIN_FETCH_DIRTYRATE_TIME_SEC ||
@@ -396,44 +520,6 @@ static bool compare_page_hash_info(struct RamblockDirtyInfo *info,
return true;
}
-static inline void record_dirtypages(DirtyPageRecord *dirty_pages,
- CPUState *cpu, bool start)
-{
- if (start) {
- dirty_pages[cpu->cpu_index].start_pages = cpu->dirty_pages;
- } else {
- dirty_pages[cpu->cpu_index].end_pages = cpu->dirty_pages;
- }
-}
-
-static void dirtyrate_global_dirty_log_start(void)
-{
- qemu_mutex_lock_iothread();
- memory_global_dirty_log_start(GLOBAL_DIRTY_DIRTY_RATE);
- qemu_mutex_unlock_iothread();
-}
-
-static void dirtyrate_global_dirty_log_stop(void)
-{
- qemu_mutex_lock_iothread();
- memory_global_dirty_log_sync();
- memory_global_dirty_log_stop(GLOBAL_DIRTY_DIRTY_RATE);
- qemu_mutex_unlock_iothread();
-}
-
-static int64_t do_calculate_dirtyrate_vcpu(DirtyPageRecord dirty_pages)
-{
- uint64_t memory_size_MB;
- int64_t time_s;
- uint64_t increased_dirty_pages =
- dirty_pages.end_pages - dirty_pages.start_pages;
-
- memory_size_MB = (increased_dirty_pages * TARGET_PAGE_SIZE) >> 20;
- time_s = DirtyStat.calc_time;
-
- return memory_size_MB / time_s;
-}
-
static inline void record_dirtypages_bitmap(DirtyPageRecord *dirty_pages,
bool start)
{
@@ -444,11 +530,6 @@ static inline void record_dirtypages_bitmap(DirtyPageRecord *dirty_pages,
}
}
-static void do_calculate_dirtyrate_bitmap(DirtyPageRecord dirty_pages)
-{
- DirtyStat.dirty_rate = do_calculate_dirtyrate_vcpu(dirty_pages);
-}
-
static inline void dirtyrate_manual_reset_protect(void)
{
RAMBlock *block = NULL;
@@ -492,71 +573,52 @@ static void calculate_dirtyrate_dirty_bitmap(struct DirtyRateConfig config)
DirtyStat.start_time = start_time / 1000;
msec = config.sample_period_seconds * 1000;
- msec = set_sample_page_period(msec, start_time);
+ msec = dirty_stat_wait(msec, start_time);
DirtyStat.calc_time = msec / 1000;
/*
- * dirtyrate_global_dirty_log_stop do two things.
+ * do two things.
* 1. fetch dirty bitmap from kvm
* 2. stop dirty tracking
*/
- dirtyrate_global_dirty_log_stop();
+ global_dirty_log_sync(GLOBAL_DIRTY_DIRTY_RATE, true);
record_dirtypages_bitmap(&dirty_pages, false);
- do_calculate_dirtyrate_bitmap(dirty_pages);
+ DirtyStat.dirty_rate = do_calculate_dirtyrate(dirty_pages, msec);
}
static void calculate_dirtyrate_dirty_ring(struct DirtyRateConfig config)
{
- CPUState *cpu;
- int64_t msec = 0;
int64_t start_time;
+ int64_t duration;
uint64_t dirtyrate = 0;
uint64_t dirtyrate_sum = 0;
- DirtyPageRecord *dirty_pages;
- int nvcpu = 0;
int i = 0;
- CPU_FOREACH(cpu) {
- nvcpu++;
- }
-
- dirty_pages = malloc(sizeof(*dirty_pages) * nvcpu);
-
- DirtyStat.dirty_ring.nvcpu = nvcpu;
- DirtyStat.dirty_ring.rates = malloc(sizeof(DirtyRateVcpu) * nvcpu);
-
- dirtyrate_global_dirty_log_start();
-
- CPU_FOREACH(cpu) {
- record_dirtypages(dirty_pages, cpu, true);
- }
+ /* start log sync */
+ global_dirty_log_change(GLOBAL_DIRTY_DIRTY_RATE, true);
start_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
DirtyStat.start_time = start_time / 1000;
- msec = config.sample_period_seconds * 1000;
- msec = set_sample_page_period(msec, start_time);
- DirtyStat.calc_time = msec / 1000;
+ /* calculate vcpu dirtyrate */
+ duration = vcpu_calculate_dirtyrate(config.sample_period_seconds * 1000,
+ start_time,
+ &DirtyStat.dirty_ring,
+ GLOBAL_DIRTY_DIRTY_RATE,
+ true);
- dirtyrate_global_dirty_log_stop();
-
- CPU_FOREACH(cpu) {
- record_dirtypages(dirty_pages, cpu, false);
- }
+ DirtyStat.calc_time = duration / 1000;
+ /* calculate vm dirtyrate */
for (i = 0; i < DirtyStat.dirty_ring.nvcpu; i++) {
- dirtyrate = do_calculate_dirtyrate_vcpu(dirty_pages[i]);
- trace_dirtyrate_do_calculate_vcpu(i, dirtyrate);
-
- DirtyStat.dirty_ring.rates[i].id = i;
+ dirtyrate = DirtyStat.dirty_ring.rates[i].dirty_rate;
DirtyStat.dirty_ring.rates[i].dirty_rate = dirtyrate;
dirtyrate_sum += dirtyrate;
}
DirtyStat.dirty_rate = dirtyrate_sum;
- free(dirty_pages);
}
static void calculate_dirtyrate_sample_vm(struct DirtyRateConfig config)
@@ -574,7 +636,7 @@ static void calculate_dirtyrate_sample_vm(struct DirtyRateConfig config)
rcu_read_unlock();
msec = config.sample_period_seconds * 1000;
- msec = set_sample_page_period(msec, initial_time);
+ msec = dirty_stat_wait(msec, initial_time);
DirtyStat.start_time = initial_time / 1000;
DirtyStat.calc_time = msec / 1000;
diff --git a/migration/dirtyrate.h b/migration/dirtyrate.h
index 69d4c5b..594a5c0 100644
--- a/migration/dirtyrate.h
+++ b/migration/dirtyrate.h
@@ -13,6 +13,8 @@
#ifndef QEMU_MIGRATION_DIRTYRATE_H
#define QEMU_MIGRATION_DIRTYRATE_H
+#include "sysemu/dirtyrate.h"
+
/*
* Sample 512 pages per GB as default.
*/
@@ -65,11 +67,6 @@ typedef struct SampleVMStat {
uint64_t total_block_mem_MB; /* size of total sampled pages in MB */
} SampleVMStat;
-typedef struct VcpuStat {
- int nvcpu; /* number of vcpu */
- DirtyRateVcpu *rates; /* array of dirty rate for each vcpu */
-} VcpuStat;
-
/*
* Store calculation statistics for each measure.
*/
--
1.8.3.1
^ permalink raw reply related [flat|nested] 22+ messages in thread
* [PATCH v14 4/7] softmmu/dirtylimit: Implement vCPU dirtyrate calculation periodically
2022-02-10 16:17 [PATCH v14 0/7] support dirty restraint on vCPU huangy81
` (2 preceding siblings ...)
2022-02-10 16:17 ` [PATCH v14 3/7] migration/dirtyrate: Refactor dirty page rate calculation huangy81
@ 2022-02-10 16:17 ` huangy81
2022-02-14 8:01 ` Peter Xu
2022-02-10 16:17 ` [PATCH v14 5/7] accel/kvm/kvm-all: Introduce kvm_dirty_ring_size function huangy81
` (2 subsequent siblings)
6 siblings, 1 reply; 22+ messages in thread
From: huangy81 @ 2022-02-10 16:17 UTC (permalink / raw)
To: qemu-devel
Cc: Eduardo Habkost, David Hildenbrand, Hyman, Juan Quintela,
Richard Henderson, Markus ArmBruster, Peter Xu,
Dr. David Alan Gilbert, Paolo Bonzini,
Philippe Mathieu-Daudé
From: Hyman Huang(黄勇) <huangy81@chinatelecom.cn>
Introduce the third method GLOBAL_DIRTY_LIMIT of dirty
tracking for calculate dirtyrate periodly for dirty page
rate limit.
Add dirtylimit.c to implement dirtyrate calculation periodly,
which will be used for dirty page rate limit.
Add dirtylimit.h to export util functions for dirty page rate
limit implementation.
Signed-off-by: Hyman Huang(黄勇) <huangy81@chinatelecom.cn>
---
include/exec/memory.h | 5 +-
include/sysemu/dirtylimit.h | 22 ++++++++
softmmu/dirtylimit.c | 120 ++++++++++++++++++++++++++++++++++++++++++++
softmmu/meson.build | 1 +
4 files changed, 147 insertions(+), 1 deletion(-)
create mode 100644 include/sysemu/dirtylimit.h
create mode 100644 softmmu/dirtylimit.c
diff --git a/include/exec/memory.h b/include/exec/memory.h
index 4d5997e..88ca510 100644
--- a/include/exec/memory.h
+++ b/include/exec/memory.h
@@ -69,7 +69,10 @@ static inline void fuzz_dma_read_cb(size_t addr,
/* Dirty tracking enabled because measuring dirty rate */
#define GLOBAL_DIRTY_DIRTY_RATE (1U << 1)
-#define GLOBAL_DIRTY_MASK (0x3)
+/* Dirty tracking enabled because dirty limit */
+#define GLOBAL_DIRTY_LIMIT (1U << 2)
+
+#define GLOBAL_DIRTY_MASK (0x7)
extern unsigned int global_dirty_tracking;
diff --git a/include/sysemu/dirtylimit.h b/include/sysemu/dirtylimit.h
new file mode 100644
index 0000000..da459f0
--- /dev/null
+++ b/include/sysemu/dirtylimit.h
@@ -0,0 +1,22 @@
+/*
+ * Dirty page rate limit common functions
+ *
+ * Copyright (c) 2022 CHINA TELECOM CO.,LTD.
+ *
+ * Authors:
+ * Hyman Huang(黄勇) <huangy81@chinatelecom.cn>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+ * See the COPYING file in the top-level directory.
+ */
+#ifndef QEMU_DIRTYRLIMIT_H
+#define QEMU_DIRTYRLIMIT_H
+
+#define DIRTYLIMIT_CALC_TIME_MS 1000 /* 1000ms */
+
+int64_t vcpu_dirty_rate_get(int cpu_index);
+void vcpu_dirty_rate_stat_start(void);
+void vcpu_dirty_rate_stat_stop(void);
+void vcpu_dirty_rate_stat_initialize(void);
+void vcpu_dirty_rate_stat_finalize(void);
+#endif
diff --git a/softmmu/dirtylimit.c b/softmmu/dirtylimit.c
new file mode 100644
index 0000000..a10ac6f
--- /dev/null
+++ b/softmmu/dirtylimit.c
@@ -0,0 +1,120 @@
+/*
+ * Dirty page rate limit implementation code
+ *
+ * Copyright (c) 2022 CHINA TELECOM CO.,LTD.
