summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2023-11-07fastboot: Depend on CMDLINETom Rini
Much of the functionality of fastboot relies on being able to run commands as defined in the environment. This means it does depend on CMDLINE being enabled. Signed-off-by: Tom Rini <[email protected]>
2023-11-07cmd: Make most commands depend on CMDLINESimon Glass
If we disable CMDLINE then we should not ask about enabling the hush parser nor any of the commands that would be run on the command line as it is no longer available. Convert the CMDLINE option into a menuconfig and make every command referenced under cmd/Kconfig depend on it. This leaves as future work moving the commands that are not under the cmd/ hierarchy as future work. Signed-off-by: Simon Glass <[email protected]> Signed-off-by: Tom Rini <[email protected]>
2023-11-07boot: Make preboot and bootcmd require CMDLINETom Rini
In order for a predefined "preboot" or "bootcmd" to be executed by the running system we must have a command line. Add CMDLINE as a dependency. Signed-off-by: Tom Rini <[email protected]>
2023-11-07bootmeth_script: Depend on CMDLINETom Rini
As this particular bootmeth requires the command line and assorted commands to function, make sure we have CMDLINE enabled. Signed-off-by: Tom Rini <[email protected]>
2023-11-07bootmeth_cros: Require bootm.o and bootm_os.oTom Rini
In order to use bootmeth_cros, at least on non-X86, we need to be able to start any type of kernel that the "bootm" code paths can handle. Add these objects to the required list for this option. Signed-off-by: Tom Rini <[email protected]>
2023-11-07boot: Move SYS_BOOTM_LEN to be by LEGACY_IMAGE_FORMATTom Rini
This particular option is required for booting all image types, regardless of if we are starting an OS via command line or something else. Move the question for SYS_BOOTM_LEN to be by the question for LEGACY_IMAGE_FORMAT, as that's where our generic OS questions start. Signed-off-by: Tom Rini <[email protected]>
2023-11-07boot: Rework BOOT_DEFAULTS to allow for CMDLINE to be disabledTom Rini
We split BOOT_DEFAULTS to have BOOT_DEFAULTS_FEATURES and BOOT_DEFAULTS_CMDS that in turn list general features or commands that we want enabled when BOOT_DEFAULTS is selected. We only select BOOT_DEFAULTS_CMDS if CMDLINE is set. Signed-off-by: Tom Rini <[email protected]>
2023-11-07boot: Make DISTRO_DEFAULTS select CMDLINETom Rini
The implementation of DISTRO_DEFAULTS is done in environment scripts and requires the command line in order to work. Because of this, select CMDLINE here. Signed-off-by: Tom Rini <[email protected]>
2023-11-07autoboot: Correct dependencies on CMDLINESimon Glass
Make AUTOBOOT depend on CMDLINE since it is mostly meaningless without it. Signed-off-by: Simon Glass <[email protected]> Reviewed-by: Tom Rini <[email protected]>
2023-11-07bootmeth: Make BOOTMETH_EFILOADER depend on CMD_BOOTEFITom Rini
Today, the bootmeth for using the EFI loader via bootefi depends on calling the bootefi command directly, so make this in turn depend on CMD_BOOTEFI. Signed-off-by: Tom Rini <[email protected]>
2023-11-07efi: Rearrange the Kconfig for CMD_BOOTEFI_BOOTMGRSimon Glass
The command should not be used to enable library functionality. Add a new BOOTEFI_BOOTMGR Kconfig for that. Adjust the conditions so that the same code is built. Signed-off-by: Simon Glass <[email protected]> Suggested-by: AKASHI Takahiro <[email protected]>
2023-11-07cli_simple: Rework this support slightlyTom Rini
The interactive portion of our non-HUSH 'simple' parser is guarded by CONFIG_CMDLINE already. Much of the code behind this simple parser is also used as "input" parser, such as from menu interfaces and so forth and not strictly command line input. To support this, always build the assorted cli object files, but guard the interactive portions of cli_simple.c with a CMDLINE check. Signed-off-by: Tom Rini <[email protected]>
