On 11/10/17 at 08:25P, Ingo Molnar wrote: > > * Rafael J. Wysocki <rafael.j.wyso...@intel.com> wrote: > > > Hi Linus, > > > > On 11/9/2017 11:38 AM, WANG Chao wrote: > > > Commit 941f5f0f6ef5 (x86: CPU: Fix up "cpu MHz" in /proc/cpuinfo) caused > > > a serious performance issue when reading from /proc/cpuinfo on system > > > with aperfmperf. > > > > > > For each cpu, arch_freq_get_on_cpu() sleeps 20ms to get its frequency. > > > On a system with 64 cpus, it takes 1.5s to finish running `cat > > > /proc/cpuinfo`, while it previously was done in 15ms. > > > > Honestly, I'm not sure what to do to address this ATM. > > > > The last requested frequency is only available in the non-HWP case, so it > > cannot be used universally. > > This is a serious regression that needs to be fixed ASAP, because the > slowdown is > utterly ridiculous on a 120 CPU system: > > fomalhaut:~> time cat /proc/cpuinfo >/dev/null > > real 0m2.689s > user 0m0.001s > sys 0m0.007s
Because 4.14 is about to release, how about adding this patch to 4.14 now? We can work on a more sophisticated solution later. Thanks, WANG Chao
