2015-11-26 16:34 GMT+08:00 Andrey Smetanin :
>
>
> On 11/26/2015 08:28 AM, Wanpeng Li wrote:
>>
>> 2015-11-25 23:20 GMT+08:00 Andrey Smetanin :
>>>
>>> Per Hyper-V specification (and as required by Hyper-V-aware guests),
>>> SynIC provides 4 per-v
2015-11-26 16:34 GMT+08:00 Andrey Smetanin :
>
>
> On 11/26/2015 08:28 AM, Wanpeng Li wrote:
>>
>> 2015-11-25 23:20 GMT+08:00 Andrey Smetanin :
>>>
>>> Per Hyper-V specification (and as required by Hyper-V-aware guests),
>>> SynIC provides 4 per-v
ynIC message slot and triggering the corresponding
> synthetic interrupt.
Could you post a link for this specification?
Regards,
Wanpeng Li
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majord...@vger.kernel.org
More majordomo info a
xposes LM bit to 32 bit guest(however it
doesn't expose syscall/sysret), so maybe we can't depend on F(LM).
Regards,
Wanpeng Li
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majord...@vger.kernel.org
More majordomo info at htt
2015-11-20 10:52 GMT+08:00 Wanpeng Li :
> Hi Paolo,
> 2015-02-09 17:03 GMT+08:00 Paolo Bonzini :
>> NoWrite instructions (e.g. cmp or test) never set the "write access"
>> bit in the error code, even if one of the operands is treated as a
>> destination.
>
&
SDM
describe above?
Regards,
Wanpeng Li
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
2015-11-19 19:05 GMT+08:00 Paolo Bonzini :
>
>
> On 19/11/2015 11:45, Wanpeng li wrote:
>> Intel cpu doesn't support syscall/sysret in non 64-bit mode which
>> is different from AMD. Expose syscall/sysret to intel 32-bit guest
>> just makes no sense and leads to
test_rdtscp(10);
> + test_rdtscp(0x100);
> + } else
> + printf("rdtscp not supported\n");
> return 0;
> }
> diff --git a/x86/unittests.cfg b/x86/unittests.cfg
> index ac652fc..fd7e9b2 100644
> --- a/x86/unittests.cf
+ test_rdtscp(10);
> + test_rdtscp(0x100);
> + } else
> + printf("rdtscp not supported\n");
> return 0;
> }
> diff --git a/x86/unittests.cfg b/x86/unittests.cfg
> index ac652fc..fd7e9b2 100644
> --- a/x86/unitt
modifies the existing call to vmcs_writel? If there is a
> bug, can you write a unit test for it? It should be possible to test
> for the problem using INIT+SIPI on an AP.
You are right, I write a INIT+SIPI kvm-unit-test and didn't trigger
any bug. Please ignore the patch.
Regards,
reset eflags register to 0002H following initialization
of the processor.
Signed-off-by: Wanpeng Li
---
v1 -> v2:
* use vmcs_writel
arch/x86/kvm/vmx.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index b680c2e..1a95ef7 100644
--- a/arch/x86/
(with 2:1 cpu overcommit). irq time accounting gets crazy. Even 'top'
shows weird statistic: 50% hi, 50% st, ~0% user, spinning processes
use ~0% cpu - that is not correct.
If this is the desired behavior or something need to improve?
Regards,
Wanpeng Li
--
To unsubscribe from this list
Ping, :-)
On 10/21/15 2:28 PM, Wanpeng Li wrote:
Reference SDM 3.4.3:
Following initialization of the processor (either by asserting the
RESET pin or the INIT pin), the state of the EFLAGS register is
0002H.
However, the eflags fixed bit is not set and other bits are also not
cleared
On 10/21/15 2:46 PM, Paolo Bonzini wrote:
On 21/10/2015 00:57, Wanpeng Li wrote:
kvm_sched_out and kvm_sched_in are part of KVM's preemption hooks. The
hooks are registered only between vcpu_load and vcpu_put, therefore they
know that the mutex is taken. The sequence will go like
set eflags register to 0002H following
initialization of the processor.
Signed-off-by: Wanpeng Li
---
arch/x86/kvm/vmx.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index b680c2e..326f6ea 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
or worse. It prefers to stuck at quarters (100% 75% 50% 25%)
Output is attached.
If the output in attachment is for VM2 only?
Regards,
Wanpeng Li
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majord...@vger.kernel.org
More majordomo info at
out
kvm_sched_in
kvm_sched_out
...
vcpu_put
Regards,
Wanpeng Li
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Ping, :-)
On 10/14/15 12:12 AM, Wanpeng Li wrote:
v2 -> v3:
* separate nested_vmx_vpid_caps and move checks to patch 3/5,
only rejoin them when reading the MSR.
v1 -> v2:
* set bit 32 of the VMX_EPT_VPID_CAP MSR
* check against the supported types in the implementation of
L2's vCPUs not sched in/out on L1.
Reviewed-by: Wincy Van
Signed-off-by: Wanpeng Li
---
arch/x86/kvm/vmx.c | 9 +++--
1 file changed, 7 insertions(+), 2 deletions(-)
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 2a54cc7..0b558ae 100644
--- a/arch/x86/kvm/vmx.c
+++ b/ar
nested VPID
kernelLinux 3.5.0-1 1.2600 1.4300 1.5600 12.7 12.9 3.49000 7.46000
vanilla
Reviewed-by: Jan Kiszka
Reviewed-by: Wincy Van
Signed-off-by: Wanpeng Li
---
arch/x86/kvm/vmx.c | 39 ---
1 file changed, 32 insertions(+), 7 deletions(-)
diff
Add the INVVPID instruction emulation.
Reviewed-by: Wincy Van
Signed-off-by: Wanpeng Li
---
arch/x86/include/asm/vmx.h | 1 +
arch/x86/kvm/vmx.c | 61 +-
2 files changed, 61 insertions(+), 1 deletion(-)
diff --git a/arch/x86/include/asm
Introduce __vmx_flush_tlb() to handle specific vpid. It will be
used by later patches, note that the "all context" variant can
be mapped to vpid_sync_vcpu_single with vpid02 as the argument
(a nice side effect of vpid02 design).
Reviewed-by: Wincy Van
Signed-off-by: Wanpeng Li
---
Adjust allocate/free_vid so that they can be reused for the nested vpid.
Reviewed-by: Wincy Van
Signed-off-by: Wanpeng Li
---
arch/x86/kvm/vmx.c | 25 -
1 file changed, 12 insertions(+), 13 deletions(-)
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index c5c2283
2.3300 5.6 2.88000
nested VPID
kernelLinux 3.5.0-1 1.2600 1.4300 1.5600 12.7 12.9 3.49000 7.46000
vanilla
Wanpeng Li (5):
KVM: VMX: adjust interface to allocate/free_vpid
KVM: VMX: introduce __vmx_flush_tlb to handle specific vpid
KVM: nVMX: emulate the INVVPID instruction
K
On 10/13/15 10:44 PM, Paolo Bonzini wrote:
On 08/10/2015 07:57, Wanpeng Li wrote:
Expose VPID capability to L1. For nested guests, we don't do anything
specific for single context invalidation. Hence, only advertise support
for global context invalidation. The major benefit of nested
On 9/29/15 6:39 PM, Paolo Bonzini wrote:
On 29/09/2015 04:55, Wanpeng Li wrote:
Expose VPID capability to L1.
Signed-off-by: Wanpeng Li
---
v1 -> v2:
* set only VMX_VPID_EXTENT_GLOBAL_CONTEXT_BIT
Thanks. I've checked more thoroughly your implementation against the
SDM now, and the
Introduce __vmx_flush_tlb() to handle specific vpid. It will be
used by later patches, note that the "all context" variant can
be mapped to vpid_sync_vcpu_single with vpid02 as the argument
(a nice side effect of vpid02 design).
Reviewed-by: Wincy Van
Signed-off-by: Wanpeng Li
---
ctxsw ctxsw ctxsw ctxsw ctxsw ctxsw ctxsw
- - -- -- -- -- -- --- ---
kernelLinux 3.5.0-1 1.2200 1.3700 1.4500 4.7800 2.3300 5.6 2.88000
nested VPID
kernelLinux 3.5.0-1 1.2600 1.4300 1.5600 12.7 12.9 3.49000 7.46000
vanilla
Adjust allocate/free_vid so that they can be reused for the nested vpid.