2023-11-07video: Don't require the font commandSimon Glass
While it is nice to have the font command, using 'select' makes it impossible to build the console code without it. Stop using 'select' and make it default if CONSOLE_TRUETYPE is enabled when asking the command. Signed-off-by: Simon Glass <[email protected]> Signed-off-by: Tom Rini <[email protected]>
2023-11-07test: Make UNIT_TEST depend on CMDLINESimon Glass
Many tests make some use of the command line, so require it for all test code. This could be teased apart, perhaps with a test flag indicating that it uses the command line. Leave that for later. Signed-off-by: Simon Glass <[email protected]> Reviewed-by: Tom Rini <[email protected]>
2023-11-07env: Move env_set() out of cmd/nvedit.c and in to env/common.cTom Rini
Inside of env/common.c we already have our helper env_set_xxx functions, and even have a comment that explains why env_set() itself wasn't moved. We now handle that move. This requires that we rename the previous _do_env_set() to env_do_env_set() and note it as an internal env function. Add comments about this function to explain why we do this when we add the prototype. Add a new function, env_inc_id() to allow for the counter to be updated by both commands and callers, and document this as well by the prototype. Signed-off-by: Tom Rini <[email protected]>
2023-11-07Kconfig: Move CONFIG_SYS_[CP]BSIZE to common/KconfigTom Rini
Move CONFIG_SYS_CBSIZE (console buffer size) and CONFIG_SYS_PBSIZE (console print buffer size) out of cmd/Kconfig and in to common/Kconfig. Create help entries for both which explain their usage and why they are both not entirely command centric. Signed-off-by: Tom Rini <[email protected]>
2023-11-07qemu: Correct CMD_QFW dependencies in KconfigTom Rini
Rather than selecting CMD_QFW, we should make the option itself by enabled by default on these platforms. Then in the board-specific Kconfig we should select the appropriate back-end as needed if the command is enabled. Signed-off-by: Tom Rini <[email protected]>
2023-11-07version: Separate our version string from the version commandTom Rini
In order to be able to disable all commands we need to construct our version string in a common file, and have the version command reference that string, like the other users of it do. Create common/version.c with just the strings. Signed-off-by: Tom Rini <[email protected]>
2023-11-07dfu: Make DFU_TFTP depend on NETDEVICESTom Rini
In order to do a DFU update over TFTP we need to have some network device available, so make this depend on NETDEVICES Signed-off-by: Tom Rini <[email protected]>
2023-11-07virtio: Make VIRTIO_NET depend on NETDEVICESTom Rini
As VIRTIO_NET is the symbol for enabling network devices, make this depend on NETDEVICES Signed-off-by: Tom Rini <[email protected]>
2023-11-07buildman: Use oldconfig when adjusting the configSimon Glass
We cannot be sure that the new config is consistent, particularly when changing a major item like CONFIG_CMDLINE. Use 'make oldconfig' to check that and avoid any such problems. Signed-off-by: Simon Glass <[email protected]>
2023-11-07Merge tag 'xilinx-for-v2024.01-rc3' of ↵Tom Rini
https://source.denx.de/u-boot/custodians/u-boot-microblaze Xilinx changes for v2024.01-rc3 xilinx: - Disable lock in mini spi configurations zynq: - DTS syncups - Kconfig updates zynqmp: - DTS syncups - Kconfig fixups versal: - Make 30MHz as default freq for spi versal net: - Enable ADMA for mmc serial: - Read baudrate from DT spi: - Put spi lock under one Kconfig - Support 64bit addresses in cadance_ospi - zynqmp_gqspi - change logging support firmware: - Handle errors in zynqmp_pm_feature() include: - Sync vsc8531 dt binding with kernel
2023-11-07MAINTAINERS: fastboot: add MattijsMattijs Korpershoek
