diff mbox

[v2,1/2] ARM: dts: rockchip: reserve unusable memory region on rk3288

Message ID 1763907.pckMEaRNPs@diego (mailing list archive)
State New, archived
Headers show

Commit Message

Heiko Stübner Aug. 6, 2015, 5:33 p.m. UTC
The all current Rockchip SoCs supporting 4GB of ram have problems accessing
the memory region 0xfe000000~0xff000000. This also seems to includes the
rk3368 arm64 soc.

All current code handling dma memory oddities I could find, seem to involve
soc-specific code (zone-dma or so) while this issue is shared between arm32
and arm64 socs from Rockchip, which would need to have this described in
the soc devicetree on both socs.

Limiting the dma-zone alone also does not solve the issue and as the
dma-masks need to be a power-of-two in the kernel, the next lower dma-mask
brings memory usable for dma down to 2GB.

So as a stop-gap block off the affected region to prevent its use by
devices with 4GB of memory, like some recent Chromebooks.

Signed-off-by: Heiko Stuebner <heiko@sntech.de>
---
changes since v1:
- expand reasons for the reserved memory

 arch/arm/boot/dts/rk3288.dtsi | 20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

Comments

Doug Anderson Aug. 7, 2015, 9:31 p.m. UTC | #1
Heiko,

On Thu, Aug 6, 2015 at 10:33 AM, Heiko Stübner <heiko@sntech.de> wrote:
> The all current Rockchip SoCs supporting 4GB of ram have problems accessing
> the memory region 0xfe000000~0xff000000. This also seems to includes the
> rk3368 arm64 soc.
>
> All current code handling dma memory oddities I could find, seem to involve
> soc-specific code (zone-dma or so) while this issue is shared between arm32
> and arm64 socs from Rockchip, which would need to have this described in
> the soc devicetree on both socs.
>
> Limiting the dma-zone alone also does not solve the issue and as the
> dma-masks need to be a power-of-two in the kernel, the next lower dma-mask
> brings memory usable for dma down to 2GB.
>
> So as a stop-gap block off the affected region to prevent its use by
> devices with 4GB of memory, like some recent Chromebooks.
>
> Signed-off-by: Heiko Stuebner <heiko@sntech.de>
> ---
> changes since v1:
> - expand reasons for the reserved memory
>
>  arch/arm/boot/dts/rk3288.dtsi | 20 ++++++++++++++++++++
>  1 file changed, 20 insertions(+)

As discussed earlier, I'm now convinced that this is an appropriate
stopgap.  ...and the description looks good to me.

Reviewed-by: Douglas Anderson <dianders@chromium.org>
diff mbox

Patch

diff --git a/arch/arm/boot/dts/rk3288.dtsi b/arch/arm/boot/dts/rk3288.dtsi
index 47a15aa..118fe74 100644
--- a/arch/arm/boot/dts/rk3288.dtsi
+++ b/arch/arm/boot/dts/rk3288.dtsi
@@ -169,6 +169,26 @@ 
 		};
 	};
 
+	reserved-memory {
+		#address-cells = <1>;
+		#size-cells = <1>;
+		ranges;
+
+		/*
+		 * The rk3288 cannot use the memory area above 0xfe000000
+		 * for dma operations for some reason. While there is
+		 * probably a better solution available somewhere, we
+		 * haven't found it yet and while devices with 2GB of ram
+		 * are not affected, this issue prevents 4GB from booting.
+		 * So to make these devices at least bootable, block
+		 * this area for the time being until the real solution
+		 * is found.
+		 */
+		dma-unusable@fe000000 {
+			reg = <0xfe000000 0x1000000>;
+		};
+	};
+
 	xin24m: oscillator {
 		compatible = "fixed-clock";
 		clock-frequency = <24000000>;