Re: [PATCH v2] blk-mq: Fix races between iterating over requests and freeing requests

On 3/24/21 2:27 AM, Hannes Reinecke wrote:

What I don't particularly like is the global blk_sched_srcu here;
can't we make it per tagset?

Hi Hannes,

Thanks for having taken a look. I'm concerned about the additional
memory required for one srcu_struct per tag set. Can you take a look at
the attached patches since the approach of these patches does not rely
on SRCU at all?

Thanks,

Bart.