Hi All,
I'd like to propose a patch which adds a tracepoint at write_tsc_offset for
tracing guests TSC offset changes. It is required for sorting the trace data of
a guest and the host in chronological order.
In a virtualization environment, it is difficult to analyze performance
problems, such as a delay of I/O request on a guest. This is because multiple
guests operate on the host. One of approaches for solving such kind of problems
is to sort trace data of guests and the host in chronological order.
Raw TSC can be chosen as a timestamp of ftrace. I think TSC is useful for
merging trace data in chronological order by two reasons. One of the reasons is
that guests can directly read raw TSC from the CPU using rdtsc operation. This
means that raw TSC value is not software clock like sched_clock, so we don't
need to consider about how the timestamp is calculated. The other is that TSC
of recent x86 CPUs is constantly incremented. This means that we don't need to
worry about pace of the timestamp. Therefore, choosing TSC as a timestamp for
tracing is reasonable to integrate trace data of guests and a host.
Here, we need to consider about just one matter for using TSC on guests. TSC
value on a guest is always the host TSC plus the guest's "TSC offset". In other
words, to merge trace data using TSC as timestamp in chronological order, we
need to consider TSC offset of the guest.
However, only the host kernel can read the TSC offset from VMCS and TSC offset
is not output in anywhere now. In other words, tools in userland cannot get
the TSC offset value, so we cannot merge trace data of guest and the host in
chronological order. Therefore, I think the TSC offset should be exported for
userland tools.
In this patch, TSC offset is exported by the tracepoint kvm_write_tsc_offset
on the host. TSC offset events will be very rare event because guests may
seldom execute write_tsc. So if we enable this event for a normal buffer of
ftrace, the events will be overwritten by other events. For a recent linux
kernel, a multiple buffer function is available. So, I recommend to enable
this event for a sub buffer of ftrace.
<Example>
We assume that wakeup-latency for a command is big on a guest. Normally
we will use ftrace's wakeup-latency tracer or event tracer on the guest, but we
may not be able to solve this problem. This is because guests often exit to
the host for several reasons. In the next, we will use TSC as ftrace's timestamp
and record the trace data on the guest and the host. Then, we get following
data:
/* guest data */
comm-3826 [000] d...49836825726903: sched_wakeup: [detail]
comm-3826 [000] d...49836832225344: sched_switch: [detail]
/* host data */
qemu-kvm-2687 [003] d...50550079203669: kvm_exit: [detail]
qemu-kvm-2687 [003] d...50550079206816: kvm_entry: [detail]
qemu-kvm-2687 [003] d...50550079240656: kvm_exit: [detail]
qemu-kvm-2687 [003] d...50550079243467: kvm_entry: [detail]
qemu-kvm-2687 [003] d...50550079256103: kvm_exit: [detail]
qemu-kvm-2687 [003] d...50550079268391: kvm_entry: [detail]
qemu-kvm-2687 [003] d...50550079280829: kvm_exit: [detail]
qemu-kvm-2687 [003] d...50550079286028: kvm_entry: [detail]
Since TSC offset is not considered, these data cannot be merged. If this trace
data is shown like as follows, we will be able to understand the reason:
qemu-kvm-2687 [003] d...50550079203669: kvm_exit: [detail]
qemu-kvm-2687 [003] d...50550079206816: kvm_entry: [detail]
comm-3826 [000] d.h.49836825726903: sched_wakeup: [detail] <=
qemu-kvm-2687 [003] d...50550079240656: kvm_exit: [detail]
qemu-kvm-2687 [003] d...50550079243467: kvm_entry: [detail]
qemu-kvm-2687 [003] d...50550079256103: kvm_exit: [detail]
qemu-kvm-2687 [003] d...50550079268391: kvm_entry: [detail]
comm-3826 [000] d...49836832225344: sched_switch: [detail] <=
qemu-kvm-2687 [003] d...50550079280829: kvm_exit: [detail]
qemu-kvm-2687 [003] d...50550079286028: kvm_entry: [detail]
In this case, we can understand wakeup-latency was big due to exit to host
twice. Getting this data sorted in chronological order is our goal.
To merge the data like previous pattern, we apply this patch. Then, we can
get TSC offset of the guest as follows:
# cat /sys/kernel/debug/tracing/instances/tsc_offset/trace
.. d...4300151845072: kvm_write_tsc_offset: vcpu=0 prev=0 next=18446739773557710924
^ ^^^^^^^^^^^^^^^^^^^^
We use this TSC offset value to a merge script and obtain the following data:
$ ./trace-merge.pl -g guest_data -h host_data -t 18446739773557710924
h qemu-kvm-2687 [003] d...50550079203669: kvm_exit: [detail]
h qemu-kvm-2687 [003] d...50550079206816: kvm_entry: [detail]
g comm-3826 [000] d.h.50550079226331: sched_wakeup: [detail] <=
h qemu-kvm-2687 [003] d...50550079240656: kvm_exit: [detail]
h qemu-kvm-2687 [003] d...50550079243467: kvm_entry: [detail]
h qemu-kvm-2687 [003] d...50550079256103: kvm_exit: [detail]
h qemu-kvm-2687 [003] d...50550079268391: kvm_entry: [detail]
g comm-3826 [000] d...50550079279266: sched_switch: [detail] <=
h qemu-kvm-2687 [003] d...50550079280829: kvm_exit: [detail]
h qemu-kvm-2687 [003] d...50550079286028: kvm_entry: [detail]
|
\----guest/host
In this summary, I suggest the patch which TSC offset for each guest can be
output on the host.
As one example, I will send a merge tool. This tool assumes the guest doesn't
execute write_tsc.
Changes in V2:
- Use tracepoint for outputting TSC offset value instead of printk() [1/1]
Changes in V3:
- Add tracepoint in svm.c [1/1]
- Record vcpu_id [1/1]
- Change "previous=" to "prev=" in output result [1/1]
Thanks!
