linux-kernel - Re: [PATCH 0/2] arm64: sunxi: a523: Enable Mali GPU

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <CAGb2v67Pu7CTJx3K=3CVSbQ9mNB-4Vrq+iaA+hUWtxdwS2aD0Q@mail.gmail.com>
Date: Tue, 8 Jul 2025 11:43:34 +0800
From: Chen-Yu Tsai <wens@...e.org>
To: Andre Przywara <andre.przywara@....com>
Cc: iuncuim <iuncuim@...il.com>, Rob Herring <robh@...nel.org>, 
	Krzysztof Kozlowski <krzk+dt@...nel.org>, Conor Dooley <conor+dt@...nel.org>, 
	Jernej Skrabec <jernej.skrabec@...il.com>, Samuel Holland <samuel@...lland.org>, 
	devicetree@...r.kernel.org, linux-arm-kernel@...ts.infradead.org, 
	linux-sunxi@...ts.linux.dev, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 0/2] arm64: sunxi: a523: Enable Mali GPU

On Tue, Jul 8, 2025 at 7:46 AM Andre Przywara <andre.przywara@....com> wrote:
>
> On Sun,  6 Jul 2025 10:56:21 +0800
> iuncuim <iuncuim@...il.com> wrote:
>
> > From: Mikhail Kalashnikov <iuncuim@...il.com>
> >
> > This patch adds gpu support on devices with a523 processor. Since there is
> > no support for image output yet, the gpu is disabled after boot up.
> >
> > $ dmesg | grep panfrost
> > [    3.826968] panfrost 1800000.gpu: clock rate = 432000000
> > [    3.832305] panfrost 1800000.gpu: bus_clock rate = 200000000
> > [    3.838353] panfrost 1800000.gpu: mali-g57 id 0x9091 major 0x0 minor 0x1
> >                status 0x0
> > [    3.846050] panfrost 1800000.gpu: features: 00000000,000019f7, issues:
> >                00000001,80000400
> > [    3.854134] panfrost 1800000.gpu: Features: L2:0x07110206 Shader:0x00000000
> >                Tiler:0x00000809 Mem:0x1 MMU:0x00002830 AS:0xff JS:0x7
> > [    3.866011] panfrost 1800000.gpu: shader_present=0x1 l2_present=0x1
> > [    3.874108] [drm] Initialized panfrost 1.3.0 for 1800000.gpu on minor 0
>
> So what confidence do we have that this really works, apart from these
> successful initialisation messages? Has this been shown to work with
> some (yet-out-of-tree) DRM patches, to produce 3D rendered output? I
> wonder if we should wait with upstreaming until we can really test this
> on some kind of screen.

I ran glmark2-es2-drm using vkms, and it seemed to run correctly.
The GPU job interrupt was triggering frequently, and the GPU mmu
interrupt was triggered twice for each run. This was with Mesa
25.0.2 from Debian bookworm backports.

So this is sort of

Tested-by: Chen-Yu Tsai <wens@...e.org>

If anyone has some idea about how to capture the output, I can give that
a try.

> Cheers,
> Andre
>
> >
> > Tested on x96qproplus and walnutpi 2b devices.
> >
> > Based on v6.16-rc4 with patches:
> > https://lore.kernel.org/all/20250628054438.2864220-1-wens@kernel.org
> > https://lore.kernel.org/linux-sunxi/20250628161608.3072968-1-wens@kernel.org
> > https://lore.kernel.org/linux-sunxi/20250627152918.2606728-1-wens@kernel.org/
> >
> >
> > iuncuim (2):
> >   arm64: dts: allwinner: a523: add Mali GPU node
> >   arm64: dts: allwinner: a523: enable Mali GPU for all boards
> >
> >  arch/arm64/boot/dts/allwinner/sun55i-a523.dtsi    | 15 +++++++++++++++
> >  .../boot/dts/allwinner/sun55i-a527-cubie-a5e.dts  |  5 +++++
> >  .../boot/dts/allwinner/sun55i-h728-x96qpro+.dts   |  5 +++++
> >  .../boot/dts/allwinner/sun55i-t527-avaota-a1.dts  |  5 +++++
> >  .../dts/allwinner/sun55i-t527-orangepi-4a.dts     |  5 +++++
> >  5 files changed, 35 insertions(+)
> >
>