linux/arch/powerpc/kvm
Paul Mackerras c7b676709c KVM: PPC: Book3S HV: Fix accounting of stolen time
Currently the code that accounts stolen time tends to overestimate the
stolen time, and will sometimes report more stolen time in a DTL
(dispatch trace log) entry than has elapsed since the last DTL entry.
This can cause guests to underflow the user or system time measured
for some tasks, leading to ridiculous CPU percentages and total runtimes
being reported by top and other utilities.

In addition, the current code was designed for the previous policy where
a vcore would only run when all the vcpus in it were runnable, and so
only counted stolen time on a per-vcore basis.  Now that a vcore can
run while some of the vcpus in it are doing other things in the kernel
(e.g. handling a page fault), we need to count the time when a vcpu task
is preempted while it is not running as part of a vcore as stolen also.

To do this, we bring back the BUSY_IN_HOST vcpu state and extend the
vcpu_load/put functions to count preemption time while the vcpu is
in that state.  Handling the transitions between the RUNNING and
BUSY_IN_HOST states requires checking and updating two variables
(accumulated time stolen and time last preempted), so we add a new
spinlock, vcpu->arch.tbacct_lock.  This protects both the per-vcpu
stolen/preempt-time variables, and the per-vcore variables while this
vcpu is running the vcore.

Finally, we now don't count time spent in userspace as stolen time.
The task could be executing in userspace on behalf of the vcpu, or
it could be preempted, or the vcpu could be genuinely stopped.  Since
we have no way of dividing up the time between these cases, we don't
count any of it as stolen.

Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
2012-10-30 10:54:57 +01:00
..
44x_emulate.c KVM: PPC: 44x: fix DCR read/write 2012-10-30 10:54:50 +01:00
44x_tlb.c KVM: do not release the error page 2012-08-06 16:04:58 +03:00
44x_tlb.h KVM: ppc: rename 44x MMU functions used in booke.c 2009-03-24 11:02:56 +02:00
44x.c KVM: PPC: 44x: Initialize PVR 2012-10-05 23:38:47 +02:00
book3s_32_mmu_host.c KVM: PPC: Book3s: PR: Add (dumb) MMU Notifier support 2012-10-05 23:38:43 +02:00
book3s_32_mmu.c KVM: PPC: Put segment registers in shared page 2010-10-24 10:52:11 +02:00
book3s_32_sr.S KVM: PPC: book3s_pr: Simplify transitions between virtual and real mode 2011-09-25 19:52:29 +03:00
book3s_64_mmu_host.c KVM: PPC: Book3s: PR: Add (dumb) MMU Notifier support 2012-10-05 23:38:43 +02:00
book3s_64_mmu_hv.c KVM: Take kvm instead of vcpu to mmu_notifier_retry 2012-10-23 13:35:43 +02:00
book3s_64_mmu.c KVM: PPC: Interpret SDR1 as HVA in PAPR mode 2011-09-25 19:52:21 +03:00
book3s_64_slb.S KVM: PPC: Book3S: PR: No isync in slbie path 2012-05-06 16:19:09 +02:00
book3s_64_vio_hv.c kvm/book3s: Make kernel emulated H_PUT_TCE available for "PR" KVM 2012-05-06 16:19:11 +02:00
book3s_64_vio.c kvm/book3s: Make kernel emulated H_PUT_TCE available for "PR" KVM 2012-05-06 16:19:11 +02:00
book3s_emulate.c KVM: PPC: Emulator: clean up SPR reads and writes 2012-05-06 16:19:13 +02:00
book3s_exports.c powerpc: various straight conversions from module.h --> export.h 2011-10-31 19:30:44 -04:00
book3s_hv_builtin.c KVM: PPC: Quieten message about allocating linear regions 2012-10-05 23:38:40 +02:00
book3s_hv_interrupts.S KVM: PPC: Work around POWER7 DABR corruption problem 2012-04-08 14:01:36 +03:00
book3s_hv_rm_mmu.c KVM: Take kvm instead of vcpu to mmu_notifier_retry 2012-10-23 13:35:43 +02:00
book3s_hv_rmhandlers.S KVM: PPC: Book3S HV: Fix some races in starting secondary threads 2012-10-30 10:54:54 +01:00
book3s_hv.c KVM: PPC: Book3S HV: Fix accounting of stolen time 2012-10-30 10:54:57 +01:00
book3s_interrupts.S powerpc: Merge VCPU_GPR 2012-07-10 19:18:06 +10:00
book3s_mmu_hpte.c KVM: PPC: Book3s: PR: Add (dumb) MMU Notifier support 2012-10-05 23:38:43 +02:00
book3s_paired_singles.c powerpc/kvm: Fallout from system.h disintegration 2012-04-02 14:00:04 +10:00
book3s_pr_papr.c powerpc/kvm: Fix "PR" KVM implementation of H_CEDE 2012-07-11 17:36:38 +02:00
book3s_pr.c KVM: PPC: Book3S: Get/set guest FP regs using the GET/SET_ONE_REG interface 2012-10-05 23:38:54 +02:00
book3s_rmhandlers.S KVM: PPC: Book3S: PR: Rework irq disabling 2012-10-05 23:38:45 +02:00
book3s_segment.S powerpc: Move and fix MTMSR_EERI definition 2012-07-10 19:18:08 +10:00
book3s.c KVM: PPC: Book3S: Get/set guest FP regs using the GET/SET_ONE_REG interface 2012-10-05 23:38:54 +02:00
booke_emulate.c KVM: PPC: BookE: Add MCSR SPR support 2012-10-05 23:38:48 +02:00
booke_interrupts.S KVM updates for the 3.6 merge window 2012-07-24 12:01:20 -07:00
booke.c KVM: PPC: set IN_GUEST_MODE before checking requests 2012-10-05 23:38:54 +02:00
booke.h KVM: PPC: Emulator: clean up SPR reads and writes 2012-05-06 16:19:13 +02:00
bookehv_interrupts.S powerpc/kvm/bookehv: Fix build regression 2012-07-27 11:42:32 +10:00
e500_emulate.c KVM: PPC: booke: Added DECAR support 2012-05-30 11:43:11 +02:00
e500_tlb.c arch/powerpc/kvm/e500_tlb.c: fix error return code 2012-10-05 23:38:55 +02:00
e500.c KVM: PPC: e500: refactor core-specific TLB code 2012-04-08 12:51:12 +03:00
e500.h KVM: PPC: E500: Remove E500_TLB_DIRTY flag 2012-10-05 23:38:48 +02:00
e500mc.c KVM: PPC: e500mc: Fix tlbilx emulation for 64-bit guests 2012-07-11 17:39:38 +02:00
emulate.c KVM: PPC: Move mtspr/mfspr emulation into own functions 2012-10-30 10:54:51 +01:00
fpu.S powerpc: Remove fpscr use from [kvm_]cvt_{fd,df} 2010-09-02 14:07:32 +10:00
Kconfig KVM: PPC: Book3s: PR: Add (dumb) MMU Notifier support 2012-10-05 23:38:43 +02:00
Makefile kvm/book3s: Make kernel emulated H_PUT_TCE available for "PR" KVM 2012-05-06 16:19:11 +02:00
powerpc.c KVM: PPC: set IN_GUEST_MODE before checking requests 2012-10-05 23:38:54 +02:00
timing.c KVM: PPC: fix partial application of "exit timing in ticks" 2011-07-12 13:16:28 +03:00
timing.h KVM: PPC: booke: category E.HV (GS-mode) support 2012-04-08 12:51:19 +03:00
trace.h KVM: PPC: Use symbols for exit trace 2012-10-05 23:38:48 +02:00