Messages in this thread | | | Date | Mon, 28 Jul 2014 10:27:39 -0700 | From | Alexei Starovoitov <> | Subject | Re: Random panic in load_balance() with 3.16-rc |
| |
On Mon, Jul 28, 2014 at 09:45:45AM -0700, Linus Torvalds wrote: > On Mon, Jul 28, 2014 at 5:26 AM, Frank Ch. Eigler <fche@redhat.com> wrote: > > > > Please note that the data produced by "-g -fvar-tracking" is consumed > > by tools like systemtap, perf, crash, and makes a significant > > difference to the observability of debug AND non-debug kernels. > > Yeah, and compared to having a buggy kernel, I care exactly this much: "". > > Besides, "significant difference" is very debatable indeed. It may be > noticeable, it's likely not actually significant.
as far as I can see in gcc code, -fno-var-tracking-assignments is the main switch for 'vartrack' pass. So -fno-var-tracking-assignments is pretty much equivalent to -fno-var-tracking Both kill majority of debug info for local variables and function arguments. The end effect for perf/systemtap will be quite noticeable.
It's not pretty, but adding it unconditionally was the right thing to do. Black listing compiler versions is too fragile. Look at the flip side: now size of build dir will be much smaller :)
For gcc it's obviously a big blow. Now 10k lines pass developed over 10+ years will not be used :(
| |