[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <BYAPR01MB381679B112EAD6B9B7379DB2F2539@BYAPR01MB3816.prod.exchangelabs.com>
Date: Tue, 11 May 2021 19:39:54 +0000
From: "Marciniszyn, Mike" <mike.marciniszyn@...nelisnetworks.com>
To: Leon Romanovsky <leon@...nel.org>
CC: Haakon Bugge <haakon.bugge@...cle.com>,
Doug Ledford <dledford@...hat.com>,
Jason Gunthorpe <jgg@...dia.com>,
"Dalessandro, Dennis" <dennis.dalessandro@...nelisnetworks.com>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
OFED mailing list <linux-rdma@...r.kernel.org>
Subject: RE: [PATCH rdma-next] RDMA/rdmavt: Decouple QP and SGE lists
allocations
>
> On Tue, May 11, 2021 at 07:15:09PM +0000, Marciniszyn, Mike wrote:
> > > >
> > > > Why not kzalloc_node() here?
> >
> > I agree here.
> >
> > Other allocations that have been promoted to the core have lost the node
> attribute in the allocation.
>
> Did you notice any performance degradation?
>
For the QP, we most certainly will.
In any case, the promotion should address not losing the node.
The allocation gets the ib_device, and it would seem to hard to add method of determining the node.
Mike
Powered by blists - more mailing lists