* [PATCH] KVM: arm64: Count guest exit due to various reasons
@ 2015-10-18 8:04 Amit
2015-10-18 19:53 ` Christoffer Dall
2015-10-22 21:16 ` Christoffer Dall
0 siblings, 2 replies; 6+ messages in thread
From: Amit @ 2015-10-18 8:04 UTC (permalink / raw)
To: kvmarm; +Cc: marc.zyngier, andre.przywara
From: Amit Singh Tomar <amittomer25@gmail.com>
This patch adds guest exit statistics to debugfs, This can be helpful
while measuring KVM performance.
Signed-off-by: Amit Singh Tomar <amittomer25@gmail.com>
---
arch/arm/include/asm/kvm_host.h | 2 ++
arch/arm/kvm/arm.c | 1 +
arch/arm/kvm/guest.c | 2 ++
arch/arm/kvm/mmio.c | 1 +
arch/arm64/include/asm/kvm_host.h | 5 +++++
arch/arm64/kvm/guest.c | 8 ++++++++
arch/arm64/kvm/handle_exit.c | 3 +++
7 files changed, 22 insertions(+)
diff --git a/arch/arm/include/asm/kvm_host.h b/arch/arm/include/asm/kvm_host.h
index c4072d9..d1f3159 100644
--- a/arch/arm/include/asm/kvm_host.h
+++ b/arch/arm/include/asm/kvm_host.h
@@ -147,6 +147,8 @@ struct kvm_vcpu_stat {
u32 halt_successful_poll;
u32 halt_attempted_poll;
u32 halt_wakeup;
+ u32 mmio_exit_stat;
+ u32 exits;
};
int kvm_vcpu_preferred_target(struct kvm_vcpu_init *init);
diff --git a/arch/arm/kvm/arm.c b/arch/arm/kvm/arm.c
index 78b2869..5d0a2c5 100644
--- a/arch/arm/kvm/arm.c
+++ b/arch/arm/kvm/arm.c
@@ -569,6 +569,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run)
ret = kvm_call_hyp(__kvm_vcpu_run, vcpu);
vcpu->mode = OUTSIDE_GUEST_MODE;
+ vcpu->stat.exits++;
/*
* Back from guest
*************************************************************/
diff --git a/arch/arm/kvm/guest.c b/arch/arm/kvm/guest.c
index 96e935b..0eea5fc 100644
--- a/arch/arm/kvm/guest.c
+++ b/arch/arm/kvm/guest.c
@@ -33,6 +33,8 @@
#define VCPU_STAT(x) { #x, offsetof(struct kvm_vcpu, stat.x), KVM_STAT_VCPU }
struct kvm_stats_debugfs_item debugfs_entries[] = {
+ VCPU_STAT(mmio_exit_stat),
+ VCPU_STAT(exits),
{ NULL }
};
diff --git a/arch/arm/kvm/mmio.c b/arch/arm/kvm/mmio.c
index 974b1c6..51d518e 100644
--- a/arch/arm/kvm/mmio.c
+++ b/arch/arm/kvm/mmio.c
@@ -184,6 +184,7 @@ int io_mem_abort(struct kvm_vcpu *vcpu, struct kvm_run *run,
}
rt = vcpu->arch.mmio_decode.rt;
+ vcpu->stat.mmio_exit_stat++;
if (is_write) {
data = vcpu_data_guest_to_host(vcpu, *vcpu_reg(vcpu, rt), len);
diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
