Porting "jump labels" to userspace (was: Re: [ltt-dev] LTTng-UST vs SystemTap userspace tracing benchmarks)

Porting "jump labels" to userspace (was: Re: [ltt-dev] LTTng-UST vs SystemTap userspace tracing benchmarks)

[Mathieu Desnoyers]

[ Adding the relevant people in CC for jump label discussion ]

* Roland McGrath (roland@redhat.com) wrote:
> Stefan was referring to #4 in your taxonomy.
> 
> It's indeed the case that what UST uses today is an always-there normal
> C code sequence that loads global variables to decide whether to make
> indirect function calls.  I don't recall off hand how many layers of
> function calls to the libust DSO and such there are in either the
> disabled or enabled cases.  At best, there is the always the overhead of
> several instructions and at least one load in the hot code path, and the
> i-cache pollution that goes with that.
> 
> It's indeed the cast that what Systemtap uses today is a
> sometimes-inserted normal breakpoint instruction, which is indeed a
> software interrupt that requires kernel mediation.  When disabled, there
> is as close to zero overhead as you can have, being a tiny placeholder
> instruction sequence (currently just one nop), so the runtime overhead
> is under a cycle and the i-cache pollution is the smallest possible unit
> (one instruction, being just one byte on x86).
> 
> The "sweet spot" between the two is to have overhead close to
> Systemtap's epsilon for a disabled probe, while having overhead close to
> UST's pure-user method when a probe is enabled.  In the in-kernel
> context, this is what the Linux kernel's latest code (still being hashed
> out, but mostly done) has for kernel tracepoints using the so-called
> "jump label" method.  That is also possible for sdt markers with some
> careful consideration and attention to machine-specific details for each
> machine architecture of concern.  It entails making the placeholder in
> the hot code path slightly larger (at least for x86, it has to be a
> "long nop", being probably neglibly more runtime overhead, and a few
> bytes more i-cache pollution), and adding some additional static code
> outside the hot path.  The work to enable or disable a probe becomes
> just as costly as the current Systemtap method, since it involves
> modifying the program text in place (inserting jump instructions rather
> than breakpoint ones).  Once enabled, the runtime work of the probes
> firing can be very much like what UST does today.

Actually, creating the equivalent of the jump labels for userspace is on our
list of "things to do" for the UST project.

If Jason and other jump label contributors agree to dual-license the jump label
generic and per-arch implementations under both GPLv2 and LGPLv2.1, we could
probably re-use a large part of this code to create a static jump label library
for userspace. The concerned authors are precisely: Jason for generic and x86
code, Steven for some improvements in generic code and David S. Miller for the
sparc port. After things settle down with the current jump label changes
proposed by Peter Zijlstra, we might also want to ask for his permission to
LGPLize his changes.

Thanks,

Mathieu

> 
> 
> Thanks,
> Roland

-- 
Mathieu Desnoyers
Operating System Efficiency R&D Consultant
EfficiOS Inc.
http://www.efficios.com

Re: Porting "jump labels" to userspace (was: Re: [ltt-dev] LTTng-UST vs SystemTap userspace tracing benchmarks)

[Roland McGrath]

IMHO there is not really so much to the in-kernel implementation that it's
worth attempting to reuse the code in userland.  Pretty much all the work
is in the details of the implementation that would naturally differ a lot
in a different context.  If you understand the mechanism and the machine
details, then implementing it well for a userland context is not a big deal
and is cleaner to do from scratch than shoe-horning kernel-centric code
into a wildly different context.


Thanks,
Roland

Re: Porting "jump labels" to userspace

[David Daney]

On 02/16/2011 12:04 PM, Roland McGrath wrote:
> IMHO there is not really so much to the in-kernel implementation that it's
> worth attempting to reuse the code in userland.  Pretty much all the work
> is in the details of the implementation that would naturally differ a lot
> in a different context.  If you understand the mechanism and the machine
> details, then implementing it well for a userland context is not a big deal
> and is cleaner to do from scratch than shoe-horning kernel-centric code
> into a wildly different context.
>

Good point.

Certainly the details of maintaining instruction cache coherency may be 
different in userspace.

David Daney

Re: Porting "jump labels" to userspace

[Mathieu Desnoyers]

* David Daney (ddaney@caviumnetworks.com) wrote:
> On 02/16/2011 12:04 PM, Roland McGrath wrote:
>> IMHO there is not really so much to the in-kernel implementation that it's
>> worth attempting to reuse the code in userland.  Pretty much all the work
>> is in the details of the implementation that would naturally differ a lot
>> in a different context.  If you understand the mechanism and the machine
>> details, then implementing it well for a userland context is not a big deal
>> and is cleaner to do from scratch than shoe-horning kernel-centric code
>> into a wildly different context.
>>
>
> Good point.
>
> Certainly the details of maintaining instruction cache coherency may be  
> different in userspace.

