[PATCH v3 0/2] MT8189 SMI SUPPORT

[PATCH v3 0/2] MT8189 SMI SUPPORT

[Congcong Yao]

Based on tag: next-20260629, linux-next/master

This patchset add mt8189 smi support.

---
Changes in v3:
- Change the clock numbers of smi-sub-common to minium 3
- Link to v2:
  https://lore.kernel.org/linux-mediatek/20260427070444.20247-1-zhengnan.chen@mediatek.com/
---

Zhengnan Chen (2):
  dt-bindings: memory-controllers: mtk-smi: Add support for mt8189
  memory: mtk-smi: Add mt8189 support

 .../mediatek,smi-common.yaml                  | 18 ++++++++
 .../memory-controllers/mediatek,smi-larb.yaml |  3 ++
 drivers/memory/mtk-smi.c                      | 44 +++++++++++++++++++
 3 files changed, 65 insertions(+)

-- 
2.43.0

[PATCH v3 1/2] dt-bindings: memory-controllers: mtk-smi: Add support for mt8189

[Congcong Yao]

From: Zhengnan Chen <zhengnan.chen@mediatek.com>

Add smi larb, common and sub-commom binding description for mt8189.

About what smi-sub-common is, please check the below diagram,
we add it in mediatek,smi-common.yaml file.

Signed-off-by: Zhengnan Chen <zhengnan.chen@mediatek.com>
Signed-off-by: Congcong Yao <congcong.yao@mediatek.com>
---
 .../mediatek,smi-common.yaml                   | 18 ++++++++++++++++++
 .../memory-controllers/mediatek,smi-larb.yaml  |  3 +++
 2 files changed, 21 insertions(+)

diff --git a/Documentation/devicetree/bindings/memory-controllers/mediatek,smi-common.yaml b/Documentation/devicetree/bindings/memory-controllers/mediatek,smi-common.yaml
index 0762e0ff66ef..4e1deeff92b1 100644
--- a/Documentation/devicetree/bindings/memory-controllers/mediatek,smi-common.yaml
+++ b/Documentation/devicetree/bindings/memory-controllers/mediatek,smi-common.yaml
@@ -25,6 +25,21 @@ description: |
   SMI generation 1 to transform the smi clock into emi clock domain, but that is
   not needed for SMI generation 2.
 
+  The smi-common connects with smi-larb and IOMMU. The maximum inputs number of
+  a smi-common is 8. In SMI generation 2, the engines number may be over 8.
+  In this case, we use a smi-sub-common to merge some larbs.
+  The block diagram something is like:
+
+                          IOMMU
+                           | |
+                       smi-common
+              ---------------------------
+               |          |           ...
+              larb0   sub-common      ...  <-max number is 8
+                   ----------------
+                    |     |    ...
+                  larb1 larbX  ...  <-max number is 8
+
 properties:
   compatible:
     oneOf:
@@ -40,6 +55,8 @@ properties:
           - mediatek,mt8186-smi-common
           - mediatek,mt8188-smi-common-vdo
           - mediatek,mt8188-smi-common-vpp
+          - mediatek,mt8189-smi-common
+          - mediatek,mt8189-smi-sub-common
           - mediatek,mt8192-smi-common
           - mediatek,mt8195-smi-common-vdo
           - mediatek,mt8195-smi-common-vpp
@@ -108,6 +125,7 @@ allOf:
         compatible:
           contains:
             enum:
+              - mediatek,mt8189-smi-sub-common
               - mediatek,mt8195-smi-sub-common
     then:
       required:
diff --git a/Documentation/devicetree/bindings/memory-controllers/mediatek,smi-larb.yaml b/Documentation/devicetree/bindings/memory-controllers/mediatek,smi-larb.yaml
index 2e7fac4b5094..9a5dafd7c07e 100644
--- a/Documentation/devicetree/bindings/memory-controllers/mediatek,smi-larb.yaml
+++ b/Documentation/devicetree/bindings/memory-controllers/mediatek,smi-larb.yaml
@@ -27,6 +27,7 @@ properties:
           - mediatek,mt8183-smi-larb
           - mediatek,mt8186-smi-larb
           - mediatek,mt8188-smi-larb
+          - mediatek,mt8189-smi-larb
           - mediatek,mt8192-smi-larb
           - mediatek,mt8195-smi-larb
 
@@ -85,6 +86,7 @@ allOf:
             - mediatek,mt8183-smi-larb
             - mediatek,mt8186-smi-larb
             - mediatek,mt8188-smi-larb
+            - mediatek,mt8189-smi-larb
             - mediatek,mt8195-smi-larb
 
     then:
@@ -119,6 +121,7 @@ allOf:
               - mediatek,mt6779-smi-larb
               - mediatek,mt8186-smi-larb
               - mediatek,mt8188-smi-larb
+              - mediatek,mt8189-smi-larb
               - mediatek,mt8192-smi-larb
               - mediatek,mt8195-smi-larb
 
-- 
2.43.0

Re: [PATCH v3 1/2] dt-bindings: memory-controllers: mtk-smi: Add support for mt8189

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
- [Medium] The newly added compatible string `mediatek,mt8189-smi-common` is omitted from the `if-then` generational conditional blocks that define strict clock constraints.
- [Low] The SoC generation list in the binding description was not updated to include mt8189.
--

commit 6c2bad5d055d1f72dd33d20414abab2cdf2654cb
Author: Zhengnan Chen <zhengnan.chen@mediatek.com>

dt-bindings: memory-controllers: mtk-smi: Add support for mt8189
    
Add smi larb, common and sub-commom binding description for mt8189.

> diff --git a/Documentation/devicetree/bindings/memory-controllers/mediatek,smi-common.yaml b/Documentation/devicetree/bindings/memory-controllers/mediatek,smi-common.yaml
> index 0762e0ff66ef0..4e1deeff92b14 100644
> --- a/Documentation/devicetree/bindings/memory-controllers/mediatek,smi-common.yaml
> +++ b/Documentation/devicetree/bindings/memory-controllers/mediatek,smi-common.yaml
> @@ -25,6 +25,21 @@ description: |
>    SMI generation 1 to transform the smi clock into emi clock domain, but that is
>    not needed for SMI generation 2.

[Severity: Low]
This isn't a bug, but the generation 2 SoC list earlier in this description
block reads: "generation 2: mt2712, mt6779, mt8167, mt8173, mt8183, mt8186,
mt8188, mt8192 and mt8195." Should mt8189 be added to this list as well?

> +  The smi-common connects with smi-larb and IOMMU. The maximum inputs number of
> +  a smi-common is 8. In SMI generation 2, the engines number may be over 8.

[ ... ]

> @@ -40,6 +55,8 @@ properties:
>            - mediatek,mt8186-smi-common
>            - mediatek,mt8188-smi-common-vdo
>            - mediatek,mt8188-smi-common-vpp
> +          - mediatek,mt8189-smi-common
> +          - mediatek,mt8189-smi-sub-common

[Severity: Medium]
Is mediatek,mt8189-smi-common missing from the generational conditional blocks
that enforce strict clock constraints?

Looking further down in mediatek,smi-common.yaml at the allOf block:

  - if:  # for gen2 HW that have gals
      properties:
        compatible:
          enum:
            - mediatek,mt6779-smi-common
            - mediatek,mt8183-smi-common
            ...

If mt8189-smi-common shares the same clock constraints as these Gen 2 devices,
omitting it from this conditional block will cause device trees with invalid
clock configurations to silently pass validation. Should it be added to the
enum list here (or the corresponding Gen2 block without gals)?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260701074533.175803-1-Congcong.Yao@mediatek.com?part=1

[PATCH v3 2/2] memory: mtk-smi: Add mt8189 support

[Congcong Yao]

From: Zhengnan Chen <zhengnan.chen@mediatek.com>

Add the necessary platform data and ostdl setting to enable support
for mt8189 smi.

Signed-off-by: Zhengnan Chen <zhengnan.chen@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Yong Wu <yong.wu@mediatek.com>
Signed-off-by: Congcong Yao <congcong.yao@mediatek.com>
---
 drivers/memory/mtk-smi.c | 44 ++++++++++++++++++++++++++++++++++++++++
 1 file changed, 44 insertions(+)

diff --git a/drivers/memory/mtk-smi.c b/drivers/memory/mtk-smi.c
index aaeba8ab211e..f2d5462af681 100644
--- a/drivers/memory/mtk-smi.c
+++ b/drivers/memory/mtk-smi.c
@@ -401,6 +401,30 @@ static const u8 mtk_smi_larb_mt8188_ostd[][SMI_LARB_PORT_NR_MAX] = {
 	[25] = {0x01},
 };
 
