On Fri, 2019-08-02 at 15:07 +0200, Juergen Gross wrote:
Today a cpu which is removed from the system is taken directly from
Pool0 to the offline state. This will conflict with the new idle
scheduler, so remove it from Pool0 first. Additionally accept
removing
a free cpu instead of requiring it to be in Pool0.
For the resume failed case we need to call the scheduler code for
that
situation after the cpupool handling, so move the scheduler code into
a function and call it from cpupool_cpu_remove_forced() and remove
the
CPU_RESUME_FAILED case from cpu_schedule_callback().
Note that we are calling now schedule_cpu_switch() in stop_machine
context so we need to switch from spinlock_irq to spinlock_irqsave.
Signed-off-by: Juergen Gross <jgross@xxxxxxxx>
---
--- a/xen/common/cpupool.c
+++ b/xen/common/cpupool.c
@@ -282,22 +282,14 @@ static int cpupool_assign_cpu_locked(struct
cpupool *c, unsigned int cpu)
return 0;
}
-static long cpupool_unassign_cpu_helper(void *info)
+static int cpupool_unassign_cpu_epilogue(struct cpupool *c)
in schedule.c, for a similar situation, we have used '_start' and
'_finish' as suffixes. What do you think about using those here too?
It's certainly a minor thing, I know, but I (personally) like them
better (especially than 'epilogue') and I think it gives us some
consistency (yes, sure, different files.. but scheduling and cpupools
are quite tightly related).
