On Mon, 2014-07-28 at 20:50 +0200, Peter Zijlstra wrote: > On Mon, Jul 28, 2014 at 10:49:35AM -0700, Jason Low wrote: > > > So if the function is inlined, perf report would show the contention > > occurring in the calling function rather than the inlined function. As > > an example, if we were to convert osq_lock() and mutex_spin_on_owner() > > to inline, perf would report all the contention from both of those > > functions occurring in just "__mutex_lock_slowpath()". > > But that's easily fixed on the tool side, right? Dwarf info (as present > in vmlinux when build with CONFIG_DEBUG_INFO) has all that.
Ok, one more reason to keep it as inline then :) -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