---
Yoshihiro YUNOMAE (1):
kvm: Add a tracepoint write_tsc_offset
arch/x86/kvm/svm.c | 10 +++++++++-
arch/x86/kvm/trace.h | 21 +++++++++++++++++++++
arch/x86/kvm/vmx.c | 7 ++++++-
arch/x86/kvm/x86.c | 1 +
4 files changed, 37 insertions(+), 2 deletions(-)
--
Yoshihiro YUNOMAE
Software Platform Research Dept. Linux Technology Center
Hitachi, Ltd., Yokohama Research Laboratory
E-mail: [email protected]
Add a tracepoint write_tsc_offset for tracing TSC offset change.
We want to merge ftrace's trace data of guest OSs and the host OS using
TSC for timestamp in chronological order. We need "TSC offset" values for
each guest when merge those because the TSC value on a guest is always the
host TSC plus guest's TSC offset. If we get the TSC offset values, we can
calculate the host TSC value for each guest events from the TSC offset and
the event TSC value. The host TSC values of the guest events are used when we
want to merge trace data of guests and the host in chronological order.
(Note: the trace_clock of both the host and the guest must be set x86-tsc in
this case)
This tracepoint also records vcpu_id which can be used to merge trace data for
SMP guests. A merge tool will read TSC offset for each vcpu, then the tool
converts guest TSC values to host TSC values for each vcpu.
TSC offset is stored in the VMCS by vmx_write_tsc_offset() or
vmx_adjust_tsc_offset(). KVM executes the former function when a guest boots.
The latter function is executed when kvm clock is updated. Only host can read
TSC offset value from VMCS, so a host needs to output TSC offset value
when TSC offset is changed.
Since the TSC offset is not often changed, it could be overwritten by other
frequent events while tracing. To avoid that, I recommend to use a special
instance for getting this event:
1. set a instance before booting a guest
# cd /sys/kernel/debug/tracing/instances
# mkdir tsc_offset
# cd tsc_offset
# echo x86-tsc > trace_clock
# echo 1 > events/kvm/kvm_write_tsc_offset/enable
2. boot a guest
Signed-off-by: Yoshihiro YUNOMAE <[email protected]>
Cc: Joerg Roedel <[email protected]>
Cc: Marcelo Tosatti <[email protected]>
Cc: Gleb Natapov <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: "H. Peter Anvin" <[email protected]>
---
arch/x86/kvm/svm.c | 10 +++++++++-
arch/x86/kvm/trace.h | 21 +++++++++++++++++++++
arch/x86/kvm/vmx.c | 7 ++++++-
arch/x86/kvm/x86.c | 1 +
4 files changed, 37 insertions(+), 2 deletions(-)
diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index a14a6ea..c0bc803 100644
--- a/arch/x86/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
@@ -1026,7 +1026,10 @@ static void svm_write_tsc_offset(struct kvm_vcpu *vcpu, u64 offset)
g_tsc_offset = svm->vmcb->control.tsc_offset -
svm->nested.hsave->control.tsc_offset;
svm->nested.hsave->control.tsc_offset = offset;
- }
+ } else
+ trace_kvm_write_tsc_offset(vcpu->vcpu_id,
+ svm->vmcb->control.tsc_offset,
+ offset);
svm->vmcb->control.tsc_offset = offset + g_tsc_offset;
@@ -1044,6 +1047,11 @@ static void svm_adjust_tsc_offset(struct kvm_vcpu *vcpu, s64 adjustment, bool ho
svm->vmcb->control.tsc_offset += adjustment;
if (is_guest_mode(vcpu))
svm->nested.hsave->control.tsc_offset += adjustment;
+ else
+ trace_kvm_write_tsc_offset(vcpu->vcpu_id,
+ svm->vmcb->control.tsc_offset - adjustment,
+ svm->vmcb->control.tsc_offset);
+
mark_dirty(svm->vmcb, VMCB_INTERCEPTS);
}
diff --git a/arch/x86/kvm/trace.h b/arch/x86/kvm/trace.h
index fe5e00e..6c82cf1 100644
--- a/arch/x86/kvm/trace.h
+++ b/arch/x86/kvm/trace.h
@@ -815,6 +815,27 @@ TRACE_EVENT(kvm_track_tsc,
__print_symbolic(__entry->host_clock, host_clocks))
);
+TRACE_EVENT(kvm_write_tsc_offset,
+ TP_PROTO(unsigned int vcpu_id, __u64 previous_tsc_offset,
+ __u64 next_tsc_offset),
+ TP_ARGS(vcpu_id, previous_tsc_offset, next_tsc_offset),
+
+ TP_STRUCT__entry(
+ __field( unsigned int, vcpu_id )
+ __field( __u64, previous_tsc_offset )
+ __field( __u64, next_tsc_offset )
+ ),
+
+ TP_fast_assign(
+ __entry->vcpu_id = vcpu_id;
+ __entry->previous_tsc_offset = previous_tsc_offset;
+ __entry->next_tsc_offset = next_tsc_offset;
+ ),
+
+ TP_printk("vcpu=%u prev=%llu next=%llu", __entry->vcpu_id,
+ __entry->previous_tsc_offset, __entry->next_tsc_offset)
+);
+
#endif /* CONFIG_X86_64 */
#endif /* _TRACE_KVM_H */
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 25a791e..eb11856 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -2096,6 +2096,8 @@ static void vmx_write_tsc_offset(struct kvm_vcpu *vcpu, u64 offset)
(nested_cpu_has(vmcs12, CPU_BASED_USE_TSC_OFFSETING) ?
