diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2020-02-16 13:01:42 -0800 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2020-02-16 13:01:42 -0800 |
commit | 44024adb4aabefd275c6f9c00cac323473447dd5 (patch) | |
tree | 386e7975f0702729eaf0cd596c9b385825d7d257 /Documentation/virt/guest-halt-polling.rst | |
parent | b982df72ef8404de99597f1d956b6137533b429a (diff) | |
parent | 120881b9e888689cbdb90a1dd1689684d8bc95f3 (diff) |
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Pull KVM fixes from Paolo Bonzini:
"Bugfixes and improvements to selftests.
On top of this, Mauro converted the KVM documentation to rst format,
which was very welcome"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (44 commits)
docs: virt: guest-halt-polling.txt convert to ReST
docs: kvm: review-checklist.txt: rename to ReST
docs: kvm: Convert timekeeping.txt to ReST format
docs: kvm: Convert s390-diag.txt to ReST format
docs: kvm: Convert ppc-pv.txt to ReST format
docs: kvm: Convert nested-vmx.txt to ReST format
docs: kvm: Convert mmu.txt to ReST format
docs: kvm: Convert locking.txt to ReST format
docs: kvm: Convert hypercalls.txt to ReST format
docs: kvm: arm/psci.txt: convert to ReST
docs: kvm: convert arm/hyp-abi.txt to ReST
docs: kvm: Convert api.txt to ReST format
docs: kvm: convert devices/xive.txt to ReST
docs: kvm: convert devices/xics.txt to ReST
docs: kvm: convert devices/vm.txt to ReST
docs: kvm: convert devices/vfio.txt to ReST
docs: kvm: convert devices/vcpu.txt to ReST
docs: kvm: convert devices/s390_flic.txt to ReST
docs: kvm: convert devices/mpic.txt to ReST
docs: kvm: convert devices/arm-vgit.txt to ReST
...
Diffstat (limited to 'Documentation/virt/guest-halt-polling.rst')
-rw-r--r-- | Documentation/virt/guest-halt-polling.rst | 84 |
1 files changed, 84 insertions, 0 deletions
diff --git a/Documentation/virt/guest-halt-polling.rst b/Documentation/virt/guest-halt-polling.rst new file mode 100644 index 000000000000..b4e747942417 --- /dev/null +++ b/Documentation/virt/guest-halt-polling.rst @@ -0,0 +1,84 @@ +================== +Guest halt polling +================== + +The cpuidle_haltpoll driver, with the haltpoll governor, allows +the guest vcpus to poll for a specified amount of time before +halting. + +This provides the following benefits to host side polling: + + 1) The POLL flag is set while polling is performed, which allows + a remote vCPU to avoid sending an IPI (and the associated + cost of handling the IPI) when performing a wakeup. + + 2) The VM-exit cost can be avoided. + +The downside of guest side polling is that polling is performed +even with other runnable tasks in the host. + +The basic logic as follows: A global value, guest_halt_poll_ns, +is configured by the user, indicating the maximum amount of +time polling is allowed. This value is fixed. + +Each vcpu has an adjustable guest_halt_poll_ns +("per-cpu guest_halt_poll_ns"), which is adjusted by the algorithm +in response to events (explained below). + +Module Parameters +================= + +The haltpoll governor has 5 tunable module parameters: + +1) guest_halt_poll_ns: + +Maximum amount of time, in nanoseconds, that polling is +performed before halting. + +Default: 200000 + +2) guest_halt_poll_shrink: + +Division factor used to shrink per-cpu guest_halt_poll_ns when +wakeup event occurs after the global guest_halt_poll_ns. + +Default: 2 + +3) guest_halt_poll_grow: + +Multiplication factor used to grow per-cpu guest_halt_poll_ns +when event occurs after per-cpu guest_halt_poll_ns +but before global guest_halt_poll_ns. + +Default: 2 + +4) guest_halt_poll_grow_start: + +The per-cpu guest_halt_poll_ns eventually reaches zero +in case of an idle system. This value sets the initial +per-cpu guest_halt_poll_ns when growing. This can +be increased from 10000, to avoid misses during the initial +growth stage: + +10k, 20k, 40k, ... (example assumes guest_halt_poll_grow=2). + +Default: 50000 + +5) guest_halt_poll_allow_shrink: + +Bool parameter which allows shrinking. Set to N +to avoid it (per-cpu guest_halt_poll_ns will remain +high once achieves global guest_halt_poll_ns value). + +Default: Y + +The module parameters can be set from the debugfs files in:: + + /sys/module/haltpoll/parameters/ + +Further Notes +============= + +- Care should be taken when setting the guest_halt_poll_ns parameter as a + large value has the potential to drive the cpu usage to 100% on a machine + which would be almost entirely idle otherwise. |