Changes to build kernel with necessary driver support for Aspeed 2700 on arm64.

[chander-nexthop]

The changes have been tested internally to compile and load fine on a ast2700 eval board.

[mssonicbld]

/azp run

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[mssonicbld]

/azp run

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[mssonicbld]

/azp run

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[mssonicbld]

/azp run

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[mssonicbld]

/azp run

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[mssonicbld]

/azp run

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[mssonicbld]

/azp run

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[mssonicbld]

/azp run

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[mssonicbld]

/azp run

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[mssonicbld]

/azp run

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[saiarcot895]

Can you merge the latest changes from master branch into your branch, for PR check purposes?

[chander-nexthop]

Will do and update the PR.

[mssonicbld]

/azp run

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[oleksandrivantsiv]

@chander-nexthop, as we discussed in the community meeting, please provide the commit hash from the aspeed branch that you used to create the patch.

[chander-nexthop]

@chander-nexthop, as we discussed in the community meeting, please provide the commit hash from the aspeed branch that you used to create the patch.

I referred to this branch - "aspeed-master-v6.12" from "https://github.com/AspeedTech-BMC/linux.git". There is no single git commit I referred to. I had to compare the sources (everything, Makefile, options, DTS and source files) there relevant to AST2720 in that branch, brought them into a checked out trixe tree, then took a diff.

[oleksandrivantsiv]

@chander-nexthop, as we discussed in the community meeting, please provide the commit hash from the aspeed branch that you used to create the patch.

I referred to this branch - "aspeed-master-v6.12" from "https://github.com/AspeedTech-BMC/linux.git". There is no single git commit I referred to. I had to compare the sources (everything, Makefile, options, DTS and source files) there relevant to AST2720 in that branch, brought them into a checked out trixe tree, then took a diff.

@chander-nexthop, what was the hash of the top commit in the aspeed-master-v6.12 branch at the time you checked it? The branch is not frozen and receives multiple commits every day. It’s important to know the commit hash to understand which commits are included in the patch you created and which are not.

[oleksandrivantsiv]

@chander-nexthop, as we discussed in the community meeting, please provide the commit hash from the aspeed branch that you used to create the patch.

I referred to this branch - "aspeed-master-v6.12" from "https://github.com/AspeedTech-BMC/linux.git". There is no single git commit I referred to. I had to compare the sources (everything, Makefile, options, DTS and source files) there relevant to AST2720 in that branch, brought them into a checked out trixe tree, then took a diff.

@chander-nexthop, what was the hash of the top commit in the aspeed-master-v6.12 branch at the time you checked it? The branch is not frozen and receives multiple commits every day. It’s important to know the commit hash to understand which commits are included in the patch you created and which are not.

Adding @saiarcot895, @yxieca to help with the question

[chander-nexthop]

@chander-nexthop, as we discussed in the community meeting, please provide the commit hash from the aspeed branch that you used to create the patch.

I referred to this branch - "aspeed-master-v6.12" from "https://github.com/AspeedTech-BMC/linux.git". There is no single git commit I referred to. I had to compare the sources (everything, Makefile, options, DTS and source files) there relevant to AST2720 in that branch, brought them into a checked out trixe tree, then took a diff.

@chander-nexthop, what was the hash of the top commit in the aspeed-master-v6.12 branch at the time you checked it? The branch is not frozen and receives multiple commits every day. It’s important to know the commit hash to understand which commits are included in the patch you created and which are not.

Adding @saiarcot895, @yxieca to help with the question

a328e733183a750824ab796f81250e05ad4df813
Author: aspeedyh yh_chung@aspeedtech.com
Date: Fri Nov 21 14:49:51 2025 +0800

dts: remove eDAF 128MB macro definition

remove eDAF 128MB macro definition since it's not used in this dts file.

On top the above, I had to introduce local change, a small one to help build on 6.12 kernel

`commit 14be35163d8606412617e3866d33192d9ec0b2fb (HEAD -> local-patches)
Author: NextHop Build System <build@nexthop.ai>
Date:   Mon Jan 12 10:43:58 2026 +0000

    Additional fixes

diff --git a/drivers/soc/aspeed/Kconfig b/drivers/soc/aspeed/Kconfig
index 01a0ab95d..60f356bbb 100644
--- a/drivers/soc/aspeed/Kconfig
+++ b/drivers/soc/aspeed/Kconfig
@@ -77,10 +77,11 @@ config ASPEED_SOCINFO
 
 config ASPEED_SBC
        bool "ASPEED Secure Boot Controller driver"
+       depends on ARM
        default MACH_ASPEED_G6
        help
          Say yes to provide information about the secure boot controller in
-         debugfs.
+         debugfs. This is only for AST2600 (ARM32).
 
 config ASPEED_XDMA
        tristate "ASPEED XDMA Engine Driver"
diff --git a/drivers/soc/aspeed/aspeed-ssp.c b/drivers/soc/aspeed/aspeed-ssp.c
index 359b3ae85..aa1bb4ccd 100644
--- a/drivers/soc/aspeed/aspeed-ssp.c
+++ b/drivers/soc/aspeed/aspeed-ssp.c
@@ -70,7 +70,7 @@ static const struct file_operations ast_ssp_fops = {
        .owner          = THIS_MODULE,
        .open           = ast_ssp_open,
        .release        = ast_ssp_release,
-       .llseek         = no_llseek,
+       .llseek         = noop_llseek,
 };
 
 struct miscdevice ast_ssp_misc = {
@@ -240,7 +240,7 @@ static int ast_ssp_probe(struct platform_device *pdev)
        return ret;
 }
 
-static int ast_ssp_remove(struct platform_device *pdev)
+static void ast_ssp_remove(struct platform_device *pdev)
 {
        struct ast2600_ssp *priv = platform_get_drvdata(pdev);
        int i;
@@ -253,8 +253,6 @@ static int ast_ssp_remove(struct platform_device *pdev)
        kfree(priv);
 
        misc_deregister((struct miscdevice *)&ast_ssp_misc);
-
-       return 0;
 }
 
 static const struct of_device_id of_ast_ssp_match_table[] = {`

[oleksandrivantsiv]

@chander-nexthop, as we discussed in the community meeting, please provide the commit hash from the aspeed branch that you used to create the patch.

I referred to this branch - "aspeed-master-v6.12" from "https://github.com/AspeedTech-BMC/linux.git". There is no single git commit I referred to. I had to compare the sources (everything, Makefile, options, DTS and source files) there relevant to AST2720 in that branch, brought them into a checked out trixe tree, then took a diff.

@chander-nexthop, what was the hash of the top commit in the aspeed-master-v6.12 branch at the time you checked it? The branch is not frozen and receives multiple commits every day. It’s important to know the commit hash to understand which commits are included in the patch you created and which are not.

Adding @saiarcot895, @yxieca to help with the question

a328e733183a750824ab796f81250e05ad4df813 Author: aspeedyh yh_chung@aspeedtech.com Date: Fri Nov 21 14:49:51 2025 +0800

dts: remove eDAF 128MB macro definition

remove eDAF 128MB macro definition since it's not used in this dts file.

On top the above, I had to introduce local change, a small one to help build on 6.12 kernel

`commit 14be35163d8606412617e3866d33192d9ec0b2fb (HEAD -> local-patches)
Author: NextHop Build System <build@nexthop.ai>
Date:   Mon Jan 12 10:43:58 2026 +0000

    Additional fixes

diff --git a/drivers/soc/aspeed/Kconfig b/drivers/soc/aspeed/Kconfig
index 01a0ab95d..60f356bbb 100644
--- a/drivers/soc/aspeed/Kconfig
+++ b/drivers/soc/aspeed/Kconfig
@@ -77,10 +77,11 @@ config ASPEED_SOCINFO
 
 config ASPEED_SBC
        bool "ASPEED Secure Boot Controller driver"
+       depends on ARM
        default MACH_ASPEED_G6
        help
          Say yes to provide information about the secure boot controller in
-         debugfs.
+         debugfs. This is only for AST2600 (ARM32).
 
 config ASPEED_XDMA
        tristate "ASPEED XDMA Engine Driver"
diff --git a/drivers/soc/aspeed/aspeed-ssp.c b/drivers/soc/aspeed/aspeed-ssp.c
index 359b3ae85..aa1bb4ccd 100644
--- a/drivers/soc/aspeed/aspeed-ssp.c
+++ b/drivers/soc/aspeed/aspeed-ssp.c
@@ -70,7 +70,7 @@ static const struct file_operations ast_ssp_fops = {
        .owner          = THIS_MODULE,
        .open           = ast_ssp_open,
        .release        = ast_ssp_release,
-       .llseek         = no_llseek,
+       .llseek         = noop_llseek,
 };
 
 struct miscdevice ast_ssp_misc = {
@@ -240,7 +240,7 @@ static int ast_ssp_probe(struct platform_device *pdev)
        return ret;
 }
 
-static int ast_ssp_remove(struct platform_device *pdev)
+static void ast_ssp_remove(struct platform_device *pdev)
 {
        struct ast2600_ssp *priv = platform_get_drvdata(pdev);
        int i;
@@ -253,8 +253,6 @@ static int ast_ssp_remove(struct platform_device *pdev)
        kfree(priv);
 
        misc_deregister((struct miscdevice *)&ast_ssp_misc);
-
-       return 0;
 }
 
 static const struct of_device_id of_ast_ssp_match_table[] = {`

Thanks!