vmcs12->tsc_offset : 0));
} else {
+ trace_kvm_write_tsc_offset(vcpu->vcpu_id,
+ vmcs_read64(TSC_OFFSET), offset);
vmcs_write64(TSC_OFFSET, offset);
}
}
@@ -2103,11 +2105,14 @@ static void vmx_write_tsc_offset(struct kvm_vcpu *vcpu, u64 offset)
static void vmx_adjust_tsc_offset(struct kvm_vcpu *vcpu, s64 adjustment, bool host)
{
u64 offset = vmcs_read64(TSC_OFFSET);
+
vmcs_write64(TSC_OFFSET, offset + adjustment);
if (is_guest_mode(vcpu)) {
/* Even when running L2, the adjustment needs to apply to L1 */
to_vmx(vcpu)->nested.vmcs01_tsc_offset += adjustment;
- }
+ } else
+ trace_kvm_write_tsc_offset(vcpu->vcpu_id, offset,
+ offset + adjustment);
}
static u64 vmx_compute_tsc_offset(struct kvm_vcpu *vcpu, u64 target_tsc)
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 05a8b1a..c942a0c 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -7264,3 +7264,4 @@ EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_nested_intr_vmexit);
EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_invlpga);
EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_skinit);
EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_nested_intercepts);
+EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_write_tsc_offset);
This tool merges trace data of a guest and a host in chronological
order. Restrictions of this tool is as follows:
- one guest (not for multiple guests)
- stable TSC (not backward TSC)
- synchronized TSC
- unchanged TSC offset (the guest does not execute write_TSC)
- How to use
1. [host] Enable kvm_write_tsc_offset before booting a guest
# cd /sys/kernel/debug/tracing/instances
# mkdir tsc_offset
# cd tsc_offset
# echo x86-tsc > trace_clock
# echo 1 > events/kvm/kvm_write_tsc_offset/enable
2. [host] Enable events you want
Note: I recommend to enable kvm_exit/entry events.
# cd /sys/kernel/debug/tracing
# echo kvm_entry >> set_event
# echo kvm_exit >> set_event
# [snip]
# echo x86-tsc > trace_clock
3. [host] Boot the guest
4. [guest] Enable events you want
# cd /sys/kernel/debug/tracing
# echo sched_wakeup >> set_event
# echo sched_switch >> set_event
# [snip]
# echo x86-tsc > trace_clock
5. [guest] Run programs
6. [guest/host] Get trace data
# echo 0 > tracing_on
# cat trace > /home/yourdir/log/guest_trace.txt (for the guest)
(cat trace > /home/yourdir/log/host_trace.txt (for the host))
# scp [host_IP]:/home/yourdir/log/guest_trace.txt (only for the guest)
7. [host] Get next TSC offset
# cat /sys/kernel/debug/tracing/instances/tsc_offset/trace
qemu-kvm-22089 [000] d...4300151845072: kvm_write_tsc_offset:
vcpu=0 prev=0 next=18446739773557710924
8. [host] Run this tool with the next TSC offset for -t option.
$ ./trace-merge.pl -g ~/log/guest_trace.txt -h ~/log/host_trace.txt \
-t 18446739773557710924
h qemu-kvm-2687 [003] d...50550079203669: kvm_exit: [detail]
h qemu-kvm-2687 [003] d...50550079206816: kvm_entry: [detail]
g comm-3826 [000] d.h.50550079226331: sched_wakeup: [detail]
h qemu-kvm-2687 [003] d...50550079240656: kvm_exit: [detail]
h qemu-kvm-2687 [003] d...50550079243467: kvm_entry: [detail]
h qemu-kvm-2687 [003] d...50550079256103: kvm_exit: [detail]
h qemu-kvm-2687 [003] d...50550079268391: kvm_entry: [detail]
g comm-3826 [000] d...50550079279266: sched_switch: [detail]
h qemu-kvm-2687 [003] d...50550079280829: kvm_exit: [detail]
h qemu-kvm-2687 [003] d...50550079286028: kvm_entry: [detail]
|
\----guest/host
Thanks,
--
Yoshihiro YUNOMAE
Software Platform Research Dept. Linux Technology Center
Hitachi, Ltd., Yokohama Research Laboratory
E-mail: [email protected]
On Wed, Jun 12, 2013 at 04:43:44PM +0900, Yoshihiro YUNOMAE wrote:
> Add a tracepoint write_tsc_offset for tracing TSC offset change.
> We want to merge ftrace's trace data of guest OSs and the host OS using
> TSC for timestamp in chronological order. We need "TSC offset" values for
> each guest when merge those because the TSC value on a guest is always the
> host TSC plus guest's TSC offset. If we get the TSC offset values, we can
> calculate the host TSC value for each guest events from the TSC offset and
> the event TSC value. The host TSC values of the guest events are used when we
> want to merge trace data of guests and the host in chronological order.
> (Note: the trace_clock of both the host and the guest must be set x86-tsc in
> this case)
>
> This tracepoint also records vcpu_id which can be used to merge trace data for
> SMP guests. A merge tool will read TSC offset for each vcpu, then the tool
> converts guest TSC values to host TSC values for each vcpu.
>
> TSC offset is stored in the VMCS by vmx_write_tsc_offset() or
> vmx_adjust_tsc_offset(). KVM executes the former function when a guest boots.
> The latter function is executed when kvm clock is updated. Only host can read
> TSC offset value from VMCS, so a host needs to output TSC offset value
> when TSC offset is changed.