index ed03968..b32c938 100644
--- a/arch/arm64/include/asm/kvm_host.h
+++ b/arch/arm64/include/asm/kvm_host.h
@@ -194,6 +194,11 @@ struct kvm_vcpu_stat {
u32 halt_successful_poll;
u32 halt_attempted_poll;
u32 halt_wakeup;
+ u32 hvc_exit_stat;
+ u32 wfx_exit_stat;
+ u32 wfi_exit_stat;
+ u32 mmio_exit_stat;
+ u32 exits;
};
int kvm_vcpu_preferred_target(struct kvm_vcpu_init *init);
diff --git a/arch/arm64/kvm/guest.c b/arch/arm64/kvm/guest.c
index d250160..b1943cd 100644
--- a/arch/arm64/kvm/guest.c
+++ b/arch/arm64/kvm/guest.c
@@ -34,7 +34,15 @@
#include "trace.h"
+#define VM_STAT(x) { #x, offsetof(struct kvm, stat.x), KVM_STAT_VM }
+#define VCPU_STAT(x) { #x, offsetof(struct kvm_vcpu, stat.x), KVM_STAT_VCPU }
+
struct kvm_stats_debugfs_item debugfs_entries[] = {
+ VCPU_STAT(hvc_exit_stat),
+ VCPU_STAT(wfx_exit_stat),
+ VCPU_STAT(wfi_exit_stat),
+ VCPU_STAT(mmio_exit_stat),
+ VCPU_STAT(exits),
{ NULL }
};
diff --git a/arch/arm64/kvm/handle_exit.c b/arch/arm64/kvm/handle_exit.c
index 68a0759..6e38c50 100644
--- a/arch/arm64/kvm/handle_exit.c
+++ b/arch/arm64/kvm/handle_exit.c
@@ -39,6 +39,7 @@ static int handle_hvc(struct kvm_vcpu *vcpu, struct kvm_run *run)
trace_kvm_hvc_arm64(*vcpu_pc(vcpu), *vcpu_reg(vcpu, 0),
kvm_vcpu_hvc_get_imm(vcpu));
+ vcpu->stat.hvc_exit_stat++;
ret = kvm_psci_call(vcpu);
if (ret < 0) {
@@ -71,9 +72,11 @@ static int kvm_handle_wfx(struct kvm_vcpu *vcpu, struct kvm_run *run)
{
if (kvm_vcpu_get_hsr(vcpu) & ESR_ELx_WFx_ISS_WFE) {
trace_kvm_wfx_arm64(*vcpu_pc(vcpu), true);
+ vcpu->stat.wfx_exit_stat++;
kvm_vcpu_on_spin(vcpu);
} else {
trace_kvm_wfx_arm64(*vcpu_pc(vcpu), false);
+ vcpu->stat.wfi_exit_stat++;
kvm_vcpu_block(vcpu);
}
--
1.9.1
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH] KVM: arm64: Count guest exit due to various reasons
2015-10-18 8:04 [PATCH] KVM: arm64: Count guest exit due to various reasons Amit
@ 2015-10-18 19:53 ` Christoffer Dall
2015-10-20 5:17 ` Wei Huang
2015-10-22 21:16 ` Christoffer Dall
1 sibling, 1 reply; 6+ messages in thread
From: Christoffer Dall @ 2015-10-18 19:53 UTC (permalink / raw)
To: Amit; +Cc: marc.zyngier, andre.przywara, kvmarm
On Sun, Oct 18, 2015 at 01:34:42PM +0530, Amit wrote:
> From: Amit Singh Tomar <amittomer25@gmail.com>
>
> This patch adds guest exit statistics to debugfs, This can be helpful
> while measuring KVM performance.
