Import DRAM timings generated by the DDR tool 3.31 which introduce assorted
tweaks to the DRAM controller settings. Furthermore, enable DBI to improve
noise resilience of the DRAM bus by reducing the number of bit changes on
the bus.
Reduce the DRAM rate to 3600 MTps to remove all remaining correctable errors
reported by EDAC . It is not entirely clear why the slightly faster setting
does produce sporadic correctable errors, while this one does not, but this
could be related to simpler PLL setting at 3600 MTps.
Enable inline ECC which is necessary to detect ECC errors and collect
statistics by the EDAC driver in Linux. This reduces the DRAM size by
64 MiB for each 512 MiB of DRAM, so for a 4 GiB device the available
DRAM size becomes 3.5 GiB and for 2 GiB device the available DRAM size
becomes 1.8 GiB.
Signed-off-by: Marek Vasut <marex@denx.de>
In case the Buck5 and Buck6 regulators which supply DRAM Vdd1 and Vdd2/Vddq
respectively operate in automatic PWM/PFM mode, the DRAM EDAC detects more
correctable errors than if the regulators operate in forced PWM only mode.
Force DRAM regulators to forced PWM mode only to stop tempting the DRAM.
Signed-off-by: Marek Vasut <marex@denx.de>
The DRAM configuration for the 1GB and 4GB imx8mp venice boards had a
bus mapping issue (channel A and B swapped) which creates an invalid
deskewing configuration during training causing the DRAM to not be able
to run at its full bus speed.
Update the various config structures to resolve this.
Signed-off-by: Tim Harvey <tharvey@gateworks.com>
Import DRAM timings generated by the DDR tool 3.31 which introduce assorted
tweaks to the DRAM controller settings. Furthermore, enable DBI to improve
noise resilience of the DRAM bus by reducing the number of bit changes on
the bus.
Reduce the DRAM rate to 3600 MTps to remove all remaining correctable errors
reported by EDAC . It is not entirely clear why the slightly faster setting
does produce sporadic correctable errors, while this one does not, but this
could be related to simpler PLL setting at 3600 MTps.
Enable inline ECC which is necessary to detect ECC errors and collect
statistics by the EDAC driver in Linux. This reduces the DRAM size by
64 MiB for each 512 MiB of DRAM, so for a 4 GiB device the available
DRAM size becomes 3.5 GiB .
Signed-off-by: Marek Vasut <marex@denx.de>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
In case the Buck5 and Buck6 regulators which supply DRAM Vdd1 and Vdd2/Vddq
respectively operate in automatic PWM/PFM mode, the DRAM EDAC detects more
correctable errors than if the regulators operate in forced PWM only mode.
Force DRAM regulators to forced PWM mode only to stop tempting the DRAM.
Signed-off-by: Marek Vasut <marex@denx.de>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
Remove Akash Gajjar <akash@openedev.com> from
MAINTAINERS as email is bouncing.
Signed-off-by: Shantur Rathore <i@shantur.com>
Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
Later versions of Colibri iMX7D V1.1B modules use a "new" SoC fusing. The
difference lies in whether we enable the boot ROM to use the eMMC reset
signal. Depending on the SoC fuse, the boot ROM configures this pin as a
GPIO output to drive the reset signal. Our eMMC vs NAND detection
currently only sets that signal to a GPIO without explicitly setting any
direction. Previously, by default, it was set as an input. As the boot ROM
now configures it as an output, we receive a value of zero instead of one,
indicating the absence of the pull-up on eMMC modules.
To fix this, set the SION bit, allowing the reading back of the value
even if it is configured as an output by the boot ROM. It's important to
note that with the new SoC fusing, we now read back what the boot ROM
drives rather than the real value caused by the pull-up resistor. However,
if it were ever driven low, the eMMC would permanently be reset.
In addition, remove hard-coded variant in the eMMC build case as since the
commit 0c39564d0281 ("toradex: colibri_imx7: Enable nand/emmc detection
and set boot variant") will anyways always get overridden by the detection
routing in board code.
Fixes: 0c39564d ("toradex: colibri_imx7: Enable nand/emmc detection and set boot variant")
Signed-off-by: Marcel Ziswiler <marcel.ziswiler@toradex.com>
Signed-off-by: Hiago De Franco <hiago.franco@toradex.com>
Initially investigating a Linux network issue causing a lot of drop and
poor network performances on a custom system based on a TQMA6A module
(based on an iMX6Q), [1st link below].
