[patch v4 00/13] branch tracing support for Atom

[Jan Kratochvil]

On Tue, 27 Nov 2012 17:16:30 +0100, Metzger, Markus T wrote:
> Looks like it wouldn't be that difficult, after all. The assumption that
> a ret returns to the last call we've seen might be a bit optimistic, though.
> And we wouldn't be able to handle those cases where the call is not included
> in the trace. The back trace would thus be rather short.

In which case the call is not present in the btrace recoding?  We also have
the recording initial PC and we can also record the backtrace at that initial
PC.

There is wrong assumption if we record code from some later dlclose-d library
but that normally should not happen.


> Btrace is very helpful for debugging stack corruptions. It can show you
> where you came from even if gdb is not able to unwind properly.

True, I forgot about this common one.


Thanks,
Jan