>
> Signed-off-by: Amit Singh Tomar <amittomer25@gmail.com>
> ---
> arch/arm/include/asm/kvm_host.h | 2 ++
> arch/arm/kvm/arm.c | 1 +
> arch/arm/kvm/guest.c | 2 ++
> arch/arm/kvm/mmio.c | 1 +
> arch/arm64/include/asm/kvm_host.h | 5 +++++
> arch/arm64/kvm/guest.c | 8 ++++++++
> arch/arm64/kvm/handle_exit.c | 3 +++
> 7 files changed, 22 insertions(+)
>
> diff --git a/arch/arm/include/asm/kvm_host.h b/arch/arm/include/asm/kvm_host.h
> index c4072d9..d1f3159 100644
> --- a/arch/arm/include/asm/kvm_host.h
> +++ b/arch/arm/include/asm/kvm_host.h
> @@ -147,6 +147,8 @@ struct kvm_vcpu_stat {
> u32 halt_successful_poll;
> u32 halt_attempted_poll;
> u32 halt_wakeup;
> + u32 mmio_exit_stat;
> + u32 exits;
> };
>
> int kvm_vcpu_preferred_target(struct kvm_vcpu_init *init);
> diff --git a/arch/arm/kvm/arm.c b/arch/arm/kvm/arm.c
> index 78b2869..5d0a2c5 100644
> --- a/arch/arm/kvm/arm.c
> +++ b/arch/arm/kvm/arm.c
> @@ -569,6 +569,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run)
> ret = kvm_call_hyp(__kvm_vcpu_run, vcpu);
>
> vcpu->mode = OUTSIDE_GUEST_MODE;
> + vcpu->stat.exits++;
> /*
> * Back from guest
> *************************************************************/
> diff --git a/arch/arm/kvm/guest.c b/arch/arm/kvm/guest.c
> index 96e935b..0eea5fc 100644
> --- a/arch/arm/kvm/guest.c
> +++ b/arch/arm/kvm/guest.c
> @@ -33,6 +33,8 @@
> #define VCPU_STAT(x) { #x, offsetof(struct kvm_vcpu, stat.x), KVM_STAT_VCPU }
>
> struct kvm_stats_debugfs_item debugfs_entries[] = {
> + VCPU_STAT(mmio_exit_stat),
> + VCPU_STAT(exits),
> { NULL }
> };
>
> diff --git a/arch/arm/kvm/mmio.c b/arch/arm/kvm/mmio.c
> index 974b1c6..51d518e 100644
> --- a/arch/arm/kvm/mmio.c
> +++ b/arch/arm/kvm/mmio.c
> @@ -184,6 +184,7 @@ int io_mem_abort(struct kvm_vcpu *vcpu, struct kvm_run *run,
> }
>
> rt = vcpu->arch.mmio_decode.rt;
> + vcpu->stat.mmio_exit_stat++;
>
> if (is_write) {
> data = vcpu_data_guest_to_host(vcpu, *vcpu_reg(vcpu, rt), len);
> diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
> index ed03968..b32c938 100644
> --- a/arch/arm64/include/asm/kvm_host.h
> +++ b/arch/arm64/include/asm/kvm_host.h
> @@ -194,6 +194,11 @@ struct kvm_vcpu_stat {
> u32 halt_successful_poll;
> u32 halt_attempted_poll;
> u32 halt_wakeup;
> + u32 hvc_exit_stat;
> + u32 wfx_exit_stat;
> + u32 wfi_exit_stat;
> + u32 mmio_exit_stat;
> + u32 exits;
> };
>
> int kvm_vcpu_preferred_target(struct kvm_vcpu_init *init);
> diff --git a/arch/arm64/kvm/guest.c b/arch/arm64/kvm/guest.c
> index d250160..b1943cd 100644
> --- a/arch/arm64/kvm/guest.c
> +++ b/arch/arm64/kvm/guest.c
> @@ -34,7 +34,15 @@
>
> #include "trace.h"
>
> +#define VM_STAT(x) { #x, offsetof(struct kvm, stat.x), KVM_STAT_VM }
> +#define VCPU_STAT(x) { #x, offsetof(struct kvm_vcpu, stat.x), KVM_STAT_VCPU }
> +
> struct kvm_stats_debugfs_item debugfs_entries[] = {
> + VCPU_STAT(hvc_exit_stat),
> + VCPU_STAT(wfx_exit_stat),
