Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <d665b1b5-cfc5-939e-7444-490e3761ab39@quicinc.com>
Date:   Mon, 15 May 2023 18:45:54 +0530
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:102.0) Gecko/20100101
 Thunderbird/102.10.0
Subject: Re: [PATCH V3 4/6] arm64: dts: qcom: ipq9574: Add PCIe PHYs and
 controller nodes
Content-Language: en-US
To:     Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
CC:     <agross@kernel.org>, <andersson@kernel.org>,
        <konrad.dybcio@linaro.org>, <lpieralisi@kernel.org>,
        <kw@linux.com>, <robh@kernel.org>, <bhelgaas@google.com>,
        <krzysztof.kozlowski+dt@linaro.org>, <mturquette@baylibre.com>,
        <sboyd@kernel.org>, <mani@kernel.org>,
        <linux-arm-msm@vger.kernel.org>, <devicetree@vger.kernel.org>,
        <linux-kernel@vger.kernel.org>, <linux-pci@vger.kernel.org>,
        <linux-clk@vger.kernel.org>, <quic_srichara@quicinc.com>,
        <quic_sjaganat@quicinc.com>, <quic_kathirav@quicinc.com>,
        <quic_arajkuma@quicinc.com>, <quic_anusha@quicinc.com>,
        <quic_ipkumar@quicinc.com>
References: <20230421124938.21974-1-quic_devipriy@quicinc.com>
 <20230421124938.21974-5-quic_devipriy@quicinc.com>
 <CAA8EJppyro1wM3KmDU3DVjKCqXH5+KaNoT_7ObVuuYNMoZKpoA@mail.gmail.com>
 <a220e00d-0559-35d5-80e9-7e11e566851a@quicinc.com>
 <e7b4afde-2cd7-0ab5-8a15-446173b42c40@linaro.org>
 <97315104-d9cc-27f8-8c52-b99080179269@quicinc.com>
 <CAA8EJpqrza+r3bygJh=TmNuNYUFspqvuV7gDFb5-QZcs4j3Dbg@mail.gmail.com>
From:   Devi Priya <quic_devipriy@quicinc.com>
In-Reply-To: <CAA8EJpqrza+r3bygJh=TmNuNYUFspqvuV7gDFb5-QZcs4j3Dbg@mail.gmail.com>
Content-Type: text/plain; charset="UTF-8"; format=flowed
Content-Transfer-Encoding: 7bit
Precedence: bulk


On 5/15/2023 3:21 PM, Dmitry Baryshkov wrote:
> On Mon, 15 May 2023 at 12:36, Devi Priya <quic_devipriy@quicinc.com> wrote:
>>
>>
>>
>> On 5/8/2023 5:10 PM, Dmitry Baryshkov wrote:
>>> On 08/05/2023 13:53, Devi Priya wrote:
>>>>
>>>>
>>>> On 4/22/2023 5:49 AM, Dmitry Baryshkov wrote:
>>>>> On Fri, 21 Apr 2023 at 15:50, Devi Priya <quic_devipriy@quicinc.com>
>>>>> wrote:
>>>>>>
>>>>>> Add PCIe0, PCIe1, PCIe2, PCIe3 (and corresponding PHY) devices
>>>>>> found on IPQ9574 platform. The PCIe0 & PCIe1 are 1-lane Gen3
>>>>>> host whereas PCIe2 & PCIe3 are 2-lane Gen3 host.
