Skip to content

cloud-kernel
cloud-kernel

openanolis / cloud-kernel
1 年多 前同步成功

通知 160
Star 36
Fork 7
cloud-kernel
cloud-kernel
切换分支/标签
  1. 08 5月, 2015 1 次提交
  3. 15 4月, 2015 1 次提交
  5. 01 4月, 2015 3 次提交
  7. 31 3月, 2015 1 次提交
    • C
      KVM: s390: enable more features that need no hypervisor changes · a3ed8dae
      由 Christian Borntraeger 提交于 
      After some review about what these facilities do, the following
facilities will work under KVM and can, therefore, be reported
to the guest if the cpu model and the host cpu provide this bit.

There are plans underway to make the whole bit thing more readable,
but its not yet finished. So here are some last bit changes and
we enhance the KVM mask with:

9 The sense-running-status facility is installed in the
  z/Architecture architectural mode.
  ---> handled by SIE or KVM

10 The conditional-SSKE facility is installed in the
   z/Architecture architectural mode.
  ---> handled by SIE. KVM will retry SIE

13 The IPTE-range facility is installed in the
   z/Architecture architectural mode.
  ---> handled by SIE. KVM will retry SIE

36 The enhanced-monitor facility is installed in the
   z/Architecture architectural mode.
  ---> handled by SIE

47 The CMPSC-enhancement facility is installed in the
   z/Architecture architectural mode.
  ---> handled by SIE

48 The decimal-floating-point zoned-conversion facility
   is installed in the z/Architecture architectural mode.
  ---> handled by SIE

49 The execution-hint, load-and-trap, miscellaneous-
   instruction-extensions and processor-assist
  ---> handled by SIE

51 The local-TLB-clearing facility is installed in the
   z/Architecture architectural mode.
  ---> handled by SIE

52 The interlocked-access facility 2 is installed.
  ---> handled by SIE

53 The load/store-on-condition facility 2 and load-and-
   zero-rightmost-byte facility are installed in the
   z/Architecture architectural mode.
  ---> handled by SIE

57 The message-security-assist-extension-5 facility is
  installed in the z/Architecture architectural mode.
  ---> handled by SIE

66 The reset-reference-bits-multiple facility is installed
  in the z/Architecture architectural mode.
  ---> handled by SIE. KVM will retry SIE

80 The decimal-floating-point packed-conversion
   facility is installed in the z/Architecture architectural
   mode.
  ---> handled by SIE
Signed-off-by: NChristian Borntraeger <borntraeger@de.ibm.com>
Tested-by: NMichael Mueller <mimu@linux.vnet.ibm.com>
Acked-by: NCornelia Huck <cornelia.huck@de.ibm.com>
      a3ed8dae
  9. 17 3月, 2015 7 次提交
  11. 11 3月, 2015 1 次提交
  13. 06 3月, 2015 6 次提交
  15. 04 3月, 2015 4 次提交
  17. 03 3月, 2015 1 次提交
  19. 09 2月, 2015 4 次提交
  21. 06 2月, 2015 1 次提交
    • P
      kvm: add halt_poll_ns module parameter · f7819512
      由 Paolo Bonzini 提交于 
      This patch introduces a new module parameter for the KVM module; when it
is present, KVM attempts a bit of polling on every HLT before scheduling
itself out via kvm_vcpu_block.

This parameter helps a lot for latency-bound workloads---in particular
I tested it with O_DSYNC writes with a battery-backed disk in the host.
In this case, writes are fast (because the data doesn't have to go all
the way to the platters) but they cannot be merged by either the host or
the guest.  KVM's performance here is usually around 30% of bare metal,
or 50% if you use cache=directsync or cache=writethrough (these
parameters avoid that the guest sends pointless flush requests, and
at the same time they are not slow because of the battery-backed cache).
The bad performance happens because on every halt the host CPU decides
to halt itself too.  When the interrupt comes, the vCPU thread is then
migrated to a new physical CPU, and in general the latency is horrible
because the vCPU thread has to be scheduled back in.

With this patch performance reaches 60-65% of bare metal and, more
important, 99% of what you get if you use idle=poll in the guest.  This
means that the tunable gets rid of this particular bottleneck, and more
work can be done to improve performance in the kernel or QEMU.

Of course there is some price to pay; every time an otherwise idle vCPUs
is interrupted by an interrupt, it will poll unnecessarily and thus
impose a little load on the host.  The above results were obtained with
a mostly random value of the parameter (500000), and the load was around
1.5-2.5% CPU usage on one of the host's core for each idle guest vCPU.

The patch also adds a new stat, /sys/kernel/debug/kvm/halt_successful_poll,
that can be used to tune the parameter.  It counts how many HLT
instructions received an interrupt during the polling period; each
successful poll avoids that Linux schedules the VCPU thread out and back
in, and may also avoid a likely trip to C1 and back for the physical CPU.

While the VM is idle, a Linux 4 VCPU VM halts around 10 times per second.
Of these halts, almost all are failed polls.  During the benchmark,
instead, basically all halts end within the polling period, except a more
or less constant stream of 50 per second coming from vCPUs that are not
running the benchmark.  The wasted time is thus very low.  Things may
be slightly different for Windows VMs, which have a ~10 ms timer tick.

The effect is also visible on Marcelo's recently-introduced latency
test for the TSC deadline timer.  Though of course a non-RT kernel has
awful latency bounds, the latency of the timer is around 8000-10000 clock
cycles compared to 20000-120000 without setting halt_poll_ns.  For the TSC
deadline timer, thus, the effect is both a smaller average latency and
a smaller variance.
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
      f7819512
  23. 23 1月, 2015 10 次提交