This series adds support for the Cadence PCIe controller on TI's K3
family of SoCs to which the J7200 SoC belongs. The driver is an adaptation
of the Linux driver (drivers/pci/controller/cadence/pci-j721e.c) and has
been implemented specifically for Root-Complex mode of operation on the
J7200 SoC. A minor set of changes will be sufficient to support other K3
SoCs as well with plans to implement it in the near future.
configs: j7200_evm_a72_defconfig: Enable configs for PCI support
TI's J7200 SoC has a single instance of PCIe Controller namely PCIe1 which
is a Cadence PCIe Controller. To support PCIe functionality with the PCIe1
instance of PCIe, enable the corresponding configs.
We experience huge problems with cache handling on Qualcomm
systems, and it appears the dcache handling in the DWC3 gadget
code is quite wrong and causes operational issues.
This serie fixes the dcache operations on unaligned data,
and properly invalidate buffers when reading back data from
hardware.
Neil Armstrong [Fri, 11 Oct 2024 14:38:26 +0000 (16:38 +0200)]
usb: dwc3: invalidate dcache on buffer used in interrupt handling
On Qualcomm systems, the setup buffer and even buffers are in
a bad state at interrupt handling, so invalidate the dcache lines
for the setup_buf and event buffer to make sure we read correct
data written by the hardware.
This fixes the following error:
dwc3-generic-peripheral usb@a600000: UNKNOWN IRQ type -1
dwc3-generic-peripheral usb@a600000: UNKNOWN IRQ type 4673109
and invalid situation in dwc3_gadget_giveback() because setup_buf content
is read at 0s and leads to fatal crash fixed by [1].
Neil Armstrong [Fri, 11 Oct 2024 14:38:25 +0000 (16:38 +0200)]
usb: dwc3: fix dcache flush range calculation
The current flush operation will omit doing a flush/invalidate on
the first and last bytes if the base address and size are not aligned
with CACHELINE_SIZE.
This causes operation failures Qualcomm platforms.
Take in account the alignment and size of the buffer and also
flush the previous and last cacheline.
Neil Armstrong [Fri, 11 Oct 2024 14:38:24 +0000 (16:38 +0200)]
usb: dwc3: allocate setup_buf with dma_alloc_coherent()
Since setup_buf is also consumed by hardware DMA, aligns it's
allocation like other hardware buffers by introduce setup_buf_addr
populated by dma_alloc_coherent(), and use it to pass the physical
address of the buffer to the hardware.
Alex Shumsky [Tue, 15 Oct 2024 20:29:31 +0000 (23:29 +0300)]
i2c: i2c-gpio: add support for i2c-gpio,sda-output-only
Some I2C slave devices are read-only and don't even answer with NACK.
For example FD65x segment LED controllers.
Make them usable with i2c-gpio,sda-output-only that are already supported
by Linux 6.3+.
* Add document describing Ethernet boot on AM62x SoC
* Fix typo in blkmap command example
UEFI:
* Avoid #ifdef in efi_setup.c
* Reduce message noisiness if ESP is missing
* Remove ERROR:, WARNING: prefixes in messages
* Use blk_create_devicef() in block device driver
cmd/upl.c:75:51: warning: format ‘%lx’ expects argument of type
‘long unsigned int’, but argument 3 has type ‘size_t’
{aka ‘unsigned int’} [-Wformat=]
75 | printf("UPL handoff written to %lx size %lx\n", addr, abuf_size(&buf));
| ~~^ ~~~~~~~~~~~~~~~
| | |
| | size_t {aka unsigned int}
| long unsigned int
| %x
Caleb Connolly [Fri, 18 Oct 2024 13:51:07 +0000 (15:51 +0200)]
lmb: notify when adjacent regions are added
lmb_add_region() returns a positive integer if the added regions causes
existing regions to be coalesced. We still want to notify the EFI
subsystem about these added regions though, so adjust lmb_add() to only
bail on errors.
This fixes EFI memory allocation on boards with adjacent memory banks as is the
case on several Qualcomm boards like the RB3 Gen 2.
Andy Shevchenko [Sat, 5 Oct 2024 19:12:00 +0000 (22:12 +0300)]
x86: cpu: Add missing header inclusion
Without asm/cpu_x86.h inclusion a compiler is not happy:
arch/x86/cpu/cpu_x86.c:14:5: warning: no previous prototype for ‘cpu_x86_bind’ [-Wmissing-prototypes]
arch/x86/cpu/cpu_x86.c:29:5: warning: no previous prototype for ‘cpu_x86_get_vendor’ [-Wmissing-prototypes]
arch/x86/cpu/cpu_x86.c:41:5: warning: no previous prototype for ‘cpu_x86_get_desc’ [-Wmissing-prototypes]
arch/x86/cpu/cpu_x86.c:55:5: warning: no previous prototype for ‘cpu_x86_get_count’ [-Wmissing-prototypes]
Andy Shevchenko [Sat, 5 Oct 2024 19:11:59 +0000 (22:11 +0300)]
x86: cpu: Add a few prototypes to the header file
The compiler is not happy to have no prototypes for the functions that
are not defined static. Add them. This helps avoiding the compiler warnings:
arch/x86/cpu/cpu.c:197:13: warning: no previous prototype for ‘board_final_init’ [-Wmissing-prototypes]
arch/x86/cpu/cpu.c:205:13: warning: no previous prototype for ‘board_final_cleanup’ [-Wmissing-prototypes]
arch/x86/cpu/cpu.c:307:5: warning: no previous prototype for ‘reserve_arch’ [-Wmissing-prototypes]
Andy Shevchenko [Sat, 5 Oct 2024 19:11:58 +0000 (22:11 +0300)]
x86: cpu: Mark a few functions static
Some functions are not used anywhere except the same file
where they are defined. Mark them static. This helps avoiding
the compiler warnings:
arch/x86/cpu/cpu.c:343:6: warning: no previous prototype for ‘detect_coreboot_table_at’ [-Wmissing-prototypes]
arch/x86/cpu/mtrr.c:90:6: warning: no previous prototype for ‘mtrr_write_all’ [-Wmissing-prototypes]
arch/x86/cpu/i386/interrupt.c:240:6: warning: no previous prototype for ‘__do_irq’ [-Wmissing-prototypes]
Andy Shevchenko [Sat, 5 Oct 2024 19:11:57 +0000 (22:11 +0300)]
x86: cpu: Use default print_cpuinfo() for all
Most of the copies of the print_cpuinfo() call the default method.
Remove all of those in order to have only the default one when
no `cpu` command is compiled.
This also helps avoiding compiler warning, e.g.:
arch/x86/cpu/tangier/tangier.c:23:5: warning: no previous prototype for ‘print_cpuinfo’ [-Wmissing-prototypes]
Simon Glass [Mon, 14 Oct 2024 22:31:58 +0000 (16:31 -0600)]
expo: Allow menu items to have values
At present menu items are stored according to their sequence number in
the menu. In some cases we may want to have holes in that sequence, or
not use a sequence at all.
Add a new 'value' property for menu items. This will be used for
reading and writing, if present. If there is no 'value' property, then
the normal sequence number will be used instead.
Simon Glass [Mon, 14 Oct 2024 22:31:57 +0000 (16:31 -0600)]
expo: Use standard numbering for save and discard
Set aside some expo IDs for 'save' and 'discard' buttons. This avoids
needing to store the IDs for these. Adjust the documentation and expo
tool for the new EXPOID_BASE_ID value.
Ignore these objects when saving and loading the cedit, since they do
not contain real data.
Adjust 'cedit run' to return failure when the user exits the expo
without saving. Update the test for this change as well.
Simon Glass [Mon, 14 Oct 2024 22:31:55 +0000 (16:31 -0600)]
expo: Place menu items to the right of all labels
At present a fixed position is used for menu items, 200 pixels to the
right of the left side of the labels. This means that a menu item with
a very long label may overlap the items.
It seems better to calculate the maximum label width and then place the
items to the right of all of them.
To implement this, add a new struct to containing arrangement
information. Calculate it before doing the actual arrangement. Add a
new style item which sets the amount of space from the right side of
the labels to left side of the items.
Simon Glass [Mon, 14 Oct 2024 22:31:54 +0000 (16:31 -0600)]
video: Avoid starting a new line to close to the bottom
When starting a new text line, an assumption is made that the current
vertical position is a multiple of the character height. When this is
not true, characters can be written after the end of the framebuffer.
This can causes crashes and strange errors from QEMU.
Adjust the scrolling check when processing a newline character, to
avoid any problems.
Add some comments to make things a little clearer.
Simon Glass [Mon, 14 Oct 2024 22:31:53 +0000 (16:31 -0600)]
video: Add a dark-grey console colour
This is useful for highlighting something with a black background, as
is needed with cedit when using a white-on-black console. Add this as
a new colour.
Marek Vasut [Sat, 5 Oct 2024 14:33:37 +0000 (16:33 +0200)]
sandbox: dts: Add missing sandbox, emul phandle to sandbox-pmic
This phandle was missing in the sandbox DT, add it, otherwise sandbox-i2c
driver cannot look up the emulator via i2c_emul_find(). This fixes the
following i2c_emul_find() error:
"
$ ./u-boot -Dc ""
...
i2c_emul_find() No emulators for device 'sandbox_pmic'
sandbox_pmic_write() write error to device: 0000000018c568d0 register: 0x0!
out_set_value() PMIC write failed: -5
i2c_emul_find() No emulators for device 'sandbox_pmic'
sandbox_pmic_write() write error to device: 0000000018c568d0 register: 0x0!
out_set_value() PMIC write failed: -5
...
"
Simon Glass [Mon, 30 Sep 2024 18:51:36 +0000 (12:51 -0600)]
u_boot_pylib: Support running coverage on selected functions
At present run_test_coverage() assumes you want code coverage for the
entire code base. This is the normal situation, but sometimes it is
useful to see the coverage provided by just a single test. Add support
for this.
Simon Glass [Wed, 25 Sep 2024 10:44:53 +0000 (12:44 +0200)]
sandbox_spl: Restrict use of UPL when not enabled
With sandbox_spl we want to use the file-based boot in CI, so that this
flow is tested. The recent UPL change enabled booting via that method,
thus overriding the file-based boot.
Correct this by using UPL only when the --upl flag is given.
Signed-off-by: Simon Glass <[email protected]> Fixes: 91fde8e1764 ("sandbox: Add an SPL loader for UPL")
binman: implement signing FIT images during image build
The patch implement new property 'fit,sign' that can be declared
at the top-level 'fit' node. If that option is declared, fit tryies
to detect private keys directory among binman include directories.
That directory than passed to mkimage using '-k' flag and that enable
signing of FIT.
binman: fix passing loadables to mkimage on first run
FIT use mkimage from BuildSectionData() to build FIT entry contents.
BuildSectionData() get called several times during building FIT image.
Currently when fit inserts loadables, it use self._loadables property that
contain loadables computed during previuos BuildSectionData() invocation.
So for the first run it use empty list and pass no loadables to mkimage.
That makes problem for adding signature to FIT image because mkimage
fails to add signature and aborts building FIT if no loadables provided.
The patch fixes described behaviour in a way that BuildSectionData() uses
recently calculated loadables value, not previosly calculated.
When BLK is not enabled but BOOTSTD is, some features of standard boot
become unavailable. Add a check for this in the only site that is
currently apparent.
Simon Glass [Sun, 1 Sep 2024 22:27:27 +0000 (16:27 -0600)]
blk: Make functions available unconditionally
Some boards still don't enable BLK but we want to be able to at least
compile the code which relies on this. For example, bootstd includes
calls to blk_...() functions, albeit with a check for BLK so that the
code is eliminated by the compiler.
Reduce the scope of the BLK #ifdef to help with this.
Simon Glass [Wed, 28 Aug 2024 01:44:26 +0000 (19:44 -0600)]
x86: Avoid timer-clock overflow
When the clock speed is above about 4GHz, e.g. on modern PC hardware,
the timer overflows, resulting in a much lower frequency than expected.
Deal with this by capping the clock speed.
It would be possible to move to a 64-bit value for the clock, but that
is a pain to deal with. A better approach might be to express the clock
in MHz but that is left for later consideration.
Tom Rini [Mon, 22 Jan 2024 14:35:12 +0000 (09:35 -0500)]
x86: Make default_print_cpuinfo be a weak alias for print_cpuinfo
While a few SoCs have a unique print_cpuinfo function, a number of them
just use default_print_cpuinfo. Make default_print_cpuinfo have a weak
alias to provie print_cpuinfo.
Simon Glass [Mon, 8 Jan 2024 00:14:55 +0000 (17:14 -0700)]
coreboot: Switch to a monospaced font
The default font is proportional, with different character widths.
Select a monospace font for coreboot so that the 'dm tree' output lines
up correctly.
The EFI block device driver is the only user of blk_create_device() outside
the block device uclass. Use blk_create_devicef() instead like other block
device drivers.
- Switch to using upstream DT on DH i.MX8MP DHCOM PDK2/PDK3.
- Add ability to build fallback DTBOs from arch/$(ARCH)/dts.
- Remove fdt_high and initrd_high env variables from imx6-dhcom.
- Add dummy clk for imx8.
- Fix DT corruption in imx8_cpu.
- Improve DDR stability on pico-imx7d.
Tom Rini [Tue, 15 Oct 2024 01:33:10 +0000 (19:33 -0600)]
Azure: Update to latest containers
Soon Azure will be removing the macOS-12 container in following their
normal support schedule. Move us to macOS-14 so we won't have problems
there for a while. At the same time, our Windows container is the oldest
supported, so move to the newer option. Finally, Ubuntu 22.04 is the
middle option currently, but 24.04 should be fine.
EFI variables can be stored in a file on the EFI system partition. If that
partition is missing we are writing two error messages per variable. This
is too noisy.
Peng Fan [Fri, 18 Oct 2024 07:34:32 +0000 (15:34 +0800)]
cpu: imx8_cpu: Avoid revision to corrupt device tree
U-Boot device tree is padded just after U-Boot proper.
After the whole stuff loaded to DRAM space, the device tree
area is conflict with BSS region before U-Boot relocation.
So any write to BSS area before reloc_fdt will corrupt the
device tree. Without the fix, there is issue that “binman_init
failed:-2” on i.MX8MP-EVK board.
Drop 'revision' and use malloc area in cpu_imx_plat->rev.
Marek Vasut [Tue, 15 Oct 2024 23:44:32 +0000 (01:44 +0200)]
ARM: dts: imx6-dhcom: Remove fdt_high and initrd_high env variables
Remove both "fdt_high" and "initrd_high" environment variables
in favor of "bootm_size" to safely contain a kernel, device tree
and initrd for relocation inside of 256 MiB region of DRAM.
Marek Vasut [Thu, 17 Oct 2024 01:19:17 +0000 (22:19 -0300)]
arm64: dts: imx: Switch to using upstream DT on DH i.MX8MP DHCOM PDK2/PDK3
Enable OF_UPSTREAM to use upstream DT and add freescale/ prefix to the
DEFAULT_DEVICE_TREE. And thereby directly build DTB from dts/upstream/src/
including *-u-boot.dtsi and DTBOs from arch/$(ARCH)/dts/ directory.
Marek Vasut [Thu, 17 Oct 2024 01:08:40 +0000 (03:08 +0200)]
dts: Add ability to build fallback DTBOs from arch/$(ARCH)/dts
Currently the enablement of OF_UPSTREAM results on the build system
searching for DTs only in dts/upstream/ . There are platforms which
use U-Boot specific DTBOs applied on top of U-Boot control DT during
SPL stage, and source DTs for these are located in arch/$(ARCH)/dtb.
Add dedicated 'dtbos' target which builds only .dtbos and not .dtbs and
in case CONFIG_OF_UPSTREAM_INCLUDE_LOCAL_FALLBACK_DTBOS is enabled, build
this target for arch/$(ARCH)/dtb to generate local U-Boot specific DTBOs.
Adjust top level Makefile so binman would search for .dtb and .dtbo in
both OF_UPSTREAM specific paths and arch/$(ARCH)/dtb for the .dtbo case
in case CONFIG_OF_UPSTREAM_INCLUDE_LOCAL_FALLBACK_DTBOS is enabled.
Alexander Dahl [Tue, 1 Oct 2024 12:07:33 +0000 (14:07 +0200)]
ARM: at91: clock: Add missing include of asm/io.h
In one inline function in this header `readl()` is used, but the
declaration was not found, so buildman gave this warning:
w+include/asm/arch/clk.h: In function 'get_h32mxdiv':
w+include/asm/arch/clk.h:65:16: warning: implicit declaration of function 'readl' [-Wimplicit-function-declaration]
The channel allocation and deallocation for dma copy was happening on every
dma transfer. This is a overhead for transactions like NAND, which does
page reads recursively for complete data.
So, moving the dma allocation to probe and implement corresponding
remove function and cleanup dma device while exiting from spl.
Enable SPL_DM_DEVICE_REMOVE, for device removal capability in SPL.
dma: ti: k3-udma: Move DMA channel[0] allocation to probe and add udma_remove()
Currently, the allocation of DMA channel[0] for memcpy is happening
in udma_transfer() for every transfer, which leads to a huge overhead
for each transfer, especially in case of nand page reads. So, move this
allocation to udma_probe(), as a result, the allocation is done once
during probe.
Introduce udma_remove() for the cleanup of allocated channel during
probe.
dma: ti: k3-udma: Move udma_probe() below all APIs
The udma_probe() function was placed above many important APIs
related to bcdma, pktdma, which restricts these APIs to be accessed
during probe. So, move udma_probe() below all of them.
While exiting from spl, remove any dma device active through
spl_board_prepare_for_boot(). This is required for cleaning up
any dma channels being used in spl and avoid issues with overlapping
channel allocation in the next stage bootloaders.
AM62x devices now support CPU throttling based on thermal alerts
with a Linux commit 10e7bfd8114c ("arm64: dts: ti: k3-am62: Enable
CPU freq throttling on thermal alert"). However, this functionality
does not work correctly across all variants of the AM62x SoCs
which have different numbers of Cortex-A CPU cores: AM62x1 (1 core),
AM62x2 (2 cores), and AM62x4 (4 cores). On single-core and dual-core
AM62x devices, the following error is observed in the Linux kernel:
OF: /thermal-zones/main0-thermal/cooling-maps/map0: could not find
phandle 94
OF: /thermal-zones/main1-thermal/cooling-maps/map0: could not find
phandle 94
This commit adds a fixup to dynamically adjust the cooling-device
nodes in the thermal zones based on the actual number of CPU cores
available. This resolves the issue of CPU throttling not working
correctly on single-core and dual-core AM62x devices, while
maintaining the functionality for AM62x quad-core devices.
A similar approach is implemented for example on i.MX8MM SoC.
Lad Prabhakar [Tue, 1 Oct 2024 09:56:47 +0000 (10:56 +0100)]
lib: fdtdec: Parse the gzip/lzo headers only when dependencies have met
It might happen that CONFIG_GZIP and CONFIG_LZO are enabled but we might
have CONFIG_MULTI_DTB_FIT_LZO enabled in this case in the code path of
uncompress_blob() we parse the gzip headers first which results in
`Error: Bad gzipped data` being printed. To avoid this parse the gzip/lzo
headers only when dependencies have met.
Jan Kiszka [Mon, 30 Sep 2024 10:20:36 +0000 (12:20 +0200)]
Extend usage for OF_OVERLAY_LIST beyond SPL
Allow to use OF_OVERLAY_LIST also for the case that the overlays just
need be built, e.g. when they will be picked up by binman as artifacts
of the final U-Boot image. The IOT2050 boards have such a need when
switching to OF_UPSTREAM.
- stm32mp: fix boot issue with OP-TEE
- stm32mp: Add script to install U-Boot from SD/eMMC to SPI NOR on DH STM32MP15xx
- stm32mp: Switch to using upstream DT on DH STM32 DHSOM
- stm32mp: Generate u-boot.itb using binman on DH STM32 DHSOM
Add new section for USB TypeC Port Manager (TCPM) support, which
is needed to figure out cable orientation of USB-C plus and to do
USB PD communication.
projects / u-boot.git / log