Jonas Karlman [Wed, 1 May 2024 16:22:22 +0000 (16:22 +0000)]
rockchip: rk3399: Sync SoC DT from Linux kernel v6.8
Sync RK3399 SoC common .dtsi-files from Linux kernel v6.8.
The ethernet0 alias is moved to rk3399-u-boot.dtsi in this patch, the
alias will be added in board specific .dts-files and finally removed
from rk3399-u-boot.dtsi in following patches.
The rng node is replaced with crypto0 and crypto1, both can supply
random numbers.
There is no other intended change with this patch.
Signed-off-by: Jonas Karlman <jonas@kwiboo.se> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
Jonas Karlman [Wed, 1 May 2024 16:22:19 +0000 (16:22 +0000)]
clk: rockchip: rk3399: Add dummy support for ACLK_VDU clock
rk3399.dtsi from linux v5.19 and newer try to set VDU clock rate to
400 MHz using an assigned-clock-rates prop of the CRU node.
U-Boot does not use or need this clock so add dummy support for getting
and setting ACLK_VDU clock rate to allow CRU driver to be loaded with an
updated rk3399.dtsi.
Signed-off-by: Jonas Karlman <jonas@kwiboo.se> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
Jonas Karlman [Tue, 30 Apr 2024 15:30:25 +0000 (15:30 +0000)]
rockchip: rk3399: Configure sdmmc regulator pinctrl in SPL
A few boards have shown to be required to properly configure pinctrl
for the fixed regulator gpio pin used by sdmmc before being able to read
from SD-cards.
Include the related gpio, regulator and pinctrl nodes and enable related
Kconfig options so that pinctrl can be configured in SPL for boards that
may be affected by such issue.
Also change to imply SPL_DM_SEQ_ALIAS for all boards because it must be
enabled for working gpio usage in SPL after a future DT sync.
Signed-off-by: Jonas Karlman <jonas@kwiboo.se> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
Jonas Karlman [Tue, 30 Apr 2024 15:30:24 +0000 (15:30 +0000)]
rockchip: rk3399: Fix loading FIT from SD-card when booting from eMMC
When RK3399 boards run SPL from eMMC and fail to load FIT from eMMC due
to it being missing or checksum validation fails there can be a fallback
to read FIT from SD-card. However, without proper pinctrl configuration
reading FIT from SD-card may fail:
U-Boot SPL 2024.04-rc4 (Mar 17 2024 - 22:54:45 +0000)
Trying to boot from MMC2
mmc_load_image_raw_sector: mmc block read error
Trying to boot from MMC2
mmc_load_image_raw_sector: mmc block read error
Trying to boot from MMC1
Card did not respond to voltage select! : -110
mmc_init: -95, time 12
spl: mmc init failed with error: -95
SPL: failed to boot from all boot devices (err=-6)
### ERROR ### Please RESET the board ###
Fix this by tagging related sdhci, sdmmc and spi flash pinctrl nodes
with bootph props. Also move bootph for common nodes shared by all
boards to the SoC u-boot.dtsi.
eMMC, SD-Card and SPI flash nodes are also changed to only be tagged
with bootph props for SPL and U-Boot pre-reloc phases.
Signed-off-by: Jonas Karlman <jonas@kwiboo.se> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
This can be resolved by including bootph prop for U-Bood pre-reloc
phase (bootph-some-ram or bootph-all). However, this has intentionally
been excluded due to including it unnecessarily slows down boot around
200-400 ms.
Also add the clock-frequency prop similar to what has been done for
other Rockchip SoCs.
Signed-off-by: Jonas Karlman <jonas@kwiboo.se> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
Jonas Karlman [Tue, 30 Apr 2024 15:30:22 +0000 (15:30 +0000)]
rockchip: rk3399-puma: Move uart0 bootph to board u-boot.dtsi
rk3399-puma is the only supported board that use uart0 for serial
console, other RK3399 boards typically use uart2 for serial console
and may use uart0 for bluetooth.
Move setting bootph prop to board u-boot.dtsi to only include the uart0
node in TPL/SPL control FDT for the rk3399-puma target.
Signed-off-by: Jonas Karlman <jonas@kwiboo.se> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
Jonas Karlman [Tue, 30 Apr 2024 15:30:18 +0000 (15:30 +0000)]
rockchip: rk3399: Add a default spl-boot-order prop
A lot of RK3399 boards use a u-boot,spl-boot-order of "same-as-spl",
&sdhci and &sdmmc.
Move this to rk3399-u-boot.dtsi and make this default for boards
currently missing a u-boot,spl-boot-order prop.
