From mboxrd@z Thu Jan 1 00:00:00 1970 From: Leon Romanovsky Subject: Re: [PATCH for-next 2/2] IB/core: Support for CMA multicast join flags Date: Tue, 5 Jul 2016 13:45:25 +0300 Message-ID: <20160705104525.GC15009@leon.nu> References: <1467550074-24061-1-git-send-email-leon@kernel.org> <1467550074-24061-3-git-send-email-leon@kernel.org> <20160704045111.GA5289@leon.nu> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="bAmEntskrkuBymla" Return-path: Content-Disposition: inline In-Reply-To: Sender: linux-rdma-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Or Gerlitz Cc: Alex Vesker , Doug Ledford , "linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" List-Id: linux-rdma@vger.kernel.org --bAmEntskrkuBymla Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Mon, Jul 04, 2016 at 03:40:10PM +0300, Or Gerlitz wrote: > On Mon, Jul 4, 2016 at 7:51 AM, Leon Romanovsky wro= te: > > On Sun, Jul 03, 2016 at 04:46:23PM +0300, Or Gerlitz wrote: > >> On Sun, Jul 3, 2016 at 3:47 PM, Leon Romanovsky wrot= e: > >> > From: Alex Vesker > >> > > >> > >> > drivers/infiniband/core/cma.c | 98 +++++++++++++++++++++++++++++= ++++++++--- > >> > drivers/infiniband/core/ucma.c | 18 ++++++-- > >> > include/rdma/ib_sa.h | 5 ++ > >> > include/rdma/rdma_cm.h | 4 +- > >> > include/uapi/rdma/rdma_user_cm.h | 9 +++- > >> > 5 files changed, 122 insertions(+), 12 deletions(-) > >> > >> > >> For the ease/robustness of review for UAPI changes, we have a long > >> time common practice > >> to break things like this one to IB core kernel only patch, and one > >> that deals the user-space >=20 > > You are right, this practice exists and we are following it as much as = it makes sense. > > This specific case doesn't need to be separated, because it introduces = one logical > > change and separate patches will be useless as standalone patches. >=20 > Leon, >=20 > The point is that you need to get people to be used to that practice, > and it seems we're not doing that. Otherwise I wouldn't have to chat > for 20m with 2-3 people that wonder why I made these comments. I think > we should require it from the developers, period and not argue on > that. B/c in bunch of other places, it is totally required, for > example, here you could just carve this small piece to be part of your > UAPI patch, what's wrong with that? >=20 > --- a/include/uapi/rdma/rdma_user_cm.h > +++ b/include/uapi/rdma/rdma_user_cm.h > @@ -244,12 +244,19 @@ struct rdma_ucm_join_ip_mcast { > __u32 id; > }; >=20 > struct rdma_ucm_join_mcast { > __u64 response; /* rdma_ucma_create_id_resp */ > __u64 uid; > __u32 id; > __u16 addr_size; > - __u16 reserved; > + __u16 join_flags; > struct sockaddr_storage addr; > }; The main issue here that new code is using this struct and new field while = the old code uses reserved field. --bAmEntskrkuBymla Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Digital signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQIcBAEBAgAGBQJXe4/FAAoJEORje4g2clinHOAQAKKVYe5BF1EoHzJYHCWb+6Wr R6RHBRTB6ERmrpU3WbOI5hhdbafMWm9+Ca7IjW81gn58FXz8vd+e8gqFIy5pyuw4 oRQQ8DPNMccfxnzCiywBPJVQkC3BTWG03U1YPM6sM/YD/XU0Wd5VowGIK0LufEV6 lbBBuSuNJWlgntT7iuU49o0DoBEwvN6p9R0Kxwl20TRZh0Wj6xoBvwD3GIvjLw+p EODjVxuzz3on9xwBZk0vXgo6+0kAvArYJ6e+B+kdwWbAtsgMHwMnXMbDoTeEhCKy 8nKB0CC8v2K1jNDc06xt/MATQw4yCGGNFS3LQdO9360EOvKVUOXDi8k/g0MnnCPY f2KHgUaAZ1faK/ODQBXokBzP6hEWfRi+FYMgWwDrct+Cr0bQKKukqpI9ul/mP/zu TtlLYpD6RCBPyrNG416MTQmnvlzkqOGY7mbQwi7ZetdOkHDG7HCd9lObyi94dQGY nyQCh8zjoO8z6XKiQS5qVywMRiaEzW6r/r+45WLGc01Vsk0NtJmQJH8lX/rhG2yB xaOceGkp6MkYHMEddpK1L6mBCgMDofh0FFrFied/evpKir3t2f0vw5HgBRm7/Zja TxbmtKmXzoqs6oFTW6Q4h95YA/u801Dqyhho91ib4BJfSzlvAwjZdDtksSn7RXNf zpNRAjoCdHwxNO03kOE0 =Rt8s -----END PGP SIGNATURE----- --bAmEntskrkuBymla-- -- 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