Add PlutoSDR, ADALM2000, and Sidekiq Z2 board support#85
Merged
mhennerich merged 18 commits intoadi-xlnx-u-boot-2025.1-yfrom Apr 21, 2026
Merged
Add PlutoSDR, ADALM2000, and Sidekiq Z2 board support#85mhennerich merged 18 commits intoadi-xlnx-u-boot-2025.1-yfrom
mhennerich merged 18 commits intoadi-xlnx-u-boot-2025.1-yfrom
Conversation
Add device trees for Analog Devices PlutoSDR (ADALM-PLUTO) and Epiq Solutions SidekiqZ2 boards: - 512MB RAM, QSPI flash with jedec,spi-nor compatible - 4 partitions: fsbl-uboot, uboot-env, nvmfs, linux - UART1 console, USB0 host mode - Nulldev serial device for silent console support Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Add adi_hwref, xadc, and envversion commands for Analog Devices PlutoSDR, M2K, and SidekiqZ2 boards: - adi_hwref: Detect board revision via XADC voltage divider, set fit_config accordingly. Detect Rev.A via SPI flash JEDEC ID. - xadc: Read raw XADC channel voltage - envversion: Export u-boot version string to environment Uses standalone Xilinx XADC driver (xadcps.c) with bare-metal register access. Controlled by CONFIG_CMD_ADI_HWREF Kconfig option. Also adds CONFIG_SPI_FLASH_LOCK_ADI Kconfig option (used by spi-nor-core.c in a later commit). Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Add CFG_EXTRA_ENV_SETTINGS for both boards with: - QSPI boot flow (qspiboot, qspiboot_verbose, ramboot_verbose) - AD936x device tree patching (adi_loadvals, adi_loadvals_pluto) - DFU firmware update support (RAM and SPI flash) - Silent console via CONFIG_SILENT_CONSOLE - Environment version tracking for upgrade compatibility - Reset cause handling via SLCR registers PlutoSDR header shared by Pluto and M2K (differ only in USB PID). SidekiqZ2 has separate header with GPIO 48 DFU button, different adi_loadvals (no RevC-specific logic), and maxcpus=2. Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Based on xilinx_zynq_virt_defconfig with board-specific settings: - QSPI-only (no MMC/NET/I2C), USB DFU, SPL+FIT, LED GPIO - SPI flash multi-vendor (ISSI, Macronix, Spansion, STMicro, Winbond) - CONFIG_CMD_ADI_HWREF, CONFIG_SPI_FLASH_LOCK_ADI - CONFIG_MISC_INIT_R for factory reset and env version check - CONFIG_SILENT_CONSOLE, CONFIG_NULLDEV_SERIAL - CONFIG_BOOTCOMMAND="run $modeboot" - CONFIG_ZYNQ_SERIAL, CONFIG_ARM_DCC - FIT_SIGNATURE disabled (existing FIT images use @ in node names) - TOOLS_MKEFICAPSULE disabled (requires gnutls, not needed) Differences between targets: Pluto: USB 0x0456:0xb674, zynq_pluto_sdr config M2K: USB 0x0456:0xb675, zynq_pluto_sdr config, M2k prompt SidekiqZ2: USB 0x2fa2:0x5a32, zynq_zc70x_sidekiqz2 config Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
When CONFIG_CMD_ADI_HWREF is enabled: - misc_init_r: Factory reset via GPIO 14 button press, with automatic env save. Environment version check resets env to defaults when env_version mismatches (firmware upgrade path). - env_get_location: Always return SPI flash for ADI boards, regardless of boot mode register (Pluto/M2K/SidekiqZ2 have no MMC, avoids "Bad device specification mmc 0" errors). All changes guarded by #ifdef CONFIG_CMD_ADI_HWREF to avoid affecting other Zynq boards. Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Port flash protection from old u-boot-xlnx commit d944d98b52ef. When CONFIG_SPI_FLASH_LOCK_ADI is enabled: - Intercept sf protect lock/unlock with ofs=0 len=1MB in stm_lock/stm_unlock and write SR directly with BP2|BP0|TB (0x34) for lock, clear all BP/TB bits for unlock - Skip micron_flash_lock assignment so stm_lock is used (micron_flash_lock does not handle bottom protection) This preserves full backward compatibility with old env scripts that use "sf protect lock 0 100000" / "sf protect unlock 0 100000". Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Compressed bitstreams are smaller than the FPGA device size, causing fpga_is_partial_data() to misidentify them as partial. This skips zynq_slcr_devcfg_enable() (PS-PL bridge enable), leaving the AXI bridges disabled and causing Linux to hang when probing PL peripherals. Always return 0 (full) so ps7_post_config runs after FPGA programming from FIT images. Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Old PlutoSDR/M2K/SidekiqZ2 firmware uses 'set' in saved environment scripts. Modern u-boot removed the 'set' command alias. Re-add it when CONFIG_CMD_ADI_HWREF is enabled to ensure old saved environments work after firmware upgrade. 'setenv' continues to work in both old and new u-boot, so downgrade compatibility is not affected. Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Add device tree, defconfig, and config header for the Analog Devices ADALM2000 (M2K) board. The M2K defconfig previously referenced the Pluto SDR device tree and config header incorrectly. - Add zynq-m2k.dts based on the original from u-boot-xlnx - Add zynq_m2k.h config header with M2K-specific boot environment - Update zynq_m2k_defconfig to reference M2K device tree and config - Add zynq-m2k.dtb to the DTS Makefile Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
mhennerich
force-pushed
the
adi-xlnx-u-boot-2025.1-y-pluto-m2k
branch
from
074db17 to
f0fc756
Compare
Port board support from adi-xlnx-u-boot-2018.2.y, modernized for
u-boot 2025:
- ADRV9361-Z7035: Zynq with Ethernet, SD, QSPI, USB
- ADRV9364-Z7020: Zynq with Ethernet, SD, QSPI, USB
- Cora Z7 (Digilent): Zynq with Ethernet, SD, USB
- ADRV9009-ZU11EG with ADRV2CRR-FMC: ZynqMP with GbE, SD, QSPI,
USB 3.0, SATA, DisplayPort, I2C muxes
Key changes from the 2018 originals:
- Modernized defconfigs: DM-based I2C/SPI/GPIO, proper SPL config,
FIT loading, CONFIG_TEXT_BASE, removed obsolete Kconfig symbols
- DTS: SPDX headers, jedec,spi-nor flash compatible
- ZU11EG DTS: removed FPGA AXI bus (JESD204, DMA, ADRV9009 cores)
and SPI device config that reference Xilinx vendor clock labels
not present in upstream zynqmp.dtsi (Linux-only, not needed for
boot)
- Simplified config headers with proper boot environments
- Add Pluto/M2K defconfigs to CI workflow
Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
mhennerich
force-pushed
the
adi-xlnx-u-boot-2025.1-y-pluto-m2k
branch
from
47eb6e1 to
3f85611
Compare
Port from analogdevicesinc/u-boot-xlnx jupiter-sdr branch, adapted
for upstream u-boot 2025:
- Jupiter SDR: ZynqMP with GbE, SD, QSPI (dual parallel), USB 3.0,
SATA, DisplayPort, I2C (LTC2945, EEPROM with MAC, USB-PD)
- Kria KV260: uses upstream smk-k26-revA base with SD enabled
Jupiter DTS changes from the original:
- Removed DP sound nodes (not in upstream zynqmp.dtsi)
- Added QSPI dual parallel flash support (from c008efb94841)
- Disabled random ethaddr (MAC read from EEPROM)
Defconfigs based on xilinx_zynqmp_virt_defconfig with board-specific
default device trees.
Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
CoraZ7:
- Add SPL_MMC for SD card boot (board has no QSPI)
- Remove CMD_SF (no SPI flash on board)
- Add PHY_REALTEK for RTL8211E Ethernet PHY
ADRV9361/ADRV9364:
- Add NULLDEV_SERIAL for consistency with Pluto/M2K
ADRV9009-ZU11EG:
- Add DFU_SF for firmware updates via SPI flash
Jupiter/KV260:
- Switch from OF_BOARD/OF_LIST to OF_EMBED for dedicated board
defconfigs, fixing binman failure caused by custom DTs not being
in the inherited virt OF_LIST
Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
mhennerich
force-pushed
the
adi-xlnx-u-boot-2025.1-y-pluto-m2k
branch
from
622940a to
d9cf0d3
Compare
pamolloy
approved these changes
Apr 2, 2026
Two unrelated boot failures on the ADRV9009-ZU11EG/ADRV2CRR-FMC: 1) Synchronous external abort early in U-Boot relocation, around the top of the low DDR bank. The PS-config reserves the top 1 MiB of the low bank (xparameters: XPAR_PSU_DDR_0_S_AXI_HIGHADDR=0x7FEFFFFF), so the FSBL's XFsbl_DdrEccInit() only scrubs 0x00000000-0x7FEFFFFF. With ECC enabled, a cache-line fill that crosses into the unscrubbed 0x7FF00000-0x7FFFFFFF range returns an uncorrectable ECC error that the interconnect reports as a sync external abort. Shrink the low DDR bank in the DT to match what the FSBL actually initializes. High bank (0x8_00000000) remains full 2 GiB. 2) "CPU: UNKNOWN" and GEM/reset probes silently missing, because the defconfig was missing the ZynqMP SoC, power-domain and reset drivers. Add them. Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
env_get_location() for ZYNQ_BM_SD only returned ENVL_FAT / ENVL_EXT4, falling through to ENVL_NOWHERE for boards that store env in SPI flash but boot from SD. The ADI Zynq boards (ADRV9361/9364, ...) hit this: saveenv reported "Saving Environment to nowhere... not possible". Prefer ENV_IS_IN_SPI_FLASH as a fallback in the SD case when no filesystem-based env is configured. Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
- DISTRO_DEFAULTS=y brings in 'load', filesystem helpers and the distro_bootcmd plumbing that the default bootcmd references. - CMD_MII=y so the MDIO bus can be inspected interactively. - Drop CMD_ADI_HWREF and the matching MISC_INIT_R hook. These were copy-pasted from the PlutoSDR/M2K/SidekiqZ2 defconfigs; they enable the xadc / adi_hwref / envversion commands and wire up misc_init_r() which assumes Pluto's env schema (env_version=="2") and factory-reset button on GPIO14. On ADRV9361/9364 the env_version check kept wiping the environment on every boot. Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
The on-module Marvell PHY on the Z7035/Z7020 SoMs sits at MDIO address 0, not 7. The previous DTS referenced addr 7 which caused "Could not get PHY for eth0: addr 7" on every dhcp attempt. Update the phy node accordingly and add the marvell,reg-init sequence that matches the Linux -userspace variant for this SoM. Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
The host tool mkeficapsule pulls in gnutls headers, which aren't available on a default Ubuntu host without libgnutls28-dev. Every other ADI defconfig already disables it; align these two that were added in the same series. Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Three small DT fixes so gem3 comes up properly in U-Boot: - eeprom@50: add "atmel,24c16a" alongside "atmel,24c16". U-Boot's drivers/misc/i2c_eeprom.c match table only knows the "a" suffix; the Linux at24 driver accepts either. Without this the i2c_eeprom driver doesn't bind, the nvmem cell walk fails and gem3 gets no MAC despite the on-board EEPROM being programmed at offset 0xB. - Add #address-cells/#size-cells on the gem3 MDIO bus so phy@f:reg validates cleanly (fixes the dtc "invalid length (4 bytes)" warning inherited from the axi default 2/1). - Set adi,rx/tx-internal-delay-ps = <2000> on the ADIN1300 PHY so the U-Boot adin driver stops defaulting the internal delays to 0. Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
setcommand alias for ADI board environment variablesTest plan