+ *
+ * Authors:
+ * Hyman Huang(黄勇) <huangy81@chinatelecom.cn>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+ * See the COPYING file in the top-level directory.
+ */
+
+#include "qemu/osdep.h"
+#include "qapi/error.h"
+#include "qemu/main-loop.h"
+#include "qapi/qapi-commands-migration.h"
+#include "sysemu/dirtyrate.h"
+#include "sysemu/dirtylimit.h"
+#include "exec/memory.h"
+#include "hw/boards.h"
+
+struct {
+ VcpuStat stat;
+ bool running;
+ QemuThread thread;
+} *vcpu_dirty_rate_stat;
+
+static void vcpu_dirty_rate_stat_collect(void)
+{
+ int64_t start_time;
+ VcpuStat stat;
+ int i = 0;
+
+ start_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
+
+ /* calculate vcpu dirtyrate */
+ vcpu_calculate_dirtyrate(DIRTYLIMIT_CALC_TIME_MS,
+ start_time,
+ &stat,
+ GLOBAL_DIRTY_LIMIT,
+ false);
+
+ for (i = 0; i < stat.nvcpu; i++) {
+ vcpu_dirty_rate_stat->stat.rates[i].id = i;
+ vcpu_dirty_rate_stat->stat.rates[i].dirty_rate =
+ stat.rates[i].dirty_rate;
+ }
+
+ free(stat.rates);
+}
+
+static void *vcpu_dirty_rate_stat_thread(void *opaque)
+{
+ rcu_register_thread();
+
+ /* start log sync */
+ global_dirty_log_change(GLOBAL_DIRTY_LIMIT, true);
+
+ while (qatomic_read(&vcpu_dirty_rate_stat->running)) {
+ vcpu_dirty_rate_stat_collect();
+ }
+
+ /* stop log sync */
+ global_dirty_log_change(GLOBAL_DIRTY_LIMIT, false);
+
+ rcu_unregister_thread();
+ return NULL;
+}
+
+int64_t vcpu_dirty_rate_get(int cpu_index)
+{
+ DirtyRateVcpu *rates = vcpu_dirty_rate_stat->stat.rates;
+ return qatomic_read(&rates[cpu_index].dirty_rate);
+}
+
+void vcpu_dirty_rate_stat_start(void)
+{
+ if (qatomic_read(&vcpu_dirty_rate_stat->running)) {
+ return;
+ }
+
+ qatomic_set(&vcpu_dirty_rate_stat->running, 1);
+ qemu_thread_create(&vcpu_dirty_rate_stat->thread,
+ "dirtyrate-stat",
+ vcpu_dirty_rate_stat_thread,
+ NULL,
+ QEMU_THREAD_JOINABLE);
+}
+
+void vcpu_dirty_rate_stat_stop(void)
+{
+ qatomic_set(&vcpu_dirty_rate_stat->running, 0);
+ qemu_mutex_unlock_iothread();
+ qemu_thread_join(&vcpu_dirty_rate_stat->thread);
+ qemu_mutex_lock_iothread();
+}
+
+void vcpu_dirty_rate_stat_initialize(void)
+{
+ MachineState *ms = MACHINE(qdev_get_machine());
+ int max_cpus = ms->smp.max_cpus;
+
+ vcpu_dirty_rate_stat =
+ g_malloc0(sizeof(*vcpu_dirty_rate_stat));
+
+ vcpu_dirty_rate_stat->stat.nvcpu = max_cpus;
+ vcpu_dirty_rate_stat->stat.rates =
+ g_malloc0(sizeof(DirtyRateVcpu) * max_cpus);
+
+ vcpu_dirty_rate_stat->running = false;
+}
+
+void vcpu_dirty_rate_stat_finalize(void)
+{
+ free(vcpu_dirty_rate_stat->stat.rates);
+ vcpu_dirty_rate_stat->stat.rates = NULL;
+
+ free(vcpu_dirty_rate_stat);
+ vcpu_dirty_rate_stat = NULL;
+}
diff --git a/softmmu/meson.build b/softmmu/meson.build
index d8e0301..95029a5 100644
--- a/softmmu/meson.build
+++ b/softmmu/meson.build
@@ -15,6 +15,7 @@ specific_ss.add(when: 'CONFIG_SOFTMMU', if_true: [files(
'vl.c',
'cpu-timers.c',
'runstate-action.c',
+ 'dirtylimit.c',
)])
specific_ss.add(when: ['CONFIG_SOFTMMU', 'CONFIG_TCG'], if_true: [files(
--
1.8.3.1
^ permalink raw reply related [flat|nested] 22+ messages in thread
* [PATCH v14 5/7] accel/kvm/kvm-all: Introduce kvm_dirty_ring_size function
2022-02-10 16:17 [PATCH v14 0/7] support dirty restraint on vCPU huangy81
` (3 preceding siblings ...)
2022-02-10 16:17 ` [PATCH v14 4/7] softmmu/dirtylimit: Implement vCPU dirtyrate calculation periodically huangy81
@ 2022-02-10 16:17 ` huangy81
2022-02-14 8:02 ` Peter Xu
2022-02-10 16:17 ` [PATCH v14 6/7] softmmu/dirtylimit: Implement virtual CPU throttle huangy81
2022-02-10 16:17 ` [PATCH v14 7/7] softmmu/dirtylimit: Implement dirty page rate limit huangy81
6 siblings, 1 reply; 22+ messages in thread
From: huangy81 @ 2022-02-10 16:17 UTC (permalink / raw)
To: qemu-devel
Cc: Eduardo Habkost, David Hildenbrand, Hyman, Juan Quintela,
Richard Henderson, Markus ArmBruster, Peter Xu,
Dr. David Alan Gilbert, Paolo Bonzini,
Philippe Mathieu-Daudé
From: Hyman Huang(黄勇) <huangy81@chinatelecom.cn>
Introduce kvm_dirty_ring_size util function to help calculate
dirty ring ful time.
Signed-off-by: Hyman Huang(黄勇) <huangy81@chinatelecom.cn>
---
accel/kvm/kvm-all.c | 5 +++++
accel/stubs/kvm-stub.c | 5 +++++
include/sysemu/kvm.h | 2 ++
3 files changed, 12 insertions(+)
diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c
index 7b06b8a..8821d80 100644
--- a/accel/kvm/kvm-all.c
+++ b/accel/kvm/kvm-all.c
@@ -2312,6 +2312,11 @@ bool kvm_dirty_ring_enabled(void)
return kvm_state->kvm_dirty_ring_size ? true : false;
}
+uint32_t kvm_dirty_ring_size(void)
+{
+ return kvm_state->kvm_dirty_ring_size;
+}
+
static int kvm_init(MachineState *ms)
{
MachineClass *mc = MACHINE_GET_CLASS(ms);
diff --git a/accel/stubs/kvm-stub.c b/accel/stubs/kvm-stub.c
index 5319573..1128cb2 100644
--- a/accel/stubs/kvm-stub.c
+++ b/accel/stubs/kvm-stub.c
@@ -152,4 +152,9 @@ bool kvm_dirty_ring_enabled(void)
{
return false;
}
+
+uint32_t kvm_dirty_ring_size(void)
+{
+ return 0;
+}
#endif
diff --git a/include/sysemu/kvm.h b/include/sysemu/kvm.h
index 6eb39a0..bc3f0b5 100644
--- a/include/sysemu/kvm.h
+++ b/include/sysemu/kvm.h
@@ -563,4 +563,6 @@ bool kvm_cpu_check_are_resettable(void);
bool kvm_arch_cpu_check_are_resettable(void);
bool kvm_dirty_ring_enabled(void);
+
+uint32_t kvm_dirty_ring_size(void);
#endif
--
1.8.3.1
^ permalink raw reply related [flat|nested] 22+ messages in thread
* [PATCH v14 6/7] softmmu/dirtylimit: Implement virtual CPU throttle
2022-02-10 16:17 [PATCH v14 0/7] support dirty restraint on vCPU huangy81
` (4 preceding siblings ...)
2022-02-10 16:17 ` [PATCH v14 5/7] accel/kvm/kvm-all: Introduce kvm_dirty_ring_size function huangy81
@ 2022-02-10 16:17 ` huangy81
2022-02-14 8:20 ` Peter Xu
2022-02-10 16:17 ` [PATCH v14 7/7] softmmu/dirtylimit: Implement dirty page rate limit huangy81
6 siblings, 1 reply; 22+ messages in thread
From: huangy81 @ 2022-02-10 16:17 UTC (permalink / raw)
To: qemu-devel
Cc: Eduardo Habkost, David Hildenbrand, Hyman, Juan Quintela,
Richard Henderson, Markus ArmBruster, Peter Xu,
Dr. David Alan Gilbert, Paolo Bonzini,
Philippe Mathieu-Daudé
From: Hyman Huang(黄勇) <huangy81@chinatelecom.cn>
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) <huangy81@chinatelecom.cn>
---
accel/kvm/kvm-all.c | 14 +-
include/hw/core/cpu.h | 6 +
include/sysemu/dirtylimit.h | 15 +++
softmmu/dirtylimit.c | 306 ++++++++++++++++++++++++++++++++++++++++++++
softmmu/trace-events | 8 ++
5 files changed, 348 insertions(+), 1 deletion(-)
diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c
index 8821d80..5ca752b 100644
--- a/accel/kvm/kvm-all.c
+++ b/accel/kvm/kvm-all.c
@@ -45,6 +45,7 @@
#include "qemu/guest-random.h"
#include "sysemu/hw_accel.h"
#include "kvm-cpus.h"
+#include "sysemu/dirtylimit.h"
#include "hw/boards.h"
@@ -476,6 +477,7 @@ int kvm_init_vcpu(CPUState *cpu, Error **errp)
cpu->kvm_state = s;
cpu->vcpu_dirty = true;
cpu->dirty_pages = 0;
+ cpu->throttle_us_per_full = 0;
mmap_size = kvm_ioctl(s, KVM_GET_VCPU_MMAP_SIZE, 0);
if (mmap_size < 0) {
@@ -1469,6 +1471,11 @@ static void *kvm_dirty_ring_reaper_thread(void *data)
*/
sleep(1);
+ /* keep sleeping so that dirtylimit not be interfered by reaper */
+ if (dirtylimit_in_service()) {
+ continue;
+ }
+
trace_kvm_dirty_ring_reaper("wakeup");
r->reaper_state = KVM_DIRTY_RING_REAPER_REAPING;
@@ -2964,8 +2971,13 @@ int kvm_cpu_exec(CPUState *cpu)
*/
trace_kvm_dirty_ring_full(cpu->cpu_index);
qemu_mutex_lock_iothread();
- kvm_dirty_ring_reap(kvm_state, NULL);
+ if (dirtylimit_in_service()) {
+ kvm_dirty_ring_reap(kvm_state, cpu);
+ } else {
+ kvm_dirty_ring_reap(kvm_state, NULL);
+ }
qemu_mutex_unlock_iothread();
+ dirtylimit_vcpu_execute(cpu);
ret = 0;
break;
case KVM_EXIT_SYSTEM_EVENT:
diff --git a/include/hw/core/cpu.h b/include/hw/core/cpu.h
index 76ab3b8..dbeb31a 100644
--- a/include/hw/core/cpu.h
+++ b/include/hw/core/cpu.h
@@ -411,6 +411,12 @@ struct CPUState {
*/
bool throttle_thread_scheduled;
+ /*
+ * Sleep throttle_us_per_full microseconds once dirty ring is full
+ * if dirty page rate limit is enabled.
