From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jason Gunthorpe Subject: Re: [PATCH] IB/qib: Support creating qps with GFP_NOIO flag Date: Mon, 11 Jan 2016 11:13:54 -0700 Message-ID: <20160111181354.GA19166@obsidianresearch.com> References: <20160111175725.743.92967.stgit@phlsvslse11.ph.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <20160111175725.743.92967.stgit-K+u1se/DcYrLESAwzcoQNrvm/XP+8Wra@public.gmane.org> Sender: linux-rdma-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Mike Marciniszyn Cc: dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org, linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-Id: linux-rdma@vger.kernel.org On Mon, Jan 11, 2016 at 12:57:25PM -0500, Mike Marciniszyn wrote: > From: Vinit Agnihotri > > The current code is problematic when the QP creation and ipoib is used to > support NFS and NFS desires to do IO for paging purposes. In that case, the > GFP_KERNEL allocation in qib_qp.c causes a deadlock in tight memory > situations. Are you trying to do swap over NFS? I didn't think that worked reliably, or has that changed? It doesn't make much sense for one driver to have a different GFP policy for some calls compared with other drivers. Are you sure the GFP agrument shouldn't be pushed up to the real caller? Jason -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html