On 14/10/2020 21:54, Peter Zijlstra wrote: > > Add CPUPRI_HIGHER above the RT99 priority to denote the CPU is in use > by higher priority tasks (specifically deadline).
sugov:X already triggers this now on our !fast-switching devices running schedutil. > XXX: we should probably drive PUSH-PULL from cpupri, that would > automagically result in an RT-PUSH when DL sets cpupri to CPUPRI_HIGHER. > > Signed-off-by: Peter Zijlstra (Intel) <pet...@infradead.org> [...] > @@ -54,6 +56,10 @@ static int convert_prio(int prio) The BUG_ON could be tightened: - BUG_ON(prio >= MAX_PRIO); + BUG_ON(prio > MAX_RT_PRIO); > case MAX_RT_PRIO-1: > cpupri = CPUPRI_NORMAL; /* 0 */ > break; > + > + case MAX_RT_PRIO: > + cpupri = CPUPRI_HIGHER; /* 100 */ > + break; > } > > return cpupri; Just saw that the comment for cpupri_set() needs changing: @@ -205,7 +208,7 @@ int cpupri_find_fitness(struct cpupri *cp, struct task_struct *p, * cpupri_set - update the CPU priority setting * @cp: The cpupri context * @cpu: The target CPU - * @newpri: The priority (INVALID-RT99) to assign to this CPU + * @newpri: The priority (INVALID-RT1-RT99-NORMAL-HIGHER) to assign to this CPU Reviewed-by: Dietmar Eggemann <dietmar.eggem...@arm.com>
