Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Subject: Re: [PATCH] make blk_mq_map_queues more friendly for cpu topology
To:     luferry <luferry@163.com>
References: <20190325152640.48120-1-luferry@163.com>
 <20190326073954.GD21617@lst.de>
 <48ca2186.fc76.169b8fef63a.Coremail.luferry@163.com>
 <20190327081619.GG20525@lst.de>
 <764f3c.12445.169be70834b.Coremail.luferry@163.com>
Cc:     Christoph Hellwig <hch@lst.de>, Jens Axboe <axboe@kernel.dk>,
        Ming Lei <ming.lei@redhat.com>, linux-block@vger.kernel.org,
        linux-kernel@vger.kernel.org
From:   Dongli Zhang <dongli.zhang@oracle.com>
Message-ID: <8e02b730-b504-50a2-64e6-72817b07e93d@oracle.com>
Date:   Wed, 27 Mar 2019 17:39:16 +0800
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101
 Thunderbird/45.3.0
MIME-Version: 1.0
In-Reply-To: <764f3c.12445.169be70834b.Coremail.luferry@163.com>
Content-Type: text/plain; charset=gbk
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


On 03/27/2019 05:17 PM, luferry wrote:
> 
> 
> 
> Actually,  I just bought one vm from public cloud provider and run into this problem.
> after reading code and compare pci device info, I reproduce this scenario.

There is lack of vector issue when number of queues is more than maxcpus as well.

https://lore.kernel.org/lkml/e4afe4c5-0262-4500-aeec-60f30734b4fc@default

I was going to send out patch to fix that specific case on virtio side.

BTW, is your VM's virtio mmio based or pci based?

As mentioned in previous email, when I was playing with firecracker by amazon,
the virtio-blk is mmio based. Not sure if that would lead to the lack of vector
issue.

Dongli Zhang