Before commit a7e69952eb6d ("rockchip: spl: Cache boot source id for
later use") it was required to include the SPI flash node in the
u-boot,spl-boot-order prop to successfully load FIT from SPI flash.
The SPI flash node reference has been dropped from spl-boot-order from
pinebook-pro, roc-pc and rockpro64 now that "same-as-spl" also gets
resolved to the SPI flash node and loading FIT from SPI flash works
without having the node explicitly referenced in spl-boot-order prop.
Signed-off-by: Jonas Karlman <jonas@kwiboo.se> Reviewed-by: Dragan Simic <dsimic@manjaro.org> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
Jonas Karlman [Tue, 30 Apr 2024 15:30:14 +0000 (15:30 +0000)]
rockchip: rk3399: Enable random generator on all boards
The RK3399 SoC contain a crypto engine block that can generate random
numbers.
Imply DM_RNG and RNG_ROCKCHIP Kconfig options to take advantage of the
random generator on all RK3399 boards. Also remove the unnecessary use
of a status = "okay" prop.
Signed-off-by: Jonas Karlman <jonas@kwiboo.se> Reviewed-by: Quentin Schulz <quentin.schulz@theobroma-systems.com> Reviewed-by: Peter Robinson <pbrobinson@gmail.com> Reviewed-by: Kever Yang <kever.yang@rock-chips.com> Reviewed-by: Dragan Simic <dsimic@manjaro.org>
Jonas Karlman [Tue, 30 Apr 2024 15:30:13 +0000 (15:30 +0000)]
rockchip: rk3399: Enable ARMv8 crypto and FIT checksum validation
The RK3399 SoC support the ARMv8 Cryptography Extensions, use of ARMv8
crypto can speed up FIT checksum validation in SPL.
Imply ARMV8_SET_SMPEN and ARMV8_CRYPTO to take advantage of the crypto
extensions for SHA256 when validating checksum of FIT images.
Imply SPL_FIT_SIGNATURE and LEGACY_IMAGE_FORMAT to enable FIT checksum
validation to almost all RK3399 boards.
The following boards have been excluded:
- chromebook_bob: SPL max size limitation of 120 KiB
- chromebook_kevin: SPL max size limitation of 120 KiB
Also imply OF_LIVE to help speed up init of U-Boot proper and disable
CONFIG_SPL_RAW_IMAGE_SUPPORT on leez-rk3399 to ensure SPL does not try
to jump to code that failed checksum validation.
Signed-off-by: Jonas Karlman <jonas@kwiboo.se> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
Jonas Karlman [Tue, 30 Apr 2024 15:30:11 +0000 (15:30 +0000)]
rockchip: rk3399-ficus: Enable TPL and use common bss and stack addr
The rk3399-ficus board is only using SPL and not TPL+SPL like all other
RK3399 boards, chromebook bob/kevin excluded. It does not seem to be any
technical reason why this board was left using only SPL.
Switch to use TPL+SPL and to use the common bss and stack addresses
introduced in commit 008ba0d56d00 ("rockchip: Add common default bss and
stack addresses"). Also add the missing DEFAULT_FDT_FILE option.
Signed-off-by: Jonas Karlman <jonas@kwiboo.se> Reviewed-by: Peter Robinson <pbrobinson@gmail.com> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
Jonas Karlman [Tue, 30 Apr 2024 15:30:10 +0000 (15:30 +0000)]
rockchip: rk3399-puma: Use common bss and stack addresses
The rk3399-puma board is currently using SPL stack and bss addr in SRAM,
the same addr typically used by TPL, this differs from most other RK3399
boards.
Switch to use the common bss and stack addresses introduced in commit 008ba0d56d00 ("rockchip: Add common default bss and stack addresses").
Signed-off-by: Jonas Karlman <jonas@kwiboo.se> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
On rk3399-puma the FIT payload is located at sector 0x200 compared
to the more Rockchip common sector 0x4000 offset:
SYS_MMCSD_RAW_MODE_U_BOOT_SECTOR=0x200
Because FIT payload is located at sector 0x200 and IDBlock is located at
sector 64, the combined size of TPL+SPL (idbloader.img) cannot take up
more than 224 KiB:
(0x200 - 64) x 512 = 0x38000 (224 KiB)
Adjust SPL_PAD_TO to match the used 0x200 sector offset.
Signed-off-by: Jonas Karlman <jonas@kwiboo.se> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
Jonas Karlman [Tue, 30 Apr 2024 15:30:08 +0000 (15:30 +0000)]
rockchip: rk3399-gru: Fix max SPL size on bob and kevin
Chromebook bob and kevin typically run coreboot as the initial boot
loader, however, U-Boot proper can be used as a secondary boot loader.
It is also possible to run U-Boot SPL and proper bare metal, with SPL
and the U-Boot payload loaded from SPI flash.
Because of this chromebook bob and kevin only use SPL and not TPL+SPL
like other RK3399 boards, this mean that SPL is loaded to and run from
SRAM instead of DRAM.
The U-Boot payload is located at 0x40000 (256 KiB) offset in SPI flash
and because the BROM only read first 2 KiB for each 4 KiB page, the size
of SPL (idbloader.img) is limited to max 128 KiB.
The chosen bss start address further limits the size of SPL to 120 KiB.
- Add SPL variant of DM_RNG so that the DM_RNG can be disabled in SPL if
necessary. This may be necessary due to e.g. size constraints of the SPL.
- dd SPL variant of SPL_FSL_CAAM_RNG so that the SPL_FSL_CAAM_RNG can be
disabled in SPL if necessary. This may be necessary due to e.g. size
constraints of the SPL.
- Differentiate between CAAM and DCP in Kconfig entry.
Marek Vasut [Thu, 25 Apr 2024 23:02:09 +0000 (01:02 +0200)]
crypto/fsl: Differentiate between CAAM and DCP in Kconfig entry
Differentiate between "Enable Random Number Generator support" and
"Enable Random Number Generator support" in Kconfig entry, mark the
first as CAAM and the second as DCP, otherwise users cannot easily
decide which of the options is which and enable the correct one.
Marek Vasut [Thu, 25 Apr 2024 23:02:08 +0000 (01:02 +0200)]
crypto/fsl: Introduce SPL_FSL_CAAM_RNG
Add SPL variant of SPL_FSL_CAAM_RNG so that the SPL_FSL_CAAM_RNG can
be disabled in SPL if necessary. This may be necessary due to e.g.
size constraints of the SPL.
Jim Liu [Tue, 23 Apr 2024 07:22:10 +0000 (15:22 +0800)]
board: arbel: Limit the dram effective size to bank0 maximal size
For 4GB dram size, the dram is divided into 2 banks and
the address space of these 2 banks are not concatenated.
Limit the gd->ram_top to not exceed bank0 top to prevent
accessing invalid memory region.
Jim Liu [Tue, 23 Apr 2024 07:22:09 +0000 (15:22 +0800)]
configs: arbel: increase u-boot mapping size
When u-boot enable CONFIG_SYS_BOOT_RAMDISK_HIGH, rootfs image relocated
from FIU address space to memory address before jump to kernel.
Since Arbel reserved memory from 0x00000000 to 0x06200000 for tip image,
and rootfs image may too large that cannot found a suitable location
before 128MB(0x8000000), so increase mapping size from 128MB to 192MB.
net: consider option-length when parsing NIS domain
When parsing option 40 (network information service domain) the
option length is in variable 'oplen' and not in 'size'.
Addresses-Coverity-ID: 492765 Uninitialized variables (UNINIT) Fixes: 8ab388bfdbcf ("net: add support to parse the NIS domain for the dhcp options") Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Commit 37cb626da25d0d ("mmc: sdhci: Add Support for ADMA2") introduced
ADMA_DESC_LEN == 16 (64 bit case), but it was never used before commit 74755c1fed1b0 ("mmc: sdhci: introduce adma_write_desc() hook to struct sdhci_ops").
"sizeof(struct sdhci_adma_desc)" (== 12 for 64bit case) was used instead.
Confusion probably originates from Linux commit 685e444bbaa0
("mmc: sdhci: Add ADMA2 64-bit addressing support for V4 mode"), but
the latter "V4 mode" was never ported to U-Boot.
