4:25 p.m.
Add/fix at least some virsh documentation for the perf events and remove the
rather useless schema-test XML.
Peter Krempa (2):
docs: virsh: Add minimal documentation for 'mbmt' and 'mbml' perf
events
tests: schema: Remove useless perf schema data
tests/domainschemadata/domain-perf-simple.xml | 20 --------------------
tools/virsh.pod | 11 ++++++++---
2 files changed, 8 insertions(+), 23 deletions(-)
delete mode 100644 tests/domainschemadata/domain-perf-simple.xml
--
2.8.3
4:25 p.m.
New subject: [libvirt] [PATCH 1/2] docs: virsh: Add minimal documentation for 'mbmt' and 'mbml' perf events
---
tools/virsh.pod | 11 ++++++++---
1 file changed, 8 insertions(+), 3 deletions(-)
diff --git a/tools/virsh.pod b/tools/virsh.pod
index 1e56660..b11e4c3 100644
--- a/tools/virsh.pod
+++ b/tools/virsh.pod
@@ -901,6 +901,8 @@ I<--interface> returns:
I<--perf> returns the statistics of all enabled perf events:
"perf.cache" - the cache usage in Byte currently used
+"perf.mbmt" - total system bandwidth from one level of cache
+"perf.mbml" - bandwidth of memory traffic for a memory controller
I<--block> returns information about disks associated with each
domain. Using the I<--backing> flag extends this information to
@@ -2180,9 +2182,12 @@ from different sources can be supported by perf.
Currently only QEMU/KVM supports this command. The I<--enable> and
I<--disable>
option combined with B<eventSpec> can be used to enabled or disable specific
performance event. B<eventSpec> is a string list of one or more events
-separated by commas. However, just "cmt" event is supported presently. CMT is
a
-PQos (Platform Qos) feature to monitor the usage of cache by applications
-running on the platform.
+separated by commas. Valid event names are "cmt", "mbmt",
"mbml".
+CMT is a PQos (Platform Qos) feature to monitor the usage of cache by
+applications running on the platform.
+
+The statistics can be retrieved using the B<domstats> command using the
+I<--perf> flag.
If I<--live> is specified, affect a running guest.
If I<--config> is specified, affect the next boot of a persistent guest.
--
2.8.3
5:49 p.m.
New subject: [libvirt] [PATCH 1/2] docs: virsh: Add minimal documentation for 'mbmt' and 'mbml' perf events
On 06/17/2016 04:25 AM, Peter Krempa wrote:
---
tools/virsh.pod | 11 ++++++++---
1 file changed, 8 insertions(+), 3 deletions(-)
diff --git a/tools/virsh.pod b/tools/virsh.pod
index 1e56660..b11e4c3 100644
--- a/tools/virsh.pod
+++ b/tools/virsh.pod
@@ -901,6 +901,8 @@ I<--interface> returns:
I<--perf> returns the statistics of all enabled perf events:
"perf.cache" - the cache usage in Byte currently used
^^^
cmt?
Like a gift that keeps on giving... Adjust it now or trivially in a pre
or follow up patch.
+"perf.mbmt" - total system bandwidth from one level of
cache
+"perf.mbml" - bandwidth of memory traffic for a memory controller
I<--block> returns information about disks associated with each
domain. Using the I<--backing> flag extends this information to
@@ -2180,9 +2182,12 @@ from different sources can be supported by perf.
Currently only QEMU/KVM supports this command. The I<--enable> and
I<--disable>
option combined with B<eventSpec> can be used to enabled or disable specific
performance event. B<eventSpec> is a string list of one or more events
-separated by commas. However, just "cmt" event is supported presently. CMT is
a
-PQos (Platform Qos) feature to monitor the usage of cache by applications
-running on the platform.
+separated by commas. Valid event names are "cmt", "mbmt",
"mbml".
+CMT is a PQos (Platform Qos) feature to monitor the usage of cache by
+applications running on the platform.
MBM (Memory Bandwidth Mpnitoring) provides a way to monitor the Total
system memory bandwidth between one level of cache and another (mbmt)
and the amount of data (bytes/s) sent through the memory controller on
the socket (mbml).
ACK w/ this (text paraphrased from commit id '90b9995d1' before I
completed my first cup of coffee)
John
+
+The statistics can be retrieved using the B<domstats> command using the
+I<--perf> flag.
If I<--live> is specified, affect a running guest.
If I<--config> is specified, affect the next boot of a persistent guest.
