]> git.ipfire.org Git - thirdparty/u-boot.git/log
thirdparty/u-boot.git
4 years agophy: adin: add driver for Analog Devices ADIN1300 PHY
Nate Drude [Fri, 8 Apr 2022 16:28:14 +0000 (11:28 -0500)] 
phy: adin: add driver for Analog Devices ADIN1300 PHY

The current implementation configures RGMII using device tree phy-mode
property and then calls genphy_config

adin_config_rgmii_mode is derived from:
https://github.com/varigit/linux-imx/blob/lf-5.10.y_var04/drivers/net/phy/adin.c#L218-L262

Signed-off-by: Nate Drude <nate.d@variscite.com>
Reviewed-by: Ramon Fried <rfried.dev@gmail.com>
4 years agonet: eth-phy: prevent undesired de-assertion of phy-reset on request
Tim Harvey [Tue, 1 Mar 2022 20:15:02 +0000 (12:15 -0800)] 
net: eth-phy: prevent undesired de-assertion of phy-reset on request

When gpio_request_by_name allocates a gpio output it by default will
de-assert the gpio which for phy-reset will take the PHY out of reset.
As this occurs before eth_phy_reset is called to assert the reset
line it can cause undesired affects if reset timings are not properly
met.

Configure the gpio with GPIOD_IS_OUT_ACTIVE so that reset is kept active
(reset asserted) to avoid this.

Cc: Sean Anderson <sean.anderson@seco.com>
Signed-off-by: Tim Harvey <tharvey@gateworks.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Reviewed-by: Ramon Fried <rfried.dev@gmail.com>
4 years agonet: fec: prevent undesired de-assertion of phy-reset on request
Tim Harvey [Tue, 1 Mar 2022 20:15:01 +0000 (12:15 -0800)] 
net: fec: prevent undesired de-assertion of phy-reset on request

When gpio_request_by_name allocates a gpio output it by default will
de-assert the gpio which for phy-reset will take the PHY out of reset.
As this occurs before fec_gpio_reset is called to assert the reset
line it can cause undesired affects if reset timings are not properly
met.

Configure the gpio with GPIOD_IS_OUT_ACTIVE so that reset is kept active
(reset asserted) to avoid this.

Cc: Sean Anderson <sean.anderson@seco.com>
Signed-off-by: Tim Harvey <tharvey@gateworks.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
Tested-by: Adam Ford <aford173@gmail.com> #imx8mm-beacon
Reviewed-by: Ramon Fried <rfried.dev@gmail.com>
4 years agonet: phy: don't require PHY interface mode during PHY creation
Marek Behún [Wed, 6 Apr 2022 22:33:08 +0000 (00:33 +0200)] 
net: phy: don't require PHY interface mode during PHY creation

Currently we require PHY interface mode to be known when
finding/creating the PHY - the functions
  * phy_connect_phy_id()
  * phy_device_create()
  * create_phy_by_mask()
  * search_for_existing_phy()
  * get_phy_device_by_mask()
  * phy_find_by_mask()
all require the interface parameter, but the only thing done with it is
that it is assigned to phydev->interface.

This makes it impossible to find a PHY device without overwriting the
set mode.

Since the interface mode is not used during .probe() and should be used
at first in .config(), drop the interface parameter from these
functions. Make the default value of phydev->interface (in
phy_device_create()) to be PHY_INTERFACE_MODE_NA. Move the interface
parameter to phy_connect_dev(), where it should be.

Change all occurrences treewide. In occurrences where we don't call
phy_connect_dev() for some reason (they only configure the PHY without
connecting it to an ethernet controller), set
  phydev->interface = value from phy_find_by_mask call.

Signed-off-by: Marek Behún <marek.behun@nic.cz>
Reviewed-by: Ramon Fried <rfried.dev@gmail.com>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
4 years agobcmgenet, sun8i_emac: Don't connect PHY two times
Marek Behún [Wed, 6 Apr 2022 22:33:07 +0000 (00:33 +0200)] 
bcmgenet, sun8i_emac: Don't connect PHY two times

The bcmgenet and sun8i_emac drivers call phy_connect(), which finds /
creates the PHY and also connects it to the eth device via
phy_connect_dev(), then set some phydev members (bcmgenet only), and
then call phy_connect_dev() explicitly again.

Drop the second phy_connect_dev(), since it is unnecesary.

Signed-off-by: Marek Behún <marek.behun@nic.cz>
Reviewed-by: Ramon Fried <rfried.dev@gmail.com>
4 years agonet: phy: use ->is_c45 instead of is_10g_interface()
Marek Behún [Wed, 6 Apr 2022 22:33:06 +0000 (00:33 +0200)] 
net: phy: use ->is_c45 instead of is_10g_interface()

Use phydev->is_c45 instead of is_10g_interface(phydev->interface) to
determine whether clause 45 protocol should be used.

