Resending beause it bounced of linux-kernel (google inbox sends everything as html) !
Thanks for your quick response, I'll elaborate on the Changelog. Regarding the global spinlock I considered adding it to task_struct and signal_struct. My reasoning not to do it, flawed or not, was that I thought the risk for congestion and cache line bouncing would be small given the following assumptions: 1. As far as I understand neither of the callers are typically called very frequently. (procfs, k_getrusage, wait_task_zombie and sys_times) 2 The and the time spent in the lock region is small. Did I have bad luck when thinking :) or do you still think it's better to add the locks to the structs above ? /Fredrik On Fri, Jun 12, 2015 at 12:16 PM, Peter Zijlstra <pet...@infradead.org> wrote: > On Fri, 2015-06-12 at 10:55 +0200, Fredrik Markstrom wrote: >> The scaling mechanism might sometimes cause top to report >100% >> (sometimes > 1000%) cpu usage for a single thread. This patch makes >> sure that stime+utime corresponds to the actual runtime of the thread. > > This Changelog is inadequate, it does not explain the actual problem. > >> +static DEFINE_SPINLOCK(prev_time_lock); > > global (spin)locks are bad. -- /Fredrik -- 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/