Reviewed-by: Wincy Van
Signed-off-by: Wanpeng Li
---
arch/x86/kvm/vmx.c | 25 -
1 file changed, 12 insertions(+), 13 deletions(-)
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 6407674
Add the INVVPID instruction emulation.
Reviewed-by: Wincy Van
Signed-off-by: Wanpeng Li
---
arch/x86/include/asm/vmx.h | 3 +++
arch/x86/kvm/vmx.c | 49 +-
2 files changed, 51 insertions(+), 1 deletion(-)
diff --git a/arch/x86/include/asm
nested VPID
kernelLinux 3.5.0-1 1.2600 1.4300 1.5600 12.7 12.9 3.49000 7.46000
vanilla
Reviewed-by: Jan Kiszka
Reviewed-by: Wincy Van
Signed-off-by: Wanpeng Li
---
arch/x86/kvm/vmx.c | 39 ---
1 file changed, 32 insertions(+), 7 deletions(-)
diff
L2's vCPUs not sched in/out on L1.
Reviewed-by: Wincy Van
Signed-off-by: Wanpeng Li
---
arch/x86/kvm/vmx.c | 36
1 file changed, 24 insertions(+), 12 deletions(-)
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 31d272e..22b4dc7 100644
---
On 9/29/15 6:39 PM, Paolo Bonzini wrote:
On 29/09/2015 04:55, Wanpeng Li wrote:
Expose VPID capability to L1.
Signed-off-by: Wanpeng Li
---
v1 -> v2:
* set only VMX_VPID_EXTENT_GLOBAL_CONTEXT_BIT
Thanks. I've checked more thoroughly your implementation against the
SDM now, and the
Expose VPID capability to L1.
Signed-off-by: Wanpeng Li
---
v1 -> v2:
* set only VMX_VPID_EXTENT_GLOBAL_CONTEXT_BIT
arch/x86/kvm/vmx.c | 22 +++---
1 file changed, 11 insertions(+), 11 deletions(-)
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 75f3ee0..c4ea
On 9/28/15 8:05 PM, Paolo Bonzini wrote:
On 24/09/2015 08:51, Wanpeng Li wrote:
/*
* For nested guests, we don't do anything specific
* for single context invalidation. Hence, only advertise
* support for global co
On 9/25/15 12:12 AM, Bandan Das wrote:
Wanpeng Li writes:
Introduce __vmx_flush_tlb() to handle specific vpid.
Signed-off-by: Wanpeng Li
---
arch/x86/kvm/vmx.c | 21 +
1 file changed, 13 insertions(+), 8 deletions(-)
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm
Expose VPID capability to L1.
Signed-off-by: Wanpeng Li
---
arch/x86/kvm/vmx.c | 22 +++---
1 file changed, 11 insertions(+), 11 deletions(-)
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index f9219ad..866045c 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
Introduce __vmx_flush_tlb() to handle specific vpid.
Signed-off-by: Wanpeng Li
---
arch/x86/kvm/vmx.c | 21 +
1 file changed, 13 insertions(+), 8 deletions(-)
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 794c529..7188c5e 100644
--- a/arch/x86/kvm/vmx.c
+++ b
On 9/23/15 4:39 PM, Paolo Bonzini wrote:
On 23/09/2015 09:59, Wanpeng Li wrote:
Add the INVVPID instruction emulation.
Reviewed-by: Wincy Van
Signed-off-by: Wanpeng Li
---
arch/x86/include/asm/vmx.h | 1 +
arch/x86/kvm/vmx.c | 23 ++-
2 files changed, 23
vpid_sync_vcpu_single() still handles vpid01 during nested
vmentry/vmexit since vmx->vpid is used for invvpid. This
patch fix it by specific the vpid02 through __vmx_flush_tlb()
to flush the right vpid.
Signed-off-by: Wanpeng Li
---
arch/x86/kvm/vmx.c | 4 ++--
1 file changed, 2 inserti
Add the INVVPID instruction emulation.
Reviewed-by: Wincy Van
Signed-off-by: Wanpeng Li
---
arch/x86/include/asm/vmx.h | 1 +
arch/x86/kvm/vmx.c | 23 ++-
2 files changed, 23 insertions(+), 1 deletion(-)
diff --git a/arch/x86/include/asm/vmx.h b/arch/x86/include
On 9/21/15 10:51 AM, Xiao Guangrong wrote:
Thanks for your report and analysis, Janusz!