I eventually correlated my observations with a contention at the NIC
level when in concurrency with the graphics pipeline. Troubleshooting
this in the kernel lead to disabling DMA bursts accesses made by the IPU
in order to avoid triggering the QoS at the interconnect level, reducing
from 50 to 10% the drop rate on eth0, [2nd link below]. The solution
worked on my setup but not on others, which still suffered from
abnormally high drop rates even with this "fix".
After looking a while into TQ Systems BSP I figured out a number of
differences in recent U-Boot out-of-tree patches they had in their
repository [3rd link]. Parsing the differences one after the other lead
me to this final solution.
The reset pad of the DDR controller was apparently misconfigured, Bit
18-19 picturing the "DDR select field". The current value b11 is
reserved. The only defined value as of version 6 of the iMX6Q manual was
b00 "DDR3 and LPDDR2 mode". In practice no register difference has been
spotted after changing this configuration but all issues tracked thus
far just vanished. All previous fixes have been proven irrelevant. Just
clearing this field solved all our network issues and the drop rate as
measured by iperf3 felt back to 0%.
Link: https://lore.kernel.org/netdev/20231012193410.3d1812cf@xps-13/
Link: https://lists.freedesktop.org/archives/dri-devel/2023-October/428251.html
Link: 15eb6abbef
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
GW73xx-F board revision switched back to the original PCIe switch due to
part availability.
Signed-off-by: Tim Harvey <tharvey@gateworks.com>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
- StarFive: Add StarFive watchdog driver
- VisionFive2: Support device tree overlay for VisionFive2 board
- Andes: Fix PLIC-SW setting
- RISC-V: Fix NVMe support by implying NVME_PCI for QEMU
- RISC-V: Fix binman for 64 bit format load address
The OrangePi Zero 3 is a small development board featuring the Allwinner
H618 SoC, shipping with up to 4GB of DRAM, Gigabit Ethernet, a micro-HDMI
connector and two USB sockets.
The board uses LPDDR4 DRAM and an X-Powers AXP313a PMIC, support for
which was recently added to U-Boot.
Add a defconfig file selecting the right drivers and DRAM options.
Since the .dts file was synced from the Linux kernel repo already, we
just need to add one line to the Makefile to actually build the .dtb.
The DRAM parameters were derived from the values found in the BSP DRAM
drivers on the SPI NOR flash.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Tested-by: Mikhail Kalashnikov <iuncuim@gmail.com>
Tested-by: Bob McChesney <bob@electricworry.net>
Tested-by: Stephen Graf <stephen.graf@gmail.com>
CONFIG_NVME=y without CONFIG_NVME_PCI=y does not provide working NVMe
support. Instead of implying CONFIG_NVME we must imply CONFIG_NVME_PCI
which will select CONFIG_NVME.
Fixes: e64db0d92e32 ("riscv: qemu: Enable e1000 and nvme support")
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Reviewed-by: Mark Kettenis <kettenis@openbsd.org>
There is a couple of issues related to SYS_CONFIG_NAME config in
axy17lte Kconfig.
1. The global SYS_CONFIG_NAME in axy17lte Kconfig overrides
SYS_CONFIG_NAME for all boards specified after this line in
arch/arm/mach-exynos/Kconfig:
source "board/samsung/axy17lte/Kconfig"
Right now it's the last 'source' line there, so the issue is not
reproducible. But once some board is moved or added after this line
the next build error will happen:
GEN include/autoconf.mk.dep
In file included from ./include/common.h:16:
include/config.h:3:10: fatal error: configs/exynos78x0-common.h.h:
No such file or directory
3 | #include <configs/exynos78x0-common.h.h>
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
compilation terminated.
That's happening because axy17lte Kconfig defines SYS_CONFIG_NAME
option in global namespace (not guarded with any "if TARGET_..."), so
it basically rewrites the correct SYS_CONFIG_NAME defined in the
hypothetical boards which might appear after axy17lte in mach-exynos
Kconfig.
2. Another side of the issue is that SYS_CONFIG_NAME is defined
incorrectly in axy17lte Kconfig:
config SYS_CONFIG_NAME
default "exynos78x0-common.h"
The .h extension should not have been specified there. It's leading
to a build error, as the generated include file has a double '.h'
extension.
3. Each target in axy17lte/Kconfig defines its own SYS_CONFIG_NAME. But
all of those in fact incorrect, as corresponding
include/configs/<CONFIG_SYS_CONFIG_NAME>.h header files don't exist.
4. The global SYS_CONFIG_NAME pretty much repeats the help description
from arch/Kconfig and doc/README.kconfig.