Fixes: 74755c1fed1b0 ("mmc: sdhci: introduce adma_write_desc() hook to struct sdhci_ops") Signed-off-by: Alexander Sverdlin <alexander.sverdlin@siemens.com> Tested-by: Jonathan Humphreys <j-humphreys@ti.com> Tested-by: Judith Mendez <jm@ti.com>
- RISC-V: cmd: Add SBI implementation ID and extension ID
- Board: Rename spl_soc_init to spl_dram_init
- Board: milkv_duo: Add SPI NOR flash, Ethernet, Sysreset support
Tom Rini [Thu, 2 May 2024 01:39:45 +0000 (19:39 -0600)]
Merge tag 'efi-2024-07-rc2' of https://source.denx.de/u-boot/custodians/u-boot-efi
Pull request efi-2024-07-rc2
Documentation:
* correct description of 'env print -e'
UEFI:
* remove superfluous efi_restore_gd after EFI_CALL
* terminate efidebug test bootmgr early on error
* do not install device-tree if bootmgr fails
* pass GUID by address to efi_dp_from_lo
* remove dead code in efi_var_mem_init()
* enable QueryVariableInfo at runtime for file backed variables
board: starfive: Rename spl_soc_init() to spl_dram_init()
Rename spl_soc_init() to spl_dram_init() because the generic function
name does not reflect what the function actually does. Also
spl_dram_init() is commonly used for dram initialization and should be
called from board_init_f().
Signed-off-by: Lukas Funke <lukas.funke@weidmueller.com> Reviewed-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
board: sifive: Rename spl_soc_init() to spl_dram_init()
Rename spl_soc_init() to spl_dram_init() because the generic function
name does not reflect what the function actually does. Also
spl_dram_init() is commonly used for dram initialization and should be
called from board_init_f().
Signed-off-by: Lukas Funke <lukas.funke@weidmueller.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
Kongyang Liu [Tue, 16 Apr 2024 07:31:05 +0000 (15:31 +0800)]
mmc: cv1800b: Add transmit tap delay config to fix write error
Currently, only the receive delay is configured while the transmit delay
is not set, which may result in errors when writing to the file. This issue
can be resolved by setting PHY_TX_SRC_INVERT to SDHCI_PHY_TX_RX_DLY.
riscv: andesv5: Set default cache line size to 64-bytes
The instruction and data cache line sizes of Andes core
are 64-byte. Select SYS_CACHE_SHIFT_6 for RISCV_NDS so
the SYS_CACHELINE_SIZE is enabled with a default value.
Signed-off-by: Yu Chien Peter Lin <peterlin@andestech.com> Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
efi_selftest: add tests for QueryVariableInfo at boottime
Previous patches added QueryVariableInfo at runtime tests and
split a common function that can be used at boottime. Weire it
up and run a similar set of tets. While at it move a test which is
checiking for 0 available storage in the common code
Add tests for
- Test QueryVariableInfo returns EFI_SUCCESS
- Test null pointers for the function arguments
- Test invalid combination of attributes
efi_loader: enable QueryVariableInfo at runtime for file backed variables
Since commit c28d32f946f0 ("efi_loader: conditionally enable SetvariableRT")
we are enabling the last bits of missing runtime services.
Add support for QueryVariableInfo which we already support at boottime
and we just need to mark some fucntions available at runtime and move
some checks around.
It's worth noting that pointer checks for maxmimum and remaining
variable storage aren't when we store variables on the RPMB, since the
Secure World backend is already performing them.
efi_loader: improve error handling in try_load_entry()
The image is not unloaded if a security violation occurs.
If efi_set_load_options() fails, we do not free the memory allocated for
the optional data. We do not unload the image.
* Unload the image if a security violation occurs.
* Free load_options if efi_set_load_options() fails.
* Unload the image if efi_set_load_options() fails.
Fixes: 53f6a5aa8626 ("efi_loader: Replace config option for initrd loading") Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com> Reviewed-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Daniel Schultz [Fri, 19 Apr 2024 15:55:39 +0000 (08:55 -0700)]
board: phytec: common: Fix eepom is empty check
The ptr variable is currently defined as int and sizeof
returns the size of the eeprom data struct as Byte (32 in total).
In case the eeprom is empty, the check, if the eeprom is empty,
will most likely stop after 8 iterations because it will continue
with the stack which should contain some data. Therefore, the
init function will detect an empty EEPROM as API0 and return with
the valid flag set to True.
Fixes: dc22188cdc8 ("board: phytec: Add common PHYTEC SoM detection") Signed-off-by: Daniel Schultz <d.schultz@phytec.de> Reviewed-by: Teresa Remmet <t.remmet@phytec.de>
For all of the functions that access the eeprom_data, make sure these
data are valid. Use the valid member of the phytec_eeprom_data struct.
This fixes a bug where only the API revision check guarded against
accessing rubbish. But if API revision was e.g. 6, eeprom setup failed
before, but phytec_get_imx8m_eth would still happily access the data.
