[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <MWHPR2101MB072916BAB0F56AA2CD8C444BCEB50@MWHPR2101MB0729.namprd21.prod.outlook.com>
Date: Thu, 19 Apr 2018 04:59:12 +0000
From: Long Li <longli@...rosoft.com>
To: "Martin K. Petersen" <martin.petersen@...cle.com>
CC: "Michael Kelley (EOSG)" <Michael.H.Kelley@...rosoft.com>,
Long Li <longli@...uxonhyperv.com>,
KY Srinivasan <kys@...rosoft.com>,
Haiyang Zhang <haiyangz@...rosoft.com>,
Stephen Hemminger <sthemmin@...rosoft.com>,
"James E . J . Bottomley" <JBottomley@...n.com>,
"devel@...uxdriverproject.org" <devel@...uxdriverproject.org>,
"linux-scsi@...r.kernel.org" <linux-scsi@...r.kernel.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: RE: [PATCH v2] storvsc: Set up correct queue depth values for IDE
devices
> Subject: Re: [PATCH v2] storvsc: Set up correct queue depth values for IDE
> devices
>
>
> Long,
>
> > Can you take a look at the following patch?
>
> >> > + max_sub_channels =
> >> > + (num_cpus - 1) / storvsc_vcpus_per_sub_channel;
>
> What happens if num_cpus = 1?
If num_cpus=1, we don't have any sub channels.
The host offers one sub channel for VM with 5 CPUs, after that it offers an additional sub channel every 4 CPUs.
The primary channel is always offered.
>
> --
> Martin K. Petersen Oracle Linux Engineering
Powered by blists - more mailing lists