On 09/19/2015 01:48 AM, Janusz wrote:
W dniu 18.09.2015 o 12:07, Laszlo Ersek pisze:
On 09/18/15 11:37, Janusz wrote:
Hello,
I am writting about this patch that was posted by Xiao:
http://www.spinics.net
eused while the value of
Looks good to me.
Regards,
Wanpeng Li
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
ems to work (goes back to
1 cpu).
The proper way might be to feedback the result of the
interrupt dequeue into the heuristics. Don't know yet how
to handle that properly.
If this can be reproduced on x86 platform?
Regards,
Wanpeng Li
--
To unsubscribe from this list: send the line "unsubsc
eserve H for VMX root operation.
This patch fix it by reintroducing reserve H for VMX root operation.
Reported-by: Wincy Van
Signed-off-by: Wanpeng Li
---
arch/x86/kvm/vmx.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 9ff6a3f..
On 9/16/15 6:04 PM, Paolo Bonzini wrote:
On 16/09/2015 11:30, Wanpeng Li wrote:
Enhance allocate/free_vid to handle shadow vpid.
Adjusting the commit message:
KVM: nVMX: adjust interface to allocate/free_vpid
Adjust allocate/free_vid so that they can be reused for the nested
0-1 1.2200 1.3700 1.4500 4.7800 2.3300 5.6 2.88000
nested VPID
kernelLinux 3.5.0-1 1.2600 1.4300 1.5600 12.7 12.9 3.49000 7.46000
vanilla
Wanpeng Li (2):
KVM: nVMX: enhance allocate/free_vpid to handle shadow vpid
KVM: nVMX: nested VPID emulation
arch/x
2.88000
nested VPID
kernelLinux 3.5.0-1 1.2600 1.4300 1.5600 12.7 12.9 3.49000 7.46000
vanilla
Reviewed-by: Jan Kiszka
Suggested-by: Wincy Van
Signed-off-by: Wanpeng Li
---
arch/x86/kvm/vmx.c | 37 +++--
1 file changed, 31 insertions(+), 6 deletions
Enhance allocate/free_vid to handle shadow vpid.
Signed-off-by: Wanpeng Li
---
arch/x86/kvm/vmx.c | 25 -
1 file changed, 12 insertions(+), 13 deletions(-)
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 9ff6a3f..f8d704d 100644
--- a/arch/x86/kvm/vmx.c
+++ b
On 9/16/15 5:11 PM, Jan Kiszka wrote:
On 2015-09-16 09:19, Wanpeng Li wrote:
Enhance allocate/free_vid to handle shadow vpid.
Signed-off-by: Wanpeng Li
---
arch/x86/kvm/vmx.c | 23 +++
1 file changed, 11 insertions(+), 12 deletions(-)
diff --git a/arch/x86/kvm/vmx.c b
.6 2.88000
nested VPID
kernelLinux 3.5.0-1 1.2600 1.4300 1.5600 12.7 12.9 3.49000 7.46000
vanilla
Wanpeng Li (2):
KVM: nVMX: enhance allocate/free_vpid to handle shadow vpid
KVM: nVMX: nested VPID emulation
arch/x86/kvm/vmx.c | 60 ++--
2.88000
nested VPID
kernelLinux 3.5.0-1 1.2600 1.4300 1.5600 12.7 12.9 3.49000 7.46000
vanilla
Reviewed-by: Jan Kiszka
Suggested-by: Wincy Van
Signed-off-by: Wanpeng Li
---
arch/x86/kvm/vmx.c | 37 +++--
1 file changed, 31 insertions(+), 6 deletions
Enhance allocate/free_vid to handle shadow vpid.
Signed-off-by: Wanpeng Li
---
arch/x86/kvm/vmx.c | 23 +++
1 file changed, 11 insertions(+), 12 deletions(-)
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 9ff6a3f..c5222b8 100644
--- a/arch/x86/kvm/vmx.c
+++ b
On 9/16/15 2:42 PM, Jan Kiszka wrote:
On 2015-09-16 05:51, Wanpeng Li wrote:
Enhance allocate/free_vid to handle shadow vpid.
