summary refs log tree commit diff stats
path: root/results/classifier/118/performance/864
diff options
context:
space:
mode:
Diffstat (limited to 'results/classifier/118/performance/864')
-rw-r--r--results/classifier/118/performance/86445
1 files changed, 0 insertions, 45 deletions
diff --git a/results/classifier/118/performance/864 b/results/classifier/118/performance/864
deleted file mode 100644
index 6db3a8b57..000000000
--- a/results/classifier/118/performance/864
+++ /dev/null
@@ -1,45 +0,0 @@
-performance: 0.910
-hypervisor: 0.866
-virtual: 0.855
-device: 0.840
-graphic: 0.777
-semantic: 0.754
-architecture: 0.661
-user-level: 0.485
-network: 0.459
-debug: 0.448
-mistranslation: 0.414
-kernel: 0.385
-permissions: 0.361
-register: 0.346
-vnc: 0.346
-risc-v: 0.300
-peripherals: 0.283
-PID: 0.273
-socket: 0.257
-ppc: 0.244
-i386: 0.220
-x86: 0.219
-boot: 0.218
-arm: 0.208
-files: 0.193
-assembly: 0.163
-TCG: 0.140
-VMM: 0.131
-KVM: 0.056
-
-HVF virtual counter diverges from CLOCK_VIRTUAL when the host sleeps
-Description of problem:
-HVF's virtual counter diverges from `CLOCK_VIRTUAL` when the host sleeps and causes the inconsistency between Linux's system counter and everything else.
-
-HVF's virtual counter apparently relies on something similar to `mach_absolute_time`, which stops when the host sleeps and resumes after it wakes up. However, `CLOCK_VIRTUAL` is implemented with `mach_continuous_time`, which continues even while the host sleeps. Linux uses the virtual counter as the source of the system counter and sees inconsistencies between the system counter and the other devices.
-Steps to reproduce:
-1. Launch Fedora.
-2. Compare the time shown at the top of the guest display and one at the top of the host display. The difference should be less than 2 minutes.
-3. Let the host sleep for 3 minutes.
-4. Compare the times again. The difference is now greater than 2 minutes.
-Additional information:
-Here are solutions I've came up with so far. There are trade-offs but any of them should be better than the current situation. I'm happy to implement one if the maintainers have decided which one is the best or figure out a superior alternative.
-- Implement `cpus_get_virtual_clock` of `AccelOpsClass` with `mach_absolute_time`. It would make HVF inconsistent with the other accelerators. Linux also expects the virtual clock is "continuous" and it leaves the divergence from the real time.
-- Request XNU `HOST_NOTIFY_CALENDAR_CHANGE` to update the virtual clock with the continuous time. The interface is undocumented.
-- Use `IORegisterForSystemPower` to update the virtual clock with the continuous time. It is undocumented that the interface handles every cases where `mach_absolute_time` and `mach_continuous_time`, but it actually does if I read XNU's source code correctly.