Message ID | 20231106-am65-dss-clk-edge-v1-1-4a959fec0e1e@ideasonboard.com |
---|---|
State | Accepted |
Commit | b57160859263c083c49482b0d083a586b1517f78 |
Headers | show |
Series | arm64: dts: ti: k3-am65-main: Fix DSS irq trigger type | expand |
diff --git a/arch/arm64/boot/dts/ti/k3-am65-main.dtsi b/arch/arm64/boot/dts/ti/k3-am65-main.dtsi index bc460033a37a..c98068b6c122 100644 --- a/arch/arm64/boot/dts/ti/k3-am65-main.dtsi +++ b/arch/arm64/boot/dts/ti/k3-am65-main.dtsi @@ -1034,7 +1034,7 @@ dss: dss@4a00000 { assigned-clocks = <&k3_clks 67 2>; assigned-clock-parents = <&k3_clks 67 5>; - interrupts = <GIC_SPI 166 IRQ_TYPE_EDGE_RISING>; + interrupts = <GIC_SPI 166 IRQ_TYPE_LEVEL_HIGH>; dma-coherent;
DSS irq trigger type is set to IRQ_TYPE_EDGE_RISING in the DT file, but the TRM says it is level triggered. For some reason triggering on rising edge results in double the amount of expected interrupts, e.g. for normal page flipping test the number of interrupts per second is 2 * fps. It is as if the IRQ triggers on both edges. There are no other side effects to this issue than slightly increased CPU & power consumption due to the extra interrupt. Switching to IRQ_TYPE_LEVEL_HIGH is correct and fixes the issue, so let's do that. Fixes: fc539b90eda2 ("arm64: dts: ti: am654: Add DSS node") Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com> --- arch/arm64/boot/dts/ti/k3-am65-main.dtsi | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) --- base-commit: ffc253263a1375a65fa6c9f62a893e9767fbebfa change-id: 20231106-am65-dss-clk-edge-7d1bdd7c0f58 Best regards,