In order to support IOMMUs in non-bypass mode we need device ops
to map and unmap DMA memory. The map operation enters a mapping
for a region specified by CPU address and size into the translation
table of the IOMMU and returns a DMA address suitable for
programming the device to do DMA. The unmap operation removes
this mapping from the translation table of the IOMMU.
Signed-off-by: Mark Kettenis <kettenis@openbsd.org>
There are no SPL/TPL variants of CONFIG_CLK_EXYNOS and
CONFIG_ARCH_APPLE, so switch from CONFIG_IS_ENABLED to
IS_ENABLED.
Signed-off-by: Mark Kettenis <kettenis@openbsd.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
Extend support for signing in auto-generated (-f auto) FIT. Previously,
it was possible to get signed 'images' subnodes in the FIT using
options -g and -o together with -f auto. This patch allows signing
'configurations' subnodes instead of 'images' ones (which are hashed),
using option -f auto-conf instead of -f auto. Adding also -K <dtb> and
-r options, will add public key to <dtb> file with required = "conf"
property.
Summary:
-f auto => FIT with crc32 images
-f auto -g ... -o ... => FIT with signed images
-f auto-conf -g ... -o ... => FIT with sha1 images and signed confs
Example: FIT with kernel, two device tree files, and signed
configurations; public key (needed to verify signatures) is
added to u-boot.dtb with required = "conf" property.
mkimage -f auto-conf -A arm -O linux -T kernel -C none -a 43e00000 \
-e 0 -d vmlinuz -b /path/to/first.dtb -b /path/to/second.dtb \
-k /folder/with/key-files -g keyname -o sha256,rsa4096 \
-K u-boot.dtb -r kernel.itb
Example: Add public key with required = "conf" property to u-boot.dtb
without needing to sign anything. This will also create a useless FIT
named unused.itb.
mkimage -f auto-conf -d /dev/null -k /folder/with/key-files \
-g keyname -o sha256,rsa4096 -K u-boot.dtb -r unused.itb
Signed-off-by: Massimo Pegorer <massimo.pegorer@vimar.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
This matches how this would be done in Linux and these functions
do the alignment for us which makes the code look cleaner.
Signed-off-by: Andrew Davis <afd@ti.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Now that single defconfig shall be used for booting J721S2 EVM and
AM68 SK, the default device tree will not work for selecting dtb for
kernel. Update the findfdt env to select right dtb based on
board_name env variable.
Signed-off-by: Sinthu Raja <sinthu.raja@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
The SK architecture comprises of baseboard and a SOM board. The
AM68 Starter Kit's baseboard contains most of the actual connectors,
power supply etc. The System on Module (SoM) is plugged on to the base
board. Therefore, add support for peripherals brought out in the base
board.
Schematics: https://www.ti.com/lit/zip/SPRR463
Signed-off-by: Sinthu Raja <sinthu.raja@ti.com>
AM68 Starter Kit (SK) is a low cost, small form factor board designed
for TI’s AM68 SoC. TI’s AM68 SoC comprises of dual core A72, high
performance vision accelerators, hardware accelerators, latest C71x
DSP, high bandwidth real-time IPs for capture and display. The SoC is
power optimized to provide best in class performance for industrial
applications.
AM68 SK supports the following interfaces:
* 16 GB LPDDR4 RAM
* x1 Gigabit Ethernet interface
* x1 USB 3.1 Type-C port
* x2 USB 3.1 Type-A ports
* x1 PCIe M.2 M Key
* 512 Mbit OSPI flash
* x2 CSI2 Camera interface (RPi and TI Camera connector)
* 40-pin Raspberry Pi GPIO header
SK's System on Module (SoM) contains the SoC, PMIC, DDR and OSPI flash.
Therefore, add support for the components present on the SoM.
Schematics: https://www.ti.com/lit/zip/SPRR463
TRM: http://www.ti.com/lit/pdf/spruj28
Signed-off-by: Sinthu Raja <sinthu.raja@ti.com>
Add the board_init_f API for SPL and run the platform-required SoC
initialization.
Add the functionality for board name-based DTB selection from FIT
within SPL. This will make it easier to utilise one defconfig for
both the EVM and the SK.