Corresponding defconfig files (a*y17lte_defconfig) fix above issues by
overriding SYS_CONFIG_NAME and correctly setting it to
"exynos78x0-common".
Fix all mentioned issues by removing the incorrect global
SYS_CONFIG_NAME and instead specifying it (correctly) in SYS_CONFIG_NAME
options for each target instead.
Signed-off-by: Sam Protsenko <semen.protsenko@linaro.org>
Fixes: 3e2095e960b4 ("board: samsung: add support for Galaxy A series of 2017 (a5y17lte)")
Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
Add the README files for the HiSilicon boards to the HTML documentation.
This required a bit of reformatting.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
To quote Andre:
The first few patches are some easy refactorings and fixes, most of them
actually don't change the generated binaries at all. Then there is a
defconfig for a new board, for which we just gained the .dts file from
the last kernel DT sync.
On top there is support for a new PMIC (AXP313), and LPDDR4 support for
the Allwinner H616 SoC, both of which are needed to support new devices
that appeared lately, especially cheap TV boxes.
While those are technically new features, they don't affect existing
boards, for instance the LPDDR4 support code is guarded by a new DRAM
type Kconfig variable. So the risk for regressions is very slim.
Gitlab CI passed, and I booted that briefly on some boards, including an
H616 and an H618 one (with LPDDR4).
On boards using the AXP313 PMIC, the DRAM rail is often not setup
correctly at reset time, so we have to program the PMIC very early in
the SPL, before running the DRAM initialisation.
Add a simple AXP313 PMIC driver that knows about DCDC2(CPU) and
DCDC3(DRAM), so that we can bump up the voltage before the DRAM init.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Acked-by: Jernej Skrabec <jernej.skrabec@gmail.com>
When building this with clang, we get a warning about having excess
parenthesis here, or that we're incorrectly using "==" when we want "=".
Correct these by using the common size macro.
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
[Andre: Use SZ_512M as per Simon's suggestion]
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
So far we have a convoluted #ifdef mesh that guards the early AXP PMIC
setup in board.c. That combination of &&, || and negations is very hard
to read, maintain and especially to extend.
Fortunately we have those same conditions already modelled in the
Kconfig file, so they are actually redundant. On top of that the real
reason we have those preprocessor guards in the first place is about the
symbols that are *conditionally* defined: without #ifdefs the build
would break because of them being undefined for many boards.
To simplify this, just change the guards to actually look at the symbols
needed, so CONFIG_AXP_xxx_VOLT instead of CONFIG_AXPyyy_POWER.
This drastically improves the readability of this code, and makes adding
PMIC support a pure Kconfig matter.
Doing this revealed one bug in Kconfig: there is no axp_set_dcdc4() for
the AXP818, even though CONFIG_AXP_DCDC4_VOLT includes that PMIC.
Since the AXP818 wasn't included when calling axp_set_dcdc4() in board.c,
this wasn't an issue, but becomes one now, so also remove the AXP818 from
the DCDC4 Kconfig symbol.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
The build instructions for the hikey960 had some broken links. Update
the links to use new vendor URLs. Also change build instructions to
reference a different file name.
Signed-off-by: Dylan Corrales <deathcamel58@gmail.com>
Cc: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
- Fix some issues Coverity has reported, update MAINTAINERS file,
another bootstd fix, typo fix in error message, gitignore fix and
update TI's URL in many places.
It's happened that glitches on the STM32_RST and STM32_BOOT lines have
put the STM32 companion microcontroller into DFU mode making it not boot
its FW, rendering it useless for the user.
Considering that the STM32 companion microcontroller is always reset on
a reboot or power cycle, resetting it once again in U-Boot SPL isn't
going to hurt it any more.
For ATtiny companion microcontroller, the situation is a bit different
because a reboot or power cycle doesn't reset it. Additionally, since it
can only be reset with a UPDI reset on the STM32_RST line, and that is
virtually impossible to mistakenly trigger, the ATtiny is unlikely to be
in unwanted reset or enter reset because U-Boot toggles STM32_RST line.
Cc: Quentin Schulz <foss+uboot@0leil.net>
Signed-off-by: Quentin Schulz <quentin.schulz@theobroma-systems.com>
Reviewed-by: Heiko Stuebner <heiko@sntech.de>
Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
QuartzPro64 is a Rockchip RK3588 based SBC by Pine64.
UART and boot over SD/eMMC/RJ45 are tested to work.
