The idea of adding a timeout parameter sounds good to me.
I think it's safe to call bdrv_drained_end() after the timeout fails,
but I'm not 100% sure. Maybe Kevin has thoughts on this.
I left comments on the patches.
Stefan
signature.asc
Description: PGP signature
Ping.
From: Zhu Yangyang
In the disable branch of qmp_block_set_io_throttle(), we call
bdrv_drained_begin().
We know that bdrv_drained_begin() is a blocking interface used to wait for all
submitted
I/O operations to complete, i.e., to wait until bs->in_flight becomes zero.
Theoretically, once we sto
