mm, memory_hotplug: remove timeout from __offline_memory
authorMichal Hocko <[email protected]>
Thu, 16 Nov 2017 01:33:38 +0000 (17:33 -0800)
committerLinus Torvalds <[email protected]>
Thu, 16 Nov 2017 02:21:02 +0000 (18:21 -0800)
commitecde0f3e7f9edf8629f56b2354385dc8d0a6a24d
tree5f82d86166ffc872db9c43ecc1da30f138fcd723
parent72b39cfc4d750e5b8c633a7a6fdd7d07927995ad
mm, memory_hotplug: remove timeout from __offline_memory

We have a hardcoded 120s timeout after which the memory offline fails
basically since the hot remove has been introduced.  This is essentially
a policy implemented in the kernel.  Moreover there is no way to adjust
the timeout and so we are sometimes facing memory offline failures if
the system is under a heavy memory pressure or very intensive CPU
workload on large machines.

It is not very clear what purpose the timeout actually serves.  The
offline operation is interruptible by a signal so if userspace wants
some timeout based termination this can be done trivially by sending a
signal.

If there is a strong usecase to do this from the kernel then we should
do it properly and have a it tunable from the userspace with the timeout
disabled by default along with the explanation who uses it and for what
purporse.

Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Michal Hocko <[email protected]>
Acked-by: Vlastimil Babka <[email protected]>
Cc: KAMEZAWA Hiroyuki <[email protected]>
Cc: Reza Arbab <[email protected]>
Cc: Yasuaki Ishimatsu <[email protected]>
Cc: Xishi Qiu <[email protected]>
Cc: Igor Mammedov <[email protected]>
Cc: Vitaly Kuznetsov <[email protected]>
Cc: Michael Ellerman <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
mm/memory_hotplug.c