>
> Since the TSC offset is not often changed, it could be overwritten by other
> frequent events while tracing. To avoid that, I recommend to use a special
> instance for getting this event:
>
> 1. set a instance before booting a guest
> # cd /sys/kernel/debug/tracing/instances
> # mkdir tsc_offset
> # cd tsc_offset
> # echo x86-tsc > trace_clock
> # echo 1 > events/kvm/kvm_write_tsc_offset/enable
>
> 2. boot a guest
>
> Signed-off-by: Yoshihiro YUNOMAE <[email protected]>
> Cc: Joerg Roedel <[email protected]>
> Cc: Marcelo Tosatti <[email protected]>
> Cc: Gleb Natapov <[email protected]>
> Cc: Thomas Gleixner <[email protected]>
> Cc: Ingo Molnar <[email protected]>
> Cc: "H. Peter Anvin" <[email protected]>
> ---
> arch/x86/kvm/svm.c | 10 +++++++++-
> arch/x86/kvm/trace.h | 21 +++++++++++++++++++++
> arch/x86/kvm/vmx.c | 7 ++++++-
> arch/x86/kvm/x86.c | 1 +
> 4 files changed, 37 insertions(+), 2 deletions(-)
>
> diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
> index a14a6ea..c0bc803 100644
> --- a/arch/x86/kvm/svm.c
> +++ b/arch/x86/kvm/svm.c
> @@ -1026,7 +1026,10 @@ static void svm_write_tsc_offset(struct kvm_vcpu *vcpu, u64 offset)
> g_tsc_offset = svm->vmcb->control.tsc_offset -
> svm->nested.hsave->control.tsc_offset;
> svm->nested.hsave->control.tsc_offset = offset;
> - }
> + } else
> + trace_kvm_write_tsc_offset(vcpu->vcpu_id,
> + svm->vmcb->control.tsc_offset,
> + offset);
>
> svm->vmcb->control.tsc_offset = offset + g_tsc_offset;
>
> @@ -1044,6 +1047,11 @@ static void svm_adjust_tsc_offset(struct kvm_vcpu *vcpu, s64 adjustment, bool ho
> svm->vmcb->control.tsc_offset += adjustment;
> if (is_guest_mode(vcpu))
> svm->nested.hsave->control.tsc_offset += adjustment;
> + else
> + trace_kvm_write_tsc_offset(vcpu->vcpu_id,
> + svm->vmcb->control.tsc_offset - adjustment,
> + svm->vmcb->control.tsc_offset);
> +
> mark_dirty(svm->vmcb, VMCB_INTERCEPTS);
> }
>
> diff --git a/arch/x86/kvm/trace.h b/arch/x86/kvm/trace.h
> index fe5e00e..6c82cf1 100644
> --- a/arch/x86/kvm/trace.h
> +++ b/arch/x86/kvm/trace.h
> @@ -815,6 +815,27 @@ TRACE_EVENT(kvm_track_tsc,
> __print_symbolic(__entry->host_clock, host_clocks))
> );
>
> +TRACE_EVENT(kvm_write_tsc_offset,
> + TP_PROTO(unsigned int vcpu_id, __u64 previous_tsc_offset,
> + __u64 next_tsc_offset),
> + TP_ARGS(vcpu_id, previous_tsc_offset, next_tsc_offset),
> +
> + TP_STRUCT__entry(
> + __field( unsigned int, vcpu_id )
> + __field( __u64, previous_tsc_offset )
> + __field( __u64, next_tsc_offset )
> + ),
> +
> + TP_fast_assign(
> + __entry->vcpu_id = vcpu_id;
> + __entry->previous_tsc_offset = previous_tsc_offset;
> + __entry->next_tsc_offset = next_tsc_offset;
> + ),
> +
> + TP_printk("vcpu=%u prev=%llu next=%llu", __entry->vcpu_id,
> + __entry->previous_tsc_offset, __entry->next_tsc_offset)
> +);
> +
> #endif /* CONFIG_X86_64 */
>
> #endif /* _TRACE_KVM_H */
> diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
> index 25a791e..eb11856 100644
> --- a/arch/x86/kvm/vmx.c
> +++ b/arch/x86/kvm/vmx.c
> @@ -2096,6 +2096,8 @@ static void vmx_write_tsc_offset(struct kvm_vcpu *vcpu, u64 offset)
> (nested_cpu_has(vmcs12, CPU_BASED_USE_TSC_OFFSETING) ?
> vmcs12->tsc_offset : 0));
> } else {
> + trace_kvm_write_tsc_offset(vcpu->vcpu_id,
> + vmcs_read64(TSC_OFFSET), offset);
> vmcs_write64(TSC_OFFSET, offset);
> }
> }
> @@ -2103,11 +2105,14 @@ static void vmx_write_tsc_offset(struct kvm_vcpu *vcpu, u64 offset)
> static void vmx_adjust_tsc_offset(struct kvm_vcpu *vcpu, s64 adjustment, bool host)
> {
> u64 offset = vmcs_read64(TSC_OFFSET);
> +
> vmcs_write64(TSC_OFFSET, offset + adjustment);
> if (is_guest_mode(vcpu)) {
> /* Even when running L2, the adjustment needs to apply to L1 */
> to_vmx(vcpu)->nested.vmcs01_tsc_offset += adjustment;
> - }
> + } else
> + trace_kvm_write_tsc_offset(vcpu->vcpu_id, offset,
> + offset + adjustment);
> }
>
> static u64 vmx_compute_tsc_offset(struct kvm_vcpu *vcpu, u64 target_tsc)
> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> index 05a8b1a..c942a0c 100644
> --- a/arch/x86/kvm/x86.c
> +++ b/arch/x86/kvm/x86.c
> @@ -7264,3 +7264,4 @@ EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_nested_intr_vmexit);
> EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_invlpga);
> EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_skinit);
> EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_nested_intercepts);
> +EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_write_tsc_offset);
ACK
On Wed, Jun 12, 2013 at 04:43:41PM +0900, Yoshihiro YUNOMAE wrote:
> Hi All,
>
Applied, thanks.
> I'd like to propose a patch which adds a tracepoint at write_tsc_offset for
> tracing guests TSC offset changes. It is required for sorting the trace data of
> a guest and the host in chronological order.
>
> In a virtualization environment, it is difficult to analyze performance
> problems, such as a delay of I/O request on a guest. This is because multiple
> guests operate on the host. One of approaches for solving such kind of problems
> is to sort trace data of guests and the host in chronological order.
>
> Raw TSC can be chosen as a timestamp of ftrace. I think TSC is useful for
> merging trace data in chronological order by two reasons. One of the reasons is
> that guests can directly read raw TSC from the CPU using rdtsc operation. This
> means that raw TSC value is not software clock like sched_clock, so we don't
> need to consider about how the timestamp is calculated. The other is that TSC
> of recent x86 CPUs is constantly incremented. This means that we don't need to
> worry about pace of the timestamp. Therefore, choosing TSC as a timestamp for
> tracing is reasonable to integrate trace data of guests and a host.
>
> Here, we need to consider about just one matter for using TSC on guests. TSC
> value on a guest is always the host TSC plus the guest's "TSC offset". In other
> words, to merge trace data using TSC as timestamp in chronological order, we
> need to consider TSC offset of the guest.
>
> However, only the host kernel can read the TSC offset from VMCS and TSC offset
> is not output in anywhere now. In other words, tools in userland cannot get
> the TSC offset value, so we cannot merge trace data of guest and the host in
> chronological order. Therefore, I think the TSC offset should be exported for
> userland tools.
