[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <4e01f121-1912-4897-98e8-cb543a501707@rock-chips.com>
Date: Fri, 9 Jan 2026 09:26:00 +0800
From: Chaoyi Chen <chaoyi.chen@...k-chips.com>
To: Heiko Stuebner <heiko@...ech.de>,
Neil Armstrong <neil.armstrong@...aro.org>,
Jessica Zhang <jessica.zhang@....qualcomm.com>,
David Airlie <airlied@...il.com>, Simona Vetter <simona@...ll.ch>,
Maarten Lankhorst <maarten.lankhorst@...ux.intel.com>,
Maxime Ripard <mripard@...nel.org>, Thomas Zimmermann <tzimmermann@...e.de>,
Rob Herring <robh@...nel.org>, Krzysztof Kozlowski <krzk+dt@...nel.org>,
Conor Dooley <conor+dt@...nel.org>, Sandy Huang <hjc@...k-chips.com>,
Andy Yan <andy.yan@...k-chips.com>, Vinod Koul <vkoul@...nel.org>,
Kishon Vijay Abraham I <kishon@...nel.org>, Chaoyi Chen <kernel@...kyi.com>
Cc: dri-devel@...ts.freedesktop.org, devicetree@...r.kernel.org,
linux-kernel@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
linux-rockchip@...ts.infradead.org, linux-phy@...ts.infradead.org
Subject: Re: [PATCH 8/9] drm/rockchip: Drop ROCKCHIP_IOMMU depend for
DRM_ROCKCHIP
Hi Heiko,
On 1/9/2026 3:03 AM, Heiko Stuebner wrote:
> Am Donnerstag, 6. November 2025, 03:06:31 Mitteleuropäische Normalzeit schrieb Chaoyi Chen:
>> From: Chaoyi Chen <chaoyi.chen@...k-chips.com>
>>
>> On the RK3506 platform, there is no iommu hardware. And even on
>> platform that have iommu hardware, it should be possible to use
>> VOP without enabling iommu. In this case, a contiguous memory
>> space like CMA should be used.
>>
>> So this patch removes the dependency on ROCKCHIP_IOMMU.
>>
>> Signed-off-by: Chaoyi Chen <chaoyi.chen@...k-chips.com>
>> ---
>> drivers/gpu/drm/rockchip/Kconfig | 3 ++-
>> 1 file changed, 2 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/gpu/drm/rockchip/Kconfig b/drivers/gpu/drm/rockchip/Kconfig
>> index b7b025814e72..a056d419190c 100644
>> --- a/drivers/gpu/drm/rockchip/Kconfig
>> +++ b/drivers/gpu/drm/rockchip/Kconfig
>> @@ -1,7 +1,8 @@
>> # SPDX-License-Identifier: GPL-2.0-only
>> config DRM_ROCKCHIP
>> tristate "DRM Support for Rockchip"
>> - depends on DRM && ROCKCHIP_IOMMU
>> + depends on DRM
>> + depends on ROCKCHIP_IOMMU || !ROCKCHIP_IOMMU
>
> I don't really understand this yes+no line :-)
>
> Can't you just
> - drop the dependency altogether
> or
> - do a depends on ROCKCHIP_IOMMU if ARM64
>
>
This trick is called optional-dependencies [0]. In addition to the
familiar depends on ROCKCHIP_IOMMU part, the newly added !ROCKCHIP_IOMMU
ensures that DRM_ROCKCHIP can still be built even when ROCKCHIP_IOMMU
is not build.
[0]: https://docs.kernel.org/kbuild/kconfig-language.html#optional-dependencies
If we just:
- drop the dependency altogether
When IOMMU is enabled, the dependency relationship cannot be handled
correctly. For example, the following configuration is possible:
ROCKCHIP_IOMMU=m, DRM_ROCKCHIP=y, which leads to a build failure.
- do a depends on ROCKCHIP_IOMMU if ARM64
This changes the semantics. On arm64 we should also be able to work
without IOMMU being enabled.
--
Best,
Chaoyi
Powered by blists - more mailing lists