Deduplicate similar DDRC configurations and LPDDR4 training patterns
by patching a single configuration.
The aim is to reduce the SPL memory footprint and simplify maintenance
of lpddr4_timing.c
Signed-off-by: Emanuele Ghidoli <emanuele.ghidoli@toradex.com>
Signed-off-by: Marcel Ziswiler <marcel.ziswiler@toradex.com>
Update LPDDR4 configuration and training using updated spreadsheet and
tools from NXP using data from previous spreadsheet and verified
toward datasheet:
- MX8M_Plus_LPDDR4_RPA_v9.xlsx
- mscale_ddr_tool_v3.30.exe
From:
https://community.nxp.com/t5/i-MX-Processors-Knowledge-Base/i-MX-8M-Family-DDR-Tool-Release/ta-p/1104467
Some register values differ due to these fixes/modifications:
- corrected calculation of T_CKPDX parameter (equal to tCKCKEH for LPDDR4)
- corrected ECC related items, none of which affect normal operation
when ECC is not enabled
- corrected formula for calculation of tRTP in cell D122
Signed-off-by: Emanuele Ghidoli <emanuele.ghidoli@toradex.com>
Signed-off-by: Marcel Ziswiler <marcel.ziswiler@toradex.com>
Change tRFCmin (tRFCab) from 280 ns to 380 ns to be compliant with
current and futures memories.
Fixes: 2bc2f817ce ("board: toradex: add verdin imx8m plus support")
Signed-off-by: Emanuele Ghidoli <emanuele.ghidoli@toradex.com>
Signed-off-by: Marcel Ziswiler <marcel.ziswiler@toradex.com>
Add support to Verdin IMX8MP V1.1B SKU which uses
MT53E1G32D2FW-046 WT:B memory.
Compared to the 8 GB memory (MT53E2G32D4NQ-046 WT:A) used on
Verdin IMX8MP V1.0A it has 16 row addresses instead of 17.
In fact, the new memory, is a 2 GB/rank memory. The 8 GB memory is a
4 GB/rank memory.
Manually tweaking Host Interface addresses vs LPDDR4 signals mapping it
is possible to have a single configuration working with both memories:
- Old configuration: HIF bit 30 -> rank, HIF bit 29 -> Row 16
- New configuration: HIF bit 29 -> rank, HIF bit 30 -> Row 16
With this change the memory space from the host processor is contiguous
for both the configurations and the correct memory size is computed
using get_ram_size() at runtime.
Support for single rank memories still works thanks to the fact
dual ranks training fails (ddr_init->ddr_cfg_phy) toward single rank
memories.
Signed-off-by: Emanuele Ghidoli <emanuele.ghidoli@toradex.com>
Signed-off-by: Marcel Ziswiler <marcel.ziswiler@toradex.com>
U-Boot can be booted from NAND without SPL by prepending the DCD header to
the actual U-Boot binary. However this requires prepending 1024 bytes to
u-boot.imx (DCD + u-boot.bin).
There is already a similar target to build spl/u-boot-nand-spl.imx, add the
same option for no-SPL boot.
Tested on i.MX6ULL.
The resulting layout of u-boot-nand.imx is:
- Offset 0x0000 (0 KiB): padding
- Offset 0x0400 (1 KiB): DCD header
- Offset 0x1000 (4 KiB): u-boot.bin
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
Enable SDP protocol support in SPL for DH i.MX6 DHSOM, now that those
components fit into the SPL due to LTO.
To start U-Boot via SDP upload on i.MX6 DHSOM based board, proceed as follows:
- Compile imx_usb [1] .
- Power off the i.MX6 DHSOM based board.
- Connect both USB-serial console and USB-OTG miniB ports to host PC.
- Switch board to USB boot mode.
- Power on the board.
- Verify using '$ dmesg' that a new device has been detected as follows:
New USB device found, idVendor=15a2, idProduct=0054, bcdDevice= 0.01
New USB device strings: Mfr=1, Product=2, SerialNumber=0
Product: SE Blank ARIK
Manufacturer: Freescale SemiConductor Inc
- Upload U-Boot SPL:
$ imx_usb u-boot-with-spl.imx
- Wait for SPL to come up, the following print ought to be the last on
UART console:
SDP: handle requests...
