Received-SPF: pass (google.com: domain of linux-kernel+bounces-37063-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45e3:2400::1 as permitted sender) client-ip=2604:1380:45e3:2400::1;
Message-ID: <9feae876-5170-4b57-8cd4-dbf35dddbefc@linaro.org>
Date: Wed, 24 Jan 2024 14:22:55 +0100
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: [PATCH 09/14] arm64: dts: qcom: sc8280xp: Drop PCIE_AUX_CLK from
 pcie_phy nodes
Content-Language: en-US
To: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>,
 Bjorn Andersson <andersson@kernel.org>, Vinod Koul <vkoul@kernel.org>,
 Kishon Vijay Abraham I <kishon@kernel.org>, Rob Herring
 <robh+dt@kernel.org>, Krzysztof Kozlowski
 <krzysztof.kozlowski+dt@linaro.org>, Conor Dooley <conor+dt@kernel.org>,
 cros-qcom-dts-watchers@chromium.org
Cc: linux-arm-msm@vger.kernel.org, linux-phy@lists.infradead.org,
 devicetree@vger.kernel.org, linux-kernel@vger.kernel.org
References: <20240124-pcie-aux-clk-fix-v1-0-d8a4852b6ba6@linaro.org>
 <20240124-pcie-aux-clk-fix-v1-9-d8a4852b6ba6@linaro.org>
From: Konrad Dybcio <konrad.dybcio@linaro.org>
In-Reply-To: <20240124-pcie-aux-clk-fix-v1-9-d8a4852b6ba6@linaro.org>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit


On 1/24/24 08:36, Manivannan Sadhasivam wrote:
> PCIe PHY hw doesn't require PCIE_AUX_CLK for functioning. This clock is
> only required by the PCIe controller. Hence drop it from pcie_phy nodes.
> 
> Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
> ---

So, I have a small phytest module that basically does this:

phy_get
phy_init
phy_set_mode_ext(PCIE_RC)
phy_power_on

To load it, I skip PCIe init (comment out the node), then wait for clk
and pmdomain unused cleanup, then load the phytest module to ensure that
I'm not piggybacking off of the RC's resources.

I tried it out on the CRD, on PCIE2A (NVMe)

Without this patch, the PHY seems to init fine

With this patch, I get:

(1) qcom-qmp-pcie-phy 1c24000.phy: phy initialization timed-out


Kicking the PCIe GDSC from the PHY, I additionally get:

(2) gcc_pcie_2a_cfg_ahb_clk status stuck at 'off'


I think we expected (2), but is (1) okay?

Konrad