Re: instrumentation and kprobes really still "EXPERIMENTAL"?

Re: instrumentation and kprobes really still "EXPERIMENTAL"?

[Abhishek Sagar]

On 7/1/07, Robert P. J. Day <rpjday@mindspring.com> wrote:
> isn't kprobes mature enough to not be considered experimental anymore?

That would vary from arch to arch.

>   in addition, while most of the KPROBES config options depend on
>
>   KALLSYMS && EXPERIMENTAL && MODULES
>
> the s390 architecture depends only on
>
>   EXPERIMENTAL && MODULES
>
> and its instrumentation support is *not* listed as experimental.
>
>   also, the avr32 entry is in the file Kconfig.debug, and depends only
> on DEBUG_KERNEL.  just an observation.

This probably stems from the episodic growth kprobes across different
archs. It can be cleaned up I suppose.

--
Abhishek Sagar

P.S: Adding systemtap in CC.
-
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

Re: instrumentation and kprobes really still "EXPERIMENTAL"?

[Robert P. J. Day]

On Mon, 2 Jul 2007, Abhishek Sagar wrote:

> On 7/1/07, Robert P. J. Day <rpjday@mindspring.com> wrote:
> > isn't kprobes mature enough to not be considered experimental anymore?
>
> That would vary from arch to arch.

fair enough.  however, at the very least, i'm thinking that the entire
"Instrumentation support" submenu can be made non-EXPERIMENTAL, while
individual entries therein can be EXPERIMENTAL on a choice-by-choice
basis or something like that.

anyway, others are welcome to play with that as they see fit.

rday
-- 
========================================================================
Robert P. J. Day
Linux Consulting, Training and Annoying Kernel Pedantry
Waterloo, Ontario, CANADA

http://fsdev.net/wiki/index.php?title=Main_Page
========================================================================

Re: instrumentation and kprobes really still "EXPERIMENTAL"?

[Abhishek Sagar]

On 7/2/07, Robert P. J. Day <rpjday@mindspring.com> wrote:
> On Mon, 2 Jul 2007, Abhishek Sagar wrote:
>
> > On 7/1/07, Robert P. J. Day <rpjday@mindspring.com> wrote:
> > > isn't kprobes mature enough to not be considered experimental anymore?
> >
> > That would vary from arch to arch.
>
> fair enough.  however, at the very least, i'm thinking that the entire
> "Instrumentation support" submenu can be made non-EXPERIMENTAL, while
> individual entries therein can be EXPERIMENTAL on a choice-by-choice
> basis or something like that.

There's not a lot in that menu. It holds oprofile on every arch I
grepped on. Oprofile might come across as being marked
non-experimental but that may very well be due to the assumption that
the menu that holds that option will enforce the EXPERIMENTAL check.
How many archs do you see where oprofile and kprobe defer on their
dependency on EXPERIMENTAL?

Removing the EXPERIMENTAL check from the instrumentation menu will
also require you to fix/double-check dependency on EXPERIMENTAL of
oprofile/kprobes, provided you know that for sure. Which would bring
you back to the same question you began with, namely, whether option
xyz is EXPERIMENTAL or not.

--
Abhishek Sagar
-
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

Re: instrumentation and kprobes really still "EXPERIMENTAL"?

[Robert P. J. Day]

On Mon, 2 Jul 2007, Abhishek Sagar wrote:

> On 7/2/07, Robert P. J. Day <rpjday@mindspring.com> wrote:
> > On Mon, 2 Jul 2007, Abhishek Sagar wrote:
> >
> > > On 7/1/07, Robert P. J. Day <rpjday@mindspring.com> wrote:
> > > > isn't kprobes mature enough to not be considered experimental anymore?
> > >
> > > That would vary from arch to arch.
> >
> > fair enough.  however, at the very least, i'm thinking that the entire
> > "Instrumentation support" submenu can be made non-EXPERIMENTAL, while
> > individual entries therein can be EXPERIMENTAL on a choice-by-choice
> > basis or something like that.
>
> There's not a lot in that menu. It holds oprofile on every arch I
> grepped on. Oprofile might come across as being marked
> non-experimental but that may very well be due to the assumption
> that the menu that holds that option will enforce the EXPERIMENTAL
> check.

consider this random Kconfig file arch/x86_64/oprofile/Kconfig:

=============================
config PROFILING
        bool "Profiling support (EXPERIMENTAL)"
        help
          Say Y here to enable the extended profiling support mechanisms used
          by profilers such as OProfile.


config OPROFILE
        tristate "OProfile system profiling (EXPERIMENTAL)"
        depends on PROFILING
        help
          OProfile is a profiling system capable of profiling the
          whole system, include the kernel, kernel modules, libraries,
          and applications.

          If unsure, say N.
==============================

  the above is a bit silly.  note that both prompts advertise
themselves as "EXPERIMENTAL" even though neither of them has such a
dependency.  they *are*, however, part of an submenu that *is*
dependent on EXPERIMENTAL (although you'd never know that just by
looking at that Kconfig file).  in short, it's just kind of ugly
hackery at the moment.

  i made a suggestion a while back about having maturity levels like
"EXPERIMENTAL" **automatically** added to the displayed prompts when
you're doing a config -- that would go a long way to avoiding all this
manual adding of a maturity level to Kconfig prompts, but that's a
non-trivial change and is something the Kconfig folks would have to
handle.

  my original point was simply that, based on its acceptance, it would
seem kprobes has progressed beyond the EXPERIMENTAL phase, that's all.

rday

-- 
========================================================================
Robert P. J. Day
Linux Consulting, Training and Annoying Kernel Pedantry
Waterloo, Ontario, CANADA

http://fsdev.net/wiki/index.php?title=Main_Page
========================================================================

Re: instrumentation and kprobes really still "EXPERIMENTAL"?

[Abhishek Sagar]

On 7/2/07, Robert P. J. Day <rpjday@mindspring.com> wrote:
> consider this random Kconfig file arch/x86_64/oprofile/Kconfig:
>
> =============================
> config PROFILING
>         bool "Profiling support (EXPERIMENTAL)"
>         help
>           Say Y here to enable the extended profiling support mechanisms used
>           by profilers such as OProfile.
>
>
> config OPROFILE
>         tristate "OProfile system profiling (EXPERIMENTAL)"
>         depends on PROFILING
>         help
>           OProfile is a profiling system capable of profiling the
>           whole system, include the kernel, kernel modules, libraries,
>           and applications.
>
>           If unsure, say N.
> ==============================
>
>   the above is a bit silly.  note that both prompts advertise
> themselves as "EXPERIMENTAL" even though neither of them has such a
> dependency.  they *are*, however, part of an submenu that *is*
> dependent on EXPERIMENTAL (although you'd never know that just by
> looking at that Kconfig file).  in short, it's just kind of ugly
> hackery at the moment.

Yes but pushing the EXPERIMENTAL check to individual menu members
shouldn't be done in a way which leads to an empty menu for any kernel
configuration. I suspect that this is why the EXPERIMENTAL check is on
the instrumentation menu in the first place.

>   my original point was simply that, based on its acceptance, it would
> seem kprobes has progressed beyond the EXPERIMENTAL phase, that's all.

It's worth a look on some archs.

--
Abhishek Sagar
-
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/