> + VCPU_STAT(wfi_exit_stat),
> + VCPU_STAT(mmio_exit_stat),
> + VCPU_STAT(exits),
> { NULL }
> };
>
> diff --git a/arch/arm64/kvm/handle_exit.c b/arch/arm64/kvm/handle_exit.c
> index 68a0759..6e38c50 100644
> --- a/arch/arm64/kvm/handle_exit.c
> +++ b/arch/arm64/kvm/handle_exit.c
> @@ -39,6 +39,7 @@ static int handle_hvc(struct kvm_vcpu *vcpu, struct kvm_run *run)
>
> trace_kvm_hvc_arm64(*vcpu_pc(vcpu), *vcpu_reg(vcpu, 0),
> kvm_vcpu_hvc_get_imm(vcpu));
> + vcpu->stat.hvc_exit_stat++;
>
> ret = kvm_psci_call(vcpu);
> if (ret < 0) {
> @@ -71,9 +72,11 @@ static int kvm_handle_wfx(struct kvm_vcpu *vcpu, struct kvm_run *run)
> {
> if (kvm_vcpu_get_hsr(vcpu) & ESR_ELx_WFx_ISS_WFE) {
> trace_kvm_wfx_arm64(*vcpu_pc(vcpu), true);
> + vcpu->stat.wfx_exit_stat++;
> kvm_vcpu_on_spin(vcpu);
> } else {
> trace_kvm_wfx_arm64(*vcpu_pc(vcpu), false);
> + vcpu->stat.wfi_exit_stat++;
> kvm_vcpu_block(vcpu);
> }
>
> --
> 1.9.1
>
I thought the idea was to not use this infrastructure anymore and
instead use tracepoints to count exit reasons?
I'm cc'ing Wei here who sent some patches to QEMU for this purpose
recently.
Thanks,
-Christoffer
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] KVM: arm64: Count guest exit due to various reasons
2015-10-18 19:53 ` Christoffer Dall
@ 2015-10-20 5:17 ` Wei Huang
0 siblings, 0 replies; 6+ messages in thread
From: Wei Huang @ 2015-10-20 5:17 UTC (permalink / raw)
To: Christoffer Dall, Amit; +Cc: marc.zyngier, andre.przywara, kvmarm
On 10/18/2015 02:53 PM, Christoffer Dall wrote:
> On Sun, Oct 18, 2015 at 01:34:42PM +0530, Amit wrote:
>> From: Amit Singh Tomar <amittomer25@gmail.com>
>>
>> This patch adds guest exit statistics to debugfs, This can be helpful
>> while measuring KVM performance.
>>
>> Signed-off-by: Amit Singh Tomar <amittomer25@gmail.com>
>> ---
>> arch/arm/include/asm/kvm_host.h | 2 ++
>> arch/arm/kvm/arm.c | 1 +
>> arch/arm/kvm/guest.c | 2 ++
>> arch/arm/kvm/mmio.c | 1 +
>> arch/arm64/include/asm/kvm_host.h | 5 +++++
>> arch/arm64/kvm/guest.c | 8 ++++++++
>> arch/arm64/kvm/handle_exit.c | 3 +++
>> 7 files changed, 22 insertions(+)
>>
>> diff --git a/arch/arm/include/asm/kvm_host.h b/arch/arm/include/asm/kvm_host.h
>> index c4072d9..d1f3159 100644
>> --- a/arch/arm/include/asm/kvm_host.h
>> +++ b/arch/arm/include/asm/kvm_host.h
>> @@ -147,6 +147,8 @@ struct kvm_vcpu_stat {
>> u32 halt_successful_poll;
>> u32 halt_attempted_poll;
>> u32 halt_wakeup;
>> + u32 mmio_exit_stat;
>> + u32 exits;
>> };
>>
>> int kvm_vcpu_preferred_target(struct kvm_vcpu_init *init);
>> diff --git a/arch/arm/kvm/arm.c b/arch/arm/kvm/arm.c
>> index 78b2869..5d0a2c5 100644
>> --- a/arch/arm/kvm/arm.c
>> +++ b/arch/arm/kvm/arm.c
>> @@ -569,6 +569,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run)
>> ret = kvm_call_hyp(__kvm_vcpu_run, vcpu);
>>
>> vcpu->mode = OUTSIDE_GUEST_MODE;
>> + vcpu->stat.exits++;
>> /*
>> * Back from guest
