[libvirt] [PATCH v4 5/9] conf: Adjust the iothreadsched expectations
John Ferlan
jferlan at redhat.com
Thu Apr 23 12:01:20 UTC 2015
On 04/23/2015 07:24 AM, Peter Krempa wrote:
> On Tue, Apr 21, 2015 at 19:31:26 -0400, John Ferlan wrote:
>> With iothreadid's allowing any 'id' value for an iothread_id, the
>> iothreadsched code needs a slight adjustment to allow for "any"
>> unsigned int value in order to create the bitmap of ids that will
>> have scheduler adjustments. Adjusted the doc description as well.
>>
>> Signed-off-by: John Ferlan <jferlan at redhat.com>
>> ---
>> docs/formatdomain.html.in | 6 +++++-
>> src/conf/domain_conf.c | 2 +-
>> .../qemuxml2argvdata/qemuxml2argv-cputune-iothreadsched-toomuch.xml | 1 +
>> 3 files changed, 7 insertions(+), 2 deletions(-)
>>
>> diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in
>> index 7af6bd7..99ce298 100644
>> --- a/docs/formatdomain.html.in
>> +++ b/docs/formatdomain.html.in
>> @@ -691,7 +691,11 @@
>> type (values <code>batch</code>, <code>idle</code>, <code>fifo</code>,
>> <code>rr</code>) for particular vCPU/IOThread threads (based on
>> <code>vcpus</code> and <code>iothreads</code>, leaving out
>> - <code>vcpus</code>/<code>iothreads</code> sets the default). For
>> + <code>vcpus</code>/<code>iothreads</code> sets the default). Valid
>> + <code>vcpus</code> values start at 0 through one less than the
>> + number of vCPU's defined for the domain. Valid <code>iothreads</code>
>> + values are described in the <code>iothreadids</code>
>> + <a href="#elementsIOThreadsAllocation"><code>description</code></a>. For
>
> <iothreadids> don't include the implicit ones that were added by
> <iothreads>, but this code should be able to set them.
>
It can, but I'll add the following to help out (at least ranges are
described now...):
If no <code>iothreadids</code> are defined, then libvirt numbers
IOThreads from 1 to the number of <code>iothreads</code> available
for the domain.
>> real-time schedulers (<code>fifo</code>, <code>rr</code>), priority must
>> be specified as well (and is ignored for non-real-time ones). The value
>> range for the priority depends on the host kernel (usually 1-99).
>> diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c
>> index 969e56f..3e8551b 100644
>> --- a/src/conf/domain_conf.c
>> +++ b/src/conf/domain_conf.c
>> @@ -14313,7 +14313,7 @@ virDomainDefParseXML(xmlDocPtr xml,
>>
>> for (i = 0; i < def->cputune.niothreadsched; i++) {
>> if (virDomainThreadSchedParse(nodes[i],
>> - 1, def->iothreads,
>> + 1, UINT_MAX,
>> "iothreads",
>> &def->cputune.iothreadsched[i]) < 0)
>> goto error;
>> diff --git a/tests/qemuxml2argvdata/qemuxml2argv-cputune-iothreadsched-toomuch.xml b/tests/qemuxml2argvdata/qemuxml2argv-cputune-iothreadsched-toomuch.xml
>> index 1540969..97a5cde 100644
>> --- a/tests/qemuxml2argvdata/qemuxml2argv-cputune-iothreadsched-toomuch.xml
>> +++ b/tests/qemuxml2argvdata/qemuxml2argv-cputune-iothreadsched-toomuch.xml
>> @@ -13,6 +13,7 @@
>> <vcpupin vcpu='1' cpuset='1'/>
>> <emulatorpin cpuset='1'/>
>> <vcpusched vcpus='0-1' scheduler='fifo' priority='1'/>
>> + <iothreadsched iothreads='1' scheduler='batch'/>
>> <iothreadsched iothreads='2' scheduler='batch'/>
>
> Both lines above are equivalent to
> <iothreadsched iothreads='1-2' scheduler='batch'/>. Change the scheduler
> type if there's a reason to have two lines.
>
OK - the reason for two lines is the "assumption" change that the
IOThreads are numbered 1..n. The existing test would have iothread_id=2
which would fail the > iothreads (of 1) check.
The following doesn't change anything,
<iothreadsched iothreads='1-2' scheduler='batch'/>
so, the following is needed:
- <iothreadsched iothreads='2' scheduler='batch'/>
+ <iothreadsched iothreads='1' scheduler='batch'/>
+ <iothreadsched iothreads='2' scheduler='fifo'/>
>> </cputune>
>> <os>
>
> Peter
>
More information about the libvir-list
mailing list