+ */
+ int64_t throttle_us_per_full;
+
bool ignore_memory_transaction_failures;
/* Used for user-only emulation of prctl(PR_SET_UNALIGN). */
diff --git a/include/sysemu/dirtylimit.h b/include/sysemu/dirtylimit.h
index da459f0..4b7effe 100644
--- a/include/sysemu/dirtylimit.h
+++ b/include/sysemu/dirtylimit.h
@@ -19,4 +19,19 @@ void vcpu_dirty_rate_stat_start(void);
void vcpu_dirty_rate_stat_stop(void);
void vcpu_dirty_rate_stat_initialize(void);
void vcpu_dirty_rate_stat_finalize(void);
+
+void dirtylimit_state_lock(void);
+void dirtylimit_state_unlock(void);
+void dirtylimit_state_initialize(void);
+void dirtylimit_state_finalize(void);
+bool dirtylimit_in_service(void);
+bool dirtylimit_vcpu_index_valid(int cpu_index);
+void dirtylimit_setup(void);
+void dirtylimit_change(bool start);
+void dirtylimit_set_vcpu(int cpu_index,
+ uint64_t quota,
+ bool enable);
+void dirtylimit_set_all(uint64_t quota,
+ bool enable);
+void dirtylimit_vcpu_execute(CPUState *cpu);
#endif
diff --git a/softmmu/dirtylimit.c b/softmmu/dirtylimit.c
index a10ac6f..8b8d8d7 100644
--- a/softmmu/dirtylimit.c
+++ b/softmmu/dirtylimit.c
@@ -18,6 +18,26 @@
#include "sysemu/dirtylimit.h"
#include "exec/memory.h"
#include "hw/boards.h"
+#include "sysemu/kvm.h"
+#include "trace.h"
+
+/*
+ * Dirtylimit stop working if dirty page rate error
+ * value less than DIRTYLIMIT_TOLERANCE_RANGE
+ */
+#define DIRTYLIMIT_TOLERANCE_RANGE 25 /* MB/s */
+/*
+ * Plus or minus vcpu sleep time linearly if dirty
+ * page rate error value percentage over
+ * DIRTYLIMIT_LINEAR_ADJUSTMENT_PCT.
+ * Otherwise, plus or minus a fixed vcpu sleep time.
+ */
+#define DIRTYLIMIT_LINEAR_ADJUSTMENT_PCT 50
+/*
+ * Max vcpu sleep time percentage during a cycle
+ * composed of dirty ring full and sleep time.
+ */
+#define DIRTYLIMIT_THROTTLE_PCT_MAX 99
struct {
VcpuStat stat;
@@ -25,6 +45,34 @@ struct {
QemuThread thread;
} *vcpu_dirty_rate_stat;
+typedef struct VcpuDirtyLimitState {
+ int cpu_index;
+ bool enabled;
+ /*
+ * Quota dirty page rate, unit is MB/s
+ * zero if not enabled.
+ */
+ uint64_t quota;
+} VcpuDirtyLimitState;
+
+typedef void (*DirtyLimitFunc)(void);
+
+struct {
+ VcpuDirtyLimitState *states;
+ /* Max cpus number configured by user */
+ int max_cpus;
+ /* Number of vcpu under dirtylimit */
+ int limited_nvcpu;
+ /* Function to implement throttle set up */
+ DirtyLimitFunc setup;
+} *dirtylimit_state;
+
+/* protect dirtylimit_state */
+static QemuMutex dirtylimit_mutex;
+
+/* dirtylimit thread quit if dirtylimit_quit is true */
+static bool dirtylimit_quit;
+
static void vcpu_dirty_rate_stat_collect(void)
{
int64_t start_time;
@@ -58,6 +106,10 @@ static void *vcpu_dirty_rate_stat_thread(void *opaque)
while (qatomic_read(&vcpu_dirty_rate_stat->running)) {
vcpu_dirty_rate_stat_collect();
+ if (dirtylimit_in_service() &&
+ dirtylimit_state->setup) {
+ dirtylimit_state->setup();
+ }
}
/* stop log sync */
@@ -90,9 +142,11 @@ void vcpu_dirty_rate_stat_start(void)
void vcpu_dirty_rate_stat_stop(void)
{
qatomic_set(&vcpu_dirty_rate_stat->running, 0);
+ dirtylimit_state_unlock();
qemu_mutex_unlock_iothread();
qemu_thread_join(&vcpu_dirty_rate_stat->thread);
qemu_mutex_lock_iothread();
+ dirtylimit_state_lock();
}
void vcpu_dirty_rate_stat_initialize(void)
@@ -118,3 +172,255 @@ void vcpu_dirty_rate_stat_finalize(void)
free(vcpu_dirty_rate_stat);
vcpu_dirty_rate_stat = NULL;
}
+
+void dirtylimit_state_lock(void)
+{
+ qemu_mutex_lock(&dirtylimit_mutex);
+}
+
+void dirtylimit_state_unlock(void)
+{
+ qemu_mutex_unlock(&dirtylimit_mutex);
+}
+
+static void
+__attribute__((__constructor__)) dirtylimit_mutex_init(void)
+{
+ qemu_mutex_init(&dirtylimit_mutex);
+}
+
+static inline VcpuDirtyLimitState *dirtylimit_vcpu_get_state(int cpu_index)
+{
+ return &dirtylimit_state->states[cpu_index];
+}
+
+void dirtylimit_state_initialize(void)
+{
+ MachineState *ms = MACHINE(qdev_get_machine());
+ int max_cpus = ms->smp.max_cpus;
+ int i;
+
+ dirtylimit_state = g_malloc0(sizeof(*dirtylimit_state));
+
+ dirtylimit_state->states =
+ g_malloc0(sizeof(VcpuDirtyLimitState) * max_cpus);
+
+ for (i = 0; i < max_cpus; i++) {
+ dirtylimit_state->states[i].cpu_index = i;
+ }
+
+ dirtylimit_state->max_cpus = max_cpus;
+ dirtylimit_state->setup = dirtylimit_setup;
+ trace_dirtylimit_state_initialize(max_cpus);
+}
+
+void dirtylimit_state_finalize(void)
+{
+ free(dirtylimit_state->states);
+ dirtylimit_state->states = NULL;
+ dirtylimit_state->setup = NULL;
+
+ free(dirtylimit_state);
+ dirtylimit_state = NULL;
+
+ trace_dirtylimit_state_finalize();
+}
+
+bool dirtylimit_in_service(void)
+{
+ return !!dirtylimit_state;
+}
+
+bool dirtylimit_vcpu_index_valid(int cpu_index)
+{
+ MachineState *ms = MACHINE(qdev_get_machine());
+
+ return !(cpu_index < 0 ||
+ cpu_index >= ms->smp.max_cpus);
+}
+
+static inline void dirtylimit_vcpu_set_quota(int cpu_index,
+ uint64_t quota,
+ bool on)
+{
+ if (on) {
+ dirtylimit_state->states[cpu_index].quota = quota;
+ if (!dirtylimit_vcpu_get_state(cpu_index)->enabled) {
+ dirtylimit_state->limited_nvcpu++;
+ }
+ } else {
+ dirtylimit_state->states[cpu_index].quota = 0;
+ if (dirtylimit_state->states[cpu_index].enabled) {
+ dirtylimit_state->limited_nvcpu--;
+ }
+ }
+
+ dirtylimit_state->states[cpu_index].enabled = on;
+}
+
+static inline int64_t dirtylimit_dirty_ring_full_time(uint64_t dirtyrate)
+{
+ static uint64_t max_dirtyrate;
+ uint32_t dirty_ring_size = kvm_dirty_ring_size();
+ uint64_t dirty_ring_size_meory_MB =
+ dirty_ring_size * TARGET_PAGE_SIZE >> 20;
+
+ if (max_dirtyrate < dirtyrate) {
+ max_dirtyrate = dirtyrate;
+ }
+
+ return dirty_ring_size_meory_MB * 1000000 / max_dirtyrate;
+}
+
+static inline bool dirtylimit_done(uint64_t quota,
+ uint64_t current)
+{
+ uint64_t min, max;
+
+ min = MIN(quota, current);
+ max = MAX(quota, current);
+
+ return ((max - min) <= DIRTYLIMIT_TOLERANCE_RANGE) ? true : false;
+}
+
+static inline bool
+dirtylimit_need_linear_adjustment(uint64_t quota,
+ uint64_t current)
+{
+ uint64_t min, max;
+
+ min = MIN(quota, current);
+ max = MAX(quota, current);
+
+ return ((max - min) * 100 / max) > DIRTYLIMIT_LINEAR_ADJUSTMENT_PCT;
+}
+
+static void dirtylimit_set_throttle(CPUState *cpu,
+ uint64_t quota,
+ uint64_t current)
+{
+ int64_t ring_full_time_us = 0;
+ uint64_t sleep_pct = 0;
+ uint64_t throttle_us = 0;
+
+ ring_full_time_us = dirtylimit_dirty_ring_full_time(current);
+
+ if (dirtylimit_need_linear_adjustment(quota, current)) {
+ if (quota < current) {
+ sleep_pct = (current - quota) * 100 / current;
+ throttle_us =
+ ring_full_time_us * sleep_pct / (double)(100 - sleep_pct);
+ cpu->throttle_us_per_full += throttle_us;
+ } else {
+ sleep_pct = (quota - current) * 100 / quota;
+ throttle_us =
+ ring_full_time_us * sleep_pct / (double)(100 - sleep_pct);
+ cpu->throttle_us_per_full -= throttle_us;
+ }
+
+ trace_dirtylimit_throttle_pct(cpu->cpu_index,
+ sleep_pct,
+ throttle_us);
+ } else {
+ if (quota < current) {
+ cpu->throttle_us_per_full += ring_full_time_us / 10;
+ } else {
+ cpu->throttle_us_per_full -= ring_full_time_us / 10;
+ }
+ }
+
+ /*
+ * TODO: in the big kvm_dirty_ring_size case (eg: 65536, or other scenario),
+ * current dirty page rate may never reach the quota, we should stop
+ * increasing sleep time?
