From mboxrd@z Thu Jan 1 00:00:00 1970 From: Doug Ledford Subject: Re: [PATCH rdma-core] redhat/spec: can't build ibumad on 32-bit arm Date: Tue, 10 Jan 2017 12:36:04 -0500 Message-ID: <1484069764.2149.18.camel@redhat.com> References: <20170109213944.31970-1-jarod@redhat.com> <20170109223108.GA10850@obsidianresearch.com> Mime-Version: 1.0 Content-Type: multipart/signed; micalg="pgp-sha256"; protocol="application/pgp-signature"; boundary="=-onPhYRNlDpoJwQIb8SAy" Return-path: In-Reply-To: Sender: linux-rdma-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Hal Rosenstock , Jason Gunthorpe , Jarod Wilson Cc: linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-Id: linux-rdma@vger.kernel.org --=-onPhYRNlDpoJwQIb8SAy Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Tue, 2017-01-10 at 11:30 -0500, Hal Rosenstock wrote: > On 1/9/2017 5:31 PM, Jason Gunthorpe wrote: > >=20 > > On Mon, Jan 09, 2017 at 04:39:44PM -0500, Jarod Wilson wrote: > > >=20 > > > Building for 32-bit arm, things fall down, due to lack of arch- > > > specific > > > memory barriers. > >=20 > > Since we now have rxe that should work on ARM I think we need to > > fix > > this upstream.. > >=20 > > Do you have time to test some patches on ARM? >=20 > Looks to me that issue was introduced by: >=20 > commit 1df0888f6a736e1612ce8b054d6c17651ebd003f > Author: Jason Gunthorpe > Date:=C2=A0=C2=A0=C2=A0Fri Sep 2 12:57:57 2016 -0600 >=20 > =C2=A0=C2=A0=C2=A0=C2=A0Remove most checks of __BYTE_ORDER >=20 > =C2=A0=C2=A0=C2=A0=C2=A0For a long time now endian.h has defined sane fix= ed with > conversion > =C2=A0=C2=A0=C2=A0=C2=A0macros, so lets just use them instead of rolling = our own. >=20 > =C2=A0=C2=A0=C2=A0=C2=A0Also, htonll is defined in this source tree under > infiniband/arch.h, > =C2=A0=C2=A0=C2=A0=C2=A0so all users of that macro can just use the heade= r. >=20 > =C2=A0=C2=A0=C2=A0=C2=A0Someday we should also get rid of all the endless= wrappers.. >=20 > =C2=A0=C2=A0=C2=A0=C2=A0Signed-off-by: Jason Gunthorpe >=20 > where byte order macros are from libibverbs arch.h which also > includes > the memory barrier stuff. >=20 > So can't arch.h be separated out into 2 headers and have the various > parts of rdma-core include one or both of these headers as needed ? No, that won't help. =C2=A0At most it would shift where the compile breaks. =C2=A0When Steve Wise did the patch to add the arm64 memory barriers, we discussed on list the fact that not having a proper memory barrier define for an arch should be a fatal error, so we also made it such that without an arch specific memory barrier define, we fail the build. =C2=A0If you split the headers, then somewhere you are going to come to a file that needs the memory barriers defined in the headers, and when you include the header, the compile error returns. =C2=A0The proper fix is just to get 32bit arm defines in place. --=20 Doug Ledford =C2=A0 =C2=A0 GPG KeyID: B826A3330E572FDD =C2=A0 =C2=A0 Key fingerprint =3D AE6B 1BDA 122B 23B4 265B =C2=A01274 B826 A333 0E57 2FDD --=-onPhYRNlDpoJwQIb8SAy Content-Type: application/pgp-signature; name="signature.asc" Content-Description: This is a digitally signed message part Content-Transfer-Encoding: 7bit -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQIcBAABCAAGBQJYdRuEAAoJELgmozMOVy/d2NUP/iQBAa8xN1JDXZt/lJdHgmNc blA468KCNwVyWV7zfG2L/XyZjnMTgVCL9ctY09qfLPyRMki6mCFPrxkg5CFRcFCP q8jAga4/XByYetb3jBkARC1yd0cr03qxKpkRz/+Exi/iNrxGSpoJs5vwvf9oPxsp 98sAAJB4bveHHu8imz75gVU2jTYnfmyiFEM0alyCp+H4AvV93S7QeofWnxUrwbFd mFw30vE6PKRBUv/ybVsxNl0a3Cy9XK1Dw+8n0pEXYRjwqsKhdeza+6e8Y2pVuNrT gtJePY9XRcO6OEjFRBE5U3SrJmv6GR+f5E/X09roqu5qGen8LAZ4XxLlT79+gcCW xC7wtRWSOmC1XcsUgNArD43qp2CebRfjhWyyQ1Erxmpym+p9sAGKDuR+CUacztuK WaQfIWqj0dIeSgeK0NbK87XhRse9lcnIupWM5ddB+6oaaliWB9PVYz0IC1u8ACR8 kfUWNaVGkRszu/LuHjDaNAMhFSsvYwvAeSq4uIvL11+Yfw3V/CVdymBf75WZzAPF OCfsd1ewbCBcSD+hQbe8bHqIw1BKUWqe9e/WC+3l6mWbXcxCWOGAIfKa0mrYJl+o TDavnNpmgNZRniHyey9rgwmCJrEWzIZYYgoTod2OioRweXJXwJaoAcjz7vg6myCG Uz/gSBbKO47sdqo0Qq6/ =OSir -----END PGP SIGNATURE----- --=-onPhYRNlDpoJwQIb8SAy-- -- 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