6:42 p.m.
New subject: [libvirt] [PATCH 1/2] docs: virsh: Add minimal documentation for 'mbmt' and 'mbml' perf events
On Fri, Jun 17, 2016 at 05:49:45 -0400, John Ferlan wrote:
On 06/17/2016 04:25 AM, Peter Krempa wrote:
> ---
> tools/virsh.pod | 11 ++++++++---
> 1 file changed, 8 insertions(+), 3 deletions(-)
>
> diff --git a/tools/virsh.pod b/tools/virsh.pod
> index 1e56660..b11e4c3 100644
> --- a/tools/virsh.pod
> +++ b/tools/virsh.pod
> @@ -901,6 +901,8 @@ I<--interface> returns:
>
> I<--perf> returns the statistics of all enabled perf events:
> "perf.cache" - the cache usage in Byte currently used
^^^
cmt?
I made that change and then overwrote the file from a different terminal
and forgot to re-do the change. Sigh. Thanks.
Like a gift that keeps on giving... Adjust it now or trivially in a pre
or follow up patch.
> +"perf.mbmt" - total system bandwidth from one level of cache
> +"perf.mbml" - bandwidth of memory traffic for a memory controller
>
> I<--block> returns information about disks associated with each
> domain. Using the I<--backing> flag extends this information to
> @@ -2180,9 +2182,12 @@ from different sources can be supported by perf.
> Currently only QEMU/KVM supports this command. The I<--enable> and
I<--disable>
> option combined with B<eventSpec> can be used to enabled or disable specific
> performance event. B<eventSpec> is a string list of one or more events
> -separated by commas. However, just "cmt" event is supported presently.
CMT is a
> -PQos (Platform Qos) feature to monitor the usage of cache by applications
> -running on the platform.
> +separated by commas. Valid event names are "cmt", "mbmt",
"mbml".
> +CMT is a PQos (Platform Qos) feature to monitor the usage of cache by
> +applications running on the platform.
MBM (Memory Bandwidth Mpnitoring) provides a way to monitor the Total
system memory bandwidth between one level of cache and another (mbmt)
and the amount of data (bytes/s) sent through the memory controller on
the socket (mbml).
Okay I'll add this.
ACK w/ this (text paraphrased from commit id '90b9995d1' before I
completed my first cup of coffee)
Thanks, I'll push it shortly.
4:25 p.m.
New subject: [libvirt] [PATCH 2/2] tests: schema: Remove useless perf schema data
We have a test case that excercises the parser and formatter now which
takes part in schema checking so remove the schema-only test.
---
tests/domainschemadata/domain-perf-simple.xml | 20 --------------------
1 file changed, 20 deletions(-)
delete mode 100644 tests/domainschemadata/domain-perf-simple.xml
diff --git a/tests/domainschemadata/domain-perf-simple.xml
b/tests/domainschemadata/domain-perf-simple.xml
deleted file mode 100644
index d7be7c9..0000000
--- a/tests/domainschemadata/domain-perf-simple.xml
+++ /dev/null
@@ -1,20 +0,0 @@
-<domain type='qemu'>
- <name>QEMUGuest1</name>
- <uuid>c7a5fdbd-edaf-9455-926a-d65c16db1809</uuid>
- <currentMemory unit='KiB'>219136</currentMemory>
- <vcpu placement='static'>1</vcpu>
- <os>
- <type arch='x86_64' machine='pc'>hvm</type>
- <boot dev='hd'/>
- </os>
- <perf>
- <event name='cmt' enabled='yes'/>
- </perf>
- <clock offset='utc'/>
- <on_poweroff>destroy</on_poweroff>
- <on_reboot>restart</on_reboot>
- <on_crash>destroy</on_crash>
- <devices>
- <emulator>/usr/bin/qemu</emulator>
- </devices>
-</domain>
--
2.8.3
5:50 p.m.
New subject: [libvirt] [PATCH 2/2] tests: schema: Remove useless perf schema data
On 06/17/2016 04:25 AM, Peter Krempa wrote:
We have a test case that excercises the parser and formatter now
which
takes part in schema checking so remove the schema-only test.
---
tests/domainschemadata/domain-perf-simple.xml | 20 --------------------
1 file changed, 20 deletions(-)
delete mode 100644 tests/domainschemadata/domain-perf-simple.xml
ACK
John
3059
days inactive
3059
days old
5 comments
2 participants
participants (2)
-
John Ferlan -
Peter Krempa