[PATCH 0/4] Describe MT8188's SCP as dual-core

[PATCH 0/4] Describe MT8188's SCP as dual-core

[Nícolas F. R. A. Prado]

This series updates the MT8188 SCP description to dual-core and prepares
its usage. Patch 3 updates the MT8188 DT to describe the SCP as a dual
core cluster and patch 4 adds the firmware-name property so the firmware
can be loaded. Patches 1 and 2 are dt-binding fixes for the SCP.

Depends on the "MediaTek MT8188 MDP3 Enablement" series [1].

[1] https://lore.kernel.org/all/20241218105320.38980-1-angelogioacchino.delregno@collabora.com

Signed-off-by: Nícolas F. R. A. Prado <nfraprado@collabora.com>
---
Nícolas F. R. A. Prado (4):
      dt-bindings: remoteproc: mediatek: Add missing minItems for 8192/8195
      dt-bindings: remoteproc: mediatek: Remove l1tcm for dual-core MT8188 SCP
      arm64: dts: mediatek: mt8188: Describe SCP as a cluster with two cores
      arm64: dts: mediatek: mt8390-genio-common: Add firmware-name for scp0

 .../devicetree/bindings/remoteproc/mtk,scp.yaml    | 20 ++++++++++--
 arch/arm64/boot/dts/mediatek/mt8188-evb.dts        |  6 +++-
 arch/arm64/boot/dts/mediatek/mt8188.dtsi           | 36 ++++++++++++++++------
 .../boot/dts/mediatek/mt8390-genio-common.dtsi     |  7 ++++-
 4 files changed, 55 insertions(+), 14 deletions(-)
---
base-commit: 04b0d54949f9660384ceefc6f707e40daba6bf3e
change-id: 20250318-scp-dual-core-mt8390-8ef1b02f2d92

Best regards,
-- 
Nícolas F. R. A. Prado <nfraprado@collabora.com>

[PATCH 1/4] dt-bindings: remoteproc: mediatek: Add missing minItems for 8192/8195

[Nícolas F. R. A. Prado]

Both MT8192 and MT8195 have an L1TCM memory, so it should be described
in reg, and therefore reg's minItems should be 3, as is already implicit
by reg-names. Override minItems to 3 for MT8192 and MT8195.

Fixes: 6b55b1e2fd7f ("dt-bindings: remoteproc: mediatek: Support MT8195 dual-core SCP")
Signed-off-by: Nícolas F. R. A. Prado <nfraprado@collabora.com>
---
 Documentation/devicetree/bindings/remoteproc/mtk,scp.yaml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/devicetree/bindings/remoteproc/mtk,scp.yaml b/Documentation/devicetree/bindings/remoteproc/mtk,scp.yaml
index adc6b3f36fde49eb3fa7ed3f08b0fa9e7f331162..ee33c233e44f3d51f0851b35697a24208c87f68a 100644
--- a/Documentation/devicetree/bindings/remoteproc/mtk,scp.yaml
+++ b/Documentation/devicetree/bindings/remoteproc/mtk,scp.yaml
@@ -185,6 +185,7 @@ allOf:
     then:
       properties:
         reg:
+          minItems: 3
           maxItems: 3
         reg-names:
           items:

-- 
2.49.0

Re: [PATCH 1/4] dt-bindings: remoteproc: mediatek: Add missing minItems for 8192/8195

[Krzysztof Kozlowski]

On 18/03/2025 23:22, Nícolas F. R. A. Prado wrote:
> Both MT8192 and MT8195 have an L1TCM memory, so it should be described
> in reg, and therefore reg's minItems should be 3, as is already implicit
> by reg-names. Override minItems to 3 for MT8192 and MT8195.

I think that was the original intention already, because reg-names are
expecting min 3 items. Basically you just correct the missing lower
constraint.

