cpuidle haltpoll driver and governor (v4)
Get series with b4
b4 am [email protected]Versions newest first
- v1 Latest5 patches 6 repliesIn review
V2 from the cover letter
- Move from x86 to generic code (Paolo/Christian)
- Add auto-tuning logic (Paolo)
- Add MSR to disable host side polling (Paolo)
Patches in v1 No applies recorded yet
| # | Subject and files | Size | Replies | Applied |
|---|---|---|---|---|
| 1/5 | drivers/cpuidle: add cpuidle-haltpoll driverarch/ drivers/ drivers/ +1 more | +76 | 2 replies | |
| 2/5 | cpuidle: add get_poll_time callbackdrivers/ drivers/ include/ | +50 | 1 reply | |
| 3/5 | cpuidle: add haltpoll governorDocumentation/ drivers/ drivers/ +1 more | +266 | 3 replies | |
| 4/5 | kvm: x86: add host poll control msrsDocumentation/ arch/ arch/ +3 more | +39 | 0 replies | |
| 5/5 | cpuidle-haltpoll: disable host side polling when kvm virtualizedarch/ arch/ arch/ +2 more | +82 | 0 replies |
Reviews trailers per patch
1 of 5 patches has a Reviewed-by or Acked-by.
| Patch | Rafael J. Wysocki |
|---|---|
| 1/5 | |
| 2/5 | A-b |
| 3/5 | |
| 4/5 | |
| 5/5 |
Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes.
Cover letter [patch 0/5] cpuidle haltpoll driver and governor (v4)
The cpuidle-haltpoll driver with 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.
Results comparing halt_poll_ns and server/client application where a small packet is ping-ponged:
host --> 31.33 halt_poll_ns=300000 / no guest busy spin --> 33.40 (93.8%) halt_poll_ns=0 / guest_halt_poll_ns=300000 --> 32.73 (95.7%)
For the SAP HANA benchmarks (where idle_spin is a parameter of the previous version of the patch, results should be the same):
hpns == halt_poll_ns
idle_spin=0/ idle_spin=800/ idle_spin=0/
hpns=200000 hpns=0 hpns=800000
DeleteC06T03 (100 thread) 1.76 1.71 (-3%) 1.78 (+1%)
InsertC16T02 (100 thread) 2.14 2.07 (-3%) 2.18 (+1.8%)
DeleteC00T01 (1 thread) 1.34 1.28 (-4.5%) 1.29 (-3.7%)
UpdateC00T03 (1 thread) 4.72 4.18 (-12%) 4.53 (-5%)V2:
- Move from x86 to generic code (Paolo/Christian)
- Add auto-tuning logic (Paolo)
- Add MSR to disable host side polling (Paolo)
V3:
- Do not be specific about HLT VM-exit in the documentation (Ankur Arora)
- Mark tuning parameters static and __read_mostly (Andrea Arcangeli)
- Add WARN_ON if host does not support poll control (Joao Martins)
- Use sched_clock and cleanup haltpoll_enter_idle (Peter Zijlstra)
- Mark certain functions in kvm.c as static (kernel test robot)
- Remove tracepoints as they use RCU from extended quiescent state (kernel test robot)
V4:
- Use a haltpoll governor, use poll_state.c poll code (Rafael J. Wysocki)