+ */
+ cpu->throttle_us_per_full = MIN(cpu->throttle_us_per_full,
+ ring_full_time_us * DIRTYLIMIT_THROTTLE_PCT_MAX);
+
+ cpu->throttle_us_per_full = MAX(cpu->throttle_us_per_full, 0);
+}
+
+static void dirtylimit_adjust_throttle(CPUState *cpu)
+{
+ uint64_t quota = 0;
+ uint64_t current = 0;
+ int cpu_index = cpu->cpu_index;
+
+ quota = dirtylimit_vcpu_get_state(cpu_index)->quota;
+ current = vcpu_dirty_rate_get(cpu_index);
+
+ if (current == 0) {
+ cpu->throttle_us_per_full = 0;
+ goto end;
+ } else if (dirtylimit_done(quota, current)) {
+ goto end;
+ } else {
+ dirtylimit_set_throttle(cpu, quota, current);
+ }
+end:
+ trace_dirtylimit_adjust_throttle(cpu_index,
+ quota, current,
+ cpu->throttle_us_per_full);
+ return;
+}
+
+void dirtylimit_setup(void)
+{
+ CPUState *cpu;
+
+ if (!qatomic_read(&dirtylimit_quit)) {
+ dirtylimit_state_lock();
+
+ if (!dirtylimit_in_service()) {
+ dirtylimit_state_unlock();
+ }
+
+ CPU_FOREACH(cpu) {
+ if (!dirtylimit_vcpu_get_state(cpu->cpu_index)->enabled) {
+ continue;
+ }
+ dirtylimit_adjust_throttle(cpu);
+ }
+ dirtylimit_state_unlock();
+ }
+}
+
+void dirtylimit_change(bool start)
+{
+ if (start) {
+ qatomic_set(&dirtylimit_quit, 0);
+ } else {
+ qatomic_set(&dirtylimit_quit, 1);
+ }
+}
+
+void dirtylimit_set_vcpu(int cpu_index,
+ uint64_t quota,
+ bool enable)
+{
+ dirtylimit_vcpu_set_quota(cpu_index, quota, enable);
+ trace_dirtylimit_set_vcpu(cpu_index, quota);
+}
+
+void dirtylimit_set_all(uint64_t quota,
+ bool enable)
+{
+ MachineState *ms = MACHINE(qdev_get_machine());
+ int max_cpus = ms->smp.max_cpus;
+ int i;
+
+ for (i = 0; i < max_cpus; i++) {
+ dirtylimit_set_vcpu(i, quota, enable);
+ }
+}
+
+void dirtylimit_vcpu_execute(CPUState *cpu)
+{
+ if (dirtylimit_in_service() &&
+ dirtylimit_vcpu_get_state(cpu->cpu_index)->enabled &&
+ cpu->throttle_us_per_full) {
+ trace_dirtylimit_vcpu_execute(cpu->cpu_index,
+ cpu->throttle_us_per_full);
+ usleep(cpu->throttle_us_per_full);
+ }
+}
diff --git a/softmmu/trace-events b/softmmu/trace-events
index 9c88887..ff441ac 100644
--- a/softmmu/trace-events
+++ b/softmmu/trace-events
@@ -31,3 +31,11 @@ runstate_set(int current_state, const char *current_state_str, int new_state, co
system_wakeup_request(int reason) "reason=%d"
qemu_system_shutdown_request(int reason) "reason=%d"
qemu_system_powerdown_request(void) ""
+
+#dirtylimit.c
+dirtylimit_state_initialize(int max_cpus) "dirtylimit state initialize: max cpus %d"
+dirtylimit_state_finalize(void)
+dirtylimit_adjust_throttle(int cpu_index, uint64_t quota, uint64_t current, int64_t time_us) "CPU[%d] throttle: quota %" PRIu64 ", current %" PRIu64 ", throttle %"PRIi64 " us"
+dirtylimit_throttle_pct(int cpu_index, uint64_t pct, int64_t time_us) "CPU[%d] throttle percent: %" PRIu64 ", throttle adjust time %"PRIi64 " us"
+dirtylimit_set_vcpu(int cpu_index, uint64_t quota) "CPU[%d] set dirty page rate limit %"PRIu64
+dirtylimit_vcpu_execute(int cpu_index, int64_t sleep_time_us) "CPU[%d] sleep %"PRIi64 " us"
--
1.8.3.1
^ permalink raw reply related [flat|nested] 22+ messages in thread
* [PATCH v14 7/7] softmmu/dirtylimit: Implement dirty page rate limit
2022-02-10 16:17 [PATCH v14 0/7] support dirty restraint on vCPU huangy81
` (5 preceding siblings ...)
2022-02-10 16:17 ` [PATCH v14 6/7] softmmu/dirtylimit: Implement virtual CPU throttle huangy81
@ 2022-02-10 16:17 ` huangy81
2022-02-11 13:40 ` Markus Armbruster
2022-02-14 8:25 ` Peter Xu
6 siblings, 2 replies; 22+ messages in thread
From: huangy81 @ 2022-02-10 16:17 UTC (permalink / raw)
To: qemu-devel
Cc: Eduardo Habkost, David Hildenbrand, Hyman, Juan Quintela,
Richard Henderson, Markus ArmBruster, Peter Xu,
Dr. David Alan Gilbert, Paolo Bonzini,
Philippe Mathieu-Daudé
From: Hyman Huang(黄勇) <huangy81@chinatelecom.cn>
Implement dirtyrate calculation periodically basing on
dirty-ring and throttle virtual CPU until it reachs the quota
dirty page rate given by user.
Introduce qmp commands "set-vcpu-dirty-limit",
"cancel-vcpu-dirty-limit", "query-vcpu-dirty-limit"
to enable, disable, query dirty page limit for virtual CPU.
Meanwhile, introduce corresponding hmp commands
"set_vcpu_dirty_limit", "cancel_vcpu_dirty_limit",
"info vcpu_dirty_limit" so the feature can be more usable.
Signed-off-by: Hyman Huang(黄勇) <huangy81@chinatelecom.cn>
---
hmp-commands-info.hx | 13 ++++
hmp-commands.hx | 32 +++++++++
include/monitor/hmp.h | 3 +
qapi/migration.json | 80 +++++++++++++++++++++
softmmu/dirtylimit.c | 194 ++++++++++++++++++++++++++++++++++++++++++++++++++
5 files changed, 322 insertions(+)
diff --git a/hmp-commands-info.hx b/hmp-commands-info.hx
index e90f20a..61b23d2 100644
--- a/hmp-commands-info.hx
+++ b/hmp-commands-info.hx
@@ -865,6 +865,19 @@ SRST
Display the vcpu dirty rate information.
ERST
+ {
+ .name = "vcpu_dirty_limit",
+ .args_type = "",
+ .params = "",
+ .help = "show dirty page limit information of all vCPU",
+ .cmd = hmp_info_vcpu_dirty_limit,
+ },
+
+SRST
+ ``info vcpu_dirty_limit``
+ Display the vcpu dirty page limit information.
+ERST
+
#if defined(TARGET_I386)
{
.name = "sgx",
diff --git a/hmp-commands.hx b/hmp-commands.hx
index 70a9136..3fef976 100644
--- a/hmp-commands.hx
+++ b/hmp-commands.hx
@@ -1744,3 +1744,35 @@ ERST
"\n\t\t\t -b to specify dirty bitmap as method of calculation)",
.cmd = hmp_calc_dirty_rate,
},
+
+SRST
+``set_vcpu_dirty_limit``
+ Set dirty page rate limit on virtual CPU, the information about all the
+ virtual CPU dirty limit status can be observed with ``info vcpu_dirty_limit``
+ command.
+ERST
+
+ {
+ .name = "set_vcpu_dirty_limit",
+ .args_type = "dirty_rate:l,cpu_index:l?",
+ .params = "dirty_rate [cpu_index]",
+ .help = "set dirty page rate limit, use cpu_index to set limit"
+ "\n\t\t\t\t on a specified virtual cpu",
+ .cmd = hmp_set_vcpu_dirty_limit,
+ },
+
+SRST
+``cancel_vcpu_dirty_limit``
+ Cancel dirty page rate limit on virtual CPU, the information about all the
+ virtual CPU dirty limit status can be observed with ``info vcpu_dirty_limit``
+ command.
+ERST
+
+ {
+ .name = "cancel_vcpu_dirty_limit",
+ .args_type = "cpu_index:l?",
+ .params = "[cpu_index]",
+ .help = "cancel dirty page rate limit, use cpu_index to cancel"
+ "\n\t\t\t\t limit on a specified virtual cpu",
+ .cmd = hmp_cancel_vcpu_dirty_limit,
+ },
diff --git a/include/monitor/hmp.h b/include/monitor/hmp.h
index 96d0148..478820e 100644
--- a/include/monitor/hmp.h
+++ b/include/monitor/hmp.h
@@ -131,6 +131,9 @@ void hmp_replay_delete_break(Monitor *mon, const QDict *qdict);
void hmp_replay_seek(Monitor *mon, const QDict *qdict);
void hmp_info_dirty_rate(Monitor *mon, const QDict *qdict);
void hmp_calc_dirty_rate(Monitor *mon, const QDict *qdict);
+void hmp_set_vcpu_dirty_limit(Monitor *mon, const QDict *qdict);
+void hmp_cancel_vcpu_dirty_limit(Monitor *mon, const QDict *qdict);
+void hmp_info_vcpu_dirty_limit(Monitor *mon, const QDict *qdict);
void hmp_human_readable_text_helper(Monitor *mon,
HumanReadableText *(*qmp_handler)(Error **));
diff --git a/qapi/migration.json b/qapi/migration.json
index 5975a0e..2ccbb92 100644
--- a/qapi/migration.json
+++ b/qapi/migration.json
@@ -1861,6 +1861,86 @@
{ 'command': 'query-dirty-rate', 'returns': 'DirtyRateInfo' }
##
+# @DirtyLimitInfo:
+#
+# Dirty page rate limit information of a virtual CPU.
+#
+# @cpu-index: index of a virtual CPU.
+#
+# @limit-rate: upper limit of dirty page rate (MB/s) for a virtual
+# CPU, 0 means unlimited.