Indeed, the arch-specific parts will need some extra care (which might, in the
worse case scenario, require to suspend a whole process during the update), but
the generic code in jump_label.c and jump_label.h could certainly be reused.

Thanks,

Mathieu

>
> David Daney

-- 
Mathieu Desnoyers
Operating System Efficiency R&D Consultant
EfficiOS Inc.
http://www.efficios.com

Re: Porting "jump labels" to userspace

[Thomas Gleixner]

On Wed, 16 Feb 2011, Mathieu Desnoyers wrote:

> * David Daney (ddaney@caviumnetworks.com) wrote:
> > On 02/16/2011 12:04 PM, Roland McGrath wrote:
> >> IMHO there is not really so much to the in-kernel implementation that it's
> >> worth attempting to reuse the code in userland.  Pretty much all the work
> >> is in the details of the implementation that would naturally differ a lot
> >> in a different context.  If you understand the mechanism and the machine
> >> details, then implementing it well for a userland context is not a big deal
> >> and is cleaner to do from scratch than shoe-horning kernel-centric code
> >> into a wildly different context.
> >>
> >
> > Good point.
> >
> > Certainly the details of maintaining instruction cache coherency may be  
> > different in userspace.
> 
> Indeed, the arch-specific parts will need some extra care (which might, in the
> worse case scenario, require to suspend a whole process during the update), but
> the generic code in jump_label.c and jump_label.h could certainly be reused.

We talk about 500 lines of code, where half of it is modules specific
and the whole thing is full of kernelims. IMNSHO, that's faster
reimplemented from scratch than writing all the mails and get the
authors to sign off on the license change.

Thanks,

	tglx

Re: Porting "jump labels" to userspace

[Peter Zijlstra]

On Wed, 2011-02-16 at 21:45 +0100, Thomas Gleixner wrote:

> We talk about 500 lines of code, where half of it is modules specific
> and the whole thing is full of kernelims. IMNSHO, that's faster
> reimplemented from scratch than writing all the mails and get the
> authors to sign off on the license change.

That and I'm not going to consent with an LGPL license, I'm an avid
GPLv2 fan as various people already know ;-)

Re: Porting "jump labels" to userspace

[Ingo Molnar]

* Peter Zijlstra <peterz@infradead.org> wrote:

> On Wed, 2011-02-16 at 21:45 +0100, Thomas Gleixner wrote:
> 
> > We talk about 500 lines of code, where half of it is modules specific
> > and the whole thing is full of kernelims. IMNSHO, that's faster
> > reimplemented from scratch than writing all the mails and get the
> > authors to sign off on the license change.
> 
> That and I'm not going to consent with an LGPL license, I'm an avid
> GPLv2 fan as various people already know ;-)

Ditto for me.

Thanks,

	Ingo

Re: Porting "jump labels" to userspace

[David Daney]

On 02/16/2011 12:00 PM, Mathieu Desnoyers wrote:
[...]>
> Actually, creating the equivalent of the jump labels for userspace is on our
> list of "things to do" for the UST project.
>
> If Jason and other jump label contributors agree to dual-license the jump label
> generic and per-arch implementations under both GPLv2 and LGPLv2.1, we could
> probably re-use a large part of this code to create a static jump label library
> for userspace. The concerned authors are precisely: Jason for generic and x86
> code, Steven for some improvements in generic code and David S. Miller for the
> sparc port. After things settle down with the current jump label changes
> proposed by Peter Zijlstra, we might also want to ask for his permission to
> LGPLize his changes.
>
FWIW:  MIPS jump label support is now in as of:

94bb0c1 (MIPS: jump label: Add MIPS support.)

I would be the author of that.

David Daney

Re: Porting "jump labels" to userspace

[Mathieu Desnoyers]

* David Daney (ddaney@caviumnetworks.com) wrote:
> On 02/16/2011 12:00 PM, Mathieu Desnoyers wrote:
> [...]>
>> Actually, creating the equivalent of the jump labels for userspace is on our
>> list of "things to do" for the UST project.
>>
>> If Jason and other jump label contributors agree to dual-license the jump label
>> generic and per-arch implementations under both GPLv2 and LGPLv2.1, we could
>> probably re-use a large part of this code to create a static jump label library
>> for userspace. The concerned authors are precisely: Jason for generic and x86
>> code, Steven for some improvements in generic code and David S. Miller for the
>> sparc port. After things settle down with the current jump label changes
>> proposed by Peter Zijlstra, we might also want to ask for his permission to
>> LGPLize his changes.
>>
> FWIW:  MIPS jump label support is now in as of:
>
> 94bb0c1 (MIPS: jump label: Add MIPS support.)
>
> I would be the author of that.

Sorry David, my tree was a bit outdated from Linus's head :)

Mathieu

>
> David Daney

-- 
Mathieu Desnoyers
Operating System Efficiency R&D Consultant
EfficiOS Inc.
http://www.efficios.com