>
> In this patch, TSC offset is exported by the tracepoint kvm_write_tsc_offset
> on the host. TSC offset events will be very rare event because guests may
> seldom execute write_tsc. So if we enable this event for a normal buffer of
> ftrace, the events will be overwritten by other events. For a recent linux
> kernel, a multiple buffer function is available. So, I recommend to enable
> this event for a sub buffer of ftrace.
>
> <Example>
> We assume that wakeup-latency for a command is big on a guest. Normally
> we will use ftrace's wakeup-latency tracer or event tracer on the guest, but we
> may not be able to solve this problem. This is because guests often exit to
> the host for several reasons. In the next, we will use TSC as ftrace's timestamp
> and record the trace data on the guest and the host. Then, we get following
> data:
>
> /* guest data */
> comm-3826 [000] d...49836825726903: sched_wakeup: [detail]
> comm-3826 [000] d...49836832225344: sched_switch: [detail]
> /* host data */
> qemu-kvm-2687 [003] d...50550079203669: kvm_exit: [detail]
> qemu-kvm-2687 [003] d...50550079206816: kvm_entry: [detail]
> qemu-kvm-2687 [003] d...50550079240656: kvm_exit: [detail]
> qemu-kvm-2687 [003] d...50550079243467: kvm_entry: [detail]
> qemu-kvm-2687 [003] d...50550079256103: kvm_exit: [detail]
> qemu-kvm-2687 [003] d...50550079268391: kvm_entry: [detail]
> qemu-kvm-2687 [003] d...50550079280829: kvm_exit: [detail]
> qemu-kvm-2687 [003] d...50550079286028: kvm_entry: [detail]
>
> Since TSC offset is not considered, these data cannot be merged. If this trace
> data is shown like as follows, we will be able to understand the reason:
>
> qemu-kvm-2687 [003] d...50550079203669: kvm_exit: [detail]
> qemu-kvm-2687 [003] d...50550079206816: kvm_entry: [detail]
> comm-3826 [000] d.h.49836825726903: sched_wakeup: [detail] <=
> qemu-kvm-2687 [003] d...50550079240656: kvm_exit: [detail]
> qemu-kvm-2687 [003] d...50550079243467: kvm_entry: [detail]
> qemu-kvm-2687 [003] d...50550079256103: kvm_exit: [detail]
> qemu-kvm-2687 [003] d...50550079268391: kvm_entry: [detail]
> comm-3826 [000] d...49836832225344: sched_switch: [detail] <=
> qemu-kvm-2687 [003] d...50550079280829: kvm_exit: [detail]
> qemu-kvm-2687 [003] d...50550079286028: kvm_entry: [detail]
>
> In this case, we can understand wakeup-latency was big due to exit to host
> twice. Getting this data sorted in chronological order is our goal.
>
> To merge the data like previous pattern, we apply this patch. Then, we can
> get TSC offset of the guest as follows:
>
> # cat /sys/kernel/debug/tracing/instances/tsc_offset/trace
> .. d...4300151845072: kvm_write_tsc_offset: vcpu=0 prev=0 next=18446739773557710924
> ^ ^^^^^^^^^^^^^^^^^^^^
>
> We use this TSC offset value to a merge script and obtain the following data:
>
> $ ./trace-merge.pl -g guest_data -h host_data -t 18446739773557710924
>
> h qemu-kvm-2687 [003] d...50550079203669: kvm_exit: [detail]
> h qemu-kvm-2687 [003] d...50550079206816: kvm_entry: [detail]
> g comm-3826 [000] d.h.50550079226331: sched_wakeup: [detail] <=
> h qemu-kvm-2687 [003] d...50550079240656: kvm_exit: [detail]
> h qemu-kvm-2687 [003] d...50550079243467: kvm_entry: [detail]
> h qemu-kvm-2687 [003] d...50550079256103: kvm_exit: [detail]
> h qemu-kvm-2687 [003] d...50550079268391: kvm_entry: [detail]
> g comm-3826 [000] d...50550079279266: sched_switch: [detail] <=
> h qemu-kvm-2687 [003] d...50550079280829: kvm_exit: [detail]
> h qemu-kvm-2687 [003] d...50550079286028: kvm_entry: [detail]
> |
> \----guest/host
>
> In this summary, I suggest the patch which TSC offset for each guest can be
> output on the host.
>
> As one example, I will send a merge tool. This tool assumes the guest doesn't
> execute write_tsc.
>
> Changes in V2:
> - Use tracepoint for outputting TSC offset value instead of printk() [1/1]
>
> Changes in V3:
> - Add tracepoint in svm.c [1/1]
> - Record vcpu_id [1/1]
> - Change "previous=" to "prev=" in output result [1/1]
>
> Thanks!
>
> ---
>
> Yoshihiro YUNOMAE (1):
> kvm: Add a tracepoint write_tsc_offset
>
>
> arch/x86/kvm/svm.c | 10 +++++++++-
> arch/x86/kvm/trace.h | 21 +++++++++++++++++++++
> arch/x86/kvm/vmx.c | 7 ++++++-
> arch/x86/kvm/x86.c | 1 +
> 4 files changed, 37 insertions(+), 2 deletions(-)
>
> --
> Yoshihiro YUNOMAE
> Software Platform Research Dept. Linux Technology Center
> Hitachi, Ltd., Yokohama Research Laboratory
> E-mail: [email protected]
--
Gleb.
Il 12/06/2013 09:43, Yoshihiro YUNOMAE ha scritto:
> Add a tracepoint write_tsc_offset for tracing TSC offset change.
> We want to merge ftrace's trace data of guest OSs and the host OS using
> TSC for timestamp in chronological order. We need "TSC offset" values for
> each guest when merge those because the TSC value on a guest is always the
> host TSC plus guest's TSC offset. If we get the TSC offset values, we can
> calculate the host TSC value for each guest events from the TSC offset and
> the event TSC value. The host TSC values of the guest events are used when we
> want to merge trace data of guests and the host in chronological order.
