On 27 January 2014 20:46, Sriram V <vshrirama@xxxxxxxxx> wrote: > Looping in linux-pm and viresh to see if they can advice. Sorry for delay, as I was away on vacation. > Hi, > > am trying to debug an random issue in which __list_del gets a prev > pointer which is LIST_POISON, But the next pointer contains a valid > value. > There seems to be a double deletion. Since the next pointer is valid - > It determines that the timer is pending and tries to delete it > resulting in a crash. > > I am trying to understand when this could happen from the cpufreq governor. > > There are a couple of new patches recently applied in the kernel that > deal with timer corruption. Not sure if they take care of this. I need more information on this as I haven't experienced it. - Which kernel you tested this on? - Can you try on latest one? i.e. 3.13? - How do you reproduce it? Guaranteed steps? -- viresh