Signed-off-by: Wanpeng Li
---
arch/x86/kvm/vmx.c | 24 +++-
1 file changed, 11 insertions(+), 13 deletions(-)
diff --git a/arch/x86/kvm/vmx.c b
On 9/16/15 1:20 PM, Jan Kiszka wrote:
On 2015-09-16 04:36, Wanpeng Li wrote:
On 9/16/15 1:32 AM, Jan Kiszka wrote:
On 2015-09-15 12:14, Wanpeng Li wrote:
On 9/14/15 10:54 PM, Jan Kiszka wrote:
Last but not least: the guest can now easily exhaust the host's pool of
vpid by simply spa
On 9/15/15 8:54 PM, Paolo Bonzini wrote:
On 15/09/2015 12:30, Wanpeng Li wrote:
+ if (!nested) {
+ vpid = find_first_zero_bit(vmx_vpid_bitmap, VMX_NR_VPIDS);
+ if (vpid < VMX_NR_VPIDS) {
vmx->vpid = vpid;
__set_bi
12.7 12.9 3.49000 7.46000
vanilla
Wanpeng Li (2):
KVM: nVMX: enhance allocate/free_vpid to handle shadow vpid
KVM: nVMX: nested VPID emulation
arch/x86/kvm/vmx.c | 61 +-
1 file changed, 42 insertions(+), 19 deletions(-)
--
1
2.88000
nested VPID
kernelLinux 3.5.0-1 1.2600 1.4300 1.5600 12.7 12.9 3.49000 7.46000
vanilla
Suggested-by: Wincy Van
Signed-off-by: Wanpeng Li
---
arch/x86/kvm/vmx.c | 37 +++--
1 file changed, 31 insertions(+), 6 deletions(-)
diff --git a/arch/x86/kvm
Enhance allocate/free_vid to handle shadow vpid.
Signed-off-by: Wanpeng Li
---
arch/x86/kvm/vmx.c | 24 +++-
1 file changed, 11 insertions(+), 13 deletions(-)
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 9ff6a3f..4956081 100644
--- a/arch/x86/kvm/vmx.c
+++ b
On 9/16/15 1:32 AM, Jan Kiszka wrote:
On 2015-09-15 12:14, Wanpeng Li wrote:
On 9/14/15 10:54 PM, Jan Kiszka wrote:
Last but not least: the guest can now easily exhaust the host's pool of
vpid by simply spawning plenty of VCPUs for L2, no? Is this acceptable
or should there be some limi
On 9/16/15 6:00 AM, David Matlack wrote:
On Tue, Sep 15, 2015 at 12:04 AM, Oliver Yang wrote:
Hi Guys,
I found below patch for KVM TSC trapping / migration support,
https://lkml.org/lkml/2011/1/6/90
It seemed the patch were not merged in Linux mainline.
So I have 3 questions here,
1. Can
xsw
- - -- -- -- -- -- --- ---
kernelLinux 3.5.0-1 1.2200 1.3700 1.4500 4.7800 2.3300 5.6 2.88000
nested VPID
kernelLinux 3.5.0-1 1.2600 1.4300 1.5600 12.7 12.9 3.49000 7.46000
vanilla
Wanpeng Li (2):
KVM: nVMX: enhance allocate/free_vpid
2.88000
nested VPID
kernelLinux 3.5.0-1 1.2600 1.4300 1.5600 12.7 12.9 3.49000 7.46000
vanilla
Suggested-by: Wincy Van
Signed-off-by: Wanpeng Li
---
arch/x86/kvm/vmx.c | 25 ++---
1 file changed, 18 insertions(+), 7 deletions(-)
diff --git a/arch/x86/kvm/vmx.c b/arch
Enhance allocate/free_vid to handle shadow vpid.
Suggested-by: Wincy Van
Signed-off-by: Wanpeng Li
---
arch/x86/kvm/vmx.c | 33 +++--
1 file changed, 27 insertions(+), 6 deletions(-)
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index da1590e..bd07d88 100644
On 9/15/15 12:08 AM, Bandan Das wrote:
Wanpeng Li writes:
VPID is used to tag address space and avoid a TLB flush. Currently L0 use
the same VPID to run L1 and all its guests. KVM flushes VPID when switching
between L1 and L2.
