From mboxrd@z Thu Jan 1 00:00:00 1970 From: Doug Ledford Subject: Re: [PATCH] RDMA/cxgb3: fail get_dma_mr if the memory footprint can exceed 32b Date: Fri, 24 Jul 2015 10:55:42 -0400 Message-ID: <55B251EE.3030806@redhat.com> References: <20150722191417.9306.17387.stgit@build2.ogc.int> <55B15D7D.3090601@redhat.com> <00d501d0c599$84afdb50$8e0f91f0$@opengridcomputing.com> <55B24F65.5040103@redhat.com> <008b01d0c620$05dce480$1196ad80$@opengridcomputing.com> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha256; protocol="application/pgp-signature"; boundary="wNQhHAGsIQbRIASecIHTKoLdB3HNloNPa" Return-path: In-Reply-To: <008b01d0c620$05dce480$1196ad80$@opengridcomputing.com> Sender: linux-rdma-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Steve Wise Cc: linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-Id: linux-rdma@vger.kernel.org This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --wNQhHAGsIQbRIASecIHTKoLdB3HNloNPa Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable On 07/24/2015 10:50 AM, Steve Wise wrote: >=20 >=20 >> -----Original Message----- >> From: linux-rdma-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org [mailto:linux-rdma-owner@vger.k= ernel.org] On Behalf Of Doug Ledford >> Sent: Friday, July 24, 2015 9:45 AM >> To: Steve Wise >> Cc: linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org >> Subject: Re: [PATCH] RDMA/cxgb3: fail get_dma_mr if the memory footpri= nt can exceed 32b >> >> On 07/23/2015 06:47 PM, Steve Wise wrote: >>>> -----Original Message----- From: Doug Ledford >>>> Should this be a static check of the pointer size versus installed >>>> memory? Would it be possible to have this work for machines with >>>> less than 4GB of physical memory even if they have 64bit pointers, >>>> or are you concerned that hotplug memory could take us over the >>>> limit after registration and cause problems? >>> >>> NFSRDMA doesn't need dma-mrs for T3 since it has FRMR + local dma >>> lkey support. And since the deficiency really can cause problems on >>> 64b systems if the memory grows > 4GB after dma-mr allocation, I >>> decided to just not allow them for potential large memory systems. >> >> Ok. I've pulled this for 4.2-rc then. >> >=20 > The problem has been there since day one, so it doesn't represent a reg= ression. It is your call, but I think 4.3 is fine. It's a long standing problem, but it's a potential memory wrap issue, so might as well fix it (especially given the simple nature of the fix). I've got a few other things heading in for 4.2-rc, so no biggie. --=20 Doug Ledford GPG KeyID: 0E572FDD --wNQhHAGsIQbRIASecIHTKoLdB3HNloNPa Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQIcBAEBCAAGBQJVslHuAAoJELgmozMOVy/dfPkQAMD2z25Jk9KsjISdspXsRMRR TG7iQq54XUh+oR4TyFvi0kjwZGb5s7EBQjHcI35/s3ngS8vv6iNcA/cZmxTeOXTP nOetX4hBl0k15pQXdhBP5tbgbmqCx9Jjxippiv20p9QuFWUUJz6TWkcBXvlIIYtX lK6wIsZra7IGen79bXfpkpKBqjxwzhemciGdYN3a8oa5JXlK7CwgCNLBLr+jL6AM QS2eMB7joz9ruwnald1ZISR85oS5PMnm8OKdvSvcd3Tokkmr5lMuDfG3VXhewyYF s5+2l8rVIvZFwB/jWoCa27+aDVl+6gm0me6awm/0h2k5Vg9FTIz/F8DdlfY+Tstg SmRSp+VmKfWGmmmpt/EIfET43d/TSUr+eH3Lz8ZdQXYjiioVpd9Mnl7C38XXgizD /R5Wq5/WIoHN+5m/ab0450hjjxv3iC7J6a68UcnMjR57/gjoq0v1+tZVnYoUxrTC +AIk2E23YkuOzVCWh5jXgpkhiZn1RFHjIWdpAtHSmmllT2th9+DTdOpJ3n2yh4wl ch3W23NG3ACbj6Jx4vEIbX+Jzun1BIgpbb8sq3hmU3sS0sAwRdNqzwtQ6rFFeYhL uGXOjIe2rurDJEeWUBQ6aXoamZ2C2leEbjN8EDV5TMJiLUAQQrI3lv5ELcEkjQ9n hGSTMvo6N5hNTWjqoOJR =FVUV -----END PGP SIGNATURE----- --wNQhHAGsIQbRIASecIHTKoLdB3HNloNPa-- -- 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