>> My profiling is 100% realiable.
>
>I would not say 100% reliable, but more reliable - there are two issues with
>-pg profiling: one is cache footprint of mcount, the other is that gcc at
>the moment calls mcount only on function entry and not exit, so that means
The profiling I implemented tell you _exactly_ how many times every not
inlined function is been recalled. This has nothing to do with the time
that the kernel has spent in such function (and as I just said probably is
not the kind of profiling needed to get numbers of different vma structure
implementation).
Andrea[s] Arcangeli
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/faq.html