[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20180516181655.GH25661@ziepe.ca>
Date: Wed, 16 May 2018 12:16:55 -0600
From: Jason Gunthorpe <jgg@...pe.ca>
To: Håkon Bugge <haakon.bugge@...cle.com>
Cc: Doug Ledford <dledford@...hat.com>,
Don Hiatt <don.hiatt@...el.com>,
Ira Weiny <ira.weiny@...el.com>,
Sean Hefty <sean.hefty@...el.com>,
OFED mailing list <linux-rdma@...r.kernel.org>,
linux-kernel@...r.kernel.org,
Hal Rosenstock <hal@....mellanox.co.il>
Subject: Re: [PATCH IB/core 2/2] IB/cm: Send authentic pkey in REQ msg and
check eligibility of the pkeys
On Wed, May 16, 2018 at 08:14:50PM +0200, Håkon Bugge wrote:
>
>
> > On 16 May 2018, at 20:01, Jason Gunthorpe <jgg@...pe.ca> wrote:
> >
> > On Wed, May 16, 2018 at 07:46:10PM +0200, Håkon Bugge wrote:
> >
> >> OK. Lets take one example. The pkey table contains 0xFFFF, 0x8001,
> >> 0x0001.
> >>
> >> The wce.pkey_index is 1 (i.e., pointing to 0x8001). Now, tell me, was
> >> BTH.PKey 0x8001 (matches 0x8001) or was it 0x0001 (also matching
> >> 0x8001) ?
> >
> > As far as the Linux core is concerned, it must have been 0x8001,
> > because the only way the pkey_index feature works properly is if
> > exact-match takes precedence over in-exact match.
>
> And now if the table only contains 0xFFFF, 0x8001, how do you tell?
It doesn't matter.
The delgation of Pkeys to VMs are on a pkey-table-index basis, so if
it matches table entry 1 and entry 1 is passed to the VM, then the
packet can be passed to the VM.
Jason
Powered by blists - more mailing lists