Jukka Ruohonen <jruoho...@iki.fi> wrote: > > Why interrupts are disabled here? Calling xc_wait() with interrupts > > disabled is, at minimum, very expensive, and I would say it is wrong. > > Note that xc_wait(9) does not necessary spin - it can block. > > Because interrupts must be disabled when APERF and MPERF are read. See the > references in the comment, particularly the AMD one.
Why they are not disabled in acpicpu_md_pstate_percent_status(), called on remote CPU, but around xcall(9) calls, which is potentially broken? Also, it has no locking. Is it single-threaded? -- Mindaugas