+static const u8 mtk_smi_larb_mt8189_ostd[][SMI_LARB_PORT_NR_MAX] = {
+	[0] = {0x8, 0x20, 0x20, 0x20, 0x20, 0x20, 0x10, 0x0,},
+	[1] = {0x8, 0x20, 0x20, 0x20, 0x20, 0x20, 0x10, 0x0,},
+	[2] = {0x7, 0x7, 0x4, 0x4, 0x0, 0x0, 0x2, 0x2, 0x7, 0x7, 0x0,},
+	[4] = {0x2F, 0x1E, 0x9, 0x1, 0x1, 0x1, 0x1, 0x2, 0x2, 0x5, 0x1, 0x17,},
+	[7] = {0x20, 0x2, 0x1, 0x1, 0x1, 0x4, 0x2, 0x1, 0x1, 0x2, 0x3, 0x2,
+	       0xA, 0xF, 0x4, 0x6, 0x5, 0x1,},
+	[9] = {0x6, 0x3, 0xC, 0x6, 0x1, 0x4, 0x3, 0x1, 0x2, 0x4, 0x5, 0x2,
+	       0x4, 0x2, 0x3, 0xB, 0x1, 0x4, 0x1, 0x1, 0x1, 0x1, 0x1, 0x1,
+	       0x1, 0x1,},
+	[11] = {0x1, 0x1, 0x1, 0x1, 0x1, 0x1, 0x1, 0x1, 0x1, 0x1, 0x1, 0x1,
+		0x1, 0x1, 0x1, 0xB, 0x1, 0x4, 0x6, 0x5, 0x6, 0x1, 0x5, 0x2,
+		0x9, 0x5,},
+	[13] = {0x2, 0x8, 0x8, 0x8, 0x4, 0x4, 0x4, 0x4, 0x4, 0xE, 0x4, 0x1,
+		0x6, 0x6, 0x2,},
+	[14] = {0x1, 0x1, 0x1, 0x20, 0xE, 0x4, 0x8, 0x8, 0x6, 0x4,},
+	[16] = {0x1E, 0xC, 0x2, 0x8, 0xE, 0x2, 0x1E, 0x10, 0x4, 0x2, 0x2, 0x2,
+		0x2, 0x2, 0x4, 0x2, 0x4,},
+	[17] = {0x1E, 0xC, 0x2, 0x8, 0xE, 0x2, 0x1E, 0x10, 0x4, 0x2, 0x2, 0x2,
+		0x2, 0x2, 0x4, 0x2, 0x4,},
+	[19] = {0x2, 0x1, 0x3, 0x1,},
+	[20] = {0x7, 0x7, 0x3, 0x3, 0x1, 0x1,},
+};
+
 static const u8 mtk_smi_larb_mt8192_ostd[][SMI_LARB_PORT_NR_MAX] = {
 	[0] = {0x2, 0x2, 0x28, 0xa, 0xc, 0x28,},
 	[1] = {0x2, 0x2, 0x18, 0x18, 0x18, 0xa, 0xc, 0x28,},
@@ -533,6 +557,13 @@ static const struct mtk_smi_larb_gen mtk_smi_larb_mt8188 = {
 	.ostd		            = mtk_smi_larb_mt8188_ostd,
 };
 
+static const struct mtk_smi_larb_gen mtk_smi_larb_mt8189 = {
+	.config_port                = mtk_smi_larb_config_port_gen2_general,
+	.flags_general	            = MTK_SMI_FLAG_THRT_UPDATE | MTK_SMI_FLAG_SW_FLAG |
+				      MTK_SMI_FLAG_SLEEP_CTL | MTK_SMI_FLAG_CFG_PORT_SEC_CTL,
+	.ostd		            = mtk_smi_larb_mt8189_ostd,
+};
+
 static const struct mtk_smi_larb_gen mtk_smi_larb_mt8192 = {
 	.config_port                = mtk_smi_larb_config_port_gen2_general,
 	.ostd			    = mtk_smi_larb_mt8192_ostd,
@@ -556,6 +587,7 @@ static const struct of_device_id mtk_smi_larb_of_ids[] = {
 	{.compatible = "mediatek,mt8183-smi-larb", .data = &mtk_smi_larb_mt8183},
 	{.compatible = "mediatek,mt8186-smi-larb", .data = &mtk_smi_larb_mt8186},
 	{.compatible = "mediatek,mt8188-smi-larb", .data = &mtk_smi_larb_mt8188},
+	{.compatible = "mediatek,mt8189-smi-larb", .data = &mtk_smi_larb_mt8189},
 	{.compatible = "mediatek,mt8192-smi-larb", .data = &mtk_smi_larb_mt8192},
 	{.compatible = "mediatek,mt8195-smi-larb", .data = &mtk_smi_larb_mt8195},
 	{}
@@ -808,6 +840,16 @@ static const struct mtk_smi_common_plat mtk_smi_common_mt8188_vpp = {
 	.init     = mtk_smi_common_mt8195_init,
 };
 
