On 2021-06-14 13:07:14 [+0200], Vlastimil Babka wrote:

quoted

+#ifdef CONFIG_PREEMPT_RT
+#define slub_get_cpu_ptr(var)	get_cpu_ptr(var)
+#define slub_put_cpu_ptr(var)	put_cpu_ptr(var)

After Mel's report and bisect pointing to this patch, I realized I got the
#ifdef wrong and it should be #ifnded

So if you got the ifdef wrong (and kept everything as-is) then you
tested the RT version on !RT. migrate_disable() behaves on !RT as on RT.
As per changelog you don't use migrate_disable() unconditionally because
it increases the overhead on !RT.
I haven't looked at the series and I have just this tiny question: why
did migrate_disable() crash for Mel on !RT and why do you expect that it
does not happen on PREEMPT_RT?

Sebastian