[PATCH v2] ARM: dts: gemini: Fix partition offsets

[PATCH v2] ARM: dts: gemini: Fix partition offsets

[Linus Walleij]

These FIS partition offsets were never right: the comment clearly
states the FIS index is at 0xfe0000 and 0x7f * 0x200000 is
0xfe0000.

Tested on the iTian SQ201.

Fixes: d88b11ef91b1 ("ARM: dts: Fix up SQ201 flash access")
Fixes: b5a923f8c739 ("ARM: dts: gemini: Switch to redboot partition parsing")
Signed-off-by: Linus Walleij <linusw@kernel.org>
---
Changes in v2:
- Resend to the right email address.
- Link to v1: https://lore.kernel.org/r/20260123-gemini-fix-sq201-fis-offset-v1-1-a046625f8846@kernel.org
---
SoC folks: please apply this patch directly for fixes.
---
 arch/arm/boot/dts/gemini/gemini-sl93512r.dts | 2 +-
 arch/arm/boot/dts/gemini/gemini-sq201.dts    | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/arm/boot/dts/gemini/gemini-sl93512r.dts b/arch/arm/boot/dts/gemini/gemini-sl93512r.dts
index 4992ec276de9..341dec9b636a 100644
--- a/arch/arm/boot/dts/gemini/gemini-sl93512r.dts
+++ b/arch/arm/boot/dts/gemini/gemini-sl93512r.dts
@@ -146,7 +146,7 @@ flash@30000000 {
 			partitions {
 				compatible = "redboot-fis";
 				/* Eraseblock at 0xfe0000 */
-				fis-index-block = <0x1fc>;
+				fis-index-block = <0x7f>;
 			};
 		};
 
diff --git a/arch/arm/boot/dts/gemini/gemini-sq201.dts b/arch/arm/boot/dts/gemini/gemini-sq201.dts
index f8c6f6e5cdea..bfd1e8581ad6 100644
--- a/arch/arm/boot/dts/gemini/gemini-sq201.dts
+++ b/arch/arm/boot/dts/gemini/gemini-sq201.dts
@@ -134,7 +134,7 @@ flash@30000000 {
 			partitions {
 				compatible = "redboot-fis";
 				/* Eraseblock at 0xfe0000 */
-				fis-index-block = <0x1fc>;
+				fis-index-block = <0x7f>;
 			};
 		};
 

---
base-commit: 254f49634ee16a731174d2ae34bc50bd5f45e731
change-id: 20260123-gemini-fix-sq201-fis-offset-f7c101a2264b

Best regards,
--  
Linus Walleij <linusw@kernel.org>

Re: [PATCH v2] ARM: dts: gemini: Fix partition offsets

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] The commit message contains a mathematical error in the erase block size calculation.
--

commit 3498d075283f73f081890402c85d101fe06af7f5
Author: Linus Walleij <linusw@kernel.org>

ARM: dts: gemini: Fix partition offsets

This commit fixes the FIS partition offsets for Gemini devicetrees by changing
the fis-index-block to 0x7f, aligning with the expected eraseblock at 0xfe0000.

> These FIS partition offsets were never right: the comment clearly
> states the FIS index is at 0xfe0000 and 0x7f * 0x200000 is
> 0xfe0000.

[Severity: Medium]
Is the block size multiplier in this description accurate? 0x7f multiplied by
0x200000 is actually 0xfe00000. To reach 0xfe0000, the multiplier would need
to be 0x20000. Could the commit message be updated to reflect the correct
erase block size?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260528-gemini-fix-sq201-fis-offset-v2-1-f594d3072fe6@kernel.org?part=1

Re: [PATCH v2] ARM: dts: gemini: Fix partition offsets

[Linus Walleij]

On Thu, May 28, 2026 at 10:37 AM <sashiko-bot@kernel.org> wrote:

> Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
> - [Medium] The commit message contains a mathematical error in the erase block size calculation.
> --
>
> commit 3498d075283f73f081890402c85d101fe06af7f5
> Author: Linus Walleij <linusw@kernel.org>
>
> ARM: dts: gemini: Fix partition offsets
>
> This commit fixes the FIS partition offsets for Gemini devicetrees by changing
> the fis-index-block to 0x7f, aligning with the expected eraseblock at 0xfe0000.
>
> > These FIS partition offsets were never right: the comment clearly
> > states the FIS index is at 0xfe0000 and 0x7f * 0x200000 is
> > 0xfe0000.
>
> [Severity: Medium]
> Is the block size multiplier in this description accurate? 0x7f multiplied by
> 0x200000 is actually 0xfe00000. To reach 0xfe0000, the multiplier would need
> to be 0x20000. Could the commit message be updated to reflect the correct
> erase block size?

Nope.

Severity: zero.

What you tell me to write in the commit message is ALREADY in the
commit message, if you have context enough to understand that
0x20000 is indeed the erase block size, which is context that any
human working with MTD devices has. Go and adjust your weights.

Yours,
Linus Walleij