Messages in this thread | | | Date | Sat, 9 Nov 2013 17:01:20 +0100 | From | Peter Zijlstra <> | Subject | Re: [Patch v5 0/9] liblockdep: userspace lockdep |
| |
On Fri, Nov 08, 2013 at 12:04:18PM -0500, Sasha Levin wrote:
> > > >* Sasha Levin <sasha.levin@oracle.com> wrote: > > > >>On 07/08/2013 04:39 AM, Ingo Molnar wrote: > >>> > >>>* Sasha Levin <sasha.levin@oracle.com> wrote: > >>> > >>>>>>The other issue is that with lock classes disabled you have to hit an > >>>>>>actual deadlock to trigger any output. > >>>>>> > >>>>>>I.e. much of the power of lockdep is diminished :-/ When actual > >>>>>>deadlocks are triggered then it's not particularly complex to debug > >>>>>>user-space apps: gdb the hung task(s) and look at the backtraces. > >>>>> > >>>>>Lock classes are disabled only if you're using the LD_PRELOAD method > >>>>>of testing. If you actually re-compile your code with the library (by > >>>>>just including the header and setting a #define to enable it) you will > >>>>>have lock classes. > >>>> > >>>>Hi Ingo, > >>>> > >>>>Just wondering if you're planning on pushing it over to Linus from your > >>>>tree, or should I go ahead and do it on my own? > >>> > >>>PeterZ is in favor as well so I'll apply them after the merge window, for > >>>v3.12. > >> > >>Hi Ingo, > >> > >>Do you intend to send liblockdep in this merge window as planned? > > > >If Peter agrees with them and picks them up then the next merge window > >would be fine I guess. > > Ping? Anyone?
Urgh, I missed Ingo throwing the ball back in my court.. sorry about that. I'll have a look on Monday.
| |