Re: [PATCH 10/14] blk-mq: initial support for multiple queue maps
From: Bart Van Assche
Date: Mon Oct 29 2018 - 16:25:36 EST
On Mon, 2018-10-29 at 14:09 -0600, Jens Axboe wrote:
+AD4 hctx-+AD4-type will be set to the value of the first type. This is all driver
+AD4 private, blk-mq could not care less what the value of the type means.
+AD4 As to the other question, it works just fine since that is the queue
+AD4 that is being accessed. There's no confusion there. I think you're
+AD4 misunderstanding how it's seutp. To use nvme as the example, type 0
+AD4 would be reads, 1 writes, and 2 pollable queues. If reads and writes
+AD4 share the same set of hardware queues, then type 1 simply doesn't
+AD4 exist in terms of -+AD4-flags+AF8-to+AF8-type() return value. This is purely
+AD4 driven by the driver. That hook is the only decider of where something
+AD4 will go. If we share hctx sets, we share the same hardware queue as
+AD4 well. There is just the one set for that case.
How about adding a comment in blk-mq.h that explains that hardware queues can
be shared among different hardware queue types? I think this is nontrivial and
deserves a comment.