+#
+# @current-rate: current dirty page rate (MB/s) for a virtual CPU.
+#
+# Since: 7.0
+#
+##
+{ 'struct': 'DirtyLimitInfo',
+ 'data': { 'cpu-index': 'int',
+ 'limit-rate': 'uint64',
+ 'current-rate': 'uint64' } }
+
+##
+# @set-vcpu-dirty-limit:
+#
+# Set the upper limit of dirty page rate for virtual CPUs.
+#
+# Requires KVM with accelerator property "dirty-ring-size" set.
+# A virtual CPU's dirty page rate is a measure of its memory load.
+# To observe dirty page rates, use @calc-dirty-rate.
+#
+# @cpu-index: index of a virtual CPU, default is all.
+#
+# @dirty-rate: upper limit of dirty page rate (MB/s) for virtual CPUs.
+#
+# Since: 7.0
+#
+# Example:
+# {"execute": "set-vcpu-dirty-limit"}
+# "arguments": { "dirty-rate": 200,
+# "cpu-index": 1 } }
+#
+##
+{ 'command': 'set-vcpu-dirty-limit',
+ 'data': { '*cpu-index': 'int',
+ 'dirty-rate': 'uint64' } }
+
+##
+# @cancel-vcpu-dirty-limit:
+#
+# Cancel the upper limit of dirty page rate for virtual CPUs.
+#
+# Cancel the dirty page limit for the vCPU which has been set with
+# set-vcpu-dirty-limit command. Note that this command requires
+# support from dirty ring, same as the "set-vcpu-dirty-limit".
+#
+# @cpu-index: index of a virtual CPU, default is all.
+#
+# Since: 7.0
+#
+# Example:
+# {"execute": "cancel-vcpu-dirty-limit"}
+# "arguments": { "cpu-index": 1 } }
+#
+##
+{ 'command': 'cancel-vcpu-dirty-limit',
+ 'data': { '*cpu-index': 'int'} }
+
+##
+# @query-vcpu-dirty-limit:
+#
+# Returns information about virtual CPU dirty page rate limits, if any.
+#
+# Since: 7.0
+#
+# Example:
+# {"execute": "query-vcpu-dirty-limit"}
+#
+##
+{ 'command': 'query-vcpu-dirty-limit',
+ 'returns': [ 'DirtyLimitInfo' ] }
+
+##
# @snapshot-save:
#
# Save a VM snapshot
diff --git a/softmmu/dirtylimit.c b/softmmu/dirtylimit.c
index 8b8d8d7..32e3575 100644
--- a/softmmu/dirtylimit.c
+++ b/softmmu/dirtylimit.c
@@ -14,8 +14,12 @@
#include "qapi/error.h"
#include "qemu/main-loop.h"
#include "qapi/qapi-commands-migration.h"
+#include "qapi/qmp/qdict.h"
+#include "qapi/error.h"
#include "sysemu/dirtyrate.h"
#include "sysemu/dirtylimit.h"
+#include "monitor/hmp.h"
+#include "monitor/monitor.h"
#include "exec/memory.h"
#include "hw/boards.h"
#include "sysemu/kvm.h"
@@ -424,3 +428,193 @@ void dirtylimit_vcpu_execute(CPUState *cpu)
usleep(cpu->throttle_us_per_full);
}
}
+
+static void dirtylimit_init(void)
+{
+ dirtylimit_state_initialize();
+ dirtylimit_change(true);
+ vcpu_dirty_rate_stat_initialize();
+ vcpu_dirty_rate_stat_start();
+}
+
+static void dirtylimit_cleanup(void)
+{
+ vcpu_dirty_rate_stat_stop();
+ vcpu_dirty_rate_stat_finalize();
+ dirtylimit_change(false);
+ dirtylimit_state_finalize();
+}
+
+void qmp_cancel_vcpu_dirty_limit(bool has_cpu_index,
+ int64_t cpu_index,
+ Error **errp)
+{
+ if (!kvm_enabled() || !kvm_dirty_ring_enabled()) {
+ return;
+ }
+
+ if (has_cpu_index && !dirtylimit_vcpu_index_valid(cpu_index)) {
+ error_setg(errp, "incorrect cpu index specified");
+ return;
+ }
+
+ if (!dirtylimit_in_service()) {
+ return;
+ }
+
+ dirtylimit_state_lock();
+
+ if (has_cpu_index) {
+ dirtylimit_set_vcpu(cpu_index, 0, false);
+ } else {
+ dirtylimit_set_all(0, false);
+ }
+
+ if (!dirtylimit_state->limited_nvcpu) {
+ dirtylimit_cleanup();
+ }
+
+ dirtylimit_state_unlock();
+}
+
+void hmp_cancel_vcpu_dirty_limit(Monitor *mon, const QDict *qdict)
+{
+ int64_t cpu_index = qdict_get_try_int(qdict, "cpu_index", -1);
+ Error *err = NULL;
+
+ qmp_cancel_vcpu_dirty_limit(!!(cpu_index != -1), cpu_index, &err);
+ if (err) {
+ hmp_handle_error(mon, err);
+ return;
+ }
+
+ monitor_printf(mon, "[Please use 'info vcpu_dirty_limit' to query "
+ "dirty limit for virtual CPU]\n");
+}
+
+void qmp_set_vcpu_dirty_limit(bool has_cpu_index,
+ int64_t cpu_index,
+ uint64_t dirty_rate,
+ Error **errp)
+{
+ if (!kvm_enabled() || !kvm_dirty_ring_enabled()) {
+ error_setg(errp, "dirty page limit feature requires KVM with"
+ " accelerator property 'dirty-ring-size' set'");
+ return;
+ }
+
+ if (has_cpu_index && !dirtylimit_vcpu_index_valid(cpu_index)) {
+ error_setg(errp, "incorrect cpu index specified");
+ return;
+ }
+
+ if (!dirty_rate) {
+ qmp_cancel_vcpu_dirty_limit(has_cpu_index, cpu_index, errp);
+ return;
+ }
+
+ dirtylimit_state_lock();
+
+ if (!dirtylimit_in_service()) {
+ dirtylimit_init();
+ }
+
+ if (has_cpu_index) {
+ dirtylimit_set_vcpu(cpu_index, dirty_rate, true);
+ } else {
+ dirtylimit_set_all(dirty_rate, true);
+ }
+
+ dirtylimit_state_unlock();
+}
+
+void hmp_set_vcpu_dirty_limit(Monitor *mon, const QDict *qdict)
+{
+ int64_t dirty_rate = qdict_get_int(qdict, "dirty_rate");
+ int64_t cpu_index = qdict_get_try_int(qdict, "cpu_index", -1);
+ Error *err = NULL;
+
+ qmp_set_vcpu_dirty_limit(!!(cpu_index != -1), cpu_index, dirty_rate, &err);
+ if (err) {
+ hmp_handle_error(mon, err);
+ return;
+ }
+
+ monitor_printf(mon, "[Please use 'info vcpu_dirty_limit' to query "
+ "dirty limit for virtual CPU]\n");
+}
+
+static struct DirtyLimitInfo *dirtylimit_query_vcpu(int cpu_index)
+{
+ DirtyLimitInfo *info = NULL;
+
+ info = g_malloc0(sizeof(*info));
+ info->cpu_index = cpu_index;
+ info->limit_rate = dirtylimit_vcpu_get_state(cpu_index)->quota;
+ info->current_rate = vcpu_dirty_rate_get(cpu_index);
+
+ return info;
+}
+
+static struct DirtyLimitInfoList *dirtylimit_query_all(void)
+{
+ int i, index;
+ DirtyLimitInfo *info = NULL;
+ DirtyLimitInfoList *head = NULL, **tail = &head;
+
+ dirtylimit_state_lock();
+
+ if (!dirtylimit_in_service()) {
+ return NULL;
+ }
+
+ for (i = 0; i < dirtylimit_state->max_cpus; i++) {
+ index = dirtylimit_state->states[i].cpu_index;
+ if (dirtylimit_vcpu_get_state(index)->enabled) {
+ info = dirtylimit_query_vcpu(index);
+ QAPI_LIST_APPEND(tail, info);
+ }
+ }
+
+ dirtylimit_state_unlock();
+
+ return head;
+}
+
+struct DirtyLimitInfoList *qmp_query_vcpu_dirty_limit(Error **errp)
+{
+ if (!dirtylimit_in_service()) {
+ error_setg(errp, "dirty page limit not enabled");
+ return NULL;
+ }
+
+ return dirtylimit_query_all();
+}
+
+void hmp_info_vcpu_dirty_limit(Monitor *mon, const QDict *qdict)
+{
+ DirtyLimitInfoList *limit, *head, *info = NULL;
+ Error *err = NULL;
+
+ if (!dirtylimit_in_service()) {
+ monitor_printf(mon, "Dirty page limit not enabled!\n");
+ return;
+ }
+
+ info = qmp_query_vcpu_dirty_limit(&err);
+ if (err) {
+ hmp_handle_error(mon, err);
+ return;
+ }
+
+ head = info;
+ for (limit = head; limit != NULL; limit = limit->next) {
+ monitor_printf(mon, "vcpu[%"PRIi64"], limit rate %"PRIi64 " (MB/s),"
+ " current rate %"PRIi64 " (MB/s)\n",
+ limit->value->cpu_index,
+ limit->value->limit_rate,
+ limit->value->current_rate);
+ }
+
+ g_free(info);
+}
--
1.8.3.1
^ permalink raw reply related [flat|nested] 22+ messages in thread
* Re: [PATCH v14 7/7] softmmu/dirtylimit: Implement dirty page rate limit
2022-02-10 16:17 ` [PATCH v14 7/7] softmmu/dirtylimit: Implement dirty page rate limit huangy81
@ 2022-02-11 13:40 ` Markus Armbruster
2022-02-14 8:25 ` Peter Xu
1 sibling, 0 replies; 22+ messages in thread
From: Markus Armbruster @ 2022-02-11 13:40 UTC (permalink / raw)
To: huangy81
Cc: Eduardo Habkost, David Hildenbrand, Juan Quintela,
Richard Henderson, qemu-devel, Peter Xu, Dr. David Alan Gilbert,
Paolo Bonzini, Philippe Mathieu-Daudé
huangy81@chinatelecom.cn writes:
> From: Hyman Huang(黄勇) <huangy81@chinatelecom.cn>
>
> Implement dirtyrate calculation periodically basing on
> dirty-ring and throttle virtual CPU until it reachs the quota
> dirty page rate given by user.
>
> Introduce qmp commands "set-vcpu-dirty-limit",
> "cancel-vcpu-dirty-limit", "query-vcpu-dirty-limit"
> to enable, disable, query dirty page limit for virtual CPU.
