This is the mail archive of the gdb-patches@sourceware.org mailing list for the GDB 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 2/2] new tracepoint downloaded MI notification.


> -----Original Message-----
> From: gdb-patches-owner@sourceware.org 
> [mailto:gdb-patches-owner@sourceware.org] On Behalf Of Pedro Alves
> Sent: Wednesday, October 31, 2012 2:03 PM
> To: dje@google.com
> Cc: Yao Qi; gdb-patches@sourceware.org
> Subject: Re: [PATCH 2/2] new tracepoint downloaded MI notification.
> 
> >  > 2012-09-27  Yao Qi  <yao@codesourcery.com>
> >  > 
> >  > 	* gdb.trace/mi-traceframe-changed.exp 
> (test_tfind_remote): Adjust.
> >  > 	* gdb.trace/mi-tracepoint-downloaded.exp: New.
> > 
> > Hi.
> > It would be useful if the reason why this notification 
> exists was specified in the code.
> > E.g, "This notification exists because frontends ... [fill 
> in the blank]."
> 
> Yes, indeed.  I'll probably upgrade that "useful" to "required".  :-)
> It is not clear for example, why would the frontend
> care about a particular tracepoint having been downloaded.  I can see
> it wanting to know when a trace run has started from the CLI, 
> for instance,
> which already implies that tracepoints have been downloaded.

In older GDB versions, when creating a tracepoint during a trace run,
that tracepoint would not be pushed to the target until the next
trace run.  The idea is that a frontend could indicate which tracespoints
were active on the target and which were not.

Now that GDB pushes new tracepoints to the target immediately, that
use-case may not apply, but I wonder if there are other situations
where some tracepoints will be on the target and other will not?

Marc


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