>>>>>>
>>>>>> Co-developed-by: Anusha Rao <quic_anusha@quicinc.com>
>>>>>> Signed-off-by: Anusha Rao <quic_anusha@quicinc.com>
>>>>>> Signed-off-by: Devi Priya <quic_devipriy@quicinc.com>
>>>>>> ---
>>>>>>    Changes in V3:
>>>>>>           - Fixed up the PCI I/O port ranges
>>>>>>
>>>>>>    arch/arm64/boot/dts/qcom/ipq9574.dtsi | 375
>>>>>> +++++++++++++++++++++++++-
>>>>>>    1 file changed, 370 insertions(+), 5 deletions(-)
>>>>>>
>>>>>> diff --git a/arch/arm64/boot/dts/qcom/ipq9574.dtsi
>>>>>> b/arch/arm64/boot/dts/qcom/ipq9574.dtsi
>>>>>> index e757b57957cf..953a839a1141 100644
>>>>>> --- a/arch/arm64/boot/dts/qcom/ipq9574.dtsi
>>>>>> +++ b/arch/arm64/boot/dts/qcom/ipq9574.dtsi
>>>>>> @@ -6,8 +6,8 @@
>>>>>>     * Copyright (c) 2023, Qualcomm Innovation Center, Inc. All rights
>>>>>> reserved.
>>>>>>     */
>>>>>>
>>>>>> -#include <dt-bindings/interrupt-controller/arm-gic.h>
>>>>>>    #include <dt-bindings/clock/qcom,ipq9574-gcc.h>
>>>>>> +#include <dt-bindings/interrupt-controller/arm-gic.h>
>>>>>>    #include <dt-bindings/reset/qcom,ipq9574-gcc.h>
>>>>>>
>>>>>>    / {
>>>>>> @@ -116,6 +116,58 @@
>>>>>>                   #size-cells = <1>;
>>>>>>                   ranges = <0 0 0 0xffffffff>;
>>>>>>
>>>>>> +               pcie0_phy: phy@84000 {
>>>>>> +                       compatible =
>>>>>> "qcom,ipq9574-qmp-gen3x1-pcie-phy";
>>>>>> +                       reg = <0x00084000 0x1000>;
>>>>>> +
>>>>>> +                       clocks = <&gcc GCC_PCIE0_AUX_CLK>,
>>>>>> +                                <&gcc GCC_PCIE0_AHB_CLK>,
>>>>>> +                                <&gcc GCC_ANOC_PCIE0_1LANE_M_CLK>,
>>>>>> +                                <&gcc GCC_SNOC_PCIE0_1LANE_S_CLK>,
>>>>>> +                                <&gcc GCC_PCIE0_PIPE_CLK>;
>>>>>> +                       clock-names = "aux", "cfg_ahb", "anoc_lane",
>>>>>> "snoc_lane", "pipe";
>>>>>> +
>>>>>> +                       assigned-clocks = <&gcc GCC_PCIE0_AUX_CLK>;
>>>>>> +                       assigned-clock-rates = <20000000>;
>>>>>> +
>>>>>> +                       resets = <&gcc GCC_PCIE0_PHY_BCR>,
>>>>>> +                                <&gcc GCC_PCIE0PHY_PHY_BCR>;
>>>>>> +                       reset-names = "phy", "common";
>>>>>> +
>>>>>> +                       #clock-cells = <0>;
>>>>>> +                       clock-output-names = "gcc_pcie0_pipe_clk_src";
>>>>>> +
>>>>>> +                       #phy-cells = <0>;
>>>>>> +                       status = "disabled";
>>>>>> +
>>>>>> +               };
>>>>>> +
>>>>>> +               pcie2_phy: phy@8c000 {
>>>>>> +                       compatible =
>>>>>> "qcom,ipq9574-qmp-gen3x2-pcie-phy";
>>>>>> +                       reg = <0x0008c000 0x2000>;
>>>>>> +
>>>>>> +                       clocks = <&gcc GCC_PCIE2_AUX_CLK>,
>>>>>> +                                <&gcc GCC_PCIE2_AHB_CLK>,
>>>>>> +                                <&gcc GCC_ANOC_PCIE2_2LANE_M_CLK>,
>>>>>> +                                <&gcc GCC_SNOC_PCIE2_2LANE_S_CLK>,
>>>>>> +                                <&gcc GCC_PCIE2_PIPE_CLK>;
>>>>>> +                       clock-names = "aux", "cfg_ahb", "anoc_lane",
>>>>>> "snoc_lane", "pipe";
>>>>>> +
>>>>>> +                       assigned-clocks = <&gcc GCC_PCIE2_AUX_CLK>;
>>>>>> +                       assigned-clock-rates = <20000000>;
>>>>>> +
