Hi Randy, On Fri, 27 Nov 2020 07:57:32 -0800 Randy Dunlap wrote: > > On 11/27/20 1:04 AM, Stephen Rothwell wrote: > > Hi all, > > > > Changes since 20201126: > > > > (This looks strange to me.) > > > on i386 or x86_64: > > # CONFIG_ACPI is not set > > ../drivers/idle/intel_idle.c: In function ‘intel_idle_init_cstates_icpu’: > ../drivers/idle/intel_idle.c:1510:7: error: implicit declaration of function ‘intel_idle_state_needs_timer_stop’; did you mean ‘intel_idle_init_cstates_icpu’? [-Werror=implicit-function-declaration] > if (intel_idle_state_needs_timer_stop(&drv->states[drv->state_count])) > ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ intel_idle_state_needs_timer_stop() is only defined when CONFIG_ACPI_PROCESSOR_CSTATE is set, but is used once where that is not necessarily set. I assume CONFIG_ACPI_PROCESSOR_CSTATE is not set in your config? Caused by commit 6e1d2bc675bd ("intel_idle: Fix intel_idle() vs tracing") from the tip tree. -- Cheers, Stephen Rothwell