On 12.11.2021 19:28, Andrew Cooper wrote: > Currently, each feature_detect() (called on CPU add) hook for both cpufreq > drivers duplicates cpu_has_aperfmperf in a per-cpu datastructure, and edits > cpufreq_driver.getavg to point at get_measured_perf(). > > As all parts of this are vendor-neutral, drop the function pointer and > duplicated boolean, and call get_measured_perf() directly. > > Signed-off-by: Andrew Cooper <andrew.coop...@citrix.com> > --- > CC: Jan Beulich <jbeul...@suse.com> > CC: Roger Pau Monné <roger....@citrix.com> > CC: Wei Liu <w...@xen.org> > > Slightly RFC. This does introduce an arch-specific call into a nominally > arch-neutral driver, but struct cpufreq_policy already had x86-specifics in it > so this is apparently ok.
That's fine for the time being; disentangling for another arch to use the driver would be quite a bit of work anyway. Reviewed-by: Jan Beulich <jbeul...@suse.com> Jan