>>>>>> +                       resets = <&gcc GCC_PCIE2_PHY_BCR>,
>>>>>> +                                <&gcc GCC_PCIE2PHY_PHY_BCR>;
>>>>>> +                       reset-names = "phy", "common";
>>>>>> +
>>>>>> +                       #clock-cells = <0>;
>>>>>> +                       clock-output-names = "gcc_pcie2_pipe_clk_src";
>>>>>> +
>>>>>> +                       #phy-cells = <0>;
>>>>>> +                       status = "disabled";
>>>>>> +
>>>>>> +               };
>>>>>> +
>>>>>>                   rng: rng@e3000 {
>>>>>>                           compatible = "qcom,prng-ee";
>>>>>>                           reg = <0x000e3000 0x1000>;
>>>>>> @@ -123,6 +175,58 @@
>>>>>>                           clock-names = "core";
>>>>>>                   };
>>>>>>
>>>>>> +               pcie3_phy: phy@f4000 {
>>>>>> +                       compatible =
>>>>>> "qcom,ipq9574-qmp-gen3x2-pcie-phy";
>>>>>> +                       reg = <0x000f4000 0x2000>;
>>>>>> +
>>>>>> +                       clocks = <&gcc GCC_PCIE3_AUX_CLK>,
>>>>>> +                                <&gcc GCC_PCIE3_AHB_CLK>,
>>>>>> +                                <&gcc GCC_ANOC_PCIE3_2LANE_M_CLK>,
>>>>>> +                                <&gcc GCC_SNOC_PCIE3_2LANE_S_CLK>,
>>>>>> +                                <&gcc GCC_PCIE3_PIPE_CLK>;
>>>>>> +                       clock-names = "aux", "cfg_ahb", "anoc_lane",
>>>>>> "snoc_lane", "pipe";
>>>>>> +
>>>>>> +                       assigned-clocks = <&gcc GCC_PCIE3_AUX_CLK>;
>>>>>> +                       assigned-clock-rates = <20000000>;
>>>>>> +
>>>>>> +                       resets = <&gcc GCC_PCIE3_PHY_BCR>,
>>>>>> +                                <&gcc GCC_PCIE3PHY_PHY_BCR>;
>>>>>> +                       reset-names = "phy", "common";
>>>>>> +
>>>>>> +                       #clock-cells = <0>;
>>>>>> +                       clock-output-names = "gcc_pcie3_pipe_clk_src";
>>>>>> +
>>>>>> +                       #phy-cells = <0>;
>>>>>> +                       status = "disabled";
>>>>>> +
>>>>>> +               };
>>>>>> +
>>>>>> +               pcie1_phy: phy@fc000 {
>>>>>> +                       compatible =
>>>>>> "qcom,ipq9574-qmp-gen3x1-pcie-phy";
>>>>>> +                       reg = <0x000fc000 0x1000>;
>>>>>> +
>>>>>> +                       clocks = <&gcc GCC_PCIE1_AUX_CLK>,
>>>>>> +                                <&gcc GCC_PCIE1_AHB_CLK>,
>>>>>> +                                <&gcc GCC_ANOC_PCIE1_1LANE_M_CLK>,
>>>>>> +                                <&gcc GCC_SNOC_PCIE1_1LANE_S_CLK>,
>>>>>> +                                <&gcc GCC_PCIE1_PIPE_CLK>;
>>>>>> +                       clock-names = "aux", "cfg_ahb", "anoc_lane",
>>>>>> "snoc_lane", "pipe";
>>>>>> +
>>>>>> +                       assigned-clocks = <&gcc GCC_PCIE1_AUX_CLK>;
>>>>>> +                       assigned-clock-rates = <20000000>;
>>>>>> +
>>>>>> +                       resets = <&gcc GCC_PCIE1_PHY_BCR>,
>>>>>> +                                <&gcc GCC_PCIE1PHY_PHY_BCR>;
>>>>>> +                       reset-names = "phy", "common";
>>>>>> +
>>>>>> +                       #clock-cells = <0>;
>>>>>> +                       clock-output-names = "gcc_pcie1_pipe_clk_src";
>>>>>> +
>>>>>> +                       #phy-cells = <0>;
>>>>>> +                       status = "disabled";
>>>>>> +
>>>>>> +               };
>>>>>> +