>
> Meanwhile, introduce corresponding hmp commands
> "set_vcpu_dirty_limit", "cancel_vcpu_dirty_limit",
> "info vcpu_dirty_limit" so the feature can be more usable.
>
> Signed-off-by: Hyman Huang(黄勇) <huangy81@chinatelecom.cn>
QAPI schema
Acked-by: Markus Armbruster <armbru@redhat.com>
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH v14 3/7] migration/dirtyrate: Refactor dirty page rate calculation
2022-02-10 16:17 ` [PATCH v14 3/7] migration/dirtyrate: Refactor dirty page rate calculation huangy81
@ 2022-02-14 7:57 ` Peter Xu
2022-02-14 8:40 ` Hyman Huang
0 siblings, 1 reply; 22+ messages in thread
From: Peter Xu @ 2022-02-14 7:57 UTC (permalink / raw)
To: huangy81
Cc: Eduardo Habkost, David Hildenbrand, Juan Quintela,
Richard Henderson, qemu-devel, Markus ArmBruster, Paolo Bonzini,
Philippe Mathieu-Daudé,
Dr. David Alan Gilbert
Mostly good, one trivial nit below:
On Fri, Feb 11, 2022 at 12:17:37AM +0800, huangy81@chinatelecom.cn wrote:
> +int64_t vcpu_calculate_dirtyrate(int64_t calc_time_ms,
> + int64_t init_time_ms,
> + VcpuStat *stat,
> + unsigned int flag,
> + bool one_shot)
> +{
> + DirtyPageRecord *records;
> + int64_t duration;
> + int64_t dirtyrate;
> + int i = 0;
> + unsigned int gen_id;
> +
> +retry:
> + cpu_list_lock();
> + gen_id = cpu_list_generation_id_get();
> + records = vcpu_dirty_stat_alloc(stat);
> + vcpu_dirty_stat_collect(stat, records, true);
> + cpu_list_unlock();
> +
> + duration = dirty_stat_wait(calc_time_ms, init_time_ms);
Is it a must to pass in init_time_ms rather than always sleep in
dirty_stat_wait()? Could we simply drop it?
--
Peter Xu
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH v14 1/7] accel/kvm/kvm-all: Refactor per-vcpu dirty ring reaping
2022-02-10 16:17 ` [PATCH v14 1/7] accel/kvm/kvm-all: Refactor per-vcpu dirty ring reaping huangy81
@ 2022-02-14 7:57 ` Peter Xu
0 siblings, 0 replies; 22+ messages in thread
From: Peter Xu @ 2022-02-14 7:57 UTC (permalink / raw)
To: huangy81
Cc: Eduardo Habkost, David Hildenbrand, Juan Quintela,
Richard Henderson, qemu-devel, Markus ArmBruster, Paolo Bonzini,
Philippe Mathieu-Daudé,
Dr. David Alan Gilbert
On Fri, Feb 11, 2022 at 12:17:35AM +0800, huangy81@chinatelecom.cn wrote:
> From: Hyman Huang(黄勇) <huangy81@chinatelecom.cn>
>
> Add a non-required argument 'CPUState' to kvm_dirty_ring_reap so
> that it can cover single vcpu dirty-ring-reaping scenario.
>
> Signed-off-by: Hyman Huang(黄勇) <huangy81@chinatelecom.cn>
Reviewed-by: Peter Xu <peterx@redhat.com>
--
Peter Xu
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH v14 2/7] cpus: Introduce cpu_list_generation_id
2022-02-10 16:17 ` [PATCH v14 2/7] cpus: Introduce cpu_list_generation_id huangy81
@ 2022-02-14 7:59 ` Peter Xu
0 siblings, 0 replies; 22+ messages in thread
From: Peter Xu @ 2022-02-14 7:59 UTC (permalink / raw)
To: huangy81
Cc: Eduardo Habkost, David Hildenbrand, Juan Quintela,
Richard Henderson, qemu-devel, Markus ArmBruster, Paolo Bonzini,
Philippe Mathieu-Daudé,
Dr. David Alan Gilbert
On Fri, Feb 11, 2022 at 12:17:36AM +0800, huangy81@chinatelecom.cn wrote:
> From: Hyman Huang(黄勇) <huangy81@chinatelecom.cn>
>
> Introduce cpu_list_generation_id to track cpu list generation so
> that cpu hotplug/unplug can be detected during measurement of
> dirty page rate.
Could you add a short paragraph on showing how the gen_id should be used?
>
> Signed-off-by: Hyman Huang(黄勇) <huangy81@chinatelecom.cn>
Reviewed-by: Peter Xu <peterx@redhat.com>
Thanks,
--
Peter Xu
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH v14 4/7] softmmu/dirtylimit: Implement vCPU dirtyrate calculation periodically
2022-02-10 16:17 ` [PATCH v14 4/7] softmmu/dirtylimit: Implement vCPU dirtyrate calculation periodically huangy81
@ 2022-02-14 8:01 ` Peter Xu
0 siblings, 0 replies; 22+ messages in thread
From: Peter Xu @ 2022-02-14 8:01 UTC (permalink / raw)
To: huangy81
Cc: Eduardo Habkost, David Hildenbrand, Juan Quintela,
Richard Henderson, qemu-devel, Markus ArmBruster, Paolo Bonzini,
Philippe Mathieu-Daudé,
Dr. David Alan Gilbert
On Fri, Feb 11, 2022 at 12:17:38AM +0800, huangy81@chinatelecom.cn wrote:
> From: Hyman Huang(黄勇) <huangy81@chinatelecom.cn>
>
> Introduce the third method GLOBAL_DIRTY_LIMIT of dirty
> tracking for calculate dirtyrate periodly for dirty page
> rate limit.
>
> Add dirtylimit.c to implement dirtyrate calculation periodly,
> which will be used for dirty page rate limit.
>
> Add dirtylimit.h to export util functions for dirty page rate
> limit implementation.
>
> Signed-off-by: Hyman Huang(黄勇) <huangy81@chinatelecom.cn>
Reviewed-by: Peter Xu <peterx@redhat.com>
--
Peter Xu
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH v14 5/7] accel/kvm/kvm-all: Introduce kvm_dirty_ring_size function
2022-02-10 16:17 ` [PATCH v14 5/7] accel/kvm/kvm-all: Introduce kvm_dirty_ring_size function huangy81
@ 2022-02-14 8:02 ` Peter Xu
0 siblings, 0 replies; 22+ messages in thread
From: Peter Xu @ 2022-02-14 8:02 UTC (permalink / raw)
To: huangy81
Cc: Eduardo Habkost, David Hildenbrand, Juan Quintela,
Richard Henderson, qemu-devel, Markus ArmBruster, Paolo Bonzini,
Philippe Mathieu-Daudé,
Dr. David Alan Gilbert
On Fri, Feb 11, 2022 at 12:17:39AM +0800, huangy81@chinatelecom.cn wrote:
> From: Hyman Huang(黄勇) <huangy81@chinatelecom.cn>
>
> Introduce kvm_dirty_ring_size util function to help calculate
> dirty ring ful time.
>
> Signed-off-by: Hyman Huang(黄勇) <huangy81@chinatelecom.cn>
Acked-by: Peter Xu <peterx@redhat.com>
--
Peter Xu
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH v14 6/7] softmmu/dirtylimit: Implement virtual CPU throttle
2022-02-10 16:17 ` [PATCH v14 6/7] softmmu/dirtylimit: Implement virtual CPU throttle huangy81
@ 2022-02-14 8:20 ` Peter Xu
2022-02-14 9:05 ` Hyman Huang
2022-02-14 9:22 ` Hyman Huang
0 siblings, 2 replies; 22+ messages in thread
From: Peter Xu @ 2022-02-14 8:20 UTC (permalink / raw)
To: huangy81
Cc: Eduardo Habkost, David Hildenbrand, Juan Quintela,
Richard Henderson, qemu-devel, Markus ArmBruster, Paolo Bonzini,
Philippe Mathieu-Daudé,
Dr. David Alan Gilbert
On Fri, Feb 11, 2022 at 12:17:40AM +0800, huangy81@chinatelecom.cn wrote:
> @@ -2964,8 +2971,13 @@ int kvm_cpu_exec(CPUState *cpu)
> */
> trace_kvm_dirty_ring_full(cpu->cpu_index);
> qemu_mutex_lock_iothread();
> - kvm_dirty_ring_reap(kvm_state, NULL);
> + if (dirtylimit_in_service()) {
> + kvm_dirty_ring_reap(kvm_state, cpu);
> + } else {
> + kvm_dirty_ring_reap(kvm_state, NULL);
> + }
Could you add some comment here on why the cpu pointer is conditionally passed
into the reaping routine? Even if we know it now, it's not immediately obvious
to all the readers.
[...]
> +struct {
> + VcpuDirtyLimitState *states;
> + /* Max cpus number configured by user */
> + int max_cpus;
> + /* Number of vcpu under dirtylimit */
> + int limited_nvcpu;
> + /* Function to implement throttle set up */
> + DirtyLimitFunc setup;
"setup" normally is used only at startup of something, but not per interval.
Perhaps "process" or "adjust"? Same across other "setup" namings across the
patch.
Again, I'd rather call the function directly..
[...]
> +static void dirtylimit_adjust_throttle(CPUState *cpu)
> +{
> + uint64_t quota = 0;
> + uint64_t current = 0;
> + int cpu_index = cpu->cpu_index;
> +
> + quota = dirtylimit_vcpu_get_state(cpu_index)->quota;
> + current = vcpu_dirty_rate_get(cpu_index);
> +
> + if (current == 0) {
> + cpu->throttle_us_per_full = 0;
> + goto end;
Can be dropped?
> + } else if (dirtylimit_done(quota, current)) {
> + goto end;
Same here. Dropping it wholely and:
} else if (!dirtylimit_done(quota, current)) {
dirtylimit_set_throttle(cpu, quota, current);
}
Would work?
> + } else {
> + dirtylimit_set_throttle(cpu, quota, current);
> + }
> +end:
Can be dropped?
> + trace_dirtylimit_adjust_throttle(cpu_index,
> + quota, current,
> + cpu->throttle_us_per_full);
> + return;
> +}
> +
> +void dirtylimit_setup(void)
> +{
> + CPUState *cpu;
> +
> + if (!qatomic_read(&dirtylimit_quit)) {
> + dirtylimit_state_lock();
> +
> + if (!dirtylimit_in_service()) {
> + dirtylimit_state_unlock();
Need to return?
> + }
> +
> + CPU_FOREACH(cpu) {
> + if (!dirtylimit_vcpu_get_state(cpu->cpu_index)->enabled) {
> + continue;
> + }
> + dirtylimit_adjust_throttle(cpu);
> + }
> + dirtylimit_state_unlock();
> + }
> +}
[...]
