On Thu, Apr 25, 2019 at 03:27:12PM +0300, Konstantin Khlebnikov wrote: > On 25.04.2019 14:45, Peter Zijlstra wrote: > > On Thu, Apr 25, 2019 at 01:46:51PM +0300, Konstantin Khlebnikov wrote: > > > Sensitive workloads like DPDK polling do not like any interrupts. > > > > > > This patch adds flag 'freq' for boot option 'isolcpu': > > > isolcpu=freq,<cpus>. > > > > > > Users like show_cpuinfo() fallback to frequency from generic cpufreq > > > policy if arch-specific requesting method returns 0. > > > > Then don't touch that sysfs file? > > What sysfs file?
/sys/devices/system/cpu/cpu*/cpufreq/scaling_cur_freq > I mean many applications read /proc/cpuinfo and this triggers IPI to all cpus. Argh, I missed it was wired into that too :/ > > I really hate having an option for just this. > > > > This could be generalized to 'isolate cpu from unnecessary IPI'. > Something like mm pagevec drain also could be avoided. Well, that should be a default feature of nohz full, no additional flags required.