> 
> Fixes: 6b55b1e2fd7f ("dt-bindings: remoteproc: mediatek: Support MT8195 dual-core SCP")
> Signed-off-by: Nícolas F. R. A. Prado <nfraprado@collabora.com>
> ---
>  Documentation/devicetree/bindings/remoteproc/mtk,scp.yaml | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/Documentation/devicetree/bindings/remoteproc/mtk,scp.yaml b/Documentation/devicetree/bindings/remoteproc/mtk,scp.yaml
> index adc6b3f36fde49eb3fa7ed3f08b0fa9e7f331162..ee33c233e44f3d51f0851b35697a24208c87f68a 100644
> --- a/Documentation/devicetree/bindings/remoteproc/mtk,scp.yaml
> +++ b/Documentation/devicetree/bindings/remoteproc/mtk,scp.yaml
> @@ -185,6 +185,7 @@ allOf:
>      then:
>        properties:
>          reg:
> +          minItems: 3
>            maxItems: 3

maxItems can be dropped.


Best regards,
Krzysztof

Re: [PATCH 1/4] dt-bindings: remoteproc: mediatek: Add missing minItems for 8192/8195

[Nícolas F. R. A. Prado]

On Wed, Mar 19, 2025 at 08:16:37AM +0100, Krzysztof Kozlowski wrote:
> On 18/03/2025 23:22, Nícolas F. R. A. Prado wrote:
> > Both MT8192 and MT8195 have an L1TCM memory, so it should be described
> > in reg, and therefore reg's minItems should be 3, as is already implicit
> > by reg-names. Override minItems to 3 for MT8192 and MT8195.
> 
> I think that was the original intention already, because reg-names are
> expecting min 3 items. Basically you just correct the missing lower
> constraint.

Yes, exactly, that's what I meant by "is already implicit by reg-names".

> 
> > 
> > Fixes: 6b55b1e2fd7f ("dt-bindings: remoteproc: mediatek: Support MT8195 dual-core SCP")
> > Signed-off-by: Nícolas F. R. A. Prado <nfraprado@collabora.com>
> > ---
> >  Documentation/devicetree/bindings/remoteproc/mtk,scp.yaml | 1 +
> >  1 file changed, 1 insertion(+)
> > 
> > diff --git a/Documentation/devicetree/bindings/remoteproc/mtk,scp.yaml b/Documentation/devicetree/bindings/remoteproc/mtk,scp.yaml
> > index adc6b3f36fde49eb3fa7ed3f08b0fa9e7f331162..ee33c233e44f3d51f0851b35697a24208c87f68a 100644
> > --- a/Documentation/devicetree/bindings/remoteproc/mtk,scp.yaml
> > +++ b/Documentation/devicetree/bindings/remoteproc/mtk,scp.yaml
> > @@ -185,6 +185,7 @@ allOf:
> >      then:
> >        properties:
> >          reg:
> > +          minItems: 3
> >            maxItems: 3
> 
> maxItems can be dropped.

Sure, I can send a patch for that in the next version.

Thanks,
Nícolas

Re: [PATCH 1/4] dt-bindings: remoteproc: mediatek: Add missing minItems for 8192/8195

[AngeloGioacchino Del Regno]

Il 19/03/25 15:25, Nícolas F. R. A. Prado ha scritto:
> On Wed, Mar 19, 2025 at 08:16:37AM +0100, Krzysztof Kozlowski wrote:
>> On 18/03/2025 23:22, Nícolas F. R. A. Prado wrote:
>>> Both MT8192 and MT8195 have an L1TCM memory, so it should be described
>>> in reg, and therefore reg's minItems should be 3, as is already implicit
>>> by reg-names. Override minItems to 3 for MT8192 and MT8195.
>>
>> I think that was the original intention already, because reg-names are
>> expecting min 3 items. Basically you just correct the missing lower
>> constraint.
> 
> Yes, exactly, that's what I meant by "is already implicit by reg-names".
> 

Nicolas, any way you can respin this patch so that we can get SCP for MT8188
in this merge window, please?

If you cannot do that, I can respin it on my own, no problem.

Thanks,
Angelo

>>
>>>
>>> Fixes: 6b55b1e2fd7f ("dt-bindings: remoteproc: mediatek: Support MT8195 dual-core SCP")
>>> Signed-off-by: Nícolas F. R. A. Prado <nfraprado@collabora.com>
>>> ---
>>>   Documentation/devicetree/bindings/remoteproc/mtk,scp.yaml | 1 +
>>>   1 file changed, 1 insertion(+)
>>>
>>> diff --git a/Documentation/devicetree/bindings/remoteproc/mtk,scp.yaml b/Documentation/devicetree/bindings/remoteproc/mtk,scp.yaml
>>> index adc6b3f36fde49eb3fa7ed3f08b0fa9e7f331162..ee33c233e44f3d51f0851b35697a24208c87f68a 100644
>>> --- a/Documentation/devicetree/bindings/remoteproc/mtk,scp.yaml
>>> +++ b/Documentation/devicetree/bindings/remoteproc/mtk,scp.yaml
>>> @@ -185,6 +185,7 @@ allOf:
>>>       then:
>>>         properties:
>>>           reg:
>>> +          minItems: 3
>>>             maxItems: 3
>>
>> maxItems can be dropped.
> 
> Sure, I can send a patch for that in the next version.
> 
> Thanks,
> Nícolas

[PATCH 2/4] dt-bindings: remoteproc: mediatek: Remove l1tcm for dual-core MT8188 SCP

[Nícolas F. R. A. Prado]

The SCP present on MT8188 does not have an L1TCM memory region, but the
binding incorrectly requires one for the dual-core description of the
MT8188 SCP. Remove that requirement. Also update the minimum number of
reg and reg-names to 1, since as this is a multi-core SCP with no
L1TCM memory, only the CFG memory region is present in the parent node.

Fixes: 91e0d560b9fd ("dt-bindings: remoteproc: mediatek: Support MT8188 dual-core SCP")
Signed-off-by: Nícolas F. R. A. Prado <nfraprado@collabora.com>
---
 .../devicetree/bindings/remoteproc/mtk,scp.yaml       | 19 ++++++++++++++++---
 1 file changed, 16 insertions(+), 3 deletions(-)

diff --git a/Documentation/devicetree/bindings/remoteproc/mtk,scp.yaml b/Documentation/devicetree/bindings/remoteproc/mtk,scp.yaml
index ee33c233e44f3d51f0851b35697a24208c87f68a..04348cfcb0424dc78ff1ddd77665285a052925e9 100644
--- a/Documentation/devicetree/bindings/remoteproc/mtk,scp.yaml
+++ b/Documentation/devicetree/bindings/remoteproc/mtk,scp.yaml
@@ -28,11 +28,11 @@ properties:
     description:
       Should contain the address ranges for memory regions SRAM, CFG, and,
       on some platforms, L1TCM.
-    minItems: 2
+    minItems: 1
     maxItems: 3
 
   reg-names:
-    minItems: 2
+    minItems: 1
     maxItems: 3
 
   clocks:
@@ -171,6 +171,7 @@ allOf:
     then:
       properties:
         reg:
+          minItems: 2
           maxItems: 2
         reg-names:
           items:
@@ -196,16 +197,28 @@ allOf:
       properties:
         compatible:
           enum:
-            - mediatek,mt8188-scp-dual
             - mediatek,mt8195-scp-dual
     then:
       properties:
         reg:
+          minItems: 2
           maxItems: 2
         reg-names:
           items:
             - const: cfg
             - const: l1tcm
+  - if:
+      properties:
+        compatible:
+          enum:
+            - mediatek,mt8188-scp-dual
+    then:
+      properties:
+        reg:
+          maxItems: 1
+        reg-names:
+          items:
+            - const: cfg
 
 additionalProperties: false
 

-- 
2.49.0

Re: [PATCH 2/4] dt-bindings: remoteproc: mediatek: Remove l1tcm for dual-core MT8188 SCP

[Krzysztof Kozlowski]

On 18/03/2025 23:22, Nícolas F. R. A. Prado wrote:
> The SCP present on MT8188 does not have an L1TCM memory region, but the
> binding incorrectly requires one for the dual-core description of the
> MT8188 SCP. Remove that requirement. Also update the minimum number of
> reg and reg-names to 1, since as this is a multi-core SCP with no
> L1TCM memory, only the CFG memory region is present in the parent node.
> 
> Fixes: 91e0d560b9fd ("dt-bindings: remoteproc: mediatek: Support MT8188 dual-core SCP")
> Signed-off-by: Nícolas F. R. A. Prado <nfraprado@collabora.com>

Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>

Best regards,
Krzysztof

Re: [PATCH 2/4] dt-bindings: remoteproc: mediatek: Remove l1tcm for dual-core MT8188 SCP