> +void dirtylimit_set_vcpu(int cpu_index,
> + uint64_t quota,
> + bool enable)
> +{
> + dirtylimit_vcpu_set_quota(cpu_index, quota, enable);
> + trace_dirtylimit_set_vcpu(cpu_index, quota);
> +}
This helper is not "help"ful.. How about wrapping the trace into
dirtylimit_vcpu_set_quota, then drop it?
Thanks,
--
Peter Xu
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH v14 7/7] softmmu/dirtylimit: Implement dirty page rate limit
2022-02-10 16:17 ` [PATCH v14 7/7] softmmu/dirtylimit: Implement dirty page rate limit huangy81
2022-02-11 13:40 ` Markus Armbruster
@ 2022-02-14 8:25 ` Peter Xu
2022-02-14 9:24 ` Hyman Huang
1 sibling, 1 reply; 22+ messages in thread
From: Peter Xu @ 2022-02-14 8:25 UTC (permalink / raw)
To: huangy81
Cc: Eduardo Habkost, David Hildenbrand, Juan Quintela,
Richard Henderson, qemu-devel, Markus ArmBruster, Paolo Bonzini,
Philippe Mathieu-Daudé,
Dr. David Alan Gilbert
On Fri, Feb 11, 2022 at 12:17:41AM +0800, huangy81@chinatelecom.cn wrote:
> +static struct DirtyLimitInfoList *dirtylimit_query_all(void)
> +{
> + int i, index;
> + DirtyLimitInfo *info = NULL;
> + DirtyLimitInfoList *head = NULL, **tail = &head;
> +
> + dirtylimit_state_lock();
> +
> + if (!dirtylimit_in_service()) {
Need to unlock?
> + return NULL;
> + }
> +
> + for (i = 0; i < dirtylimit_state->max_cpus; i++) {
> + index = dirtylimit_state->states[i].cpu_index;
> + if (dirtylimit_vcpu_get_state(index)->enabled) {
> + info = dirtylimit_query_vcpu(index);
> + QAPI_LIST_APPEND(tail, info);
> + }
> + }
> +
> + dirtylimit_state_unlock();
> +
> + return head;
> +}
--
Peter Xu
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH v14 3/7] migration/dirtyrate: Refactor dirty page rate calculation
2022-02-14 7:57 ` Peter Xu
@ 2022-02-14 8:40 ` Hyman Huang
2022-02-14 9:47 ` Peter Xu
0 siblings, 1 reply; 22+ messages in thread
From: Hyman Huang @ 2022-02-14 8:40 UTC (permalink / raw)
To: Peter Xu
Cc: Eduardo Habkost, David Hildenbrand, Juan Quintela,
Richard Henderson, qemu-devel, Markus ArmBruster, Paolo Bonzini,
Philippe Mathieu-Daudé,
Dr. David Alan Gilbert
在 2022/2/14 15:57, Peter Xu 写道:
> Mostly good, one trivial nit below:
>
> On Fri, Feb 11, 2022 at 12:17:37AM +0800, huangy81@chinatelecom.cn wrote:
>> +int64_t vcpu_calculate_dirtyrate(int64_t calc_time_ms,
>> + int64_t init_time_ms,
>> + VcpuStat *stat,
>> + unsigned int flag,
>> + bool one_shot)
>> +{
>> + DirtyPageRecord *records;
>> + int64_t duration;
>> + int64_t dirtyrate;
>> + int i = 0;
>> + unsigned int gen_id;
>> +
>> +retry:
>> + cpu_list_lock();
>> + gen_id = cpu_list_generation_id_get();
>> + records = vcpu_dirty_stat_alloc(stat);
>> + vcpu_dirty_stat_collect(stat, records, true);
>> + cpu_list_unlock();
>> +
>> + duration = dirty_stat_wait(calc_time_ms, init_time_ms);
>
> Is it a must to pass in init_time_ms rather than always sleep in
> dirty_stat_wait()? Could we simply drop it?
>
Indeed, the parameter 'init_time_ms' seems kind of weird :(, we
introduce 'init_time_ms' just becasue the calculate_dirtyrate_dirty_ring
will call the function, see the following block:
static void calculate_dirtyrate_dirty_ring(struct DirtyRateConfig config)
{
- CPUState *cpu;
- int64_t msec = 0;
int64_t start_time;
+ int64_t duration;
uint64_t dirtyrate = 0;
uint64_t dirtyrate_sum = 0;
- DirtyPageRecord *dirty_pages;
- int nvcpu = 0;
int i = 0;
- CPU_FOREACH(cpu) {
- nvcpu++;
- }
-
- dirty_pages = malloc(sizeof(*dirty_pages) * nvcpu);
-
- DirtyStat.dirty_ring.nvcpu = nvcpu;
- DirtyStat.dirty_ring.rates = malloc(sizeof(DirtyRateVcpu) * nvcpu);
-
- dirtyrate_global_dirty_log_start();
-
- CPU_FOREACH(cpu) {
- record_dirtypages(dirty_pages, cpu, true);
- }
+ /* start log sync */
+ global_dirty_log_change(GLOBAL_DIRTY_DIRTY_RATE, true);
start_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
DirtyStat.start_time = start_time / 1000;
The reason why we introduce the 'init_time_ms' is wanting to store the
start_time info and display.
Dropping this parameter is fine from my point view if we ignore the
duration error result from the delay between caller and callee of
'vcpu_calculate_dirtyrate’
- msec = config.sample_period_seconds * 1000;
- msec = set_sample_page_period(msec, start_time);
- DirtyStat.calc_time = msec / 1000;
+ /* calculate vcpu dirtyrate */
+ duration = vcpu_calculate_dirtyrate(config.sample_period_seconds *
1000,
+ start_time,
+ &DirtyStat.dirty_ring,
+ GLOBAL_DIRTY_DIRTY_RATE,
+ true);
- dirtyrate_global_dirty_log_stop();
-
- CPU_FOREACH(cpu) {
- record_dirtypages(dirty_pages, cpu, false);
- }
+ DirtyStat.calc_time = duration / 1000;
--
Best regard
Hyman Huang(黄勇)
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH v14 6/7] softmmu/dirtylimit: Implement virtual CPU throttle
2022-02-14 8:20 ` Peter Xu
@ 2022-02-14 9:05 ` Hyman Huang
2022-02-14 9:51 ` Peter Xu
2022-02-14 9:22 ` Hyman Huang
1 sibling, 1 reply; 22+ messages in thread
From: Hyman Huang @ 2022-02-14 9:05 UTC (permalink / raw)
To: Peter Xu
Cc: Eduardo Habkost, David Hildenbrand, Juan Quintela,
Richard Henderson, qemu-devel, Markus ArmBruster, Paolo Bonzini,
Philippe Mathieu-Daudé,
Dr. David Alan Gilbert
在 2022/2/14 16:20, Peter Xu 写道:
> On Fri, Feb 11, 2022 at 12:17:40AM +0800, huangy81@chinatelecom.cn wrote:
>> @@ -2964,8 +2971,13 @@ int kvm_cpu_exec(CPUState *cpu)
>> */
>> trace_kvm_dirty_ring_full(cpu->cpu_index);
>> qemu_mutex_lock_iothread();
>> - kvm_dirty_ring_reap(kvm_state, NULL);
>> + if (dirtylimit_in_service()) {
>> + kvm_dirty_ring_reap(kvm_state, cpu);
>> + } else {
>> + kvm_dirty_ring_reap(kvm_state, NULL);
>> + }
>
> Could you add some comment here on why the cpu pointer is conditionally passed
> into the reaping routine? Even if we know it now, it's not immediately obvious
> to all the readers.
Sure.
>
> [...]
>
>> +struct {
>> + VcpuDirtyLimitState *states;
>> + /* Max cpus number configured by user */
>> + int max_cpus;
>> + /* Number of vcpu under dirtylimit */
>> + int limited_nvcpu;
>> + /* Function to implement throttle set up */
>> + DirtyLimitFunc setup;
>
> "setup" normally is used only at startup of something, but not per interval.
> Perhaps "process" or "adjust"? Same across other "setup" namings across the
> patch.
Ok, 'adjust' is fine.
>
> Again, I'd rather call the function directly..
Um, maybe using the function pointer is more extensible.
[...]
static void *vcpu_dirty_rate_stat_thread(void *opaque)
{
rcu_register_thread();
/* start log sync */
global_dirty_log_change(GLOBAL_DIRTY_LIMIT, true);
while (qatomic_read(&vcpu_dirty_rate_stat->running)) {
vcpu_dirty_rate_stat_collect();
if (dirtylimit_in_service() &&
dirtylimit_state->setup) {
dirtylimit_state->setup();
}
}
/* stop log sync */
global_dirty_log_change(GLOBAL_DIRTY_LIMIT, false);
rcu_unregister_thread();
return NULL;
}
[...]
Function pointer makes the 'dirtyrate-stat' logic and 'dirtylimit' logic
kind of decoupled.
But i'm ok if you insist because it's just about how to call the
'dirtylimit' and doesn't affect the whole logic.
>
> [...]
>
>> +static void dirtylimit_adjust_throttle(CPUState *cpu)
>> +{
>> + uint64_t quota = 0;
>> + uint64_t current = 0;
>> + int cpu_index = cpu->cpu_index;
>> +
>> + quota = dirtylimit_vcpu_get_state(cpu_index)->quota;
>> + current = vcpu_dirty_rate_get(cpu_index);
>> +
>> + if (current == 0) {
>> + cpu->throttle_us_per_full = 0;
>> + goto end;
>
> Can be dropped?
>
>> + } else if (dirtylimit_done(quota, current)) {
>> + goto end;
>
> Same here. Dropping it wholely and:
>
> } else if (!dirtylimit_done(quota, current)) {
> dirtylimit_set_throttle(cpu, quota, current);
> }
>
> Would work?
>
>> + } else {
>> + dirtylimit_set_throttle(cpu, quota, current);
>> + }
>> +end:
>
> Can be dropped?
>
>> + trace_dirtylimit_adjust_throttle(cpu_index,
>> + quota, current,
>> + cpu->throttle_us_per_full);
>> + return;
>> +}
>> +
>> +void dirtylimit_setup(void)
>> +{
>> + CPUState *cpu;
>> +
>> + if (!qatomic_read(&dirtylimit_quit)) {
>> + dirtylimit_state_lock();
>> +
>> + if (!dirtylimit_in_service()) {
>> + dirtylimit_state_unlock();
>
> Need to return?
>
>> + }
>> +
>> + CPU_FOREACH(cpu) {
>> + if (!dirtylimit_vcpu_get_state(cpu->cpu_index)->enabled) {
>> + continue;
>> + }
>> + dirtylimit_adjust_throttle(cpu);
>> + }
>> + dirtylimit_state_unlock();
>> + }
>> +}
>
> [...]
