[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <uk7ooezo3c3jiz2ayvfqatudpvzx6ofooc2vtpgzbembpg4y66@7tuow5vkxf55>
Date: Sat, 27 Jul 2024 13:32:40 +0300
From: Dmitry Baryshkov <dmitry.baryshkov@...aro.org>
To: Manivannan Sadhasivam <manivannan.sadhasivam@...aro.org>
Cc: lpieralisi@...nel.org, kw@...ux.com, robh@...nel.org,
bhelgaas@...gle.com, linux-arm-msm@...r.kernel.org, linux-pci@...r.kernel.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH] PCI: qcom-ep: Do not enable resources during probe()
On Sat, Jul 27, 2024 at 02:36:04PM GMT, Manivannan Sadhasivam wrote:
> Starting from commit 869bc5253406 ("PCI: dwc: ep: Fix DBI access failure
> for drivers requiring refclk from host"), all the hardware register access
> (like DBI) were moved to dw_pcie_ep_init_registers() which gets called only
> in qcom_pcie_perst_deassert() i.e., only after the endpoint received refclk
> from host.
>
> So there is no need to enable the endpoint resources (like clk, regulators,
> PHY) during probe(). Hence, remove the call to qcom_pcie_enable_resources()
> helper from probe(). This was added earlier because dw_pcie_ep_init() was
> doing DBI access, which is not done now.
... moreover his makes PCIe EP fail on some of the platforms as powering
on PHY requires refclk from the RC side, which is not enabled at the
probe time.
Tested-by: Dmitry Baryshkov <dmitry.baryshkov@...aro.org>
> While at it, let's also call dw_pcie_ep_deinit() in err path to deinit the
> EP controller in the case of failure.
>
> Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@...aro.org>
> ---
> drivers/pci/controller/dwc/pcie-qcom-ep.c | 14 ++++----------
> 1 file changed, 4 insertions(+), 10 deletions(-)
--
With best wishes
Dmitry
Powered by blists - more mailing lists