>> *************************************************************/
>> diff --git a/arch/arm/kvm/guest.c b/arch/arm/kvm/guest.c
>> index 96e935b..0eea5fc 100644
>> --- a/arch/arm/kvm/guest.c
>> +++ b/arch/arm/kvm/guest.c
>> @@ -33,6 +33,8 @@
>> #define VCPU_STAT(x) { #x, offsetof(struct kvm_vcpu, stat.x), KVM_STAT_VCPU }
>>
>> struct kvm_stats_debugfs_item debugfs_entries[] = {
>> + VCPU_STAT(mmio_exit_stat),
>> + VCPU_STAT(exits),
>> { NULL }
>> };
>>
>> diff --git a/arch/arm/kvm/mmio.c b/arch/arm/kvm/mmio.c
>> index 974b1c6..51d518e 100644
>> --- a/arch/arm/kvm/mmio.c
>> +++ b/arch/arm/kvm/mmio.c
>> @@ -184,6 +184,7 @@ int io_mem_abort(struct kvm_vcpu *vcpu, struct kvm_run *run,
>> }
>>
>> rt = vcpu->arch.mmio_decode.rt;
>> + vcpu->stat.mmio_exit_stat++;
>>
>> if (is_write) {
>> data = vcpu_data_guest_to_host(vcpu, *vcpu_reg(vcpu, rt), len);
>> diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
>> index ed03968..b32c938 100644
>> --- a/arch/arm64/include/asm/kvm_host.h
>> +++ b/arch/arm64/include/asm/kvm_host.h
>> @@ -194,6 +194,11 @@ struct kvm_vcpu_stat {
>> u32 halt_successful_poll;
>> u32 halt_attempted_poll;
>> u32 halt_wakeup;
>> + u32 hvc_exit_stat;
>> + u32 wfx_exit_stat;
>> + u32 wfi_exit_stat;
>> + u32 mmio_exit_stat;
>> + u32 exits;
>> };
>>
>> int kvm_vcpu_preferred_target(struct kvm_vcpu_init *init);
>> diff --git a/arch/arm64/kvm/guest.c b/arch/arm64/kvm/guest.c
>> index d250160..b1943cd 100644
>> --- a/arch/arm64/kvm/guest.c
>> +++ b/arch/arm64/kvm/guest.c
>> @@ -34,7 +34,15 @@
>>
>> #include "trace.h"
>>
>> +#define VM_STAT(x) { #x, offsetof(struct kvm, stat.x), KVM_STAT_VM }
>> +#define VCPU_STAT(x) { #x, offsetof(struct kvm_vcpu, stat.x), KVM_STAT_VCPU }
>> +
>> struct kvm_stats_debugfs_item debugfs_entries[] = {
>> + VCPU_STAT(hvc_exit_stat),
>> + VCPU_STAT(wfx_exit_stat),
>> + VCPU_STAT(wfi_exit_stat),
>> + VCPU_STAT(mmio_exit_stat),
>> + VCPU_STAT(exits),
>> { NULL }
>> };
>>
>> diff --git a/arch/arm64/kvm/handle_exit.c b/arch/arm64/kvm/handle_exit.c
>> index 68a0759..6e38c50 100644
>> --- a/arch/arm64/kvm/handle_exit.c
>> +++ b/arch/arm64/kvm/handle_exit.c
>> @@ -39,6 +39,7 @@ static int handle_hvc(struct kvm_vcpu *vcpu, struct kvm_run *run)
>>
>> trace_kvm_hvc_arm64(*vcpu_pc(vcpu), *vcpu_reg(vcpu, 0),
>> kvm_vcpu_hvc_get_imm(vcpu));
>> + vcpu->stat.hvc_exit_stat++;
>>
>> ret = kvm_psci_call(vcpu);
>> if (ret < 0) {
>> @@ -71,9 +72,11 @@ static int kvm_handle_wfx(struct kvm_vcpu *vcpu, struct kvm_run *run)
>> {
>> if (kvm_vcpu_get_hsr(vcpu) & ESR_ELx_WFx_ISS_WFE) {
>> trace_kvm_wfx_arm64(*vcpu_pc(vcpu), true);
>> + vcpu->stat.wfx_exit_stat++;
>> kvm_vcpu_on_spin(vcpu);
>> } else {
>> trace_kvm_wfx_arm64(*vcpu_pc(vcpu), false);
>> + vcpu->stat.wfi_exit_stat++;
>> kvm_vcpu_block(vcpu);
>> }
>>
>> --
>> 1.9.1
>>
>
> I thought the idea was to not use this infrastructure anymore and
> instead use tracepoints to count exit reasons?
