Mirror of the gdb mailing list
 help / color / mirror / Atom feed
* Tracing all 'call' instructions in a gdb session
@ 2005-06-27 22:01 Thomas Glanzmann
  2005-06-27 22:03 ` Daniel Jacobowitz
  0 siblings, 1 reply; 4+ messages in thread
From: Thomas Glanzmann @ 2005-06-27 22:01 UTC (permalink / raw)
  To: gdb

Hello,
I like to trace every 'call instruction' in a executable. How do I
obtain this with gdb or are there other tools to obtain that?

Like tracepoint every call instruction and gather eip and 'jump to'
address?

	Thomas


^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: Tracing all 'call' instructions in a gdb session
  2005-06-27 22:01 Tracing all 'call' instructions in a gdb session Thomas Glanzmann
@ 2005-06-27 22:03 ` Daniel Jacobowitz
  2005-06-27 22:10   ` Thomas Glanzmann
  0 siblings, 1 reply; 4+ messages in thread
From: Daniel Jacobowitz @ 2005-06-27 22:03 UTC (permalink / raw)
  To: Thomas Glanzmann, gdb

On Tue, Jun 28, 2005 at 12:01:22AM +0200, Thomas Glanzmann wrote:
> Hello,
> I like to trace every 'call instruction' in a executable. How do I
> obtain this with gdb or are there other tools to obtain that?
> 
> Like tracepoint every call instruction and gather eip and 'jump to'
> address?

Sorry, GDB can't help you with this.  Not today, anyway.  You'd need
both tracepoints and a more flexible facility for setting breakpoints
automatically.

A profiler might be a better tool.  I bet you can get valgrind to do
this somehow.

-- 
Daniel Jacobowitz
CodeSourcery, LLC


^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: Tracing all 'call' instructions in a gdb session
  2005-06-27 22:03 ` Daniel Jacobowitz
@ 2005-06-27 22:10   ` Thomas Glanzmann
  2005-06-27 22:32     ` Daniel Jacobowitz
  0 siblings, 1 reply; 4+ messages in thread
From: Thomas Glanzmann @ 2005-06-27 22:10 UTC (permalink / raw)
  To: gdb

Hello,

> Sorry, GDB can't help you with this.  Not today, anyway.  You'd need
> both tracepoints and a more flexible facility for setting breakpoints
> automatically.

What about using objdump and grep setting a few hunderd tracepoints?
Would that work?

> A profiler might be a better tool.  I bet you can get valgrind to do
> this somehow.

I thought exactly the same. I am going to ask this on the valgrind ml.
;-) But it shouldn't be a problem if valgrind simulates the call
instruction.

	Thomas


^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: Tracing all 'call' instructions in a gdb session
  2005-06-27 22:10   ` Thomas Glanzmann
@ 2005-06-27 22:32     ` Daniel Jacobowitz
  0 siblings, 0 replies; 4+ messages in thread
From: Daniel Jacobowitz @ 2005-06-27 22:32 UTC (permalink / raw)
  To: Thomas Glanzmann, gdb

On Tue, Jun 28, 2005 at 12:10:10AM +0200, Thomas Glanzmann wrote:
> Hello,
> 
> > Sorry, GDB can't help you with this.  Not today, anyway.  You'd need
> > both tracepoints and a more flexible facility for setting breakpoints
> > automatically.
> 
> What about using objdump and grep setting a few hunderd tracepoints?
> Would that work?

Except native GDB doesn't support tracepoints, and I don't know how
it's going to cope with a breakpoint list many thousands long.

-- 
Daniel Jacobowitz
CodeSourcery, LLC


^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2005-06-27 22:32 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2005-06-27 22:01 Tracing all 'call' instructions in a gdb session Thomas Glanzmann
2005-06-27 22:03 ` Daniel Jacobowitz
2005-06-27 22:10   ` Thomas Glanzmann
2005-06-27 22:32     ` Daniel Jacobowitz

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox