On 5/25/2021 6:23 AM, Maxime Ripard wrote:
From: Dom Cobley popcornmix@gmail.com
Enable NO_WAIT_RESP, DMA_WIDE_SOURCE, DMA_WIDE_DEST, and bump the DMA panic and AXI priorities to avoid any DMA transfer error with HBR audio (8 channel, 192Hz).
Signed-off-by: Dom Cobley popcornmix@gmail.com Signed-off-by: Maxime Ripard maxime@cerno.tech
arch/arm/boot/dts/bcm2711.dtsi | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/arm/boot/dts/bcm2711.dtsi b/arch/arm/boot/dts/bcm2711.dtsi index 720beec54d61..9d1dde973680 100644 --- a/arch/arm/boot/dts/bcm2711.dtsi +++ b/arch/arm/boot/dts/bcm2711.dtsi @@ -344,7 +344,7 @@ hdmi0: hdmi@7ef00700 { interrupt-names = "cec-tx", "cec-rx", "cec-low", "wakeup", "hpd-connected", "hpd-removed"; ddc = <&ddc0>;
dmas = <&dma 10>;
dmas = <&dma (10 | (1 << 27) | (1 << 24)| (15 << 20) | (10 << 16))>;
This uses DT as a configuration language rather than a description here, but this is most certainly an established practice that the bcm283-dma.c supports, with no validation of the various arguments.. great.
Is there at least an option to move the meaning of this bitfields into include/dt-bindings/dma/bcm2835-dma.h or something like that?