Linux commits from next-20231013:
8152d3d070a9 ("arm64: dts: rockchip: Add QuartzPro64 SBC device tree")
Signed-off-by: Tom Fitzhenry <tom@tom-fitzhenry.me.uk>
Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
Cc: Eugen Hristev <eugen.hristev@collabora.com>
Cc: Jonas Karlman <jonas@kwiboo.se>
Cc: Ondrej Jirman <megi@xff.cz>
This adds support for the Inteno XG6846 board based on the
Broadcom MIPS 6328 SoC.
The default boot will read a uImage from flash and boot it.
Reviewed-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Move ASUS Transformers, Grouper, P895 and P880 config fragments into
their respective device directory in /board/../configs/
Signed-off-by: Svyatoslav Ryhel <clamor95@gmail.com>
Remove nvidia_board_late_init calls from board since this setup is
performed in board2 of mach-tegra. Call of nvidia_board_late_init
from within the board does not provide any additional data.
Tested-by: Andreas Westman Dorcsak <hedmoo@yahoo.com> # ASUS Transformer T30
Tested-by: Svyatoslav Ryhel <clamor95@gmail.com> # LG P895 T30
Signed-off-by: Svyatoslav Ryhel <clamor95@gmail.com>
Default configuration matches values which are set in the board
so this configuration is not required.
Signed-off-by: Svyatoslav Ryhel <clamor95@gmail.com>
JTAG pins are mapped as below. To access the JTAG pins, we need to
control the GPIO pins from SPL which seems to be the earliest stage for
JTAG.
- JTAG nTRST: GPIO36 / Input
- JTAG TDI: GPIO61 / Input
- JTAG TMS: GPIO63 / Input
- JTAG TCK: GPIO60 / Input
- JTAG TDO: GPIO44 / Output
Signed-off-by: Chanho Park <chanho61.park@samsung.com>
Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
1. Fix incorrect ram size of 4GB dram with ECC enabled
2. Fix wrong place to set dram bank size
- The dram bank size should be set in dram_init_banksize
- Dram_init should not access gd->bd because the board info
struct is not reserved yet.
Signed-off-by: Jim Liu <JJLIU0@nuvoton.com>
[trini: Rework slightly]
Signed-off-by: Tom Rini <trini@konsulko.com>
Similar change was done by commit b4c2c151b14b ("Kconfig: Remove all
default n/no options") and again sync is required.
default n/no doesn't need to be specified. It is default option anyway.
Signed-off-by: Michal Simek <michal.simek@amd.com>
Reviewed-by: Svyatoslav Ryhel <clamor95@gmail.com> # tegra
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Angelo Dureghello <angelo@kernel-space.org>
The SynQuacer Developerbox, in EFI mode, supports A/B capsule
updates and single image ones. The flash layout in the latter case is
outdated, update it with the new offsets and images
Signed-off-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Tested-By: Masahisa Kojima <masahisa.kojima@linaro.org>
The "simpler" the logic, the higher the probability to not test and get
things wrong, again: The absence of a "-PG2" suffix is not sufficient to
derive that we are on PG1. There is also "IOT2050-ADVANCED-M2".
Finally fix that by exactly matching against the two PG1 device names.
While changing this, we can also drop the not really needed check for
!board_is_sr1 in board_is_m2 and call the boards by their names
("board_is_pg1").
Reported-and-tested-by: Bao Cheng Su <baocheng.su@siemens.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Extend Turris Mox board code to support CZ.NIC's RIPE Atlas Probe.
Signed-off-by: Marek Behún <kabel@kernel.org>
Reviewed-by: Stefan Roese <sr@denx.de>
We can remove common.h from most cases of the code here, and only a few
places need an additional header instead.
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Rick Chen <rick@andestech.com>
We can remove common.h from most cases of the code here, and only a few
places need an additional header instead.
Signed-off-by: Tom Rini <trini@konsulko.com>
Acked-by: Michal Simek <michal.simek@amd.com>
We can remove common.h from most cases of the code here, and only a few
places need an additional header instead.
Signed-off-by: Tom Rini <trini@konsulko.com>
Acked-by: Alexey Brodkin <abrodkin@synopsys.com>
- Add Board: rk3588 NanoPC-T6, Orange Pi 5, Orange Pi 5 Plus;
- clk driver fix for rk3568 and rk3588;
- rkmtd cmd support for rockchip nand device;
- dts update and sync from linux;
Xunlong Orange Pi 5 Plus is a single-board computer based on the
Rockchip RK3588 SoC. The board provides abundant interfaces, including
two HDMI output ports, one HDMI input port, two 2.5G Ethernet ports,
M.2 M-Key slot, M.2 E-Key slot, two USB 3.0, two USB 2.0, and two Type-C.
