From mboxrd@z Thu Jan 1 00:00:00 1970 From: Sekhar Nori Subject: Re: [PATCH v2 6/6] ARM: davinci: da850: configure system configuration chip(CFGCHIP3) for emac Date: Mon, 4 Feb 2013 10:58:46 +0530 Message-ID: <510F470E.5040205@ti.com> References: <1359380879-26306-1-git-send-email-prabhakar.lad@ti.com> <1359380879-26306-7-git-send-email-prabhakar.lad@ti.com> <510E5BCF.9080006@ti.com> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <510E5BCF.9080006@ti.com> Sender: netdev-owner@vger.kernel.org To: Sekhar Nori Cc: Prabhakar Lad , davinci-linux-open-source@linux.davincidsp.com, netdev@vger.kernel.org, devicetree-discuss@lists.ozlabs.org, linux-kernel@vger.kernel.org, Heiko Schocher , linux-arm-kernel@lists.infradead.org List-Id: devicetree@vger.kernel.org On 2/3/2013 6:15 PM, Sekhar Nori wrote: > On 1/28/2013 7:17 PM, Prabhakar Lad wrote: >> From: Lad, Prabhakar > There are couple of issues with this implementation. > > 1) da8xx_config_emac() is specific to DA850 EVM, but masquerades as > generic for da8xx. Looks like you need two functions, one for soc > specific configuration and one board specific. > 2) da8xx_config_emac() goes through all the time, whether the > particular board has emac module or not. Shouldn't > da8xx_config_emac() check if emac is actually enabled in the passed > dtb and only the do the configuration? > 3) The function assumes mii is used always, you can use the rmii_en dt > property to check if rmii/mii is enabled and configure the soc/board > accordingly. > 4) If the same function can work both for da850 and da830, then it can > be implemented outside of CONFIG_ARCH_DAVINCI_DA850. Another thing I forgot to mention: 5) This patch replicates a lot of code form board-da850-evm.c. This needs to be avoided. May be create function to configure emac in da850.c. This can then be called in board-da850-evm.c as well as in the dt case. Thanks, Sekhar