Signed-off-by: Sinthu Raja <sinthu.raja@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
Although the board_init_f API initialises the SoC, the API name is
incorrectly specified and misleads the functionality. This file should
only include k3-specific functionality. Change the API's name to something
more K3-specific and separate the function to make it more modular.
Signed-off-by: Sinthu Raja <sinthu.raja@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
Update the board_fit_config_name_match() to choose the right dtb
based on the board name read from EEPROM.
Also restrict multpile EEPROM reads by verifying if EEPROM is already
read
Signed-off-by: Sinthu Raja <sinthu.raja@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
J721S2 EVM has EEPROM populated at 0x50. AM68 SK has EEPROM populated at
next address 0x51 in order to be compatible with RPi. So start looking
for TI specific EEPROM at 0x50, if not found look for EEPROM at 0x51.
Signed-off-by: Sinthu Raja <sinthu.raja@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
Update setup_board_eeprom_env() to choose the right board name
for am68-sk.
Signed-off-by: Sinthu Raja <sinthu.raja@ti.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
J721S2 EVM used to store env on eMMC, since EVM and SK uses same
defconfig and there is no eMMC on SK, we need to keep env in an
interface which available on both EVM and SK. So, save env in FAT
partition of MMC SD Card.
Enable defconfigs relevant for storing env on FAT partition of MMC.
Signed-off-by: Sinthu Raja <sinthu.raja@ti.com>
Enable configs for building multiple dtbs into a single fit image
and load the right dtb for next stage. Add k3-am68-sk-base-board
dtb along with evm dtb inside DTB FIT image. This helps to use same
defconfig for both EVM and SK
Signed-off-by: Sinthu Raja <sinthu.raja@ti.com>
Enable configs for building multiple dtbs into a single fit image
and load the right dtb for next stage. This will help to use same
defconfig for both EVM and SK.
Signed-off-by: Sinthu Raja <sinthu.raja@ti.com>
Add the related include files to the power MAINTAINERS entry.
Signed-off-by: John Keeping <john@metanate.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
On a second and third look, a recent patch seems to be writing to the
wrong place - updating offsets from the address of the pointer instead
of what the pointer points to.
Fix it.
Signed-off-by: Simon Glass <sjg@chromium.org>
Fixes: 2d1b2ac13f ("tool: ifwitool: Fix buffer overflow")
Acked-by: Sean Anderson <seanga2@gmail.com>
Don't define an empty version for pinconfig_post_bind(). Just guard the
call and let the linker garbage collection do the rest. This way, we
also don't have to do any guesswork.
Signed-off-by: Michael Walle <michael@walle.cc>
Reviewed-by: Marek Vasut <marex@denx.de>
If CONFIG_PINCTRL_FULL is enabled, never fall back to the simple
implementation. pinctrl_select_state() is called for each device and it
is expected to fail. A fallback to the simple imeplementation doesn't
make much sense.
To keep the return code consistent, we need to change the -EINVAL (which
was ignored before) to -ENOSYS.
