From: "Heiko Stübner" <heiko@sntech.de>
To: tomeu@tomeuvizoso.net, robh@kernel.org, krzk+dt@kernel.org,
conor+dt@kernel.org, joro@8bytes.org, will@kernel.org,
robin.murphy@arm.com, ulfh@kernel.org, p.zabel@pengutronix.de,
ogabbay@kernel.org, zhangqing@rock-chips.com,
Jiaxing Hu <gahing@gahingwoo.com>
Cc: royalnet026@gmail.com, abel.vesa@oss.qualcomm.com,
sebastian.reichel@collabora.com, sidong.yang@furiosa.ai,
u.kleine-koenig@baylibre.com, chaoyi.chen@rock-chips.com,
diederik@cknow-tech.com, alchark@flipper.net, leonard@lausen.nl,
jeff.hugo@oss.qualcomm.com, rfoss@kernel.org,
dri-devel@lists.freedesktop.org,
linux-rockchip@lists.infradead.org, iommu@lists.linux.dev,
linux-pm@vger.kernel.org, devicetree@vger.kernel.org,
linux-arm-kernel@lists.infradead.org,
linux-kernel@vger.kernel.org, Jiaxing Hu <gahing@gahingwoo.com>,
chaoyi.chen@rock-chips.com
Subject: Re: [PATCH v15 14/15] arm64: dts: rockchip: add NPU (RKNN) nodes to rk3576
Date: Wed, 04 Nov 2026 11:53:47 +0100 [thread overview]
Message-ID: <2915499.MsWZr2WtbB@diego> (raw)
In-Reply-To: <20261101082307.151957-15-gahing@gahingwoo.com>
Am Sonntag, 1. November 2026, 09:23:06 Mitteleuropäische Normalzeit schrieb Jiaxing Hu:
> Add the two RKNN cores and their IOMMUs, disabled, and label PD_NPU so a
> board can give it the NPU rail. Each core lists both core domains, its own
> first, as tested.
>
> Assign CLK_RKNN_DSU0 594 MHz. Nothing in mainline sets its rate or the NPU
> rail, and at the 786 MHz it comes up at, on the 750 mV a ROCK 4D boots
> with, two cores running together write wrong words, 11 to 25 in each pass
> of 5400 rows. 594 MHz is clean, and Rockchip's OPP table asks only 725 mV
> of its 500 and 600 MHz steps.
>
> Signed-off-by: Jiaxing Hu <gahing@gahingwoo.com>
> ---
> arch/arm64/boot/dts/rockchip/rk3576.dtsi | 74 +++++++++++++++++++++++-
> 1 file changed, 73 insertions(+), 1 deletion(-)
>
> diff --git a/arch/arm64/boot/dts/rockchip/rk3576.dtsi b/arch/arm64/boot/dts/rockchip/rk3576.dtsi
> index ea9229b6e6f5..80943d748bfa 100644
> --- a/arch/arm64/boot/dts/rockchip/rk3576.dtsi
> +++ b/arch/arm64/boot/dts/rockchip/rk3576.dtsi
> @@ -1057,7 +1057,7 @@ power: power-controller {
> #address-cells = <1>;
> #size-cells = <0>;
>
> - power-domain@RK3576_PD_NPU {
> + pd_npu: power-domain@RK3576_PD_NPU {
> reg = <RK3576_PD_NPU>;
> #power-domain-cells = <1>;
> #address-cells = <1>;
> @@ -1284,6 +1284,78 @@ power-domain@RK3576_PD_VO1 {
> };
> };
>
> + rknn_core_0: npu@27700000 {
> + compatible = "rockchip,rk3576-rknn-core";
> + reg = <0x0 0x27700000 0x0 0x1000>,
> + <0x0 0x27701000 0x0 0x1000>,
> + <0x0 0x27703000 0x0 0x1000>;
> + reg-names = "pc", "cna", "core";
> + interrupts = <GIC_SPI 247 IRQ_TYPE_LEVEL_HIGH>;
> + clocks = <&cru ACLK_RKNN0>, <&cru HCLK_RKNN_ROOT>,
> + <&cru CLK_RKNN_DSU0>, <&cru PCLK_NPUTOP_ROOT>,
> + <&cru ACLK_RKNN_CBUF>, <&cru HCLK_RKNN_CBUF>;
> + clock-names = "aclk", "hclk", "npu", "pclk",
> + "aclk_cbuf", "hclk_cbuf";
> + assigned-clocks = <&cru CLK_RKNN_DSU0>;
> + assigned-clock-rates = <594000000>;
> + resets = <&cru SRST_A_RKNN0>;
> + reset-names = "srst_a";
> + power-domains = <&power RK3576_PD_NPU0>, <&power RK3576_PD_NPU1>;
Why do you need both power-domains here?
The naming does suggest each belongs to one core - similar to the rk3588?
If each core really needs both domains, please state this explicitly.