>>>>>>                   tlmm: pinctrl@1000000 {
>>>>>>                           compatible = "qcom,ipq9574-tlmm";
>>>>>>                           reg = <0x01000000 0x300000>;
>>>>>> @@ -146,10 +250,10 @@
>>>>>>                           reg = <0x01800000 0x80000>;
>>>>>>                           clocks = <&xo_board_clk>,
>>>>>>                                    <&sleep_clk>,
>>>>>> -                                <0>,
>>>>>> -                                <0>,
>>>>>> -                                <0>,
>>>>>> -                                <0>,
>>>>>> +                                <&pcie0_phy>,
>>>>>> +                                <&pcie1_phy>,
>>>>>> +                                <&pcie2_phy>,
>>>>>> +                                <&pcie3_phy>,
>>>>>>                                    <0>;
>>>>>>                           #clock-cells = <1>;
>>>>>>                           #reset-cells = <1>;
>>>>>> @@ -478,6 +582,267 @@
>>>>>>                                   status = "disabled";
>>>>>>                           };
>>>>>>                   };
>>>>>> +
>>>>>> +               pcie1: pci@10000000 {
>>>>>> +                       compatible = "qcom,pcie-ipq9574";
>>>>>> +                       reg =  <0x10000000 0xf1d>,
>>>>>> +                              <0x10000F20 0xa8>,
>>>>>> +                              <0x10001000 0x1000>,
>>>>>> +                              <0x000F8000 0x4000>,
>>>>>> +                              <0x10100000 0x1000>;
>>>>>> +                       reg-names = "dbi", "elbi", "atu", "parf",
>>>>>> "config";
>>>>>> +                       device_type = "pci";
>>>>>> +                       linux,pci-domain = <2>;
>>>>>> +                       bus-range = <0x00 0xff>;
>>>>>> +                       num-lanes = <1>;
>>>>>> +                       #address-cells = <3>;
>>>>>> +                       #size-cells = <2>;
>>>>>> +
>>>>>> +                       ranges = <0x01000000 0x0 0x00000000
>>>>>> 0x10200000 0x0 0x100000>,  /* I/O */
>>>>>> +                                <0x02000000 0x0 0x10300000
>>>>>> 0x10300000 0x0 0x7d00000>; /* MEM */
>>>>>> +
>>>>>> +                       #interrupt-cells = <1>;
>>>>>> +                       interrupt-map-mask = <0 0 0 0x7>;
>>>>>> +                       interrupt-map = <0 0 0 1 &intc 0 35
>>>>>> IRQ_TYPE_LEVEL_HIGH>, /* int_a */
>>>>>> +                                       <0 0 0 2 &intc 0 49
>>>>>> IRQ_TYPE_LEVEL_HIGH>, /* int_b */
>>>>>> +                                       <0 0 0 3 &intc 0 84
>>>>>> IRQ_TYPE_LEVEL_HIGH>, /* int_c */
>>>>>> +                                       <0 0 0 4 &intc 0 85
>>>>>> IRQ_TYPE_LEVEL_HIGH>; /* int_d */
>>>>>> +
>>>>>
>>>>> No iommu-map?
>>>> We do not enable the IOMMU stage1 translation for PCIe and the registers
>>>> have secure access only from TrustZone (It enables only stage2 for
>>>> Access control)
>>>
>>> So, no SMMU protection for PCIe transactions? This sounds like a step
>>> backwards.
>> Yes, we are not using stage1 translations.
> 
> We = software or we = hardware? If there is a hardware interface to
> SMMU, please describe it here.
> 
Trustzone software protects all non-secure access to any SMMU register.
Hence it is not possible to enable stage 1 translation from HLOS.
HLOS touching any SMMU register would result in a 'secure access violation'

Thanks,
Devi Priya