[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20230223130258.GA6422@workstation>
Date: Thu, 23 Feb 2023 18:32:58 +0530
From: Manivannan Sadhasivam <manivannan.sadhasivam@...aro.org>
To: Krzysztof Kozlowski <krzysztof.kozlowski@...aro.org>
Cc: andersson@...nel.org, lpieralisi@...nel.org, robh@...nel.org,
kw@...ux.com, krzysztof.kozlowski+dt@...aro.org, vkoul@...nel.org,
konrad.dybcio@...aro.org, bhelgaas@...gle.com, kishon@...nel.org,
linux-arm-msm@...r.kernel.org, linux-phy@...ts.infradead.org,
linux-kernel@...r.kernel.org, linux-pci@...r.kernel.org,
devicetree@...r.kernel.org
Subject: Re: [PATCH 02/11] dt-bindings: PCI: qcom: Add iommu properties
On Thu, Feb 23, 2023 at 10:37:27AM +0100, Krzysztof Kozlowski wrote:
> On 22/02/2023 16:32, Manivannan Sadhasivam wrote:
> > Most of the PCIe controllers require iommu support to function properly.
> > So let's add them to the binding.
> >
>
> If most of them require iommu, why not adding it as a required property
> to respective (or new) "if:then:" part?
>
Well, I thought about it but then followed the convention of
"dma-coherent" property. I asked this same question while adding that
property but I didn't get a clear answer (or maybe I missed something).
So if you want me to add iommu properties to individual SoCs, then please
explain why the same cannot be done for "dma-coherent" as not all SoCs
support dma coherency for PCIe controllers.
Thanks,
Mani
> Best regards,
> Krzysztof
>
Powered by blists - more mailing lists