Fastboot has been marked as orphaned since 2021. Since I'm interested in maintaining this, assign myself. Signed-off-by: Mattijs Korpershoek <[email protected]> Acked-by: Marek Vasut <[email protected]> Acked-by: Lukasz Majewski <[email protected]>
2023-11-07MAINTAINERS: usb gadget: add MattijsMattijs Korpershoek
It seems that Lukasz and Marek could get some help in maintaining the usb gadget drivers. Assign myself as maintainer. Signed-off-by: Mattijs Korpershoek <[email protected]> Acked-by: Marek Vasut <[email protected]> Acked-by: Lukasz Majewski <[email protected]>
2023-11-07dt-bindings: Remove VSC8531 specific RGMII delay definitionsMichal Simek
Based on Linux upstream discussion value enumeration shouldn't be used. Instead of it delay in pS should be used that's why remove it from the header. Link: https://lore.kernel.org/all/YNsm%[email protected]/ Signed-off-by: Michal Simek <[email protected]> Link: https://lore.kernel.org/r/bbd343474856a67252f3b31d22b3b5a80ad04043.1698851109.git.michal.simek@amd.com
2023-11-07arm64: zynqmp: Add description for nvmem efuse layoutMichal Simek
Based on discussion with DT folks at link below there is not going to be any name restrictions for child names. That's why add description for current nvmem layout. Link: https://lore.kernel.org/lkml/[email protected]/ Signed-off-by: Michal Simek <[email protected]> Link: https://lore.kernel.org/r/af81299cafc2bd13ed30dcd69bdf6efb5fbb7f68.1698840373.git.michal.simek@amd.com
2023-11-07ARM: zynq: Add partition descriptionMichal Simek
Xilinx is using standard mtd partition layout for quite a long time. It is used for testing purpose on evaluation boards. Also #address/size-cells shouldn't be present without nodes which should use them that's why move them from zynq-7000.dtsi to nand/nor nodes directly. The patch was tested on zc706 and zedboard(with also increasing max frequency and rx bus width). Signed-off-by: Michal Simek <[email protected]> Link: https://lore.kernel.org/r/4c3348981bba32d3892194420d78fe8621c47534.1698837725.git.michal.simek@amd.com
2023-11-07xilinx: Enable SPI_FLASH_MTD by defaultMichal Simek
Provide access to qspi flash layout via mtd command. Signed-off-by: Michal Simek <[email protected]> Link: https://lore.kernel.org/r/b3c4a3eddb71aab9535b034380d0dbae770828d4.1698837725.git.michal.simek@amd.com
2023-11-07arm64: zynqmp: Comment all smmu entriesMichal Simek
SMMU is disabled by default and not all masters can be enabled at the same time because of limited number of entries. That's why comment all iommu properties but keep them for reference in DT. In XEN case they should be added back and Xen should have SMMU enabled by default. Also add IDs for DP and DPDMA. Signed-off-by: Michal Simek <[email protected]> Link: https://lore.kernel.org/r/e868c27c52ded5d8ef25f75ba394b1ab3b31b80a.1698825657.git.michal.simek@amd.com
2023-11-07arm64: versal-net: Add DTSes for mini qspi/ospi configurationMichal Simek
Mini U-Boot is running out of OCM and it's only purpose is to program non volatile memories. There are different configurations which ospi/qspi can be that's why describe them via DT. DT binding is already approved that's why there is no reason not to add it. Signed-off-by: Michal Simek <[email protected]> Link: https://lore.kernel.org/r/a99a8d72201a782fc811715942dea97fb5ab583b.1698329087.git.michal.simek@amd.com
2023-11-07arm64: versal: Add DTSes for mini qspi/ospi configurationMichal Simek
Mini U-Boot is running out of OCM and it's only purpose is to program non volatile memories. There are different configurations which ospi/qspi can be that's why describe them via DT. DT binding is already approved that's why there is no reason not to add it. Signed-off-by: Michal Simek <[email protected]> Link: https://lore.kernel.org/r/9518ab1c4299a45e800b8611172edd78c9243132.1698329087.git.michal.simek@amd.com
2023-11-07ARM: zynq: Add DTSes for mini qspi configurationsMichal Simek
