On Wed, Aug 18, 2021 at 10:44:25PM +0800, Ming Lei wrote:

Hi,

blk_mq_alloc_request_hctx() is used by NVMe fc/rdma/tcp/loop to connect
io queue. Also the sw ctx is chosen as the 1st online cpu in hctx->cpumask.
However, all cpus in hctx->cpumask may be offline.

This usage model isn't well supported by blk-mq which supposes allocator is
always done on one online CPU in hctx->cpumask. This assumption is
related with managed irq, which also requires blk-mq to drain inflight
request in this hctx when the last cpu in hctx->cpumask is going to
offline.

However, NVMe fc/rdma/tcp/loop don't use managed irq, so we should allow
them to ask for request allocation when the specified hctx is inactive
(all cpus in hctx->cpumask are offline). Fix blk_mq_alloc_request_hctx() by
allowing to allocate request when all CPUs of this hctx are offline.

Wen Xiong has verified V4 in her nvmef test.

V7:
	- move blk_mq_hctx_use_managed_irq() into block/blk-mq.c, 3/3

Hello Jens,

NVMe TCP and others have been a bit popular recent days, and the kernel panic
of blk_mq_alloc_request_hctx() has annoyed people for a bit long.

Any chance to pull the three patches in so we can fix them in 5.15?


Thanks,
Ming