arch/arm64/boot/dts/qcom/monaco.dtsi | 49 ++++++++++++++++++++++++++++ 1 file changed, 49 insertions(+)
For some reason we ended up adding only 4 out of 11 compute cb's for
CDSP, add the missing compute cb. This will also improve the end
user-experience by enabling running multiple AI usecases in parallel.
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@oss.qualcomm.com>
---
arch/arm64/boot/dts/qcom/monaco.dtsi | 49 ++++++++++++++++++++++++++++
1 file changed, 49 insertions(+)
diff --git a/arch/arm64/boot/dts/qcom/monaco.dtsi b/arch/arm64/boot/dts/qcom/monaco.dtsi
index 10e799dd4a78..884a49c0b019 100644
--- a/arch/arm64/boot/dts/qcom/monaco.dtsi
+++ b/arch/arm64/boot/dts/qcom/monaco.dtsi
@@ -7739,6 +7739,55 @@ compute-cb@4 {
<&apps_smmu 0x1964 0x0400>;
dma-coherent;
};
+
+ compute-cb@5 {
+ compatible = "qcom,fastrpc-compute-cb";
+ reg = <5>;
+ iommus = <&apps_smmu 0x19c5 0x0400>;
+ dma-coherent;
+ };
+
+ compute-cb@6 {
+ compatible = "qcom,fastrpc-compute-cb";
+ reg = <6>;
+ iommus = <&apps_smmu 0x19c6 0x0400>;
+ dma-coherent;
+ };
+
+ compute-cb@7 {
+ compatible = "qcom,fastrpc-compute-cb";
+ reg = <7>;
+ iommus = <&apps_smmu 0x19c7 0x0400>;
+ dma-coherent;
+ };
+
+ compute-cb@8 {
+ compatible = "qcom,fastrpc-compute-cb";
+ reg = <8>;
+ iommus = <&apps_smmu 0x19c8 0x0400>;
+ dma-coherent;
+ };
+
+ compute-cb@9 {
+ compatible = "qcom,fastrpc-compute-cb";
+ reg = <9>;
+ iommus = <&apps_smmu 0x19c9 0x0400>;
+ dma-coherent;
+ };
+
+ compute-cb@b {
+ compatible = "qcom,fastrpc-compute-cb";
+ reg = <0xb>;
+ iommus = <&apps_smmu 0x19cb 0x0400>;
+ dma-coherent;
+ };
+
+ compute-cb@c {
+ compatible = "qcom,fastrpc-compute-cb";
+ reg = <0xc>;
+ iommus = <&apps_smmu 0x19cc 0x000>;
+ dma-coherent;
+ };
};
};
};
--
2.47.3
On Thu, 26 Mar 2026 12:58:34 +0000, Srinivas Kandagatla wrote:
> For some reason we ended up adding only 4 out of 11 compute cb's for
> CDSP, add the missing compute cb. This will also improve the end
> user-experience by enabling running multiple AI usecases in parallel.
>
>
Applied, thanks!
[1/1] arm64: dts: monaco: extend fastrpc compute cb
commit: af0d19ea5a31261e1e3aebbe7c7e45c57c4df999
Best regards,
--
Bjorn Andersson <andersson@kernel.org>
On Thu, Mar 26, 2026 at 12:58:34PM +0000, Srinivas Kandagatla wrote:
> diff --git a/arch/arm64/boot/dts/qcom/monaco.dtsi b/arch/arm64/boot/dts/qcom/monaco.dtsi
[..]
> +
> + compute-cb@b {
That doesn't match the patternProperties "(compute-)?cb@[0-9]*$".
Is the reg an index, or an "address"? Should we use base 10, or should
we fix the yaml?
Regards,
Bjorn
> + compatible = "qcom,fastrpc-compute-cb";
> + reg = <0xb>;
> + iommus = <&apps_smmu 0x19cb 0x0400>;
> + dma-coherent;
> + };
> +
> + compute-cb@c {
> + compatible = "qcom,fastrpc-compute-cb";
> + reg = <0xc>;
> + iommus = <&apps_smmu 0x19cc 0x000>;
> + dma-coherent;
> + };
> };
> };
> };
> --
> 2.47.3
>
On 3/26/26 1:24 PM, Bjorn Andersson wrote:
> On Thu, Mar 26, 2026 at 12:58:34PM +0000, Srinivas Kandagatla wrote:
>> diff --git a/arch/arm64/boot/dts/qcom/monaco.dtsi b/arch/arm64/boot/dts/qcom/monaco.dtsi
> [..]
>> +
>> + compute-cb@b {
>
> That doesn't match the patternProperties "(compute-)?cb@[0-9]*$".
>
> Is the reg an index, or an "address"? Should we use base 10, or should
> we fix the yaml?
Ah.. sorry for this, will fix this in next spin,
--srini
>
> Regards,
> Bjorn
>
>> + compatible = "qcom,fastrpc-compute-cb";
>> + reg = <0xb>;
>> + iommus = <&apps_smmu 0x19cb 0x0400>;
>> + dma-coherent;
>> + };
>> +
>> + compute-cb@c {
>> + compatible = "qcom,fastrpc-compute-cb";
>> + reg = <0xc>;
>> + iommus = <&apps_smmu 0x19cc 0x000>;
>> + dma-coherent;
>> + };
>> };
>> };
>> };
>> --
>> 2.47.3
>>
© 2016 - 2026 Red Hat, Inc.