>
> I'm cc'ing Wei here who sent some patches to QEMU for this purpose
> recently.
The patches I submitted were for fixing architecture-related issues
while running kvm_stat on ARM platforms. I don't think tracepoint is
meant to replace debugfs for kvm profiling. Both methods have their own
benefits; and they are supported on x86 & ppc platforms. From this
perspective I don't mind taking this patch (after review).
Thanks,
-Wei
>
> Thanks,
> -Christoffer
>
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] KVM: arm64: Count guest exit due to various reasons
2015-10-18 8:04 [PATCH] KVM: arm64: Count guest exit due to various reasons Amit
2015-10-18 19:53 ` Christoffer Dall
@ 2015-10-22 21:16 ` Christoffer Dall
2015-10-26 12:14 ` Amit Tomer
1 sibling, 1 reply; 6+ messages in thread
From: Christoffer Dall @ 2015-10-22 21:16 UTC (permalink / raw)
To: Amit; +Cc: marc.zyngier, andre.przywara, kvmarm
On Sun, Oct 18, 2015 at 01:34:42PM +0530, Amit wrote:
> From: Amit Singh Tomar <amittomer25@gmail.com>
>
> This patch adds guest exit statistics to debugfs, This can be helpful
> while measuring KVM performance.
>
> Signed-off-by: Amit Singh Tomar <amittomer25@gmail.com>
> ---
> arch/arm/include/asm/kvm_host.h | 2 ++
> arch/arm/kvm/arm.c | 1 +
> arch/arm/kvm/guest.c | 2 ++
> arch/arm/kvm/mmio.c | 1 +
> arch/arm64/include/asm/kvm_host.h | 5 +++++
> arch/arm64/kvm/guest.c | 8 ++++++++
> arch/arm64/kvm/handle_exit.c | 3 +++
> 7 files changed, 22 insertions(+)
>
> diff --git a/arch/arm/include/asm/kvm_host.h b/arch/arm/include/asm/kvm_host.h
> index c4072d9..d1f3159 100644
> --- a/arch/arm/include/asm/kvm_host.h
> +++ b/arch/arm/include/asm/kvm_host.h
> @@ -147,6 +147,8 @@ struct kvm_vcpu_stat {
> u32 halt_successful_poll;
> u32 halt_attempted_poll;
> u32 halt_wakeup;
> + u32 mmio_exit_stat;
I think it would be even more useful to distinguish between mmio exits
handled in the kernel vs. those going to userspace.
