On Sat, 2018-08-04 at 18:23 +0800, Ming Lei wrote:

From user view, it isn't reasonable to prevent runtime suspend from h=

appening

during queue freeze. The period can be a bit long, and it should be o=

ne perfect

opportunity to suspend device during the period since no any IO is po=

ssible.

Hello Ming,

I will look into reducing the scope of the code that is protected by
blk_pm_runtime_lock() / blk_pm_runtime_unlock() to =
the code in blk_freeze_queue_start()
under "if (freeze_depth == 1)".

Bart.