[AngeloGioacchino Del Regno]

Il 18/03/25 23:22, Nícolas F. R. A. Prado ha scritto:
> The SCP present on MT8188 does not have an L1TCM memory region, but the
> binding incorrectly requires one for the dual-core description of the
> MT8188 SCP. Remove that requirement. Also update the minimum number of
> reg and reg-names to 1, since as this is a multi-core SCP with no
> L1TCM memory, only the CFG memory region is present in the parent node.
> 
> Fixes: 91e0d560b9fd ("dt-bindings: remoteproc: mediatek: Support MT8188 dual-core SCP")
> Signed-off-by: Nícolas F. R. A. Prado <nfraprado@collabora.com>

As far as I know, the L1TCM is present on MT8188, and it's at 0x1070000 len 0x8000
exactly like MT8195.

Moudy, are you able to please confirm or deny that?

Cheers,
Angelo

Re: [PATCH 2/4] dt-bindings: remoteproc: mediatek: Remove l1tcm for dual-core MT8188 SCP

[Moudy Ho (何宗原)]

On Wed, 2025-03-19 at 10:34 +0100, AngeloGioacchino Del Regno wrote:
> 
> External email : Please do not click links or open attachments until
> you have verified the sender or the content.
> 
> 
> Il 18/03/25 23:22, Nícolas F. R. A. Prado ha scritto:
> > The SCP present on MT8188 does not have an L1TCM memory region, but
> > the
> > binding incorrectly requires one for the dual-core description of
> > the
> > MT8188 SCP. Remove that requirement. Also update the minimum number
> > of
> > reg and reg-names to 1, since as this is a multi-core SCP with no
> > L1TCM memory, only the CFG memory region is present in the parent
> > node.
> > 
> > Fixes: 91e0d560b9fd ("dt-bindings: remoteproc: mediatek: Support
> > MT8188 dual-core SCP")
> > Signed-off-by: Nícolas F. R. A. Prado <nfraprado@collabora.com>
> 
> As far as I know, the L1TCM is present on MT8188, and it's at
> 0x1070000 len 0x8000
> exactly like MT8195.
> 
> Moudy, are you able to please confirm or deny that?
> 
> Cheers,
> Angelo
> 
> 

Hi Angelo,

Unfortunately, during the transition from MT8195 to MT8188, the SCP
removed the L1TCM, retaining only the SRAM and config register.
For detailed hardware changes, you can refer to Tinghan's previous
description at the following link.

ref:
https://lore.kernel.org/all/20220715051821.30707-3-tinghan.shen@mediatek.com/

Sincerely,
Moudy Ho

Re: [PATCH 2/4] dt-bindings: remoteproc: mediatek: Remove l1tcm for dual-core MT8188 SCP

[AngeloGioacchino Del Regno]

Il 20/03/25 02:28, Moudy Ho (何宗原) ha scritto:
> On Wed, 2025-03-19 at 10:34 +0100, AngeloGioacchino Del Regno wrote:
>>
>> External email : Please do not click links or open attachments until
>> you have verified the sender or the content.
>>
>>
>> Il 18/03/25 23:22, Nícolas F. R. A. Prado ha scritto:
>>> The SCP present on MT8188 does not have an L1TCM memory region, but
>>> the
>>> binding incorrectly requires one for the dual-core description of
>>> the
>>> MT8188 SCP. Remove that requirement. Also update the minimum number
>>> of
>>> reg and reg-names to 1, since as this is a multi-core SCP with no
>>> L1TCM memory, only the CFG memory region is present in the parent
>>> node.
>>>
>>> Fixes: 91e0d560b9fd ("dt-bindings: remoteproc: mediatek: Support
>>> MT8188 dual-core SCP")
>>> Signed-off-by: Nícolas F. R. A. Prado <nfraprado@collabora.com>
>>
>> As far as I know, the L1TCM is present on MT8188, and it's at
>> 0x1070000 len 0x8000
>> exactly like MT8195.
>>
>> Moudy, are you able to please confirm or deny that?
>>
>> Cheers,
>> Angelo
>>
>>
> 
> Hi Angelo,
> 
> Unfortunately, during the transition from MT8195 to MT8188, the SCP
> removed the L1TCM, retaining only the SRAM and config register.
> For detailed hardware changes, you can refer to Tinghan's previous
> description at the following link.
> 
> ref:
> https://lore.kernel.org/all/20220715051821.30707-3-tinghan.shen@mediatek.com/
> 
Thanks Moudy!

Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>

[PATCH 3/4] arm64: dts: mediatek: mt8188: Describe SCP as a cluster with two cores

[Nícolas F. R. A. Prado]

The SCP is currently described in the Devicetree as a single-core
processor, but really it is a cluster with two cores. Describe the full
cluster but enable only core0 on the current mt8188 platforms since
that's the only one usable with the upstream firmware.

Co-developed-by: Tinghan Shen <tinghan.shen@mediatek.com>
Signed-off-by: Tinghan Shen <tinghan.shen@mediatek.com>
Co-developed-by: Jason Chen <jason-ch.chen@mediatek.corp-partner.google.com>
Signed-off-by: Jason Chen <jason-ch.chen@mediatek.corp-partner.google.com>
Signed-off-by: Nícolas F. R. A. Prado <nfraprado@collabora.com>
---
 arch/arm64/boot/dts/mediatek/mt8188-evb.dts        |  6 +++-
 arch/arm64/boot/dts/mediatek/mt8188.dtsi           | 36 ++++++++++++++++------
 .../boot/dts/mediatek/mt8390-genio-common.dtsi     |  6 +++-
 3 files changed, 37 insertions(+), 11 deletions(-)

diff --git a/arch/arm64/boot/dts/mediatek/mt8188-evb.dts b/arch/arm64/boot/dts/mediatek/mt8188-evb.dts
index f89835ac36f36f86e2054ae0c332172be97b049a..f4c207d65b877e1eefaa26540446c3c06369ca21 100644
--- a/arch/arm64/boot/dts/mediatek/mt8188-evb.dts
+++ b/arch/arm64/boot/dts/mediatek/mt8188-evb.dts
@@ -331,7 +331,11 @@ &pmic {
 	interrupts-extended = <&pio 222 IRQ_TYPE_LEVEL_HIGH>;
 };
 
-&scp {
+&scp_cluster {
+	status = "okay";
+};
+
+&scp_c0 {
 	memory-region = <&scp_mem_reserved>;
 	status = "okay";
 };
diff --git a/arch/arm64/boot/dts/mediatek/mt8188.dtsi b/arch/arm64/boot/dts/mediatek/mt8188.dtsi
index 4437b1820f2685281c3a373bf626f72b53503e62..bea3cee9b8e8f0dc656edd85e63f4f6f6645607f 100644
--- a/arch/arm64/boot/dts/mediatek/mt8188.dtsi
+++ b/arch/arm64/boot/dts/mediatek/mt8188.dtsi
@@ -1382,12 +1382,30 @@ gce1: mailbox@10330000 {
 			clocks = <&infracfg_ao CLK_INFRA_AO_GCE2>;
 		};
 
-		scp: scp@10500000 {
-			compatible = "mediatek,mt8188-scp";
-			reg = <0 0x10500000 0 0x100000>,
-			      <0 0x10720000 0 0xe0000>;
-			reg-names = "sram", "cfg";
-			interrupts = <GIC_SPI 462 IRQ_TYPE_LEVEL_HIGH 0>;
+		scp_cluster: scp@10720000 {
+			compatible = "mediatek,mt8188-scp-dual";
+			reg = <0 0x10720000 0 0xe0000>;
+			reg-names = "cfg";
+			#address-cells = <1>;
+			#size-cells = <1>;
+			ranges = <0 0 0x10500000 0x100000>;
+			status = "disabled";
+
+			scp_c0: scp@0 {
+				compatible = "mediatek,scp-core";
+				reg = <0x0 0xd0000>;
+				reg-names = "sram";
+				interrupts = <GIC_SPI 462 IRQ_TYPE_LEVEL_HIGH 0>;
+				status = "disabled";
+			};
+
+			scp_c1: scp@d0000 {
+				compatible = "mediatek,scp-core";
+				reg = <0xd0000 0x2f000>;
+				reg-names = "sram";
+				interrupts = <GIC_SPI 463 IRQ_TYPE_LEVEL_HIGH 0>;
+				status = "disabled";
+			};
 		};
 
