From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 80B336ABF for ; Sun, 24 Sep 2023 13:57:29 +0000 (UTC) Received: by smtp.kernel.org (Postfix) with ESMTPSA id E65BEC433C8; Sun, 24 Sep 2023 13:57:24 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1695563849; bh=tB7kkagYAAKovDZa1W4Av2bMf3MSyMZ6VNzkF6Na6l0=; h=Date:From:To:Cc:Subject:References:In-Reply-To:From; b=bu7DxLoxXxb5GQ6uUu7fI93gJdTMLGG4A7FFxqMQLezspwoud8ckTV18QLPsfybjZ EBnsZReKEoN+Bx3kBeT3MEUU5kI1/bAbH0L1VLN/j6uqGp811lYIuW2PwwFTMvsDGG NecpUwrBQ4FSn2/HGczwayQHMG50LvzF8lI+JnZ2eRNfc6ZKRwmRW6mBM1GMq1SjYC BnqIPVtBhRHAzaxApryqOwdripeca+u9+JEAJo3oE77F26la1P3AAY5FoMOA5mgaGS P+OpA95CBhzogp7UiIWkgIctkglSlD8Py2+EiMlaIwCi0pewzYtzkg33lHYSS024SZ +mWU6j63JLaDA== Date: Sun, 24 Sep 2023 21:57:16 +0800 From: Shawn Guo To: Marek Vasut Cc: linux-arm-kernel@lists.infradead.org, Conor Dooley , Fabio Estevam , Krzysztof Kozlowski , NXP Linux Team , Pengutronix Kernel Team , Rob Herring , Sascha Hauer , devicetree@vger.kernel.org Subject: Re: [PATCH] arm64: dts: imx8mp: Switch PCIe to HSIO PLL on i.MX8MP DHCOM PDK2 and generate clock from SoC Message-ID: <20230924135716.GK7231@dragon> References: <20230822005007.128571-1-marex@denx.de> Precedence: bulk X-Mailing-List: devicetree@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20230822005007.128571-1-marex@denx.de> On Tue, Aug 22, 2023 at 02:50:07AM +0200, Marek Vasut wrote: > The PDK2 carrier board had to be manually patched to obtain working PCIe > with the i.MX8MP DHCOM SoM so far, because the PCIe clock generator has > not been connected to the PCIe block REF_PAD_CLK inputs. > > Switch to use of HSIO PLL as the clock source for the PCIe block instead, > and use the REF_PAD_CLK as outputs to generate PCIe clock from the SoC. > This way, it is not necessary to patch the PDK2 in any way to obtain a > working PCIe. > > Note that PDK3 has PCIe clock generator always connected to REF_PAD_CLK > and is not affected. > > Signed-off-by: Marek Vasut Applied, thanks!