> (Note: the trace_clock of both the host and the guest must be set x86-tsc in
> this case)
>
> This tracepoint also records vcpu_id which can be used to merge trace data for
> SMP guests. A merge tool will read TSC offset for each vcpu, then the tool
> converts guest TSC values to host TSC values for each vcpu.
>
> TSC offset is stored in the VMCS by vmx_write_tsc_offset() or
> vmx_adjust_tsc_offset(). KVM executes the former function when a guest boots.
> The latter function is executed when kvm clock is updated. Only host can read
> TSC offset value from VMCS, so a host needs to output TSC offset value
> when TSC offset is changed.
>
> Since the TSC offset is not often changed, it could be overwritten by other
> frequent events while tracing. To avoid that, I recommend to use a special
> instance for getting this event:
>
> 1. set a instance before booting a guest
> # cd /sys/kernel/debug/tracing/instances
> # mkdir tsc_offset
> # cd tsc_offset
> # echo x86-tsc > trace_clock
> # echo 1 > events/kvm/kvm_write_tsc_offset/enable
>
> 2. boot a guest
>
> Signed-off-by: Yoshihiro YUNOMAE <[email protected]>
> Cc: Joerg Roedel <[email protected]>
> Cc: Marcelo Tosatti <[email protected]>
> Cc: Gleb Natapov <[email protected]>
> Cc: Thomas Gleixner <[email protected]>
> Cc: Ingo Molnar <[email protected]>
> Cc: "H. Peter Anvin" <[email protected]>
> ---
> arch/x86/kvm/svm.c | 10 +++++++++-
> arch/x86/kvm/trace.h | 21 +++++++++++++++++++++
> arch/x86/kvm/vmx.c | 7 ++++++-
> arch/x86/kvm/x86.c | 1 +
> 4 files changed, 37 insertions(+), 2 deletions(-)
>
> diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
> index a14a6ea..c0bc803 100644
> --- a/arch/x86/kvm/svm.c
> +++ b/arch/x86/kvm/svm.c
> @@ -1026,7 +1026,10 @@ static void svm_write_tsc_offset(struct kvm_vcpu *vcpu, u64 offset)
> g_tsc_offset = svm->vmcb->control.tsc_offset -
> svm->nested.hsave->control.tsc_offset;
> svm->nested.hsave->control.tsc_offset = offset;
> - }
> + } else
> + trace_kvm_write_tsc_offset(vcpu->vcpu_id,
> + svm->vmcb->control.tsc_offset,
> + offset);
>
> svm->vmcb->control.tsc_offset = offset + g_tsc_offset;
>
> @@ -1044,6 +1047,11 @@ static void svm_adjust_tsc_offset(struct kvm_vcpu *vcpu, s64 adjustment, bool ho
> svm->vmcb->control.tsc_offset += adjustment;
> if (is_guest_mode(vcpu))
> svm->nested.hsave->control.tsc_offset += adjustment;
> + else
> + trace_kvm_write_tsc_offset(vcpu->vcpu_id,
> + svm->vmcb->control.tsc_offset - adjustment,
> + svm->vmcb->control.tsc_offset);
> +
> mark_dirty(svm->vmcb, VMCB_INTERCEPTS);
> }
>
> diff --git a/arch/x86/kvm/trace.h b/arch/x86/kvm/trace.h
> index fe5e00e..6c82cf1 100644
> --- a/arch/x86/kvm/trace.h
> +++ b/arch/x86/kvm/trace.h
> @@ -815,6 +815,27 @@ TRACE_EVENT(kvm_track_tsc,
> __print_symbolic(__entry->host_clock, host_clocks))
> );
>
> +TRACE_EVENT(kvm_write_tsc_offset,
> + TP_PROTO(unsigned int vcpu_id, __u64 previous_tsc_offset,
> + __u64 next_tsc_offset),
> + TP_ARGS(vcpu_id, previous_tsc_offset, next_tsc_offset),
> +
> + TP_STRUCT__entry(
> + __field( unsigned int, vcpu_id )
> + __field( __u64, previous_tsc_offset )
> + __field( __u64, next_tsc_offset )
> + ),
> +
> + TP_fast_assign(
> + __entry->vcpu_id = vcpu_id;
> + __entry->previous_tsc_offset = previous_tsc_offset;
> + __entry->next_tsc_offset = next_tsc_offset;
> + ),
> +
> + TP_printk("vcpu=%u prev=%llu next=%llu", __entry->vcpu_id,
> + __entry->previous_tsc_offset, __entry->next_tsc_offset)
> +);
> +
> #endif /* CONFIG_X86_64 */
>
> #endif /* _TRACE_KVM_H */
> diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
> index 25a791e..eb11856 100644
> --- a/arch/x86/kvm/vmx.c
> +++ b/arch/x86/kvm/vmx.c
> @@ -2096,6 +2096,8 @@ static void vmx_write_tsc_offset(struct kvm_vcpu *vcpu, u64 offset)
> (nested_cpu_has(vmcs12, CPU_BASED_USE_TSC_OFFSETING) ?
> vmcs12->tsc_offset : 0));
> } else {
> + trace_kvm_write_tsc_offset(vcpu->vcpu_id,
> + vmcs_read64(TSC_OFFSET), offset);
> vmcs_write64(TSC_OFFSET, offset);
> }
> }
> @@ -2103,11 +2105,14 @@ static void vmx_write_tsc_offset(struct kvm_vcpu *vcpu, u64 offset)
> static void vmx_adjust_tsc_offset(struct kvm_vcpu *vcpu, s64 adjustment, bool host)
> {
> u64 offset = vmcs_read64(TSC_OFFSET);
> +
> vmcs_write64(TSC_OFFSET, offset + adjustment);
> if (is_guest_mode(vcpu)) {
> /* Even when running L2, the adjustment needs to apply to L1 */
> to_vmx(vcpu)->nested.vmcs01_tsc_offset += adjustment;
> - }
> + } else
> + trace_kvm_write_tsc_offset(vcpu->vcpu_id, offset,
> + offset + adjustment);
> }
>
> static u64 vmx_compute_tsc_offset(struct kvm_vcpu *vcpu, u64 target_tsc)
> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> index 05a8b1a..c942a0c 100644
> --- a/arch/x86/kvm/x86.c
> +++ b/arch/x86/kvm/x86.c
> @@ -7264,3 +7264,4 @@ EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_nested_intr_vmexit);
> EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_invlpga);
> EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_skinit);
> EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_nested_intercepts);
> +EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_write_tsc_offset);
>
Looks good for 3.11.
Reviewed-by: Paolo Bonzini <[email protected]>
Fix build error caused by an undefinition of the kvm_write_tsc_offset
tracepoint for x86_32.
Since the tracepoint in trace.h was defined for CONFIG_X86_64, kernel build
on i386 was failed.
Signed-off-by: Yoshihiro YUNOMAE <[email protected]>
Cc: Joerg Roedel <[email protected]>
Cc: Gleb Natapov <[email protected]>
Cc: Paolo Bonzini <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: "H. Peter Anvin" <[email protected]>
---
arch/x86/kvm/trace.h | 42 +++++++++++++++++++++---------------------
1 file changed, 21 insertions(+), 21 deletions(-)
diff --git a/arch/x86/kvm/trace.h b/arch/x86/kvm/trace.h
index 6c82cf1..545245d 100644
--- a/arch/x86/kvm/trace.h
+++ b/arch/x86/kvm/trace.h
@@ -756,6 +756,27 @@ TRACE_EVENT(
__entry->gpa_match ? "GPA" : "GVA")
);
+TRACE_EVENT(kvm_write_tsc_offset,
+ TP_PROTO(unsigned int vcpu_id, __u64 previous_tsc_offset,
+ __u64 next_tsc_offset),
+ TP_ARGS(vcpu_id, previous_tsc_offset, next_tsc_offset),
+
+ TP_STRUCT__entry(
+ __field( unsigned int, vcpu_id )
+ __field( __u64, previous_tsc_offset )
+ __field( __u64, next_tsc_offset )
+ ),
+
+ TP_fast_assign(
+ __entry->vcpu_id = vcpu_id;
+ __entry->previous_tsc_offset = previous_tsc_offset;
+ __entry->next_tsc_offset = next_tsc_offset;
+ ),
+
+ TP_printk("vcpu=%u prev=%llu next=%llu", __entry->vcpu_id,
+ __entry->previous_tsc_offset, __entry->next_tsc_offset)
+);
+
#ifdef CONFIG_X86_64
#define host_clocks \
@@ -815,27 +836,6 @@ TRACE_EVENT(kvm_track_tsc,
__print_symbolic(__entry->host_clock, host_clocks))
);
-TRACE_EVENT(kvm_write_tsc_offset,
- TP_PROTO(unsigned int vcpu_id, __u64 previous_tsc_offset,
- __u64 next_tsc_offset),
- TP_ARGS(vcpu_id, previous_tsc_offset, next_tsc_offset),
-
- TP_STRUCT__entry(
- __field( unsigned int, vcpu_id )
- __field( __u64, previous_tsc_offset )
- __field( __u64, next_tsc_offset )
- ),
-
- TP_fast_assign(
- __entry->vcpu_id = vcpu_id;
- __entry->previous_tsc_offset = previous_tsc_offset;
- __entry->next_tsc_offset = next_tsc_offset;
- ),
-
- TP_printk("vcpu=%u prev=%llu next=%llu", __entry->vcpu_id,
- __entry->previous_tsc_offset, __entry->next_tsc_offset)
-);
-
#endif /* CONFIG_X86_64 */
#endif /* _TRACE_KVM_H */
Hi Gleb,
Thank you for applying my patch.
I received a mail of build error for i386, so I fixed the problem.
Since the tracepoint was defined only for x86_64, build was failed in
x86_32.
Would you apply this patch to kvm-tree?
Sorry for the inconvenience this may cause.
Thank you,
Yoshihiro YUNOMAE
(2013/06/25 19:16), Yoshihiro YUNOMAE wrote:
> Fix build error caused by an undefinition of the kvm_write_tsc_offset
> tracepoint for x86_32.
> Since the tracepoint in trace.h was defined for CONFIG_X86_64, kernel build
> on i386 was failed.
>
> Signed-off-by: Yoshihiro YUNOMAE <[email protected]>
> Cc: Joerg Roedel <[email protected]>
> Cc: Gleb Natapov <[email protected]>
> Cc: Paolo Bonzini <[email protected]>
> Cc: Thomas Gleixner <[email protected]>
> Cc: Ingo Molnar <[email protected]>
> Cc: "H. Peter Anvin" <[email protected]>
> ---
> arch/x86/kvm/trace.h | 42 +++++++++++++++++++++---------------------
> 1 file changed, 21 insertions(+), 21 deletions(-)
>
> diff --git a/arch/x86/kvm/trace.h b/arch/x86/kvm/trace.h
> index 6c82cf1..545245d 100644
> --- a/arch/x86/kvm/trace.h
> +++ b/arch/x86/kvm/trace.h
> @@ -756,6 +756,27 @@ TRACE_EVENT(
> __entry->gpa_match ? "GPA" : "GVA")
> );
>
> +TRACE_EVENT(kvm_write_tsc_offset,
> + TP_PROTO(unsigned int vcpu_id, __u64 previous_tsc_offset,
> + __u64 next_tsc_offset),
> + TP_ARGS(vcpu_id, previous_tsc_offset, next_tsc_offset),
> +
> + TP_STRUCT__entry(
> + __field( unsigned int, vcpu_id )
> + __field( __u64, previous_tsc_offset )
> + __field( __u64, next_tsc_offset )
> + ),
> +
> + TP_fast_assign(
> + __entry->vcpu_id = vcpu_id;
> + __entry->previous_tsc_offset = previous_tsc_offset;
> + __entry->next_tsc_offset = next_tsc_offset;
> + ),
> +
> + TP_printk("vcpu=%u prev=%llu next=%llu", __entry->vcpu_id,
> + __entry->previous_tsc_offset, __entry->next_tsc_offset)
> +);
> +
> #ifdef CONFIG_X86_64
>
> #define host_clocks \
> @@ -815,27 +836,6 @@ TRACE_EVENT(kvm_track_tsc,
> __print_symbolic(__entry->host_clock, host_clocks))
> );
>
> -TRACE_EVENT(kvm_write_tsc_offset,
> - TP_PROTO(unsigned int vcpu_id, __u64 previous_tsc_offset,
> - __u64 next_tsc_offset),
> - TP_ARGS(vcpu_id, previous_tsc_offset, next_tsc_offset),
> -
> - TP_STRUCT__entry(
> - __field( unsigned int, vcpu_id )
> - __field( __u64, previous_tsc_offset )
> - __field( __u64, next_tsc_offset )
> - ),
> -
> - TP_fast_assign(
> - __entry->vcpu_id = vcpu_id;
> - __entry->previous_tsc_offset = previous_tsc_offset;
> - __entry->next_tsc_offset = next_tsc_offset;
> - ),
> -
> - TP_printk("vcpu=%u prev=%llu next=%llu", __entry->vcpu_id,
> - __entry->previous_tsc_offset, __entry->next_tsc_offset)
> -);
> -
> #endif /* CONFIG_X86_64 */
>
> #endif /* _TRACE_KVM_H */
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
>
--
Yoshihiro YUNOMAE
Software Platform Research Dept. Linux Technology Center
Hitachi, Ltd., Yokohama Research Laboratory
E-mail: [email protected]
On Tue, Jun 25, 2013 at 07:18:35PM +0900, Yoshihiro YUNOMAE wrote:
> Hi Gleb,
>
> Thank you for applying my patch.
