Devicetree
 help / color / mirror / Atom feed
From: sashiko-bot@kernel.org
To: "Xilin Wu" <sophon@radxa.com>
Cc: linux-pci@vger.kernel.org, dri-devel@lists.freedesktop.org,
	devicetree@vger.kernel.org, linux-gpio@vger.kernel.org,
	robh@kernel.org, conor+dt@kernel.org
Subject: Re: [PATCH v4 1/4] arm64: dts: qcom: sc8280xp: enable PCIe ECAM and 64-bit memory
Date: Thu, 29 Oct 2026 10:30:10 +0000	[thread overview]
Message-ID: <sashiko-outbox-189134@kernel.org> (raw)
In-Reply-To: <20261029-q8b-dts-v4-1-e8e7438f5545@radxa.com>

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] arm64: dts: qcom: sa8540p: missing reg and ranges on pcie2b, pcie3b, pcie4
--

commit 835660c7a888130654e80196f354e12277da4a3f
Author: Xilin Wu <sophon@radxa.com>

arm64: dts: qcom: sc8280xp: enable PCIe ECAM and 64-bit memory

This commit updates the SC8280XP PCIe nodes to use 64-bit ECAM configuration
spaces and adds 64-bit memory windows, while preserving the 32-bit address
map on SA8540P for PCIe2a.

> diff --git a/arch/arm64/boot/dts/qcom/sa8540p.dtsi b/arch/arm64/boot/dts/qcom/sa8540p.dtsi
> index 23888029cc117..e5393a237497c 100644
> --- a/arch/arm64/boot/dts/qcom/sa8540p.dtsi
> +++ b/arch/arm64/boot/dts/qcom/sa8540p.dtsi
> @@ -180,6 +180,15 @@ &gpu_smmu {
>  
>  &pcie2a {
>  	compatible = "qcom,pcie-sa8540p";
> +	reg = <0x0 0x01c20000 0x0 0x3000>,
> +	      <0x0 0x3c000000 0x0 0xf1d>,
> +	      <0x0 0x3c000f20 0x0 0xa8>,
> +	      <0x0 0x3c001000 0x0 0x1000>,
> +	      <0x0 0x3c100000 0x0 0x100000>,
> +	      <0x0 0x01c23000 0x0 0x1000>;
> +
> +	ranges = <0x01000000 0x0 0x00000000 0x0 0x3c200000 0x0 0x100000>,
> +		 <0x02000000 0x0 0x3c300000 0x0 0x3c300000 0x0 0x1d00000>;

[Severity: High]
Should the legacy 32-bit reg and ranges properties also be preserved on
&pcie2b, &pcie3b, and &pcie4 in arch/arm64/boot/dts/qcom/sa8540p.dtsi?

This patch updates all five PCIe controllers in
arch/arm64/boot/dts/qcom/sc8280xp.dtsi (pcie4, pcie3b, pcie3a, pcie2b, and
pcie2a) to use the 64-bit ECAM config and MMIO address windows, but
arch/arm64/boot/dts/qcom/sa8540p.dtsi only preserves the 32-bit windows for
&pcie2a (and &pcie3a, which already overrides reg and ranges):

arch/arm64/boot/dts/qcom/sa8540p.dtsi:
    &pcie2b {
        compatible = "qcom,pcie-sa8540p";

        linux,pci-domain = <1>;

        interrupts = <GIC_SPI 263 IRQ_TYPE_LEVEL_HIGH>;
        interrupt-names = "msi";
    };
    ...
    &pcie3b {
        compatible = "qcom,pcie-sa8540p";

        linux,pci-domain = <3>;

        interrupts = <GIC_SPI 565 IRQ_TYPE_LEVEL_HIGH>;
        interrupt-names = "msi";
    };

    &pcie4 {
        compatible = "qcom,pcie-sa8540p";

        linux,pci-domain = <4>;

        interrupts = <GIC_SPI 518 IRQ_TYPE_LEVEL_HIGH>;
        interrupt-names = "msi";
    };

Meanwhile, arch/arm64/boot/dts/qcom/sa8295p-adp.dts includes sa8540p.dtsi and
enables both &pcie3b and &pcie4 without overriding reg or ranges:

arch/arm64/boot/dts/qcom/sa8295p-adp.dts:
    &pcie3b {
        pinctrl-names = "default";
        pinctrl-0 = <&pcie3b_default>;

        status = "okay";
    };
    ...
    &pcie4 {
        pinctrl-names = "default";
        pinctrl-0 = <&pcie4_default>;

        status = "okay";
    };

As noted in commit b8bf63f8eb72 ("arm64: dts: qcom: sa8540p-ride: enable PCIe
support"), accessing unconfigured PCIe address windows on SA8540P/SA8295P
TrustZone firmware causes the board to crash into ramdump mode. When
qcom_pcie_probe() -> dw_pcie_host_init() runs on SA8295P-ADP (or any SA8540P
board enabling pcie2b, pcie3b, or pcie4), won't those controllers inherit the
SC8280XP 64-bit address ranges (0x4_00000000..0x4_ffffffff) and trigger a
crash at boot?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261029-q8b-dts-v4-0-e8e7438f5545@radxa.com?part=1

  reply	other threads:[~2026-10-29 10:30 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-10-29 10:13 [PATCH v4 0/4] arm64: dts: qcom: Add support for Radxa Dragon Q8B Xilin Wu
2026-10-29 10:13 ` [PATCH v4 1/4] arm64: dts: qcom: sc8280xp: enable PCIe ECAM and 64-bit memory Xilin Wu
2026-10-29 10:30   ` sashiko-bot [this message]
2026-10-29 10:13 ` [PATCH v4 2/4] dt-bindings: arm: qcom: document Radxa Dragon Q8B Xilin Wu
2026-10-29 10:13 ` [PATCH v4 3/4] arm64: dts: qcom: sc8280xp: add " Xilin Wu
2026-10-29 10:44   ` sashiko-bot
2026-10-29 10:13 ` [PATCH v4 4/4] firmware: qcom: scm: Allow QSEECOM for " Xilin Wu

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=sashiko-outbox-189134@kernel.org \
    --to=sashiko-bot@kernel.org \
    --cc=conor+dt@kernel.org \
    --cc=devicetree@vger.kernel.org \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=linux-gpio@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=robh@kernel.org \
    --cc=sashiko-reviews@lists.linux.dev \
    --cc=sophon@radxa.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox