Sync include/ec_commands.h from upstream commit 4f3d17aa34
("skywalker: set SLEEP_TIMEOUT_MS to 50 seconds"). The new file makes
two build assumptions that do not hold for U-Boot.

It hides '<stdint.h>' from __KERNEL__ builds, leaving UINT16_MAX
(used by EC_RES_MAX) undefined for U-Boot; widen the gate to
'!defined(__KERNEL__) || defined(__UBOOT__)'

It gates '<linux/limits.h>' on '#ifdef __KERNEL__'; the matching
'#else' branch defines BIT()/BIT_ULL()/GENMASK()/GENMASK_ULL()
locally, assuming kernel headers provide those macros otherwise.
U-Boot defines __KERNEL__ too but has no <linux/limits.h>. Nest a
'!defined(__UBOOT__)' check around the include so the __UBOOT__ path
stays in the __KERNEL__ branch (no local BIT/GENMASK defines), which
avoids redefinition warnings against U-Boot's linux/bitops.h. Pull
in linux/bitops.h up front for U-Boot so the file's own BIT() and
GENMASK() uses still resolve.

Adapt callers to two interface changes. The 'ec_current_image' enum
tag is now 'ec_image' (EC_IMAGE_* constants unchanged); rename it in
affected files to match. The VBNV-context interface was dropped
upstream, but it still used in lab Chromebooks; keep those constants and
structs in cros_ec.h

Likewise, MEC_EMI_BASE and MEC_EMI_SIZE are a U-Boot-local addition to
ec_commands.h that the upstream sync removes; preserve them in cros_ec.h
next to the VBNV block, and switch the only consumer
(arch/x86/cpu/apollolake/cpu_spl.c) to include cros_ec.h

Signed-off-by: Simon Glass <sjg@chromium.org>

If the LED is in the ON state, it is briefly set to OFF
then to ON immediately due to falling-through in the default
case.
This commit ensures that no fall-through occurs and thus
a LED initially in the ON state is turned off before blinking.

Signed-off-by: Francois Berder <fberder@outlook.fr>
Fixes: 9e3d83301e4f ("led: toggle LED on initial SW blink")
Acked-by: Quentin Schulz <quentin.schulz@cherry.de>
Reviewed-by: Simon Glass <sjg@chromium.org>

gpio status -a does not have labels: the existing path walks
the per-bank requested label table.
Issue: The boards that populate the standard gpio-line-names
property in their device tree end up with anonymous entries,
which is not logic with the purpose of having those names in the DT.

No impact with boards that does not set gpio-line-names.

Signed-off-by: Vincent Jardin <vjardin@free.fr>

Prepare v2026.07-rc3

AMD/Xilinx/FPGA changes for v2026.07-rc3

versal/fpga:
- Fix unaligned buffer handling

versal2:
- Fix buffer overflow in SOC name array

Our official domain is now u-boot-project.org, so update all in-tree
references to use the correct domain.

Reviewed-by: Tony Dinh <mibodhi@gmail.com>
Reviewed-by: Peter Robinson <pbrobinson@gmail.com>
Signed-off-by: Tom Rini <trini@konsulko.com>

When fpga load is called with a misaligned buffer address, the
versal_align_dma_buffer() function shifts the pointer forward to the
next aligned boundary and uses memcpy() to copy the data. Since the
destination is ahead of the source and the regions overlap, memcpy()
produces undefined behavior; in practice U-Boot's generic memcpy()
copies forward, repeating the first ARCH_DMA_MINALIGN-aligned chunk
throughout the buffer.

Replace memcpy() with memmove() which correctly handles overlapping
regions by copying backwards when the destination is ahead of the
source.

Fixes: 26e054c943a7 ("arm64: versal: fpga: Add PL bit stream load support")
Signed-off-by: Pranav Tilak <pranav.vinaytilak@amd.com>
Signed-off-by: Michal Simek <michal.simek@amd.com>
Link: https://lore.kernel.org/r/20260507113359.3665220-1-pranav.vinaytilak@amd.com

Currently we are trying to work out if the vendor id is from
a virtio-mmio device and then casting a u32 to a char* and using
it as a C-string. By chance there is usually a zero after the u32
and it works.

Since the vendor id we are trying to convert to a string is QEMU's
just define a value for the QEMU vendor id, check if the vendor
id matches and then use a predefined string for "QEMU".

I don't think we should have been assumming all virtio-mmio vendor
ids are printable ASCII chars in the first place so do this special
casing just for QEMU. If the vendor id isn't QEMU print the hex
value of it.

Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Kuan-Wei Chiu <visitorckw@gmail.com>
Signed-off-by: Daniel Palmer <daniel@thingy.jp>

The m68k QEMU virt machine doesn't use devicetree, yet, so
allow it to create virtio-mmio instances via platform data.

Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Kuan-Wei Chiu <visitorckw@gmail.com>
Reviewed-by: Angelo Dureghello <angelo@kernel-space.org>
Signed-off-by: Daniel Palmer <daniel@thingy.jp>

The Goldfish timer registers are native endian, so they act as
big-endian on the m68k virt machine. Currently, this driver uses
readl(), which works by luck because it's currently broken on m68k.

Use __raw_readl() instead to avoid breaking this driver when the
endianness of readl() is fixed.

Signed-off-by: Kuan-Wei Chiu <visitorckw@gmail.com>
Tested-by: Daniel Palmer <daniel@thingy.jp>
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Daniel Palmer <daniel@thingy.jp>