> I received a mail of build error for i386, so I fixed the problem.
> Since the tracepoint was defined only for x86_64, build was failed in
> x86_32.
>
> Would you apply this patch to kvm-tree?
> Sorry for the inconvenience this may cause.
>
No inconvenience. I amended your previous commit since it was still on
the branch that can be rebased.
> Thank you,
> Yoshihiro YUNOMAE
>
> (2013/06/25 19:16), Yoshihiro YUNOMAE wrote:
> >Fix build error caused by an undefinition of the kvm_write_tsc_offset
> >tracepoint for x86_32.
> >Since the tracepoint in trace.h was defined for CONFIG_X86_64, kernel build
> >on i386 was failed.
> >
> >Signed-off-by: Yoshihiro YUNOMAE <[email protected]>
> >Cc: Joerg Roedel <[email protected]>
> >Cc: Gleb Natapov <[email protected]>
> >Cc: Paolo Bonzini <[email protected]>
> >Cc: Thomas Gleixner <[email protected]>
> >Cc: Ingo Molnar <[email protected]>
> >Cc: "H. Peter Anvin" <[email protected]>
> >---
> > arch/x86/kvm/trace.h | 42 +++++++++++++++++++++---------------------
> > 1 file changed, 21 insertions(+), 21 deletions(-)
> >
> >diff --git a/arch/x86/kvm/trace.h b/arch/x86/kvm/trace.h
> >index 6c82cf1..545245d 100644
> >--- a/arch/x86/kvm/trace.h
> >+++ b/arch/x86/kvm/trace.h
> >@@ -756,6 +756,27 @@ TRACE_EVENT(
> > __entry->gpa_match ? "GPA" : "GVA")
> > );
> >
> >+TRACE_EVENT(kvm_write_tsc_offset,
> >+ TP_PROTO(unsigned int vcpu_id, __u64 previous_tsc_offset,
> >+ __u64 next_tsc_offset),
> >+ TP_ARGS(vcpu_id, previous_tsc_offset, next_tsc_offset),
> >+
> >+ TP_STRUCT__entry(
> >+ __field( unsigned int, vcpu_id )
> >+ __field( __u64, previous_tsc_offset )
> >+ __field( __u64, next_tsc_offset )
> >+ ),
> >+
> >+ TP_fast_assign(
> >+ __entry->vcpu_id = vcpu_id;
> >+ __entry->previous_tsc_offset = previous_tsc_offset;
> >+ __entry->next_tsc_offset = next_tsc_offset;
> >+ ),
> >+
> >+ TP_printk("vcpu=%u prev=%llu next=%llu", __entry->vcpu_id,
> >+ __entry->previous_tsc_offset, __entry->next_tsc_offset)
> >+);
> >+
> > #ifdef CONFIG_X86_64
> >
> > #define host_clocks \
> >@@ -815,27 +836,6 @@ TRACE_EVENT(kvm_track_tsc,
> > __print_symbolic(__entry->host_clock, host_clocks))
> > );
> >
> >-TRACE_EVENT(kvm_write_tsc_offset,
> >- TP_PROTO(unsigned int vcpu_id, __u64 previous_tsc_offset,
> >- __u64 next_tsc_offset),
> >- TP_ARGS(vcpu_id, previous_tsc_offset, next_tsc_offset),
> >-
> >- TP_STRUCT__entry(
> >- __field( unsigned int, vcpu_id )
> >- __field( __u64, previous_tsc_offset )
> >- __field( __u64, next_tsc_offset )
> >- ),
> >-
> >- TP_fast_assign(
> >- __entry->vcpu_id = vcpu_id;
> >- __entry->previous_tsc_offset = previous_tsc_offset;
> >- __entry->next_tsc_offset = next_tsc_offset;
> >- ),
> >-
> >- TP_printk("vcpu=%u prev=%llu next=%llu", __entry->vcpu_id,
> >- __entry->previous_tsc_offset, __entry->next_tsc_offset)
> >-);
> >-
> > #endif /* CONFIG_X86_64 */
> >
> > #endif /* _TRACE_KVM_H */
> >
> >--
> >To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> >the body of a message to [email protected]
> >More majordomo info at http://vger.kernel.org/majordomo-info.html
> >Please read the FAQ at http://www.tux.org/lkml/
> >
>
> --
> Yoshihiro YUNOMAE
> Software Platform Research Dept. Linux Technology Center
> Hitachi, Ltd., Yokohama Research Laboratory
> E-mail: [email protected]
>
--
Gleb.