This is the mail archive of the systemtap@sourceware.org mailing list for the systemtap project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH v2 2.6.38-rc8-tip 0/20] 0: Inode based uprobes


On Tue, 2011-03-15 at 21:09 +0100, Thomas Gleixner wrote:
> On Tue, 15 Mar 2011, Steven Rostedt wrote:

> I didn't say that ptrace rocks.
> 
> All I'm saying is that we want a better argument than a single user
> which is - and yes i tried it more than once - assbackwards beyond all
> imagination.
> 
> If gdb, perf, trace can and will make use of it then we have sensible

I'm more interested in the perf/trace than gdb, as the way gdb is mostly
used (at least now) to debug problems in the code with a big hammer
(single step, look at registers/variables). That is, gdb is usually very
interactive and its best to "stop the code" from running to examine what
has happened. gdb is not something you will run on an application that
is being used by others.

With perf/trace things are different, as you want the task to be as
little affected by the tracer as it runs, perhaps even in a production
environment. This is a completely different paradigm.

If gdb uses it, great, but I don't think we should bend over backwards
to make it usable by gdb. Debugging and tracing are different, with
different requirements and needs.

> arguments enough to go there. If systemtap can use it as well then I
> have no problem with that..

Fair enough.

-- Steve



Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]