Code Coverage with rr?
benjaminking at web.de
Sat Feb 2 10:28:53 UTC 2019
On Sat, Feb 02, 2019 at 10:29:18AM +1300, Robert O'Callahan wrote:
>On Sat, Feb 2, 2019 at 3:28 AM Benjamin King <benjaminking at web.de> wrote:
> would it be feasible to emit code coverage information based on a rr
> Yes and no.
Ok, thanks for your thoughts on this. I figured it was a bit easier than
singlestepping because rr already maintains a record of retired conditional
branches, but of course that is not everything there is to it.
My company is investigating options to assess code coverage and rr's low
overhead during record is attractive.
I'll check kcov, PIN and maybe Intels Processor Trace.
More information about the rr-dev