Mini U-Boot is running out of OCM and it's only purpose is to program non volatile memories. There are different configurations which qspi can be that's why describe them via DT. DT binding is already approved that's why there is no reason not to add it. Signed-off-by: Michal Simek <[email protected]> Link: https://lore.kernel.org/r/e7d31a9d9c4a76e171eefc619f31fabd0831a614.1698329087.git.michal.simek@amd.com
2023-11-07ARM: zynq: Add DTSes for mini qspi configurationsMichal Simek
Mini U-Boot is running out of OCM and it's only purpose is to program non volatile memories. There are different configurations which qspi can be that's why describe them via DT. DT binding is already approved that's why there is no reason not to add it. Signed-off-by: Michal Simek <[email protected]> Link: https://lore.kernel.org/r/28b3cdd7e91b2b4c3c36d0bf65aa5bac042f248c.1698329087.git.michal.simek@amd.com
2023-11-07xilinx: versal: Setup 30MHz as default spi frequencyMichal Simek
Align default SPI configuration with ZynqMP/Versal NET. There is no reason to run on lower frequencies. Signed-off-by: Michal Simek <[email protected]> Link: https://lore.kernel.org/r/c1d6ebd659f3002649b1200c926f8b9ed3132085.1698749448.git.michal.simek@amd.com
2023-11-07arm64: versal-net: enable CONFIG_MMC_SDHCI_ADMAVenkatesh Yadav Abbarapu
The Standard Host Controller Interface (SDHCI) specification version 3.00 adds support for Advanced DMA (ADMA) for both 64 and 32 bit widths of DMA. This significantly improves read and write throughput. Signed-off-by: Venkatesh Yadav Abbarapu <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Michal Simek <[email protected]>
2023-11-07arm64: zynqmp: Fix Kconfig entry indentationMichal Simek
Use tabs instead of space for entry indentation which is standard coding style. Signed-off-by: Michal Simek <[email protected]> Link: https://lore.kernel.org/r/ff28e719de82258c066f1fedae87f88597f367b5.1698302068.git.michal.simek@amd.com
2023-11-07arm: xilinx: Add missing dual parallel flash descriptionMichal Simek
Describe flash memories based on the latest DT binding. Signed-off-by: Michal Simek <[email protected]> Link: https://lore.kernel.org/r/cddf2909d0445eba08b998d42ffc31c1fa3132b9.1698045694.git.michal.simek@amd.com
2023-11-07arm64: zynqmp: Disable Tri-state for MIO38 PinTejas Bhumkar
gpio38 is used in SOM's kv260 to reset the Ethernet PHY. At present, HW reset is not working properly as Tri-state  is enabled for MIO38, causing inappropriate PHY register reads. Disabled Tri-state for MIO38 to make HW reset work. Tri-state disable : ZynqMP> md 0xFF180208 2 ff180208: 00bfe7a3 00000540 Tri-state enable : ZynqMP> md 0xFF180208 2 ff180208: 00bfe7e3 00000540 Signed-off-by: Tejas Bhumkar <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Michal Simek <[email protected]>
2023-11-07zynqmp: migrate gqspi debug to loggingIbai Erkiaga
The following patch migrates the usage of debug and printf functions to the relevant logging function as per U-Boot DM guidelines. Additionally some of the debugging statements have been rearanged for a more meaningfull debug experience. aarch64-linux-gnu-size reports 229 bytes less when debug is enabled at file level, while is just 5bytes more when disabled. Signed-off-by: Ibai Erkiaga <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Michal Simek <[email protected]>
2023-11-07arm64: dts: zynqmp: make hw-ecc as the default ecc modeAmit Kumar Mahapatra
Except for Linux no other component (i.e., u-boot, fsbl or BootRom) of the software stack supports software ecc engine. So, make hw-ecc as the default ecc mode. Signed-off-by: Amit Kumar Mahapatra <[email protected]> Signed-off-by: Michal Simek <[email protected]> Link: https://lore.kernel.org/r/f47b95616eb40d3a9908ca60df94ec6e873b071c.1697119098.git.michal.simek@amd.com
2023-11-07ARM: zynq: Disable the config CONFIG_SPI_FLASH_USE_4K_SECTORSVenkatesh Yadav Abbarapu
