Messages in this thread | | | Date | Wed, 19 Aug 2020 11:04:02 +0800 | From | Feng Tang <> | Subject | Re: [LKP] Re: [x86/mce] 1de08dccd3: will-it-scale.per_process_ops -14.1% regression |
| |
On Wed, Aug 19, 2020 at 10:23:11AM +0800, Luck, Tony wrote: > 0000000000019260 D pqr_state > > Do you have /sys/fs/resctrl mounted? This variable is read on every context switch. > If your benchmark does a lot of context switching and this now shares a cache line > with something different (especially something that is sometimes modified from another > CPU) that could cause some cache line bouncing.
No, the kernel doesn't has /sys/fs/resctrl mounted.
And the test case is not context switch intensive. It just forks a task for each CPU which loops running malloc/free and counts the number of complted operations for 300 seconds.
Thanks, Feng
| |