+static const struct mtk_smi_common_plat mtk_smi_common_mt8189 = {
+	.type     = MTK_SMI_GEN2,
+	.bus_sel  = F_MMU1_LARB(1) | F_MMU1_LARB(3) | F_MMU1_LARB(5) |
+		    F_MMU1_LARB(7),
+};
+
+static const struct mtk_smi_common_plat mtk_smi_sub_common_mt8189 = {
+	.type     = MTK_SMI_GEN2_SUB_COMM,
+};
+
 static const struct mtk_smi_common_plat mtk_smi_common_mt8192 = {
 	.type     = MTK_SMI_GEN2,
 	.has_gals = true,
@@ -852,6 +894,8 @@ static const struct of_device_id mtk_smi_common_of_ids[] = {
 	{.compatible = "mediatek,mt8186-smi-common", .data = &mtk_smi_common_mt8186},
 	{.compatible = "mediatek,mt8188-smi-common-vdo", .data = &mtk_smi_common_mt8188_vdo},
 	{.compatible = "mediatek,mt8188-smi-common-vpp", .data = &mtk_smi_common_mt8188_vpp},
+	{.compatible = "mediatek,mt8189-smi-common", .data = &mtk_smi_common_mt8189},
+	{.compatible = "mediatek,mt8189-smi-sub-common", .data = &mtk_smi_sub_common_mt8189},
 	{.compatible = "mediatek,mt8192-smi-common", .data = &mtk_smi_common_mt8192},
 	{.compatible = "mediatek,mt8195-smi-common-vdo", .data = &mtk_smi_common_mt8195_vdo},
 	{.compatible = "mediatek,mt8195-smi-common-vpp", .data = &mtk_smi_common_mt8195_vpp},
-- 
2.43.0

Re: [PATCH v3 2/2] memory: mtk-smi: Add mt8189 support

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
- [High] Implicitly sized array mtk_smi_larb_mt8189_ostd may be accessed out of bounds for LARB IDs >= 21.
- [High] Explicit 0x0 values in mtk_smi_larb_mt8189_ostd cause premature loop termination.
--

commit 34a5ce3e55a813425dbbd1c5607a1ed17951862b
Author: Zhengnan Chen <zhengnan.chen@mediatek.com>

memory: mtk-smi: Add mt8189 support
    
Add the necessary platform data and ostdl setting to enable support
for mt8189 smi.

> diff --git a/drivers/memory/mtk-smi.c b/drivers/memory/mtk-smi.c
> index aaeba8ab211e9..f2d5462af6817 100644
> --- a/drivers/memory/mtk-smi.c
> +++ b/drivers/memory/mtk-smi.c
> @@ -401,6 +401,30 @@ static const u8 mtk_smi_larb_mt8188_ostd[][SMI_LARB_PORT_NR_MAX] = {
>  	[25] = {0x01},
>  };
>  
> +static const u8 mtk_smi_larb_mt8189_ostd[][SMI_LARB_PORT_NR_MAX] = {
> +	[0] = {0x8, 0x20, 0x20, 0x20, 0x20, 0x20, 0x10, 0x0,},
> +	[1] = {0x8, 0x20, 0x20, 0x20, 0x20, 0x20, 0x10, 0x0,},
> +	[2] = {0x7, 0x7, 0x4, 0x4, 0x0, 0x0, 0x2, 0x2, 0x7, 0x7, 0x0,},

[Severity: High]
Does the inclusion of 0x0 values in this array cause valid hardware ports
to be left uninitialized?

The initialization loop in mtk_smi_larb_config_port_gen2_general() treats
a 0 value as the end of the array:

    for (i = 0; i < SMI_LARB_PORT_NR_MAX && larbostd && !!larbostd[i]; i++)
        writel_relaxed(larbostd[i], larb->base + SMI_LARB_OSTDL_PORTx(i));

When it encounters the 0x0 at the fifth element of LARB 2, the loop will
terminate immediately, ignoring the remaining configurations like 0x2 and 0x7.

[ ... ]

> +	[19] = {0x2, 0x1, 0x3, 0x1,},
> +	[20] = {0x7, 0x7, 0x3, 0x3, 0x1, 0x1,},
> +};

[Severity: High]
Will this implicitly sized array lead to an out-of-bounds read if the
device tree probes a LARB with an ID of 21 or 22?

Because the largest index provided is 20, the array has exactly 21 elements.
However, mtk_smi_larb_config_port_gen2_general() blindly accesses the array
using the LARB ID:

    const u8 *larbostd = larb->larb_gen->ostd ?
                         larb->larb_gen->ostd[larb->larbid] : NULL;

If MT8189 defines LARB IDs up to at least 22, accessing index 21 or 22 will
read past the end of the array and could write garbage data from read-only
memory into hardware limit registers.

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260701074533.175803-1-Congcong.Yao@mediatek.com?part=2