This patch advertises VPID to the L1 hypervisor, then address
On 9/14/15 10:54 PM, Jan Kiszka wrote:
On 2015-09-14 14:52, Wanpeng Li wrote:
VPID is used to tag address space and avoid a TLB flush. Currently L0 use
the same VPID to run L1 and all its guests. KVM flushes VPID when switching
between L1 and L2.
This patch advertises VPID to the L1 hypervisor
flush when swithing between L1 and
L2. This patch gets ~3x performance improvement for lmbench 8p/64k ctxsw.
Signed-off-by: Wanpeng Li
---
arch/x86/kvm/vmx.c | 39 ---
1 file changed, 32 insertions(+), 7 deletions(-)
diff --git a/arch/x86/kvm/vmx.c b/arch/x86
ng
when polling is disabled.
Reported-by: Christian Borntraeger
Signed-off-by: Wanpeng Li
---
virt/kvm/kvm_main.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 4662a88..f756cac0 100644
--- a/virt/kvm/kvm_main.c
+++ b/
On 9/10/15 3:13 PM, Christian Borntraeger wrote:
Am 10.09.2015 um 03:55 schrieb Wanpeng Li:
On 9/9/15 9:39 PM, Christian Borntraeger wrote:
Am 03.09.2015 um 16:07 schrieb Wanpeng Li:
v6 -> v7:
* explicit signal (set a bool)
* fix the tracepoint
v5 -> v6:
* fix wait_ns and p
On 9/9/15 9:39 PM, Christian Borntraeger wrote:
Am 03.09.2015 um 16:07 schrieb Wanpeng Li:
v6 -> v7:
* explicit signal (set a bool)
* fix the tracepoint
v5 -> v6:
* fix wait_ns and poll_ns
v4 -> v5:
* set base case 10us and max poll time 500us
* handle short/long halt,
On 9/6/15 10:32 PM, Paolo Bonzini wrote:
On 05/09/2015 00:38, Wanpeng Li wrote:
@@ -1940,11 +1975,16 @@ void kvm_vcpu_block(struct kvm_vcpu *vcpu)
* arrives.
*/
if (kvm_vcpu_check_block(vcpu) < 0) {
+polled = t
Hi Paolo,
On 9/3/15 10:07 PM, Wanpeng Li wrote:
[...]
static int kvm_vcpu_check_block(struct kvm_vcpu *vcpu)
{
if (kvm_arch_vcpu_runnable(vcpu)) {
@@ -1928,7 +1962,8 @@ void kvm_vcpu_block(struct kvm_vcpu *vcpu)
{
ktime_t start, cur;
DEFINE_WAIT(wait);
- bool
On 9/4/15 9:16 AM, Wanpeng Li wrote:
[...]
+
static int kvm_vcpu_check_block(struct kvm_vcpu *vcpu)
{
if (kvm_arch_vcpu_runnable(vcpu)) {
@@ -1929,6 +1963,7 @@ void kvm_vcpu_block(struct kvm_vcpu *vcpu)
ktime_t start, cur;
DEFINE_WAIT(wait);
bool waited
On 9/3/15 2:09 AM, David Matlack wrote:
On Wed, Sep 2, 2015 at 12:29 AM, Wanpeng Li wrote:
There is a downside of always-poll since poll is still happened for idle
vCPUs which can waste cpu usage. This patch adds the ability to adjust
halt_poll_ns dynamically, to grow halt_poll_ns when shot
On 9/4/15 12:07 AM, David Matlack wrote:
On Thu, Sep 3, 2015 at 2:23 AM, Wanpeng Li wrote:
How about something like:
@@ -1941,10 +1976,14 @@ void kvm_vcpu_block(struct kvm_vcpu *vcpu)
*/
if (kvm_vcpu_check_block(vcpu) < 0) {
++v
Tracepoint for dynamic halt_pool_ns, fired on every potential change.
