Skip to contentSkip to conversation

LTTng 0.87 improved page fault tracing

DISCUSSION Mathieu Desnoyers 2 participants 4 messages Jan 27, 2009 – Jan 29, 2009

lore
Share & tools
lore
#0 lore raw

Hi,

I just combined the 4 page fault handler events that were in the tracing hot path of LTTng into 2 :

kernel page_fault_entry
kernel page_fault_exit

They take as parameter the combination of what was available in the trap_entry/exit events and handle_mm_fault entry/exit events. This should lessen the performance impact of the tracer when it's active. I did the related modifications in LTTV 0.12.8.

Mathieu

Signature · 2 lines
Mathieu Desnoyers
OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F  BA06 3F25 A8FE 3BAE 9A68
#1 lore raw

Hi

interesting..

Quoted 10 lines

I just combined the 4 page fault handler events that were in the tracing hot path of LTTng into 2 :

kernel page_fault_entry
kernel page_fault_exit

They take as parameter the combination of what was available in the trap_entry/exit events and handle_mm_fault entry/exit events. This should lessen the performance impact of the tracer when it's active. I did the related modifications in LTTV 0.12.8.

Just question.

As far as I know, customer has two different requeremtn of the page fault.
(1) collect number of all page fault

      -> if it is too large, too many interrupt decrease performance.

(2) collect number of major page fault
-> major page fault indicate to increase random access I/O,
then, some customer want to collect major page fault (don't include minor page fault)

Is this patch fill (2) requirement?

#2 lore raw
Quoted 10 lines · * KOSAKI Motohiro ([email protected]) wrote:

Hi

interesting..

Quoted 6 lines

I just combined the 4 page fault handler events that were in the tracing hot path of LTTng into 2 :

kernel page_fault_entry
kernel page_fault_exit

Quoted 8 lines
Quoted 4 lines

They take as parameter the combination of what was available in the trap_entry/exit events and handle_mm_fault entry/exit events. This should lessen the performance impact of the tracer when it's active. I did the related modifications in LTTV 0.12.8.

Just question.

As far as I know, customer has two different requeremtn of the page fault.

    1 a)
Quoted 2 lines

(1) collect number of all page fault

      -> if it is too large, too many interrupt decrease performance.

(a single event is required for this)

1 b) the user may also want to know the time spent in the page fault handler to service those faults, therefore involving page fault entry and exit events.

Quoted 4 lines

(2) collect number of major page fault
-> major page fault indicate to increase random access I/O,
then, some customer want to collect major page fault (don't include minor page fault)

Yes, the trace_page_fault_exit takes the "fault" parameter returned by handle_mm_fault (which is recorded to the trace as the "res" event field). Using

res & VM_FAULT_MAJOR

will give only the major page faults. Note that some knowledge of the bitmask is required to interpret the "res" bitfield. This could be done by a specific analysis module. I would ideally like to create a LTTng module to export tables including those bitfields so we can keep the bitfield interpretation in sync with the kernel code changes more or less automatically.

Quoted 3 lines

Is this patch fill (2) requirement?

Yes.

Mathieu

Quoted 2 lines
Signature · 2 lines
Mathieu Desnoyers
OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F  BA06 3F25 A8FE 3BAE 9A68
#3 lore raw
Quoted 20 lines · * KOSAKI Motohiro ([email protected]) wrote:
Quoted 10 lines

Hi

interesting..

Quoted 6 lines

I just combined the 4 page fault handler events that were in the tracing hot path of LTTng into 2 :

kernel page_fault_entry
kernel page_fault_exit

Quoted 8 lines
Quoted 4 lines

They take as parameter the combination of what was available in the trap_entry/exit events and handle_mm_fault entry/exit events. This should lessen the performance impact of the tracer when it's active. I did the related modifications in LTTV 0.12.8.

Just question.

As far as I know, customer has two different requeremtn of the page fault.

Quoted 8 lines
    1 a)
Quoted 2 lines

(1) collect number of all page fault

      -> if it is too large, too many interrupt decrease performance.

(a single event is required for this)

1 b) the user may also want to know the time spent in the page fault handler to service those faults, therefore involving page fault entry and exit events.

Indeed.
that's great.

Quoted 9 lines
Quoted 4 lines

(2) collect number of major page fault
-> major page fault indicate to increase random access I/O,
then, some customer want to collect major page fault (don't include minor page fault)

Yes, the trace_page_fault_exit takes the "fault" parameter returned by handle_mm_fault (which is recorded to the trace as the "res" event field). Using

Quoted 8 lines

res & VM_FAULT_MAJOR

will give only the major page faults. Note that some knowledge of the bitmask is required to interpret the "res" bitfield. This could be done by a specific analysis module. I would ideally like to create a LTTng module to export tables including those bitfields so we can keep the bitfield interpretation in sync with the kernel code changes more or less automatically.

Agreed.
thanks for great improvement.

Quoted 15 lines
Quoted 4 lines

Is this patch fill (2) requirement?

Yes.

Mathieu

Quoted 2 lines
Signature · 2 lines
Mathieu Desnoyers
OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F  BA06 3F25 A8FE 3BAE 9A68