Signed-off-by: Michael Walle <michael@walle.cc>
Reviewed-by: Marek Vasut <marex@denx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
makefile:
- Add multi_dtb_fit dependency
clk:
- Handle error cases
microblaze:
- Disable falcon mode and cleanup code around
xilinx:
- Enable regular expression matching in board_fit_config_name_match()
- Fix FRU handling for 0xC1 format
- Fix Xilinx legacy format eeprom parsing
zynqmp:
- Some DT updates/cleanups
- Fix IDcode for xck24
- Remove empty mini config files
- Add support for k24
versal:
- Remove empty mini config files
versal_net:
- Setup timer when runs in EL3
- Build u-boot.elf for mini configurations
zynq-gem:
- Add support for new compatible strings
- Remove support for Avnet Ultrazedev SOM
- Handle SGMII with PCS phy
spi:
- Add support for gigadevice parts
misc:
- Remove CONFIG_TARGET_VENUS ifdef
- Add missing headers to remove sparse warnings
-----BEGIN PGP SIGNATURE-----
iF0EABECAB0WIQQbPNTMvXmYlBPRwx7KSWXLKUoMIQUCY9PmzwAKCRDKSWXLKUoM
IYWxAJwNgiXtFT17pBy+1M0JvSWgyvWF3QCfTQG2Krnw0+d65PF5A7UFTDL7SZM=
=cMrc
-----END PGP SIGNATURE-----
Merge tag 'xilinx-for-v2023.04-rc1' of https://source.denx.de/u-boot/custodians/u-boot-microblaze
Xilinx chnages for v2023.04-rc1
makefile:
- Add multi_dtb_fit dependency
clk:
- Handle error cases
microblaze:
- Disable falcon mode and cleanup code around
xilinx:
- Enable regular expression matching in board_fit_config_name_match()
- Fix FRU handling for 0xC1 format
- Fix Xilinx legacy format eeprom parsing
zynqmp:
- Some DT updates/cleanups
- Fix IDcode for xck24
- Remove empty mini config files
- Add support for k24
versal:
- Remove empty mini config files
versal_net:
- Setup timer when runs in EL3
- Build u-boot.elf for mini configurations
zynq-gem:
- Add support for new compatible strings
- Remove support for Avnet Ultrazedev SOM
- Handle SGMII with PCS phy
spi:
- Add support for gigadevice parts
misc:
- Remove CONFIG_TARGET_VENUS ifdef
- Add missing headers to remove sparse warnings
FRU data is not displayed properly in case of 0xc1 fields.
The 0xC1 can be used in two cases.
1. Char record type 8-bit ASCII + Latin 1 with length of 1.
(For example board revision 'A')
2. C1h (type/length byte encoded to indicate no more info fields).
which can follow by 00h to fill all remaining unused space
Hence removed the check end-of-the field c1 to allow c1 fields.
"ASCII+LATIN1" is defined as the printable characters from the
first set of 256 characters of Unicode 6.2 (U+0000h through U+00FFh,
inclusive) expressed as an eight-bit value. (Unicode follows ISO/IEC
8859-1 in the layout of printable characters up to U+00FFh).
So, print only printable chars and limit range from 0x20 ' ' to 0x7e '-'
which will be also indication if 0xc1 behaves as record with one char or
end of record.
Signed-off-by: Algapally Santosh Sagar <santoshsagar.algapally@amd.com>
Signed-off-by: Michal Simek <michal.simek@amd.com>
Link: https://lore.kernel.org/r/4198d73de600627872c80a5b07e5068502c589d7.1674648379.git.michal.simek@amd.com
When eeprom has random content printing random chars can stuck U-Boot.
That's why update legacy eeprom format decoding algorithm to copy only
maximum amount of chars allocated for fields.
And also print them directly from desc structure.
Previous algorithm was printing strings first directly from eeprom content
and then copy them to desc structure.
Signed-off-by: Michal Simek <michal.simek@amd.com>
Link: https://lore.kernel.org/r/42065fcbb1a10581f9f4f091d64b43c01fe595c6.1674573561.git.michal.simek@amd.com
In our system using ZynqMP with an external SGMII PHY it's necessary
to wait for the PCS link and auto negotiation to finish before the xfer
starts. Otherwise the first packet(s) might get dropped, resulting in a
delay at the start of the ethernet transfers.
This is only done when the PHY link is already up, which is done in
phy_startup(). As waiting for the PHY link bits via pcsstatus does not
make much sense, when the link is not available in general (e.g. no
cable connected).
This patch adds the necessary code including a minimal delay of 1 ms
which fixes problems of dropped first packages.
Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Michal Simek <michal.simek@amd.com>
Cc: Katakam Harini <harini.katakam@amd.com>
Cc: Ramon Fried <rfried.dev@gmail.com>
Cc: Sean Anderson <sean.anderson@seco.com>
Link: https://lore.kernel.org/r/20230125070908.1343256-1-sr@denx.de
Signed-off-by: Michal Simek <michal.simek@amd.com>
tuning_loop_counter is of char type, which is not capable of handling
the entire data range of this variable. This is pointed by below sparse
warning. Change datatype to int to fix this.
warning: comparison is always false due to limited range of data type.
Signed-off-by: Algapally Santosh Sagar <santoshsagar.algapally@amd.com>
Signed-off-by: Ashok Reddy Soma <ashok.reddy.soma@amd.com>
Link: https://lore.kernel.org/r/20230120053617.32463-5-ashok.reddy.soma@amd.com
Signed-off-by: Michal Simek <michal.simek@amd.com>
There is a unused variable ret, due to which we are getting sparse warning
as below.
warning: variable 'ret' set but not used [-Wunused-but-set-variable].
Return ret incase of error.
Signed-off-by: Algapally Santosh Sagar <santoshsagar.algapally@amd.com>
Signed-off-by: Ashok Reddy Soma <ashok.reddy.soma@amd.com>
Link: https://lore.kernel.org/r/20230120053617.32463-3-ashok.reddy.soma@amd.com
Signed-off-by: Michal Simek <michal.simek@amd.com>
Prototype is missing for board_get_usable_ram_top, which is pointed by
below sparse warning. Include init.h header file to fix this.
warning: no previous prototype for 'board_get_usable_ram_top'
[-Wmissing-prototypes].
Signed-off-by: Algapally Santosh Sagar <santoshsagar.algapally@amd.com>
Signed-off-by: Ashok Reddy Soma <ashok.reddy.soma@amd.com>
Link: https://lore.kernel.org/r/20230120053617.32463-2-ashok.reddy.soma@amd.com
Signed-off-by: Michal Simek <michal.simek@amd.com>
U-Boot is composing u-boot.bin from u-boot-nodtb.bin with appended
dts/dt.dtb. It means U-Boot doesn't have DTB inside. When REMAKE_ELF is
enabled make will also create u-boot.elf which is recreated from
u-boot.bin. Below is build output for mini configuration how ELF is
recreated.
cat arch/arm/dts/versal-net-mini.dtb > dts/dt.dtb
cat u-boot-nodtb.bin dts/dt.dtb > u-boot-dtb.bin
cp dts/dt.dtb u-boot.dtb
cp u-boot-dtb.bin u-boot.bin
aarch64-linux-gnu-objcopy -I binary -B aarch64 -O elf64-littleaarch64
u-boot.bin u-boot-elf.o
aarch64-linux-gnu-ld.bfd u-boot-elf.o -o u-boot.elf -EL -T u-boot-elf.lds
--defsym="_start"=0xBBF00000 -Ttext=0xBBF00000
It is useful to have u-boot.elf present because Xilinx XSDB debugger can
load ELF file and user doesn't need to specify loading address for
u-boot.bin.
Signed-off-by: Michal Simek <michal.simek@amd.com>
Link: https://lore.kernel.org/r/ca550c5c4bf908e757a49a83fdfab0e7100de45e.1674121617.git.michal.simek@amd.com
Drop falcon mode configs from microblaze-generic_defconfig, so that a
defconfig build will still boot into u-boot proper.
Signed-off-by: Ovidiu Panait <ovpanait@gmail.com>
Link: https://lore.kernel.org/r/20230125164157.1638680-3-ovpanait@gmail.com
Signed-off-by: Michal Simek <michal.simek@amd.com>
Drop boot_linux variable as it is not assigned anywhere. Now that there is
no variable controlling linux boot in spl_start_uboot(), make this
function always return false when falcon mode is enabled.
Signed-off-by: Ovidiu Panait <ovpanait@gmail.com>
Link: https://lore.kernel.org/r/20230125164157.1638680-2-ovpanait@gmail.com
Signed-off-by: Michal Simek <michal.simek@amd.com>
Make spl_start_uboot() available only if CONFIG_SPL_OS_BOOT is enabled,
since it is only used for falcon mode.