Signed-off-by: Wanpeng Li
---
include/trace/events/kvm.h | 30 ++
virt/kvm/kvm_main.c| 8 ++--
2 files changed, 36 insertions(+), 2 deletions(-)
diff --git a/include/trace/events
ggested-by: David Matlack
Signed-off-by: Wanpeng Li
---
virt/kvm/kvm_main.c | 63 +
1 file changed, 59 insertions(+), 4 deletions(-)
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index c06e57c..d5e07e9 100644
--- a/virt/kvm/kvm_main.c
+++
Change halt_poll_ns into per-VCPU variable, seeded from module parameter,
to allow greater flexibility.
Signed-off-by: Wanpeng Li
---
include/linux/kvm_host.h | 1 +
virt/kvm/kvm_main.c | 5 +++--
2 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/include/linux/kvm_host.h b
sing the dynamic-poll. The savings
should be even higher for higher frequency ticks.
Wanpeng Li (3):
KVM: make halt_poll_ns per-vCPU
KVM: dynamic halt-polling
KVM: trace kvm_halt_poll_ns grow/shrink
include/linux/kvm_host.h | 1 +
include/trace/eve
you want, you can leave this if() out and save some indentation.
Then we will miss the tracepoint.
Regards,
Wanpeng Li
+ if (block_ns <= vcpu->halt_poll_ns)
+ ;
+ /* we had a long block, shrink polling */
+ else if (vcpu->
us to loop calling
kvm_vcpu_block and starve the waiting task (at least until need_resched()),
which would break the "only hog the cpu when idle" aspect of halt-polling.
That's definitely a bug, yes.
Ok, I will send out v7 to fix this in this sunday since there is
vacation in my country
On 9/3/15 2:09 AM, David Matlack wrote:
On Wed, Sep 2, 2015 at 12:42 AM, Wanpeng Li wrote:
Tracepoint for dynamic halt_pool_ns, fired on every potential change.
Signed-off-by: Wanpeng Li
---
include/trace/events/kvm.h | 30 ++
virt/kvm/kvm_main.c| 8
Tracepoint for dynamic halt_pool_ns, fired on every potential change.
Signed-off-by: Wanpeng Li
---
include/trace/events/kvm.h | 30 ++
virt/kvm/kvm_main.c| 8 ++--
2 files changed, 36 insertions(+), 2 deletions(-)
diff --git a/include/trace/events
Suggested-by: David Matlack
Signed-off-by: Wanpeng Li
---
virt/kvm/kvm_main.c | 61 ++---
1 file changed, 58 insertions(+), 3 deletions(-)
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index c06e57c..3cff02f 100644
--- a/virt/kvm/kvm_ma
s. (250HZ) means the guest was ticking at 250HZ.
The big win is with ticking operating systems. Running the linux guest
with nohz=off (and HZ=250), we save 3.4%~12.8% CPUs/second and get close
to no-polling overhead levels by using the dynamic-poll. The savings
should be even higher for higher fre
Change halt_poll_ns into per-vCPU variable, seeded from module parameter,
to allow greater flexibility.
Signed-off-by: Wanpeng Li
---
include/linux/kvm_host.h | 1 +
virt/kvm/kvm_main.c | 5 +++--
2 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/include/linux/kvm_host.h b
The savings
should be even higher for higher frequency ticks.
Wanpeng Li (3):
KVM: make halt_poll_ns per-VCPU
KVM: dynamic halt_poll_ns adjustment
KVM: trace kvm_halt_poll_ns grow/shrink
include/linux/kvm_host.h | 1 +
include/trac
frequency ticks.
Suggested-by: David Matlack
Signed-off-by: Wanpeng Li
---
virt/kvm/kvm_main.c | 60 ++---
1 file changed, 57 insertions(+), 3 deletions(-)
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index c06e57c..2206cb0 10
Tracepoint for dynamic halt_pool_ns, fired on every potential change.
Signed-off-by: Wanpeng Li
---
include/trace/events/kvm.h | 30 ++
virt/kvm/kvm_main.c| 8 ++--
2 files changed, 36 insertions(+), 2 deletions(-)
diff --git a/include/trace/events
Change halt_poll_ns into per-vCPU variable, seeded from module parameter,
to allow greater flexibility.