Lock size for the flashes will be in terms of sector size, so disable the CONFIG_SPI_FLASH_USE_4K_SECTORS and read it from the flash itself. Signed-off-by: Venkatesh Yadav Abbarapu <[email protected]> Signed-off-by: Michal Simek <[email protected]> Link: https://lore.kernel.org/r/8e5c2bf3b7e5f366c4e261e8055ea254bda53aa9.1697117993.git.michal.simek@amd.com
2023-11-07arm64: zynqmp: Disable the lock option for mini qspiVenkatesh Yadav Abbarapu
As mini configs are required only for flashing the images, so disabling the lock config which will save nearly 6KB of memory. Signed-off-by: Venkatesh Yadav Abbarapu <[email protected]> Signed-off-by: Michal Simek <[email protected]> Link: https://lore.kernel.org/r/374d6b3a46d19b7ee171dfe8071676098db93e25.1697117869.git.michal.simek@amd.com
2023-11-07arm64: zynqmp: remove snps, xhci-stream-quirk property for usbPiyush Mehta
To sync up with the upstream bulk-stream feature, removed 'snps,xhci-stream-quirk' DT property for usb. Signed-off-by: Piyush Mehta <[email protected]> Signed-off-by: Michal Simek <[email protected]> Link: https://lore.kernel.org/r/1f4ecfe3ea6a4d0d8d8de324f5dffd3efc86656a.1697115523.git.michal.simek@amd.com
2023-11-07arm64: zynqmp: Remove address/size-cells from ams nodeMichal Simek
Remove unused address/size-cells which is also done upstream that's why this is pretty much sync patch with upstream. Signed-off-by: Michal Simek <[email protected]> Link: https://lore.kernel.org/r/0ca8d8fe245ad7cb665f5333202d83f70acfc11f.1697115523.git.michal.simek@amd.com
2023-11-07Revert "arm64: zynqmp: Add power domain description for PL"Michal Simek
This reverts commit d59fac2f3f247470708a1aed1af96802a05e0e61. This power domain shouldn't be enabled by default. Power domain behavior should be handled on case by case basis. Adding this property to zynqmp.dtsi is breaking some suspend/resume cases that's why remove it from this file. Signed-off-by: Michal Simek <[email protected]> Link: https://lore.kernel.org/r/7ed2a46383c6918fbbaca2d618459b1ee58f865c.1697115523.git.michal.simek@amd.com
2023-11-07arm64: zynqmp: Remove xlnx,zynqmp-aes nodeMichal Simek
AES can be discovered via firmware interface that's why remove node for it. Signed-off-by: Michal Simek <[email protected]> Link: https://lore.kernel.org/r/eaf575a6ca92f8c10cefb447c08c1292025deb74.1697115523.git.michal.simek@amd.com
2023-11-07arm64: zynqmp: Use mdio node by vp-x-a2785-00-revA and vpk120-revAMichal Simek
All boards have been converted to use mdio node that's why move ethernet phys under mdio node too. Signed-off-by: Michal Simek <[email protected]> Link: https://lore.kernel.org/r/6c60f5d29b9d9992bd0130fd263c8ed13cb8166c.1697115523.git.michal.simek@amd.com
2023-11-07arm64: zynqmp: Remove fclk driverMichal Simek
fclk will never go upstream that's why removing this node from DT. All PL (programmable logic) based IPs should handle clocks self without using this workaround. Signed-off-by: Michal Simek <[email protected]> Link: https://lore.kernel.org/r/fbb8665b8a58dbe96349abfe5492a509939e165b.1697098930.git.michal.simek@amd.com
2023-11-07spi: cadence_ospi_versal: Add support for 64-bit addressVenkatesh Yadav Abbarapu
When 64-bit address is passed only lower 32-bit address is getting updated. Program the upper 32-bit address in the DMA destination memory address MSBs register. Signed-off-by: Venkatesh Yadav Abbarapu <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Michal Simek <[email protected]>
2023-11-07drivers: firmware: Handle error case in the zynqmp_pm_featureVenkatesh Yadav Abbarapu
Unhandled error coming from xilinx_pm_request() but return value is not read back that's why getting sparse warning as below: warning: variable 'ret' set but not used [-Wunused-but-set-variable]. In case of error return the "ret" value. Signed-off-by: Venkatesh Yadav Abbarapu <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Michal Simek <[email protected]>
generated by cgit v1.3.1 (git 2.54.0) at 2026-06-27 00:53:31 +0000