From mboxrd@z Thu Jan 1 00:00:00 1970 From: Florian Fainelli Subject: Re: [PATCH] stmmac: Don't exit mdio registration when mdio subnode is not found in the DTS Date: Sat, 2 Jan 2016 09:52:06 -0800 Message-ID: <56880E46.4050206@gmail.com> References: <1451397935-23643-1-git-send-email-romain.perier@gmail.com> <0E6F98A1-B9F1-48B0-B5CA-A215803E827F@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: In-Reply-To: Sender: netdev-owner@vger.kernel.org To: Romain Perier Cc: peppe.cavallaro@st.com, netdev , "open list:ARM/Rockchip SoC..." List-Id: linux-rockchip.vger.kernel.org Le 02/01/2016 07:11, Romain Perier a =C3=A9crit : >>> Signed-off-by: Romain Perier >> >> Please include a Fixes tag to help keep track of changes. >=20 > Sure, sorry for my ignorance, but I did not know that we're supposed > to add a Fixes tag in this situation. It is a good practice and definitively helps making sure that if regression entered a merge window, and a fix came in another merge window, people can easily backport the fix, moving on.. [snip] >> This looks reasonable, does it work if we just assign mdio_node to t= he pdev->dev.of_node to be backwards compatible with DTSes which have n= ot yet been updated? >> >=20 > I don't understand what you mean. Instead of defining mdio_node with > NULL, you propose to assign it to pdev->dev.of_node as its default > value... then it would be reassigned with "child_nod" if a subnode > with the compatible string "snps,dwmac-mdio" is found, or kept with > its default value otherwise... and in this case we could call > of_mdiobus_register(new_bus, mdio_node) all the time ? (it removes th= e > conditonnal branch) Your understanding is correct, this is what I was suggesting, thanks --=20 =46lorian