 		afe: audio-controller@10b10000 {
@@ -2249,7 +2267,7 @@ dma-controller@14001000 {
 			mediatek,gce-client-reg = <&gce1 SUBSYS_1400XXXX 0x1000 0x1000>;
 			mediatek,gce-events = <CMDQ_EVENT_VPP0_MDP_RDMA_SOF>,
 					      <CMDQ_EVENT_VPP0_MDP_RDMA_FRAME_DONE>;
-			mediatek,scp = <&scp>;
+			mediatek,scp = <&scp_c0>;
 		};
 
 		display@14002000 {
@@ -2704,7 +2722,7 @@ video_decoder: video-decoder@18000000 {
 			iommus = <&vpp_iommu M4U_PORT_L23_HW_VDEC_UFO_ENC_EXT>;
 			#address-cells = <2>;
 			#size-cells = <2>;
-			mediatek,scp = <&scp>;
+			mediatek,scp = <&scp_c0>;
 
 			video-codec@10000 {
 				compatible = "mediatek,mtk-vcodec-lat";
@@ -2828,7 +2846,7 @@ video_encoder: video-encoder@1a020000 {
 				 <&vdo_iommu M4U_PORT_L19_VENC_SUB_W_LUMA>,
 				 <&vdo_iommu M4U_PORT_L19_VENC_SUB_R_LUMA>;
 			power-domains = <&spm MT8188_POWER_DOMAIN_VENC>;
-			mediatek,scp = <&scp>;
+			mediatek,scp = <&scp_c0>;
 		};
 
 		jpeg_encoder: jpeg-encoder@1a030000 {
diff --git a/arch/arm64/boot/dts/mediatek/mt8390-genio-common.dtsi b/arch/arm64/boot/dts/mediatek/mt8390-genio-common.dtsi
index 60139e6dffd8e0e326690d922f3360d829ed026b..65952bcf5d03e3b22984dd3e997e7107e74dac72 100644
--- a/arch/arm64/boot/dts/mediatek/mt8390-genio-common.dtsi
+++ b/arch/arm64/boot/dts/mediatek/mt8390-genio-common.dtsi
@@ -1055,7 +1055,11 @@ power-key {
 	};
 };
 
-&scp {
+&scp_cluster {
+	status = "okay";
+};
+
+&scp_c0 {
 	memory-region = <&scp_mem>;
 	status = "okay";
 };

-- 
2.49.0

[PATCH 4/4] arm64: dts: mediatek: mt8390-genio-common: Add firmware-name for scp0

[Nícolas F. R. A. Prado]

Add the firmware-name property for SCP core0 so the firmware can be
loaded from its canonical location in the linux-firmware repository.

Signed-off-by: Nícolas F. R. A. Prado <nfraprado@collabora.com>
---
 arch/arm64/boot/dts/mediatek/mt8390-genio-common.dtsi | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/arm64/boot/dts/mediatek/mt8390-genio-common.dtsi b/arch/arm64/boot/dts/mediatek/mt8390-genio-common.dtsi
index 65952bcf5d03e3b22984dd3e997e7107e74dac72..53951817df468ee27d20db1f1028c21a1832b880 100644
--- a/arch/arm64/boot/dts/mediatek/mt8390-genio-common.dtsi
+++ b/arch/arm64/boot/dts/mediatek/mt8390-genio-common.dtsi
@@ -1060,6 +1060,7 @@ &scp_cluster {
 };
 
 &scp_c0 {
+	firmware-name = "mediatek/mt8188/scp.img";
 	memory-region = <&scp_mem>;
 	status = "okay";
 };

-- 
2.49.0

Re: [PATCH 4/4] arm64: dts: mediatek: mt8390-genio-common: Add firmware-name for scp0

[AngeloGioacchino Del Regno]

Il 18/03/25 23:22, Nícolas F. R. A. Prado ha scritto:
> Add the firmware-name property for SCP core0 so the firmware can be
> loaded from its canonical location in the linux-firmware repository.
> 
> Signed-off-by: Nícolas F. R. A. Prado <nfraprado@collabora.com>

Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>