Features tested on a Orange Pi 5 Plus 4GB v1.2:
- SD-card boot
- eMMC boot
- SPI Flash boot
- PCIe/NVMe
- USB 2.0 host
- Ethernet
Device tree is imported from linux v6.7-rockchip-dts64-1 tag.
Signed-off-by: Jonas Karlman <jonas@kwiboo.se>
Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
Xunlong Orange Pi 5 is a single-board computer based on the Rockchip
RK3588S SoC. The board provides abundant interfaces, HDMI output, GPIO
interface, M.2 PCIe2.0, Type-C, Gigabit LAN port, 2*USB2.0, 1*USB3.0,
etc.
Features tested on a Orange Pi 5 4GB v1.2:
- SD-card boot
- SPI Flash boot
- PCIe/NVMe
- USB 2.0 host
- Ethernet
Device tree is imported from linux v6.7-rockchip-dts64-1 tag.
Signed-off-by: Jonas Karlman <jonas@kwiboo.se>
Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
-----BEGIN PGP SIGNATURE-----
iQFQBAABCgA6FiEEqxhEmNJ6d7ZdeFLIHrMeAg6sL8gFAmU2fxwcHGV1Z2VuLmhy
aXN0ZXZAY29sbGFib3JhLmNvbQAKCRAesx4CDqwvyKr+B/42+WkPiAiwt79R6Kh5
UKo404oN1na28D3V12EvikiCH2QWubi38EV7VfcB3jqEpxj4j+rY0cSLZv+u7d4/
bfQN3FPrcTwYnXYLbYkUKC+MAVzGLiVzFBBhIt5cPrT4iVXctwYY6LJ58aX8uRo6
KU6CwXYc3cnmIUqpXE4EYbbQU4TAp55SV2n8QLFFI9DHcy2kELfdrZduFn+Hm9Ij
9sQHH5JxbrGHv5XpcjJkARW3M4K3K4TPzweFGN61VMJwWqzD3L4YPshwIGDfhRui
LECKCdr/rWkavumXCfw2pD3RYFTqnqnUzaEN/3ueL9FCJ2uFH1EyaoSuXFEn3+Ul
m1LE
=CMMd
-----END PGP SIGNATURE-----
Merge tag 'u-boot-at91-2024.01-b' of https://source.denx.de/u-boot/custodians/u-boot-at91
Second set of u-boot-at91 features for the 2024.01 cycle
This feature set a new board named Conclusive KSTR sama5d27 with some
small prerequisites patches.
This is mostly about support for the Allwinner R528/T113s SoC, which is
reportedly the same die as the Allwinner D1, but with the two
Arm Cortex-A7 cores activated instead of the RISC-V one.
Using sunxi code outside of arch/arm proved to be difficult, so apart
from enabling this Arm SoC, the patches also prepare for more refactoring
to get the D1 nicely supported some day:
- We get rid of some Kconfig (hard-)coded GPIO pins, responsible for
enabling regulators.
- The GPIO code is moved out of arch/arm, into drivers/gpio.
- Some definitions are moved out of header files under asm/arch.
- Some T113s/D1 specific definitions are guarded by a generic Kconfig
symbol (CONFIG_SUNXI_GEN_NCAT2).
- The DRAM controller initialisation code is located under drivers/ram.
- The base SoC .dtsi files are shared (under arch/riscv, as in Linux).
Of course there are also the usual new SoC specific patches, like clock
and pinmux descriptions, alongside a rework of the pinctrl code, since
Allwinner changed the GPIO register layout, for the first time since
sunxi's inception.
On top of this the PSCI code sees some update, to provide SMP services
for R528/T113s boards. Many thanks to Sam for providing this code and
staying strong through the review cycles.
The final patch enables support for one popular board, I hope to see
more DTs and defconfigs contributed in the future!
Many thanks to all the various contributors, testers and reviewers,
that series was a real team effort!
Introduce support for Conclusive KSTR-SAMA5D27 Single Board Computer.
Co-developed-by: Jakub Klama <jakub@conclusive.pl>
Signed-off-by: Jakub Klama <jakub@conclusive.pl>
Co-developed-by: Marcin Jabrzyk <marcin@conclusive.pl>
Signed-off-by: Marcin Jabrzyk <marcin@conclusive.pl>
Signed-off-by: Artur Rojek <artur@conclusive.pl>