Signed-off-by: Wanpeng Li
---
include/linux/kvm_host.h | 1 +
virt/kvm/kvm_main.c | 5 +++--
2 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/include/linux/kvm_host.h b
On 9/2/15 9:49 AM, David Matlack wrote:
On Tue, Sep 1, 2015 at 5:29 PM, Wanpeng Li wrote:
On 9/2/15 7:24 AM, David Matlack wrote:
On Tue, Sep 1, 2015 at 3:58 PM, Wanpeng Li wrote:
Why this can happen?
Ah, probably because I'm missing 9c8fd1ba220 (KVM: x86: optimize delivery
o
On 9/2/15 7:24 AM, David Matlack wrote:
On Tue, Sep 1, 2015 at 3:58 PM, Wanpeng Li wrote:
On 9/2/15 6:34 AM, David Matlack wrote:
On Tue, Sep 1, 2015 at 3:30 PM, Wanpeng Li wrote:
On 9/2/15 5:45 AM, David Matlack wrote:
On Thu, Aug 27, 2015 at 2:47 AM, Wanpeng Li
wrote:
v3 ->
On 9/2/15 6:34 AM, David Matlack wrote:
On Tue, Sep 1, 2015 at 3:30 PM, Wanpeng Li wrote:
On 9/2/15 5:45 AM, David Matlack wrote:
On Thu, Aug 27, 2015 at 2:47 AM, Wanpeng Li
wrote:
v3 -> v4:
* bring back grow vcpu->halt_poll_ns when interrupt arrives and shrinks
when idle V
On 9/2/15 5:45 AM, David Matlack wrote:
On Thu, Aug 27, 2015 at 2:47 AM, Wanpeng Li wrote:
v3 -> v4:
* bring back grow vcpu->halt_poll_ns when interrupt arrives and shrinks
when idle VCPU is detected
v2 -> v3:
* grow/shrink vcpu->halt_poll_ns by *halt_pol
On 8/31/15 3:44 PM, Wanpeng Li wrote:
On 8/30/15 6:26 AM, Peter Kieser wrote:
Thanks, Wanpeng. Applied this to Linux 3.18 and seeing much higher CPU
usage (200%) for qemu 2.4.0 process on a Windows 10 x64 guest. qemu
parameters:
Interesting. I test this against latest kvm tree and stable qemu
% |
+-++---+
Regards,
Wanpeng Li
qemu-system-x86_64 -enable-kvm -name arwan-20150704 -S -machine
pc-q35-2.2,accel=kvm,usb=off -cpu
Haswell,hv_time,hv_relaxed,hv_vapic,hv_spinlocks=0x1000 -m 8192
-realtime mlock=off -smp 4,sockets=4,cores=1,threads=1 -uuid
7c2fc02d-2798-4fc9-ad04-db5f1af92723
On 8/30/15 8:13 AM, Peter Kieser wrote:
On 2015-08-29 4:55 PM, Wanpeng Li wrote:
On 8/30/15 6:26 AM, Peter Kieser wrote:
Thanks, Wanpeng. Applied this to Linux 3.18 and seeing much higher
CPU usage (200%) for qemu 2.4.0 process on a Windows 10 x64 guest.
qemu parameters:
Thanks for the
applied on your 3.18? Btw, do you test the linux guest?
Regards,
Wanpeng Li
qemu-system-x86_64 -enable-kvm -name arwan-20150704 -S -machine
pc-q35-2.2,accel=kvm,usb=off -cpu
Haswell,hv_time,hv_relaxed,hv_vapic,hv_spinlocks=0x1000 -m 8192
-realtime mlock=off -smp 4,sockets=4,cores=1,threads=1
Hi Peter,
On 8/30/15 5:18 AM, Peter Kieser wrote:
Hi Wanpeng,
Do I need to set any module parameters to use your patch, or should
halt_poll_ns automatically tune with just your patch series applied?
You don't need any module parameters.
Regards,
Wanpeng Li
Thanks.
On 2015-08-27 2:
On 8/28/15 12:25 AM, David Matlack wrote:
On Thu, Aug 27, 2015 at 2:59 AM, Wanpeng Li wrote:
Hi David,
On 8/26/15 1:19 AM, David Matlack wrote:
Thanks for writing v2, Wanpeng.
On Mon, Aug 24, 2015 at 11:35 PM, Wanpeng Li
wrote:
There is a downside of halt_poll_ns since poll is still happen
1 - 100 of 320 matches
Mail list logo