> + u32 exits;
> };
>
> int kvm_vcpu_preferred_target(struct kvm_vcpu_init *init);
> diff --git a/arch/arm/kvm/arm.c b/arch/arm/kvm/arm.c
> index 78b2869..5d0a2c5 100644
> --- a/arch/arm/kvm/arm.c
> +++ b/arch/arm/kvm/arm.c
> @@ -569,6 +569,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run)
> ret = kvm_call_hyp(__kvm_vcpu_run, vcpu);
>
> vcpu->mode = OUTSIDE_GUEST_MODE;
> + vcpu->stat.exits++;
> /*
> * Back from guest
> *************************************************************/
> diff --git a/arch/arm/kvm/guest.c b/arch/arm/kvm/guest.c
> index 96e935b..0eea5fc 100644
> --- a/arch/arm/kvm/guest.c
> +++ b/arch/arm/kvm/guest.c
> @@ -33,6 +33,8 @@
> #define VCPU_STAT(x) { #x, offsetof(struct kvm_vcpu, stat.x), KVM_STAT_VCPU }
>
> struct kvm_stats_debugfs_item debugfs_entries[] = {
> + VCPU_STAT(mmio_exit_stat),
> + VCPU_STAT(exits),
> { NULL }
> };
>
> diff --git a/arch/arm/kvm/mmio.c b/arch/arm/kvm/mmio.c
> index 974b1c6..51d518e 100644
> --- a/arch/arm/kvm/mmio.c
> +++ b/arch/arm/kvm/mmio.c
> @@ -184,6 +184,7 @@ int io_mem_abort(struct kvm_vcpu *vcpu, struct kvm_run *run,
> }
>
> rt = vcpu->arch.mmio_decode.rt;
> + vcpu->stat.mmio_exit_stat++;
>
> if (is_write) {
> data = vcpu_data_guest_to_host(vcpu, *vcpu_reg(vcpu, rt), len);
> diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
> index ed03968..b32c938 100644
> --- a/arch/arm64/include/asm/kvm_host.h
> +++ b/arch/arm64/include/asm/kvm_host.h
> @@ -194,6 +194,11 @@ struct kvm_vcpu_stat {
> u32 halt_successful_poll;
> u32 halt_attempted_poll;
> u32 halt_wakeup;
> + u32 hvc_exit_stat;
> + u32 wfx_exit_stat;
> + u32 wfi_exit_stat;
> + u32 mmio_exit_stat;
> + u32 exits;
why more exit stats on arm64 than on arm?
why u32? Would it not make sense to make these u64 or just unsigned
long?
> };
>
> int kvm_vcpu_preferred_target(struct kvm_vcpu_init *init);
> diff --git a/arch/arm64/kvm/guest.c b/arch/arm64/kvm/guest.c
> index d250160..b1943cd 100644
> --- a/arch/arm64/kvm/guest.c
> +++ b/arch/arm64/kvm/guest.c
> @@ -34,7 +34,15 @@
>
> #include "trace.h"
>
> +#define VM_STAT(x) { #x, offsetof(struct kvm, stat.x), KVM_STAT_VM }
> +#define VCPU_STAT(x) { #x, offsetof(struct kvm_vcpu, stat.x), KVM_STAT_VCPU }
> +
> struct kvm_stats_debugfs_item debugfs_entries[] = {
> + VCPU_STAT(hvc_exit_stat),
> + VCPU_STAT(wfx_exit_stat),
> + VCPU_STAT(wfi_exit_stat),
> + VCPU_STAT(mmio_exit_stat),
> + VCPU_STAT(exits),
> { NULL }
> };
>
> diff --git a/arch/arm64/kvm/handle_exit.c b/arch/arm64/kvm/handle_exit.c
> index 68a0759..6e38c50 100644
> --- a/arch/arm64/kvm/handle_exit.c
> +++ b/arch/arm64/kvm/handle_exit.c
> @@ -39,6 +39,7 @@ static int handle_hvc(struct kvm_vcpu *vcpu, struct kvm_run *run)
>
> trace_kvm_hvc_arm64(*vcpu_pc(vcpu), *vcpu_reg(vcpu, 0),
> kvm_vcpu_hvc_get_imm(vcpu));
> + vcpu->stat.hvc_exit_stat++;
>
> ret = kvm_psci_call(vcpu);
> if (ret < 0) {
> @@ -71,9 +72,11 @@ static int kvm_handle_wfx(struct kvm_vcpu *vcpu, struct kvm_run *run)
> {
> if (kvm_vcpu_get_hsr(vcpu) & ESR_ELx_WFx_ISS_WFE) {
> trace_kvm_wfx_arm64(*vcpu_pc(vcpu), true);
> + vcpu->stat.wfx_exit_stat++;
either you should use one called wfx_exit_stat or you should distinguish
between wfi and wfe; this is just a weird way of doing it.