Rockchip sadly does not provide RKNN documentation the first
two parts of the TRM.
Heiko
_______________________________________________
Linux-rockchip mailing list
Linux-rockchip@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-rockchip
next prev parent reply other threads:[~2026-11-04 10:54 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-01 8:22 [PATCH v15 00/15] accel/rocket: RK3576 NPU (RKNN) enablement Jiaxing Hu
2026-11-01 8:22 ` [PATCH v15 01/15] accel/rocket: request the core clocks by name Jiaxing Hu
2026-11-03 22:42 ` Heiko Stübner
2026-11-04 10:21 ` Heiko Stübner
2026-11-01 8:22 ` [PATCH v15 02/15] accel/rocket: take the completion register writes under job_lock Jiaxing Hu
2026-11-03 22:45 ` Heiko Stübner
2026-11-01 8:22 ` [PATCH v15 03/15] accel/rocket: wait for a running IRQ handler before resetting a core Jiaxing Hu
2026-11-03 22:47 ` Heiko Stübner
2026-11-03 22:50 ` Heiko Stübner
2026-11-04 12:55 ` Tomeu Vizoso
2026-11-01 8:22 ` [PATCH v15 04/15] accel/rocket: let the core suspend after a reset Jiaxing Hu
2026-11-04 10:25 ` Heiko Stübner
2026-11-01 8:22 ` [PATCH v15 05/15] accel/rocket: factor the completion tail out of the IRQ handler Jiaxing Hu
2026-11-04 10:27 ` Heiko Stübner
2026-11-01 8:22 ` [PATCH v15 06/15] dt-bindings: npu: rockchip: add rockchip,rk3576-rknn-core Jiaxing Hu
2026-11-01 8:22 ` [PATCH v15 07/15] dt-bindings: power: rockchip: allow resets in a power domain node Jiaxing Hu
2026-11-01 8:23 ` [PATCH v15 08/15] dt-bindings: iommu: rockchip: describe the RK3576 NPU MMU Jiaxing Hu
2026-11-04 11:02 ` Heiko Stübner
2026-11-01 8:23 ` [PATCH v15 09/15] pmdomain: rockchip: add optional per-domain power-on settle delay Jiaxing Hu
2026-11-03 22:34 ` Heiko Stübner
2026-11-01 8:23 ` [PATCH v15 10/15] pmdomain: rockchip: cycle an optional power-domain reset on power-on Jiaxing Hu
2026-11-03 22:36 ` Heiko Stübner
2026-11-01 8:23 ` [PATCH v15 11/15] accel/rocket: select the per-core clock and reset counts from match data Jiaxing Hu
2026-11-02 1:52 ` Chaoyi Chen
2026-11-01 8:23 ` [PATCH v15 12/15] accel/rocket: add RK3576 NPU (RKNN) support Jiaxing Hu
2026-11-01 8:23 ` [PATCH v15 13/15] arm64: dts: rockchip: add NPU core domain clocks and resets to rk3576 Jiaxing Hu
2026-11-01 8:23 ` [PATCH v15 14/15] arm64: dts: rockchip: add NPU (RKNN) nodes " Jiaxing Hu
2026-11-04 10:53 ` Heiko Stübner [this message]
2026-11-01 8:23 ` [PATCH v15 15/15] arm64: dts: rockchip: enable the NPU on rk3576-rock-4d Jiaxing Hu
2026-11-03 22:39 ` [PATCH v15 00/15] accel/rocket: RK3576 NPU (RKNN) enablement Heiko Stübner
2026-11-04 10:09 ` Tomeu Vizoso
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=2915499.MsWZr2WtbB@diego \
--to=heiko@sntech.de \
--cc=abel.vesa@oss.qualcomm.com \
--cc=alchark@flipper.net \
--cc=chaoyi.chen@rock-chips.com \
--cc=conor+dt@kernel.org \
--cc=devicetree@vger.kernel.org \
--cc=diederik@cknow-tech.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=gahing@gahingwoo.com \
--cc=iommu@lists.linux.dev \
--cc=jeff.hugo@oss.qualcomm.com \
--cc=joro@8bytes.org \
--cc=krzk+dt@kernel.org \
--cc=leonard@lausen.nl \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pm@vger.kernel.org \
--cc=linux-rockchip@lists.infradead.org \
--cc=ogabbay@kernel.org \
--cc=p.zabel@pengutronix.de \
--cc=rfoss@kernel.org \
--cc=robh@kernel.org \
--cc=robin.murphy@arm.com \
--cc=royalnet026@gmail.com \
--cc=sebastian.reichel@collabora.com \
--cc=sidong.yang@furiosa.ai \
--cc=tomeu@tomeuvizoso.net \
--cc=u.kleine-koenig@baylibre.com \
--cc=ulfh@kernel.org \
--cc=will@kernel.org \
--cc=zhangqing@rock-chips.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