Fixes: dc22188cdc8 ("board: phytec: Add common PHYTEC SoM detection") Signed-off-by: Yannic Moog <y.moog@phytec.de> Signed-off-by: Daniel Schultz <d.schultz@phytec.de> Reviewed-by: Teresa Remmet <t.remmet@phytec.de>
board: phytec: introduce eeprom struct member 'valid'
Add a new nember to the eeprom_data that indicates whether the
associated data is valid or not. Make use of this new member in the
phytec_eeprom_data_init function by setting the valid value
appropriately.
Move the eeprom data to a new struct payload that holds
the payload of the eeprom.
The phycore-rk3288 can be migrated to OF_UPSTREAM.
Drop redundant device tree files, update MAINTAINERS and
defconfig's DEFAULT_DEVICE_TREE for rockchip vendor prefix accordingly.
Move device tree nodes required for u-boot to rk3288-phycore-rdk-u-boot.dtsi
such as chosen, i2c_eeprom_id and it's alias.
The phycore-am64x can be migrated to OF_UPSTREAM.
Drop redundant device tree files, update MAINTAINERS and
defconfig's DEFAULT_DEVICE_TREE for ti vendor prefix accordingly.
While at it, drop the redundant definitions of AM642_PHYBOARD_ELECTRA_DTB
& SPL_AM642_PHYBOARD_ELECTRA_DTB from the binman dtsi file.
The phycore-am62x can be migrated to OF_UPSTREAM.
Drop redundant device tree files, update MAINTAINERS and
defconfig's DEFAULT_DEVICE_TREE for ti vendor prefix accordingly.
At HS400 mode the ITAPDLY value is that from High Speed mode
which is incorrect and may cause boot failures.
The ITAPDLY for HS400 speed mode should be the same as ITAPDLY
as HS200 timing after tuning is executed. Add the functionality
to save ITAPDLY from HS200 tuning and save as HS400 ITAPDLY.
Fixes: c964447ea3d6 ("mmc: am654_sdhci: Add support for input tap delay") Signed-off-by: Judith Mendez <jm@ti.com>
According to the device datasheet [0], ENDLL=1 for
DDR52 mode, so call am654_sdhci_setup_dll() and write
itapdly after since we do not carry out tuning.
[0] https://www.ti.com/lit/ds/symlink/am62p.pdf
Fixes: c964447ea3d6 ("mmc: am654_sdhci: Add support for input tap delay") Signed-off-by: Judith Mendez <jm@ti.com> Reviewed-by: Jaehoon Chung <jh80.chung@samsung.com>
mmc: am654_sdhci: Add itap_del_ena[] to store itapdlyena bit
Set itap_del_ena if ITAPDLY is found in DT or if the tuning
algorithm was executed and found the optimal ITAPDLY. Add the
functionality to save ITAPDLYENA that can be referenced later
by storing the bit in array itap_del_ena[].
mmc: am654_sdhci: Add tuning algorithm for delay chain
Currently the sdhci_am654 driver only supports one tuning
algorithm which should be used only when DLL is enabled. The
ITAPDLY is selected from the largest passing window and the
buffer is viewed as a circular buffer.
The new tuning algorithm should be used when the delay chain
is enabled; the ITAPDLY is selected from the largest passing
window and the buffer is not viewed as a circular buffer.
This implementation is based off of the following paper: [1].
Also add support for multiple failing windows.
[1] https://www.ti.com/lit/an/spract9/spract9.pdf
Fixes: a759abf569d4 ("mmc: am654_sdhci: Add support for software tuning") Signed-off-by: Judith Mendez <jm@ti.com>
Tom Rini [Mon, 29 Apr 2024 16:55:17 +0000 (10:55 -0600)]
Merge patch series "upstream DT compatibility"
Caleb Connolly <caleb.connolly@linaro.org> says:
This is a subset of [1]. With more platform maintainers switching to
OF_UPSTREAM I didn't want to get in the way (it has also proven more
difficult than I hoped to remove only the fully compatible header
files).
This series removes only the dt-bindings headers which contain generic
data like GPIO flags, input event codes, etc.
It then implements support for building all DTBs for a vendor with
OF_UPSTREAM_BUILD_VENDOR. This removes the need to maintain a set list
of DTBs that can be built by U-Boot and opens up the possibility of new
boards becoming supported "by default" just by landing their DT
upstream.
dts: support building all dtb files for a specific vendor
This adjusts OF_UPSTREAM to behave more like the kernel by allowing for
all the devicetree files for a given vendor to be compiled. This is
useful for Qualcomm in particular as most boards are supported by a
single U-Boot build just provided with a different DT.
Signed-off-by: Caleb Connolly <caleb.connolly@linaro.org> Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org> Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on Amlogic boards builds
projects / u-boot.git / log