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 v4 00/13] branch tracing support for Atom


On Nov 27, 2012, at 6:59 PM, Jan Kratochvil <jan.kratochvil@redhat.com> wrote:

> On Tue, 27 Nov 2012 18:30:03 +0100, Metzger, Markus T wrote:
>> Btrace uses a ring buffer into which the h/w writes from-to records for each
>> branch. We will only see the tail of the execution trace.
> 
> OK, I see only 168 entries here.  Originally I expected kernel traps and
> records the buffer each time.  arch/x86/ even sets bts_interrupt_threshold,
> why is the recorded buffer so short?
> 
> With such short recorded buffer I understand the backtracing feature is not
> much useful.  Although I do not understand why only so few entries got
> recorded.

I hard-coded one page of trace (or, rather, perf_event) buffer per thread. We could make this configurable. I have not done this, yet, since I wanted to keep the patch series small and simple until I get feedback.

regards,
markus.


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