- Upload U-Boot proper:
$ imx_usb u-boot.img
[1] https://github.com/boundarydevices/imx_usb_loader.git
Signed-off-by: Marek Vasut <marex@denx.de>
---------------------------------
CI: https://source.denx.de/u-boot/custodians/u-boot-imx/-/pipelines/15819
i.MX patches queued for next:
- Conversions to DM_SERIAL
- Fixes for Toradex boards
- Gateworks Boards
- i.MX8ULP
- EQoS support / fixes, changes in boards
-----BEGIN PGP SIGNATURE-----
iG0EABECAC0WIQS2TmnA27QKhpKSZe309WXkmmjvpgUCZCaU0Q8cc2JhYmljQGRl
bnguZGUACgkQ9PVl5Jpo76YCKgCgk4OueoIggmes+Ccj8bu1BcQbrfYAoJAgMeKE
It0hrVAifD62TPb2rmHt
=hFVG
-----END PGP SIGNATURE-----
Merge tag 'u-boot-imx-next-20230331' of https://gitlab.denx.de/u-boot/custodians/u-boot-imx into next
u-boot-imx-next-20230331 for next
---------------------------------
CI: https://source.denx.de/u-boot/custodians/u-boot-imx/-/pipelines/15819
i.MX patches queued for next:
- Conversions to DM_SERIAL
- Fixes for Toradex boards
- Gateworks Boards
- i.MX8ULP
- EQoS support / fixes, changes in boards
Ensure that a freshly written fat file with a lower case filename which
fits into the upper case 8.3 short filename is not mangeled with a tilde
and number.
Signed-off-by: Stefan Herbrechtsmeier <stefan.herbrechtsmeier@weidmueller.com>
Remove need of dts binding for button keyboard since it reuses
gpio-keys binding. Select gpio-keys driver if button keyboard
is selected since button keyboard can not operate on its own.
Tested-by: Svyatoslav Ryhel <clamor95@gmail.com> # HTC One X T30
Signed-off-by: Svyatoslav Ryhel <clamor95@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
This patch introduses config which allows interrupt run of usb
mass storage with any key. This is especially useful on devices
with limited input capabilities like tablets and smatphones which
have only gpio keys in direct access.
Signed-off-by: Svyatoslav Ryhel <clamor95@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Fix warnings after adding printf-like attribute format for
run_commandf():
warning: too many arguments for format [-Wformat-extra-args]
Signed-off-by: Evgeny Bachinin <EABachinin@sberdevices.ru>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
Fix warnings both for 32bit and 64bit architecture after adding
printf-like attribute format for run_commandf():
warning: format ‘%x’ expects argument of type ‘unsigned int’, but
argument 2 has type ‘ulong {aka long unsigned int}’ [-Wformat=]
ret = run_commandf("fdt addr -c %08x", addr);
^
Signed-off-by: Evgeny Bachinin <EABachinin@sberdevices.ru>
Cc: Marek Vasut <marek.vasut+renesas@mailbox.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
[trini: Fixup testcases added since patch was posted]
Signed-off-by: Tom Rini <trini@konsulko.com>
As run_commandf() is variadic version of run_command() and just a wrapper,
hence apply similar run_command's test-cases.
Let's avoid warning about empty string passing:
warning: zero-length gnu_printf format string [-Wformat-zero-length]
assert(run_commandf("") == 0);
Signed-off-by: Evgeny Bachinin <EABachinin@sberdevices.ru>
Reviewed-by: Simon Glass <sjg@chromium.org>
* vsnprintf() can truncate cmd, hence it makes no sense to launch such
command (it's broken). Moreover, it's better to signalize to the caller
about such case (for facilitating debugging or bug hunting).
* Fix kernel-doc warnings:
include/command.h:264: info: Scanning doc for run_commandf
include/command.h:268: warning: contents before sections
include/command.h:271: warning: No description found for return value
of 'run_commandf'
* Add printf-like format attribute to validate at compile-time the format
string against parameters's type.
* Fix compilation error in case of -Wall, -Werror, -Wextra:
error: variable ‘i’ set but not used [-Werror=unused-but-set-variable]
* Drop extra ret variable.
Signed-off-by: Evgeny Bachinin <EABachinin@sberdevices.ru>
Reviewed-by: Simon Glass <sjg@chromium.org>
Adding support for Analog Devices MAX313XX series RTCs.
This is ported from the Linux driver and adapted for use in u-boot.
Notable differences are
- handling of tm_year and tm_mon differ
- clock source support is omitted
- hwmon support for the MAX31328 and MAX31343 is omitted
- rtc_ops->reset is added
Signed-off-by: Chris Packham <judge.packham@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
The find_closest() macro can be used to find an element in a sorted
array that is closest to an input value. Bring in this macro from
Linux v6.3-rc1-2-g8ca09d5fa354.
Signed-off-by: Chris Packham <judge.packham@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reduce the duplicated code slightly by using a helper function to handle
the common code.
This reduces the code size very slightly.
Signed-off-by: Simon Glass <sjg@chromium.org>
The API is more convenient to use if one doesn't have to know upfront
which gpio controller has a line with the name one is searching for,
and arrange to look that device up somehow. Or implement this loop
oneself.
Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
Reviewed-by: Simon Glass <sjg@chromium.org>
These operations are required by dm_rtc_read and
dm_bootcount_get helpers.
Signed-off-by: Thomas Perrot <thomas.perrot@bootlin.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
The deletion process handles special case for symlinks whose target are
small enough that it fits in struct ext2_inode.b.symlink. So no block had
been allocated. But the check of file type wrongly considered regular
files as symlink. So, no block was freed. So, the EXT4 partition could be
corrupted because of no free block available.
Signed-off-by: Corentin GUILLEVIC <corentin.guillevic@smile.fr>
Do not mangle lower or mixed case filenames which fit into the upper
case 8.3 short filename. This ensures FAT standard compatible short
filenames (SFN) to support systems without long filename (LFN) support
like boot roms (ex. SFN BOOT.BIN instead of BOOT~1.BIN for LFN
boot.bin).
Signed-off-by: Stefan Herbrechtsmeier <stefan.herbrechtsmeier@weidmueller.com>
Adds a test for the new pci_mps command to ensure that it can set the
Maximum Payload Size (MPS) of all devices to 256 bytes in the sandbox
environment. Enables the pci_mps command in the sandbox environment so
that this test can be run.
Signed-off-by: Stephen Carlson <stcarlso@linux.microsoft.com>
Reports the sandbox swapcase PCI Express device to support a 256 byte
Maximum Payload Size for MPS tuning tests.
Signed-off-by: Stephen Carlson <stcarlso@linux.microsoft.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Enable tuning of the PCI Express MPS (Maximum Payload Size) of
each device. The Maximum Read Request Size is not altered.
The SAFE method uses the largest MPS value supported by all devices in the
system for each device. This method is the same algorithm as used by Linux
pci=pcie_bus_safe.
The PEER2PEER method sets all devices to the minimal (128 byte) MPS, which
allows hot plug of devices later that might only support the minimum size,
and ensures compatibility of DMA between two devices on the bus.
Signed-off-by: Stephen Carlson <stcarlso@linux.microsoft.com>
There can certainly be a lot more elements in the "revisions" (and
"names") arrays than there are gpios used to form the trinary number
we're searching for; we simply don't know the array size up-front.
Nor do we need to, because the loop body already knows to recognize
-EOVERFLOW as "not that many elements present" (and we have a test
that specifically ensures that dev_read_u32_index() returns exactly
that). So just drop the i < priv->gpio_num condition.
While in here, fix the weird placement of the default: keyword.
Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Sean Anderson <sean.anderson@seco.com>
Instead of calling env_get(from) up to three times, just do it once,
computing the value we will put into 'to' and error out if that is
NULL (i.e. no 'from' variable and no default provided).
No functional change.
Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
Reviewed-by: Simon Glass <sjg@chromium.org>
Kconfig settings that are related to the API for standalone applications
should be in the API sub-menu and not on the top level.
CONFIG_STANDALONE_LOAD_ADDR is only relevant if standalone example
applications are built.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Sort the defaults list in BUILD_TARGET Kconfig option. No functional change.
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
In case fitImage support is disabled, and image_locate_script() is
passed a fitImage, then the 'data' variable is used uninitialized.
Drop into the default: branch of the switch-case statement and do
not return the uninitialized data, and do not modify the return
pointer either, just print an error message.
Reported by clang build:
"
$ make HOSTCC=clang CC=clang KCFLAGS=-Werror sandbox64_defconfig && make HOSTCC=clang CC=clang KCFLAGS=-Werror
...
boot/image-board.c:1006:7: error: variable 'data' is used uninitialized whenever switch case is taken [-Werror,-Wsometimes-uninitialized]
case IMAGE_FORMAT_LEGACY:
^~~~~~~~~~~~~~~~~~~
include/image.h:608:29: note: expanded from macro 'IMAGE_FORMAT_LEGACY'
^~~~
boot/image-board.c:1128:19: note: uninitialized use occurs here
*datap = (char *)data;
^~~~
boot/image-board.c:1001:11: note: initialize the variable 'data' to silence this warning
u32 *data;
^
= NULL
"
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
Reviewed-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
The FEC interface mode is now configured in common board_interface_eth_init()
and called by FEC MAC driver when appropriate. Drop the board side duplicates
if the same functionality.
Signed-off-by: Marek Vasut <marex@denx.de>
The EQoS interface mode is now configured in common board_interface_eth_init()
and called by EQoS MAC driver when appropriate. Drop the board side duplicates
if the same functionality.
Signed-off-by: Marek Vasut <marex@denx.de>
The assigned-clock no longer have to be dropped, the clock are now
defined in clk-imx8mp.c and used by DWMAC driver to configure the
DWMAC clock. Drop the workarounds from U-Boot specific DT extras.
Signed-off-by: Marek Vasut <marex@denx.de>
Set, previously unset, CONFIG_SPL_SYS_MALLOC_F_LEN to 0x4000 whose
default value is 0x10000.
Early malloc() uses CRAM_S at 0x184000 (CFG_MALLOC_F_ADDR), this ram
area end at 0x188000.
Fixes: 2bc2f817ce ("board: toradex: add verdin imx8m plus support")
Signed-off-by: Emanuele Ghidoli <emanuele.ghidoli@toradex.com>
Signed-off-by: Francesco Dolcini <francesco.dolcini@toradex.com>
Acked-by: Marcel Ziswiler <marcel.ziswiler@toradex.com>
In order for i2c_num==4 and 5 to stay invalid for non-imx8mp SOCs, the
i2c_ccgr[] array must be sized by the number of initializers present,
not with a hard-coded 6 which would implicitly initialize the last two
elements with zeroes.
Also, the bounds check is off-by-one.
Fixes: c92c3a4453 "ARM: imx: imx8mp: Enable support for i2c5 and i2c6 on i.MX8MP"
Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
We have a few places here that the function declarations do not
match their prototypes, correct them.
Signed-off-by: Ying-Chun Liu (PaulLiu) <paulliu@debian.org>
Cc: Tom Rini <trini@konsulko.com>
Cc: Stefano Babic <sbabic@denx.de>
Cc: Fabio Estevam <festevam@gmail.com>
Cc: NXP i.MX U-Boot Team <uboot-imx@nxp.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
Reported-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
Implement common board_interface_eth_init() and call it from the FEC
driver to configure IOMUXC GPR[1] register according to the PHY mode
obtained from DT. This supports all three interface modes supported by
the i.MX8M Mini/Nano/Plus FEC and supersedes the current board-side
configuration of the same IOMUX GPR[1] duplicated in the board files.
Signed-off-by: Marek Vasut <marex@denx.de>
The FEC ref clock frequency on i.MX8M Mini/Nano/Plus was so far configured
via ad-hoc board code. Replace that with DM clock clk_set_rate() instead.
This way, the driver claims all its required clock and sets the ref clock
rate, without any need of architecture specific register fiddling.
Signed-off-by: Marek Vasut <marex@denx.de>
Implement common board_interface_eth_init() and call it from the DWMAC
driver to configure IOMUXC GPR[1] register according to the PHY mode
obtained from DT. This supports all three interface modes supported by
the i.MX8M Plus DWMAC and supersedes current board-side configuration
of the same IOMUX GPR[1] duplicated in the board files.
Reviewed-by: Ramon Fried <rfried.dev@gmail.com>
Signed-off-by: Marek Vasut <marex@denx.de>
With DM clock support in place, it is easy to add RMII support into the
MAC driver. The RMII cannot operate at 1000 Mbps and at 100 and 10 Mbps
the clock frequency is 50 MHz and 5 MHz instead of 25 MHz and 2.5 MHz.
The board DT requires the following adjustments to EQoS node:
phy-mode = "rmii";
assigned-clock-parents = <&clk IMX8MP_SYS_PLL1_266M>,
<&clk IMX8MP_SYS_PLL2_100M>,
<&clk IMX8MP_SYS_PLL2_50M>;
assigned-clock-rates = <0>, <100000000>, <50000000>;
Reviewed-by: Ramon Fried <rfried.dev@gmail.com>
Signed-off-by: Marek Vasut <marex@denx.de>
The DWMAC clock in i.MX8M Plus were so far configured via ad-hoc
architecture code. Replace that with DM clock instead. This way,
the driver claims all its required clock, enables and disables
them, and even gets the CSR clock rate and sets the TX clock rate,
without any need of architecture specific register fiddling. Drop
the architecture specific code while at it too.
The adjustment here is modeled after STM32MP15xx clock handling
in this driver.
Signed-off-by: Marek Vasut <marex@denx.de>
The driver currently only waits for DMA_MODE SWR bit to clear itself.
This is insufficient e.g. on i.MX8M Plus, where the MAC must be reset
before IOMUX GPR[1] content is latched into the MAC and used. Without
the proper reset, the i.MX8M Plus MAC variant does not take the value
in IOMUX GPR[1] into account, which makes it impossible e.g. to switch
interface mode from RGMII to any other.
Since proper reset is desired in general to put the block into defined
state, always assert the DMA_MODE SWR bit before waiting for the bit
to clear itself.
Reviewed-by: Ramon Fried <rfried.dev@gmail.com>
Signed-off-by: Marek Vasut <marex@denx.de>
