Initialize adi-u-boot-2026.01-y

[pamolloy]

@gmalysa's series was just applied by Tom! Congrats! 🥳 🙏

That means instead of incorporating changes into Greg's series we can just submit fixes directly mainline, which is honestly a bit easier to manage. So the following commits on adi-u-boot-2025.10.y need to be sent mainline:

• e6756cc configs: sc598-ezkit: enable required commands
  • Exclude the CONFIG_ENV_IS_IN_FLASH change
• 601b087 arm: dts: sc598: disable Octal SPI flash
• fa9fd14 arm: dts: sc598: add missing GPIO hogs for Rev D
• bc6aef5 arm: dts: sc5xx: drop SPI flash from common dtsi
• 6d02cf0 arm: dts: rename GPIO expanders
• ARM: dts: adi: add Rev E support to SC598-SOM linux#3056 (comment)

At least the commit messages should be refactored to do so. So let's do that in new PRs against this new branch. This PR adds Greg's series, CI/CD and a random upstream patch.

[ozan956]

Lets go 🚀

Some points to mention:

This PR should also be added:
#53

This fix should also be reflected in U-boot:
analogdevicesinc/linux#3056 (comment)

[gmalysa]

It's probably for the best that Tom approved v3 before v4 was submitted, because this way it is better to reflect the total number contributors (having 30 signed-off-bys on a device tree raises some eyebrows for sure, but having each improvement be correctly attributed looks good and ensures everyone has their authorship reflected).

These commits look good to me so I'll add a reviewed-by tag to the mailing list when you submit, apart from 601b087 ("arm: dts: sc598: disable Octal SPI flash") which causes spi2 flash to stop working for me, so I think that one should be reexamined ahead of submission.

[ozan956]

It's probably for the best that Tom approved v3 before v4 was submitted, because this way it is better to reflect the total number contributors (having 30 signed-off-bys on a device tree raises some eyebrows for sure, but having each improvement be correctly attributed looks good and ensures everyone has their authorship reflected).

These commits look good to me so I'll add a reviewed-by tag to the mailing list when you submit, apart from 601b087 ("arm: dts: sc598: disable Octal SPI flash") which causes spi2 flash to stop working for me, so I think that one should be reexamined ahead of submission.

also from: 601b087#commitcomment-175338696

With this change, u-boot does not probe the regular spi flash successfully:

U-Boot 2026.01-00735-g6b0279a7084d-dirty (Jan 21 2026 - 22:20:59 -0500)

CPU:   ADSP ADSP-SC598-0.0 (JTAG/BOOTROM boot)
Model: ADI SC598-SOM-EZKIT
DRAM:  224 MiB
Core:  143 devices, 24 uclasses, devicetree: embed
MMC:   mmc@310C7000: 0
Loading Environment from SPIFlash... jedec_spi_nor is25lp512@1: unrecognized JEDEC id bytes: c3, e5, 3b
*** Warning - spi_flash_probe_bus_cs() failed, using default environment

In:    serial@0x31003000
Out:   serial@0x31003000
Err:   serial@0x31003000
Net:   eth0: eth0
Hit any key to stop autoboot: 0
=> 

The octal flash on the ezkit cannot be used while the uart is enabled, so it does not probe either

Interestingly, thats not the case for me with https://github.com/analogdevicesinc/u-boot/releases/tag/v2025.10-2

U-Boot 2025.10 (Jan 26 2026 - 09:39:06 +0100)

CPU:   ADSP ADSP-SC598-0.0 (QSPI Master boot)
Model: ADI SC598-SOM-EZKIT
DRAM:  224 MiB
ADP558x Detected: Rev 4, Rev ID 4
Core:  143 devices, 24 uclasses, devicetree: embed
MMC:   mmc@310C7000: 0
Loading Environment from SPIFlash... SF: Detected is25lp01g with page size 256 Bytes, erase sB
*** Warning - bad CRC, using default environment

U-Boot 2025.10 (Jan 26 2026 - 09:52:17 +0100)

CPU:   ADSP ADSP-SC598-0.0 (JTAG/BOOTROM boot)
Model: ADI SC598-SOM-EZKIT
DRAM:  224 MiB
ADP558x Detected: Rev 4, Rev ID 4
Core:  143 devices, 24 uclasses, devicetree: embed
MMC:   mmc@310C7000: 0
Loading Environment from SPIFlash... SF: Detected is25lp01g with page size 256 Bytes, erase size 4 KiB, total 1B
*** Warning - bad CRC, using default environment

let me compare adi release with the latest u-boot, to see whats changed.

[ozan956]

Also for kernel, without disabling octal spi we can not access spi2 on som:

[    1.125138] sram_mmap sram-mmap@0: assigned reserved memory node sram1-reserved@20040000
[    1.153879] spi-nor spi2.0: BFPT parsing failed. Please consider using SPI_NOR_SKIP_SFDP wh
[    1.173276] spi-nor spi2.0: probe with driver spi-nor failed with error -95
[    1.189476] adi-spi3 31030000.spi: registered ADI SPI controller spi2
[    1.202459] tun: Universal TUN/TAP device driver, 1.6

[gmalysa]

ADP558x Detected: Rev 4, Rev ID 4

Is the polarity of the octal chip select potentially related to the som revision? The ADP558x is only on the rev E boards.

[ozan956]

I think it (octal chip select) should be same for both versions, since in both revisions these spi flashes both on som and crr directs to SPI2_SSb and signals are active low. I will investigate potential causes further.

Are you trying with rev D SoM? I can only test with rev E.

[gmalysa]

Yes I only have rev D and ezkit available.