From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tom Talpey Subject: Re: [PATCH RFC 1/3] IB/core: Expose a device attribute for rdma_read access flags Date: Tue, 10 Nov 2015 07:36:25 -0500 Message-ID: <5641E4C9.7000206@talpey.com> References: <1447152255-28231-1-git-send-email-sagig@mellanox.com> <1447152255-28231-2-git-send-email-sagig@mellanox.com> <1447156270.7089.3.camel@opteya.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: In-Reply-To: <1447156270.7089.3.camel-RlY5vtjFyJ3QT0dZR+AlfA@public.gmane.org> Sender: linux-rdma-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Yann Droneaud , Sagi Grimberg Cc: linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-Id: linux-rdma@vger.kernel.org On 11/10/2015 6:51 AM, Yann Droneaud wrote: > Le mardi 10 novembre 2015 =C3=A0 12:44 +0200, Sagi Grimberg a =C3=A9c= rit : >> Also, set rdma_read_access_flags in the relevant device drivers: >> mlx4, mlx5, qib, ocrdma, nes: IB_ACCESS_LOCAL_WRITE >> cxgb3, cxgb4, hfi: IB_ACCESS_LOCAL_WRITE | IB_ACCESS_REMOTE_WRITE >> > > Why were those hw providers not modified to > enforce IB_ACCESS_REMOTE_WRITE when needed, instead of asking users t= o > set it for them ? Because the Linux verbs don't tell the driver whether the registration is for an RDMA Read sink buffer. Sagi, the Windows NDKPI has an NDK_MR_FLAG_RDMA_READ_SINK attribute which the upper layer can use to convey this information, I've mentione= d it here before. https://msdn.microsoft.com/en-us/library/windows/hardware/hh439908(v=3D= vs.85).aspx When this approach is used, the upper layer doesn't have to be aware at all of the lower layer's details. Tom. -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" i= n the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html