github/acrn-hypervisor
1
0
Fork 0
mirror of https://github.com/projectacrn/acrn-hypervisor.git synced 2025-06-24 14:33:38 +00:00
Code Issues Projects Releases Wiki Activity

Doc: Content edit to cpu-sharing page

Browse Source 
Signed-off-by: Deb Taylor <deb.taylor@intel.com>
This commit is contained in:
Deb Taylor 2019-12-21 09:59:55 -05:00 committed by deb-intel
parent fa5922c8bf
commit 3e45d5e301
1 changed files with 24 additions and 24 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
doc/tutorials/cpu_sharing.rst
View File

@ -34,6 +34,7 @@ Here is an example for affinity:
- VM0: 2 vCPUs, pinned to pCPU0 and pCPU1 - VM0: 2 vCPUs, pinned to pCPU0 and pCPU1
- VM1: 2 vCPUs, pinned to pCPU2 and pCPU3 - VM1: 2 vCPUs, pinned to pCPU2 and pCPU3
- VM2: 2 vCPUs, pinned to pCPU2 and pCPU3 - VM2: 2 vCPUs, pinned to pCPU2 and pCPU3
.. figure:: images/cpu_sharing_affinity.png .. figure:: images/cpu_sharing_affinity.png
:align: center :align: center
@ -61,10 +62,10 @@ Below block diagram shows the basic concept for the scheduler. There are two kin
IORR (IO sensitive round-robin) scheduler is implemented with per-pCPU runqueue and per-pCPU tick timer, it supports more than one vCPU running on a pCPU. The scheduler schedule thread objects in round-robin policy basically, and support preemption by timeslice counting. IORR (IO sensitive round-robin) scheduler is implemented with per-pCPU runqueue and per-pCPU tick timer, it supports more than one vCPU running on a pCPU. The scheduler schedule thread objects in round-robin policy basically, and support preemption by timeslice counting.
+ Every thread object has an initial timeslice (ex: 10ms) - Every thread object has an initial timeslice (ex: 10ms)
+ timeslice is consumed with time and be counted in context switch and tick handler - Timeslice is consumed with time and be counted in context switch and tick handler
+ If timeslice is positive or zero, then switch out current thread object and put it to tail of runqueue.Then, pick next runnable one from runqueue to run. - If timeslice is positive or zero, then switch out current thread object and put it to tail of runqueue.Then, pick next runnable one from runqueue to run.
+ Threads who has IO request will preempt current running thread on the same pCPU. - Threads who has IO request will preempt current running thread on the same pCPU.
Scheduler configuration Scheduler configuration
*********************** ***********************
@ -74,19 +75,18 @@ There are two place in the code decide the usage for scheduler.
* The option in Kconfig decides the only scheduler used in runtime. * The option in Kconfig decides the only scheduler used in runtime.
``hypervisor/arch/x86/Kconfig`` ``hypervisor/arch/x86/Kconfig``
.. literalinclude:: ../hypervisor/arch/x86/Kconfig .. literalinclude:: ../../../../hypervisor/arch/x86/Kconfig
:name: Kconfig for Scheduler :name: Kconfig for Scheduler
:caption: Kconfig for Scheduler :caption: Kconfig for Scheduler
:linenos: :linenos:
:lines: 40-58 :lines: 40-58
:emphasize-lines: 42
:language: c :language: c
The default scheduler is **SCHED_NOOP**. To use the IORR, only need to set **SCHED_IORR** in **ACRN Scheduler**. The default scheduler is **SCHED_NOOP**. To use the IORR, only need to set **SCHED_IORR** in **ACRN Scheduler**.
* The affinity for VMs are set in ``hypervisor/scenarios/<scenario_name>/vm_configurations.h`` * The affinity for VMs are set in ``hypervisor/scenarios/<scenario_name>/vm_configurations.h``
.. literalinclude:: ../hypervisor/scenarios/industry/vm_configurations.h .. literalinclude:: ../../../..//hypervisor/scenarios/industry/vm_configurations.h
:name: Affinity for VMs :name: Affinity for VMs
:caption: Affinity for VMs :caption: Affinity for VMs
:linenos: :linenos: