strn(cat|cpy) has a bad habit of not nul-terminating the destination,
resulting in constructions like

	strncpy(foo, bar, sizeof(foo) - 1);
	foo[sizeof(foo) - 1] = '\0';

However, it is very easy to forget about this behavior and accidentally
leave a string unterminated. This has shown up in some recent coverity
scans [1, 2] (including code recently touched by yours truly).

Fortunately, the guys at OpenBSD came up with strl(cat|cpy), which always
nul-terminate strings. These functions are already in U-Boot, so we should
encourage new code to use them instead of strn(cat|cpy).

[1] https://lists.denx.de/pipermail/u-boot/2021-March/442888.html
[2] https://lists.denx.de/pipermail/u-boot/2021-January/438073.htmlSigned-off-by: NSean Anderson <seanga2@gmail.com>
Reviewed-by: NSimon Glass <sjg@chromium.org>

This fixes several uses of strn(cpy|cat) which did not terminate their
destinations properly.

Fixes de1728ce ("fastboot: Allow u-boot-style partitions")

Reported-by: Coverity Scan
Signed-off-by: NSean Anderson <seanga2@gmail.com>

This test is adapted from glibc, which is very concerned about alignment.
It also tests strlcpy by dependency.
Signed-off-by: NSean Anderson <seanga2@gmail.com>
Reviewed-by: NSimon Glass <sjg@chromium.org>

This introduces strlcat, which provides a safer interface than strncat. It
never copies more than its size bytes, including the terminating nul. In
addition, it never reads past dest[size - 1], even if dest is not
nul-terminated.

This also removes the stub for dwc3 now that we have a proper
implementation.
Signed-off-by: NSean Anderson <seanga2@gmail.com>
Reviewed-by: NSimon Glass <sjg@chromium.org>

strlcpy should always return the number of bytes copied. We were
accidentally missing the nul-terminator. We also always used to return a
non-zero value, even if we did not actually copy anything.

Fixes: 23cd1385 ("Integrate USB gadget layer and USB CDC driver layer")
Signed-off-by: NSean Anderson <seanga2@gmail.com>

Right now the error messages when optee has a version mismatch or shared
memory is not configured are done with a debug().
That's not very convenient since you have to enable debugging to figure
out what's going on, although this is an actual error.

So let's switch the debug() -> dev_err() and report those explicitly.
Signed-off-by: NIlias Apalodimas <ilias.apalodimas@linaro.org>
Reviewed-by: NPatrick Delaunay <patrick.delaunay@foss.st.com>

serial_reinit_all() is only available if CONFIG_SERIAL is defined (i.e.
!CONFIG_DM_SERIAL).
Signed-off-by: NAsherah Connor <ashe@kivikakk.ee>
Reviewed-by: NSimon Glass <sjg@chromium.org>

stdio_dev methods have taken a pointer to themselves since 709ea543
(nearly 7 years ago).
Signed-off-by: NAsherah Connor <ashe@kivikakk.ee>
Reviewed-by: NSimon Glass <sjg@chromium.org>

Add support for rtc3028 rtc from microcrystal.
based on linux dirver:
commit a38fd8748464: ("Linux 5.12-rc2")
Signed-off-by: NHeiko Schocher <hs@denx.de>
Reviewed-by: NStefan Roese <sr@denx.de>

Add local variables agent0/agent1 to refer to SCMI sandbox context
agent and ease readability of the test.

For consistency, rename regul_dev to regul0_dev and remove sandbox_voltd
in dm_test_scmi_voltage_domains().
Signed-off-by: NEtienne Carriere <etienne.carriere@linaro.org>
Reviewed-by: NSimon Glass <sjg@chromium.org>

Fix inline comments and empty line in scmi driver and test files.

Remove test on IS_ENABLED(CONFIG_*_SCMI) in test/dm/scmi.c since these
configuration are expected enabled when CONFIG_FIRMWARE_SCMI is enabled
in sandbox configuration.
Signed-off-by: NEtienne Carriere <etienne.carriere@linaro.org>
Reviewed-by: NSimon Glass <sjg@chromium.org>
Reviewed-by: NJaehoon Chung <jh80.chung@samsung.com>

Implement sandbox regulator devices for SCMI voltage domains
and test them in DM scmi tests.
Signed-off-by: NEtienne Carriere <etienne.carriere@linaro.org>
Reviewed-by: NSimon Glass <sjg@chromium.org>
Reviewed-by: NJaehoon Chung <jh80.chung@samsung.com>

Implement voltage regulators interfaced by the SCMI voltage domain
protocol. The DT bindings are defined in the Linux kernel since
SCMI voltage domain and regulators patches [1] and [2] integration
in v5.11-rc7.

Link: [1] https://github.com/torvalds/linux/commit/0f80fcec08e9c50b8d2992cf26495673765ebaba
Link: [2] https://github.com/torvalds/linux/commit/2add5cacff3531e54c50b0832128299faa9f0563Signed-off-by: NEtienne Carriere <etienne.carriere@linaro.org>
Reviewed-by: NJaehoon Chung <jh80.chung@samsung.com>

The gpt command require the GPT backup header at the standard location
at the end of the device. Check the alternate LBA value before reading
the GPT backup header from the last usable LBA of the device.
Signed-off-by: NStefan Herbrechtsmeier <stefan.herbrechtsmeier@weidmueller.com>
Reviewed-by: NHeinrich Schuchardt <xypron.glpk@gmx.de>

This variable is a boolean, not a string.
Signed-off-by: NSean Anderson <sean.anderson@seco.com>
Reviewed-by: NHeinrich Schuchardt <xypron.glpk@gmx.de>

Just a bit more info to the reader.
Signed-off-by: NPhil Sutter <phil@nwl.cc>
Reviewed-by: NStefan Roese <sr@denx.de>

If there were no variable substitutions in a command, then initial
assignments would be misinterpreted as commands, instead of being skipped
over. This is demonstrated by the following example:

	=> foo=bar echo baz
	Unknown command 'foo=bar' - try 'help'
Signed-off-by: NSean Anderson <seanga2@gmail.com>

For all other erase failures, the fail_addr is updated with the
failing address. Only in the case of erase failure due to bad block
detection, the fail_addr is not updated. This change simply updates
the fail_addr for this specific scenario so that it is consistent with
the rest of the code.
Signed-off-by: NFarhan Ali <farhan.ali@broadcom.com>

It's an old bringup board with out upstream Linux or L4T support
Signed-off-by: NPeter Robinson <pbrobinson@gmail.com>
Cc: Stephen Warren <swarren@nvidia.com>
Cc: Tom Warren <twarren@nvidia.com>
Acked-by: NStephen Warren <swarren@nvidia.com>

On ARMv8 systems

    load mmc 0:1 $loadaddr  vmlinuz-5.10.0-3-arm64
    booti

leads to a hanging system requiring to physically reset the system:

    FDT and ATAGS support not compiled in - hanging
    ### ERROR ### Please RESET the board ###

For systems where physical access is difficult hanging is a poor choice.
It is preferable to reset the system when U-Boot reaches a state that is
not recoverable.
Signed-off-by: NHeinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: NSimon Glass <sjg@chromium.org>

Similar to support for SHA1 and SHA256, allow the use of hardware hashing
engine by enabling the algorithm and setting  CONFIG_SHA_HW_ACCEL /
CONFIG_SHA_PROG_HW_ACCEL.
Signed-off-by: NJoel Stanley <joel@jms.id.au>

Checkpatch complains about:

 ERROR: "foo * bar" should be "foo *bar"

and

 CHECK: Alignment should match open parenthesis
Signed-off-by: NJoel Stanley <joel@jms.id.au>

The optee_copy_fdt_nodes is only used to copy op-tee nodes
of U-Boot device tree (from gd->fdt_blob when OF_LIVE is not activated)
to external device tree but it is not compatible with OF_LIVE.

This patch migrates all used function fdt_ functions to read node on
old_blob to ofnode functions, compatible with OF_LIVE and remove this
parameter "old_blob".

The generated "device tree" is checked on stm32mp platform with OF_LIVE
activated.
Signed-off-by: NPatrick Delaunay <patrick.delaunay@foss.st.com>

Remove pmic_max77696.c file.
The maintaining pmic_max77696.c file is useless.
Signed-off-by: NJaehoon Chung <jh80.chung@samsung.com>

Add power_max77696_init() function.
Since warp doesn't support DM, the keeping its code in board file is
better than maintainig the file of driver.
Signed-off-by: NJaehoon Chung <jh80.chung@samsung.com>

- Remove a large number of boards that have not migrated to DM_MMC, for
  which the migration deadline with 2 years ago at v2019.04.

 - rk3399 eDP support
 - pwm backlight without a known period_ns
 - add Chrome OS EC PWM driver
 - Kconfig SIMPLE_PANEL DM_GPIO dependency
 - remove mb862xx driver remnants
 - fix KiB format in reserve_video() debug trace
 - fix tegra124 sor CSTM LVDS_EN_ENABLE/DISABLE config
 - fix line padding calculation for 16 and 24 BPP bitmaps

Pull request for efi-2021-07-rc1

Bug fixes:

* support EFI, HOST, VIRTIO in fsinfo command
* simplify efi_get_device_path_text()
* add missing EFI_UNACCEPTED_MEMORY_TYPE
* mkeficapsule: improve online help
* avoid several build warnings

Documentation:

* UEFI documentation for initrd loading options
* describe building OP-TEE with for UEFI variables
* mmc man-page

This board has not been converted to CONFIG_DM_MMC by the deadline.
Remove it.

Cc: Priyanka Jain <priyanka.jain@nxp.com>
Signed-off-by: NTom Rini <trini@konsulko.com>

Each row in the pixel array in the bitmap file is padded
if necessary so the row size is always a multiple of 4 bytes.
In current code the complement of row size to a multiple of
4 bytes is further unnecessarily multiplied by the pixel size.
This results in incorrect displaying of bitmaps having row size
that is not a multiple of 4 bytes. Fix this by removing
the unnecessary multiplication.

Tested with 24BPP bitmap and XRGB32 display.
Signed-off-by: NSylwester Nawrocki <s.nawrocki@samsung.com>
Tested-by: NJaehoon Chung <jh80.chung@samsung.com>

Bitwise OR has a higher operator precedence than the ternary conditional.
Add the missing parentheses.
Signed-off-by: NHeinrich Schuchardt <xypron.glpk@gmx.de>

Update the debug trace for the reserved video memory to KB as indicated
in the message with "%luk"; before the patch the computed size
gd->relocaddr - addr is in bytes.

This patch aligns the debug trace in reserve_video() with others
functions, for example on stm32mp157c-dk2:

  - Reserving 3080192k for video at: dfd00000
  + Reserving 3008k for video at: dfd00000
    Reserving 873k for U-Boot at: dfc25000
    Reserving 32776k for malloc() at: ddc23000
    Reserving 72 Bytes for Board Info at: ddc22fb0
    Reserving 280 Bytes for Global Data at: ddc22e90
    Reserving 119072 Bytes for FDT at: ddc05d70
    Reserving 0x278 Bytes for bootstage at: ddc05af0

Fixes: 5630d2fb ("board: Show memory for frame buffers")
Signed-off-by: NPatrick Delaunay <patrick.delaunay@foss.st.com>
Reviewed-by: NSimon Glass <sjg@chromium.org>

drivers/video/mb862xx.c was removed in commit
9c1e098f from December 2020, however, this
last little remnant in drivers/video/cfb_console.c remained.
Signed-off-by: NTrevor Woerner <twoerner@gmail.com>

SIMPLE_PANEL currently only depends on PANEL && BACKLIGHT, but the code
makes references to dm_gpio_set_value and gpio_request_by_name.  These
are defined in drivers/gpio/gpio-uclass.c, so a dependency on DM_GPIO
corrects these link errors:

aarch64-linux-gnu-ld.bfd: drivers/built-in.o: in function `simple_panel_set_backlight':
/home/kameliya/u-boot/drivers/video/simple_panel.c:42: undefined reference to `dm_gpio_set_value'
aarch64-linux-gnu-ld.bfd: drivers/built-in.o: in function `simple_panel_enable_backlight':
/home/kameliya/u-boot/drivers/video/simple_panel.c:27: undefined reference to `dm_gpio_set_value'
aarch64-linux-gnu-ld.bfd: drivers/built-in.o: in function `simple_panel_of_to_plat':
/home/kameliya/u-boot/drivers/video/simple_panel.c:72: undefined reference to `gpio_request_by_name'

This issue is only exposed if you have a board which enables
CONFIG_DM_VIDEO without CONFIG_DM_GPIO; so far, none do, but soon a QEMU
board may.
Signed-off-by: NAsherah Connor <ashe@kivikakk.ee>

This PWM is used in rk3399-gru-bob and rk3399-gru-kevin to control
the display brightness. We can only change the duty cycle, so on
set_config() we just try to match the duty cycle that dividing duty_ns
by period_ns gives us. To disable, we set the duty cycle to zero while
keeping the old value for when we want to re-enable it.

The cros_ec_set_pwm_duty() function is taken from Depthcharge's
cros_ec_set_bl_pwm_duty() but modified to use the generic pwm type.
The driver itself is very loosely based on rk_pwm.c for the general pwm
driver structure.

The devicetree binding file is from Linux, before it was converted to
YAML at 5df5a577a6b4 ("dt-bindings: pwm: Convert google,cros-ec-pwm.txt
to YAML format") in their repo.
Signed-off-by: NAlper Nebi Yasak <alpernebiyasak@gmail.com>
Reviewed-by: NSimon Glass <sjg@chromium.org>

The PWM device provided by Chrome OS EC doesn't really support anything
other than setting a relative duty cycle. To support it as a backlight,
this patch makes the PWM period optional in the device tree and pretends
the valid brightness range is its period_ns.

Also adds a sandbox test for a PWM channel that has a fixed period,
checking that the resulting duty_cycle matches on a set_config() even if
the requested period_ns can't be set.
Signed-off-by: NAlper Nebi Yasak <alpernebiyasak@gmail.com>
Reviewed-by: NSimon Glass <sjg@chromium.org>

These boards have not been converted to CONFIG_DM_MMC by the deadline.
Remove them.  As the P5020 is the last ARCH_P5020 platform, remove that
support as well.

Cc: Andy Fleming <afleming@gmail.com>
Cc: Priyanka Jain <priyanka.jain@nxp.com>
Signed-off-by: NTom Rini <trini@konsulko.com>
Reviewed-by: NPriyanka Jain <priyanka.jain@nxp.com>

With the last of the ARCH_P1022 platforms removed, finish removing the
rest of the platform support.

Cc: Priyanka Jain <priyanka.jain@nxp.com>
Signed-off-by: NTom Rini <trini@konsulko.com>
Reviewed-by: NPriyanka Jain <priyanka.jain@nxp.com>

These boards have not been converted to CONFIG_DM_MMC by the deadline.
Remove them.

Cc: Mario Six <mario.six@gdsys.cc>
Cc: Dirk Eibach <dirk.eibach@gdsys.cc>
Signed-off-by: NTom Rini <trini@konsulko.com>

These boards have not been converted to CONFIG_DM_MMC, along with other
DM conversions, by the deadline.  Remove them.

Cc: Dirk Eibach <dirk.eibach@gdsys.cc>
Signed-off-by: NTom Rini <trini@konsulko.com>