Signed-off-by: Ovidiu Panait <ovpanait@gmail.com>
Link: https://lore.kernel.org/r/20230125164157.1638680-1-ovpanait@gmail.com
Signed-off-by: Michal Simek <michal.simek@amd.com>
Binman positioning by ELF symbol
-----BEGIN PGP SIGNATURE-----
iQFFBAABCgAvFiEEslwAIq+Gp8wWVbYnfxc6PpAIreYFAmPTNLgRHHNqZ0BjaHJv
bWl1bS5vcmcACgkQfxc6PpAIreagiwgAr1Q+8qPghgomUK9cVjBuNzG2i88YWy7/
oitshEE73lm92kP4YIQrSm1ZBy6mm0A0wyy4pRLC0fFrovzAWq1o7xfUEkuhxyTV
pR7BiLPGUsHfL1cP4EuGSNMgrfX0QOddQgZTns4s7k4fbrFBLmTTa/+1jrU3AJNW
1FVfVSc4eMqHM5gD7mfqSHRsxrbZDHpzbJEKIMS1xkVy/BSQbDWtF1f6Lri3M9VT
Q7eNxFH5OptRAttcQreVNSNu28z00x0TuJNKLORXJa5AAxGW4yvNWgDFegU10NZc
ADTKaNDsDz4D//Ar3PSt+10eGWWKqrY9ClQwls4lJmJJTIqldf+QDw==
=l2A4
-----END PGP SIGNATURE-----
Merge tag 'dm-pull-26jan23' of https://source.denx.de/u-boot/custodians/u-boot-dm
FIT improvements with split-elf, especially for Rockchip
Binman positioning by ELF symbol
Unfortunately a recent patch snuck through without the require test
coverage. Fix it.
Signed-off-by: Simon Glass <sjg@chromium.org>
Fixes: 571bc4e67d ("binman: Support positioning an entry by and ELF symbol")
Use a more accurate check for determining if the full format string will
be handled correctly, since SPL_USE_TINY_PRINTF can be disabled.
Signed-off-by: Samuel Holland <samuel@sholland.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
This property sets the minimum size of an entry, including padding but
not alignment. It can be used to reserve space for growth of an entry,
or to enforce a minimum offset for later entries in the section.
Signed-off-by: Samuel Holland <samuel@sholland.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
The FIT generated after the switch to using binman is using different
values for firmware and loadables properties compared to the old script.
With the old script:
firmware = "atf-1";
loadables = "u-boot", "atf-2", ...;
After switch to binman:
firmware = "u-boot";
loadables = "atf-1", "atf-2", ...;
This change result in SPL jumping directly into U-Boot proper instead of
initializing TF-A.
With this patch the properties change back to:
firmware = "atf-1";
loatables = "u-boot", "atf-2", ...;
Fixes: e0c0efff2a ("rockchip: Support building the all output files in binman")
Signed-off-by: Jonas Karlman <jonas@kwiboo.se>
Reviewed-by: Simon Glass <sjg@chromium.org>
In some cases it is desired for SPL to start TF-A instead of U-Boot
proper. Add support for a new property fit,firmware that picks a
valid entry and prepends the remaining valid entries to the
loadables list generated by the split-elf generator.
Signed-off-by: Jonas Karlman <jonas@kwiboo.se>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
Add sha256 hash to FIT images when CONFIG_SPL_FIT_SIGNATURE=y.
Signed-off-by: Jonas Karlman <jonas@kwiboo.se>
Reviewed-by: Simon Glass <sjg@chromium.org>
Special nodes, hash and signature, is not being added to the nodes
generated for each segment in split-elf operation.
Copy the subnode logic used in _gen_fdt_nodes to _gen_split_elf to
ensure special nodes are added to the generated nodes.
Signed-off-by: Jonas Karlman <jonas@kwiboo.se>
Reviewed-by: Simon Glass <sjg@chromium.org>
SPL load FIT images by reading the data aligned to block length.
Block length aligned image data is read directly to the load address.
Unaligned image data is written to an offset of the load address and
then the data is memcpy to the load address.
This adds a small overhead of having to memcpy unaligned data, something
that normally is not an issue.
However, TF-A may have a segment that should be loaded into SRAM, e.g.
vendor TF-A for RK3568 has a 8KiB segment that should be loaded into the
8KiB PMU SRAM. Having the image data for such segment unaligned result
in segment being written to and memcpy from beyond the SRAM boundary, in
the end this results in invalid data in SRAM.
Aligning the FIT and its external data to MMC block length to work
around such issue.
Signed-off-by: Jonas Karlman <jonas@kwiboo.se>
Reviewed-by: Simon Glass <sjg@chromium.org>
Add support to indicate what alignment to use for the FIT and its
external data. Pass the alignment to mkimage via the -B flag.
Signed-off-by: Jonas Karlman <jonas@kwiboo.se>
Reviewed-by: Simon Glass <sjg@chromium.org>