>
>> +void dirtylimit_set_vcpu(int cpu_index,
>> + uint64_t quota,
>> + bool enable)
>> +{
>> + dirtylimit_vcpu_set_quota(cpu_index, quota, enable);
>> + trace_dirtylimit_set_vcpu(cpu_index, quota);
>> +}
>
> This helper is not "help"ful.. How about wrapping the trace into
> dirtylimit_vcpu_set_quota, then drop it?
>
> Thanks,
>
--
Best regard
Hyman Huang(黄勇)
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH v14 6/7] softmmu/dirtylimit: Implement virtual CPU throttle
2022-02-14 8:20 ` Peter Xu
2022-02-14 9:05 ` Hyman Huang
@ 2022-02-14 9:22 ` Hyman Huang
1 sibling, 0 replies; 22+ messages in thread
From: Hyman Huang @ 2022-02-14 9:22 UTC (permalink / raw)
To: Peter Xu
Cc: Eduardo Habkost, David Hildenbrand, Juan Quintela,
Richard Henderson, qemu-devel, Markus ArmBruster, Paolo Bonzini,
Philippe Mathieu-Daudé,
Dr. David Alan Gilbert
在 2022/2/14 16:20, Peter Xu 写道:
> On Fri, Feb 11, 2022 at 12:17:40AM +0800, huangy81@chinatelecom.cn wrote:
>> @@ -2964,8 +2971,13 @@ int kvm_cpu_exec(CPUState *cpu)
>> */
>> trace_kvm_dirty_ring_full(cpu->cpu_index);
>> qemu_mutex_lock_iothread();
>> - kvm_dirty_ring_reap(kvm_state, NULL);
>> + if (dirtylimit_in_service()) {
>> + kvm_dirty_ring_reap(kvm_state, cpu);
>> + } else {
>> + kvm_dirty_ring_reap(kvm_state, NULL);
>> + }
>
> Could you add some comment here on why the cpu pointer is conditionally passed
> into the reaping routine? Even if we know it now, it's not immediately obvious
> to all the readers.
>
> [...]
>
>> +struct {
>> + VcpuDirtyLimitState *states;
>> + /* Max cpus number configured by user */
>> + int max_cpus;
>> + /* Number of vcpu under dirtylimit */
>> + int limited_nvcpu;
>> + /* Function to implement throttle set up */
>> + DirtyLimitFunc setup;
>
> "setup" normally is used only at startup of something, but not per interval.
> Perhaps "process" or "adjust"? Same across other "setup" namings across the
> patch.
>
> Again, I'd rather call the function directly..
>
> [...]
>
>> +static void dirtylimit_adjust_throttle(CPUState *cpu)
>> +{
>> + uint64_t quota = 0;
>> + uint64_t current = 0;
>> + int cpu_index = cpu->cpu_index;
>> +
>> + quota = dirtylimit_vcpu_get_state(cpu_index)->quota;
>> + current = vcpu_dirty_rate_get(cpu_index);
>> +
>> + if (current == 0) {
>> + cpu->throttle_us_per_full = 0;
>> + goto end;
>
> Can be dropped?
Ok, i'll move this block into dirtylimit_set_throttle.
>
>> + } else if (dirtylimit_done(quota, current)) {
>> + goto end;
>
> Same here. Dropping it wholely and:
>
> } else if (!dirtylimit_done(quota, current)) {
> dirtylimit_set_throttle(cpu, quota, current);
> }
>
> Would work?
Yes.
>
>> + } else {
>> + dirtylimit_set_throttle(cpu, quota, current);
>> + }
>> +end:
>
> Can be dropped?Ok
>
>> + trace_dirtylimit_adjust_throttle(cpu_index,
>> + quota, current,
>> + cpu->throttle_us_per_full);
>> + return;
>> +}
>> +
>> +void dirtylimit_setup(void)
>> +{
>> + CPUState *cpu;
>> +
>> + if (!qatomic_read(&dirtylimit_quit)) {
>> + dirtylimit_state_lock();
>> +
>> + if (!dirtylimit_in_service()) {
>> + dirtylimit_state_unlock();
>
> Need to return?
My fault. :(
>
>> + }
>> +
>> + CPU_FOREACH(cpu) {
>> + if (!dirtylimit_vcpu_get_state(cpu->cpu_index)->enabled) {
>> + continue;
>> + }
>> + dirtylimit_adjust_throttle(cpu);
>> + }
>> + dirtylimit_state_unlock();
>> + }
>> +}
>
> [...]
>
>> +void dirtylimit_set_vcpu(int cpu_index,
>> + uint64_t quota,
>> + bool enable)
>> +{
>> + dirtylimit_vcpu_set_quota(cpu_index, quota, enable);
>> + trace_dirtylimit_set_vcpu(cpu_index, quota);
>> +}
>
> This helper is not "help"ful.. How about wrapping the trace into
> dirtylimit_vcpu_set_quota, then drop it?
>
Ok.
> Thanks,
>
--
Best regard
Hyman Huang(黄勇)
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH v14 7/7] softmmu/dirtylimit: Implement dirty page rate limit
2022-02-14 8:25 ` Peter Xu
@ 2022-02-14 9:24 ` Hyman Huang
0 siblings, 0 replies; 22+ messages in thread
From: Hyman Huang @ 2022-02-14 9:24 UTC (permalink / raw)
To: Peter Xu
Cc: Eduardo Habkost, David Hildenbrand, Juan Quintela,
Richard Henderson, qemu-devel, Markus ArmBruster, Paolo Bonzini,
Philippe Mathieu-Daudé,
Dr. David Alan Gilbert
在 2022/2/14 16:25, Peter Xu 写道:
> On Fri, Feb 11, 2022 at 12:17:41AM +0800, huangy81@chinatelecom.cn wrote:
>> +static struct DirtyLimitInfoList *dirtylimit_query_all(void)
>> +{
>> + int i, index;
>> + DirtyLimitInfo *info = NULL;
>> + DirtyLimitInfoList *head = NULL, **tail = &head;
>> +
>> + dirtylimit_state_lock();
>> +
>> + if (!dirtylimit_in_service()) {
>
> Need to unlock?
Yes, i'll fix it next version.
>
>> + return NULL;
>> + }
>> +
>> + for (i = 0; i < dirtylimit_state->max_cpus; i++) {
>> + index = dirtylimit_state->states[i].cpu_index;
>> + if (dirtylimit_vcpu_get_state(index)->enabled) {
>> + info = dirtylimit_query_vcpu(index);
>> + QAPI_LIST_APPEND(tail, info);
>> + }
>> + }
>> +
>> + dirtylimit_state_unlock();
>> +
>> + return head;
>> +}
>
--
Best regard
Hyman Huang(黄勇)
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH v14 3/7] migration/dirtyrate: Refactor dirty page rate calculation
2022-02-14 8:40 ` Hyman Huang
@ 2022-02-14 9:47 ` Peter Xu
0 siblings, 0 replies; 22+ messages in thread
From: Peter Xu @ 2022-02-14 9:47 UTC (permalink / raw)
To: Hyman Huang
Cc: Eduardo Habkost, David Hildenbrand, Juan Quintela,
Richard Henderson, qemu-devel, Markus ArmBruster, Paolo Bonzini,
Philippe Mathieu-Daudé,
Dr. David Alan Gilbert
On Mon, Feb 14, 2022 at 04:40:31PM +0800, Hyman Huang wrote:
> Dropping this parameter is fine from my point view if we ignore the duration
> error result from the delay between caller and callee of
> 'vcpu_calculate_dirtyrate’
Agreed, then let's drop it.
--
Peter Xu
^ permalink raw reply [flat|nested] 22+ messages in thread
* Re: [PATCH v14 6/7] softmmu/dirtylimit: Implement virtual CPU throttle
2022-02-14 9:05 ` Hyman Huang
@ 2022-02-14 9:51 ` Peter Xu
0 siblings, 0 replies; 22+ messages in thread
From: Peter Xu @ 2022-02-14 9:51 UTC (permalink / raw)
To: Hyman Huang
Cc: Eduardo Habkost, David Hildenbrand, Juan Quintela,
Richard Henderson, qemu-devel, Markus ArmBruster, Paolo Bonzini,
Philippe Mathieu-Daudé,
Dr. David Alan Gilbert
On Mon, Feb 14, 2022 at 05:05:38PM +0800, Hyman Huang wrote:
> But i'm ok if you insist because it's just about how to call the
> 'dirtylimit' and doesn't affect the whole logic.
If you don't have plan to add e.g. another adjust() method then it's pointless.
The hook can be easily added on top when necessary.
But I don't insist - your call. :) Not really a big deal.
--
Peter Xu
^ permalink raw reply [flat|nested] 22+ messages in thread
end of thread, other threads:[~2022-02-14 9:53 UTC | newest]
Thread overview: 22+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-02-10 16:17 [PATCH v14 0/7] support dirty restraint on vCPU huangy81
2022-02-10 16:17 ` [PATCH v14 1/7] accel/kvm/kvm-all: Refactor per-vcpu dirty ring reaping huangy81
2022-02-14 7:57 ` Peter Xu
2022-02-10 16:17 ` [PATCH v14 2/7] cpus: Introduce cpu_list_generation_id huangy81
2022-02-14 7:59 ` Peter Xu
2022-02-10 16:17 ` [PATCH v14 3/7] migration/dirtyrate: Refactor dirty page rate calculation huangy81
2022-02-14 7:57 ` Peter Xu
2022-02-14 8:40 ` Hyman Huang
2022-02-14 9:47 ` Peter Xu
2022-02-10 16:17 ` [PATCH v14 4/7] softmmu/dirtylimit: Implement vCPU dirtyrate calculation periodically huangy81
2022-02-14 8:01 ` Peter Xu
2022-02-10 16:17 ` [PATCH v14 5/7] accel/kvm/kvm-all: Introduce kvm_dirty_ring_size function huangy81
2022-02-14 8:02 ` Peter Xu
2022-02-10 16:17 ` [PATCH v14 6/7] softmmu/dirtylimit: Implement virtual CPU throttle huangy81
2022-02-14 8:20 ` Peter Xu
2022-02-14 9:05 ` Hyman Huang
2022-02-14 9:51 ` Peter Xu
2022-02-14 9:22 ` Hyman Huang
2022-02-10 16:17 ` [PATCH v14 7/7] softmmu/dirtylimit: Implement dirty page rate limit huangy81
2022-02-11 13:40 ` Markus Armbruster
2022-02-14 8:25 ` Peter Xu
2022-02-14 9:24 ` Hyman Huang
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.