> kvm_vcpu_on_spin(vcpu);
> } else {
> trace_kvm_wfx_arm64(*vcpu_pc(vcpu), false);
> + vcpu->stat.wfi_exit_stat++;
> kvm_vcpu_block(vcpu);
> }
>
> --
> 1.9.1
>
Thanks,
-Christoffer
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] KVM: arm64: Count guest exit due to various reasons
2015-10-22 21:16 ` Christoffer Dall
@ 2015-10-26 12:14 ` Amit Tomer
2015-10-26 15:44 ` Christoffer Dall
0 siblings, 1 reply; 6+ messages in thread
From: Amit Tomer @ 2015-10-26 12:14 UTC (permalink / raw)
To: Christoffer Dall; +Cc: marc.zyngier, Andre Przywara, kvmarm
Thanks for the Review.
> I think it would be even more useful to distinguish between mmio exits
> handled in the kernel vs. those going to userspace.
Ok, I would work on it.
> why more exit stats on arm64 than on arm?
Intention is to just add these stats only for ARM64 but I had to do it for arm,
so that things don't break.
> why u32? Would it not make sense to make these u64 or just unsigned
> long?
Ok, You mean to say, for ARM we should use u32 and for ARM64 we should
u64, right?
> either you should use one called wfx_exit_stat or you should distinguish
> between wfi and wfe; this is just a weird way of doing it.
Sorry, didn't just get your point here :(
Thanks
Amit.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] KVM: arm64: Count guest exit due to various reasons
2015-10-26 12:14 ` Amit Tomer
@ 2015-10-26 15:44 ` Christoffer Dall
0 siblings, 0 replies; 6+ messages in thread
From: Christoffer Dall @ 2015-10-26 15:44 UTC (permalink / raw)
To: Amit Tomer; +Cc: marc.zyngier, Andre Przywara, kvmarm
On Mon, Oct 26, 2015 at 05:44:42PM +0530, Amit Tomer wrote:
> Thanks for the Review.
>
>
> > I think it would be even more useful to distinguish between mmio exits
> > handled in the kernel vs. those going to userspace.
>
> Ok, I would work on it.
>
>
> > why more exit stats on arm64 than on arm?
>
> Intention is to just add these stats only for ARM64 but I had to do it for arm,
> so that things don't break.
please keep both architectures up to date as much as possible for these
kinds of changes.
>
> > why u32? Would it not make sense to make these u64 or just unsigned
> > long?
>
> Ok, You mean to say, for ARM we should use u32 and for ARM64 we should
> u64, right?
loss of context here, so I'm not sure. Is there a reason why you
cannot/shouldn't use u64? I'm just thinking that overflows could happen
fairly easy on u32, but otoh. there are some existing defines using u32
for some reason...
>
> > either you should use one called wfx_exit_stat or you should distinguish
> > between wfi and wfe; this is just a weird way of doing it.
>
> Sorry, didn't just get your point here :(
>
wfx means "either a wfi or wfe happened", so you either define two
separate trace points, one for wfi and one for wfe, or you define a
single tracepoint for wfx.
I'll look forward to v2.
-Christoffer
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2015-10-26 15:40 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2015-10-18 8:04 [PATCH] KVM: arm64: Count guest exit due to various reasons Amit
2015-10-18 19:53 ` Christoffer Dall
2015-10-20 5:17 ` Wei Huang
2015-10-22 21:16 ` Christoffer Dall
2015-10-26 12:14 ` Amit Tomer
2015-10-26 15:44 ` Christoffer Dall
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.