Signed-off-by: Marek Behún <marek.behun@nic.cz>
Reviewed-by: Ramon Fried <rfried.dev@gmail.com>
4 years agonet: phy: xilinx: Check interface type in ->config(), not ->probe()
Marek Behún [Wed, 6 Apr 2022 22:33:05 +0000 (00:33 +0200)] 
net: phy: xilinx: Check interface type in ->config(), not ->probe()

We want to be able to have phydev->interface uninitialized during
->probe(). We should assume that phydev->interface is initialized only
before ->config().

Signed-off-by: Marek Behún <marek.behun@nic.cz>
Reviewed-by: Ramon Fried <rfried.dev@gmail.com>
4 years agophy: Move PHY_INTERFACE_MODE_NA to the beginning of the enum definition
Marek Behún [Wed, 6 Apr 2022 22:33:04 +0000 (00:33 +0200)] 
phy: Move PHY_INTERFACE_MODE_NA to the beginning of the enum definition

Move PHY_INTERFACE_MODE_NA to the beginning of the enum definition to
make it have zero value. This makes it possible (although not
encouraged) to test for invalid/nonexistent interface mode with !val
instead of val == PHY_INTERFACE_MODE_NA.

The comment near the definition says "Must be last", because when the
constant was introduced in commit 5f184715ecd3 ("Create PHY Lib for
U-Boot"), it was used as the maximum value when interating over the
constants. But this is no longer true - we use PHY_INTERFACE_MODE_MAX
for that now, and so we can move it.

Signed-off-by: Marek Behún <marek.behun@nic.cz>
Reviewed-by: Stefan Roese <sr@denx.de>
Reviewed-by: Ramon Fried <rfried.dev@gmail.com>
4 years agotreewide: Rename PHY_INTERFACE_MODE_NONE to PHY_INTERFACE_MODE_NA
Marek Behún [Wed, 6 Apr 2022 22:33:03 +0000 (00:33 +0200)] 
treewide: Rename PHY_INTERFACE_MODE_NONE to PHY_INTERFACE_MODE_NA

Rename constant PHY_INTERFACE_MODE_NONE to PHY_INTERFACE_MODE_NA to make
it compatible with Linux' naming.

Signed-off-by: Marek Behún <marek.behun@nic.cz>
Reviewed-by: Stefan Roese <sr@denx.de>
Reviewed-by: Ramon Fried <rfried.dev@gmail.com>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
4 years agotreewide: Rename PHY_INTERFACE_MODE_COUNT to PHY_INTERFACE_MODE_MAX
Marek Behún [Wed, 6 Apr 2022 22:33:02 +0000 (00:33 +0200)] 
treewide: Rename PHY_INTERFACE_MODE_COUNT to PHY_INTERFACE_MODE_MAX

Rename constant PHY_INTERFACE_MODE_COUNT to PHY_INTERFACE_MODE_MAX to
make it compatible with Linux' naming.

Signed-off-by: Marek Behún <marek.behun@nic.cz>
Reviewed-by: Stefan Roese <sr@denx.de>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
4 years agonet: introduce helpers to get PHY interface mode from a device/ofnode
Marek Behún [Wed, 6 Apr 2022 22:33:01 +0000 (00:33 +0200)] 
net: introduce helpers to get PHY interface mode from a device/ofnode

Add helpers ofnode_read_phy_mode() and dev_read_phy_mode() to parse the
"phy-mode" / "phy-connection-type" property. Add corresponding UT test.

Use them treewide.

This allows us to inline the phy_get_interface_by_name() into
ofnode_read_phy_mode(), since the former is not used anymore.

Signed-off-by: Marek Behún <marek.behun@nic.cz>
Reviewed-by: Ramon Fried <rfried.dev@gmail.com>
Tested-by: Patrice Chotard <patrice.chotard@foss.st.com>
4 years agonet: phy: fix parsing wrong property
Marek Behún [Wed, 6 Apr 2022 22:33:00 +0000 (00:33 +0200)] 
net: phy: fix parsing wrong property

The "phy-interface-type" property should be "phy-connection-type".

Signed-off-by: Marek Behún <marek.behun@nic.cz>
Reviewed-by: Ramon Fried <rfried.dev@gmail.com>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
4 years agotreewide: use dm_mdio_read/write/reset() wrappers
Marek Behún [Wed, 6 Apr 2022 22:32:59 +0000 (00:32 +0200)] 
treewide: use dm_mdio_read/write/reset() wrappers

Use the new dm_mdio_read/write/reset() wrappers treewide, instead of
always getting and dereferencing MDIO operations structure pointer.

Signed-off-by: Marek Behún <marek.behun@nic.cz>
Reviewed-by: Ramon Fried <rfried.dev@gmail.com>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
4 years agonet: mdio-uclass: add wrappers for read/write/reset operations
Marek Behún [Wed, 6 Apr 2022 22:32:58 +0000 (00:32 +0200)] 
net: mdio-uclass: add wrappers for read/write/reset operations

Add wrappers dm_mdio_read(), dm_mdio_write() and dm_mdio_reset() for
DM MDIO's .read(), .write() and .reset() operations.

Signed-off-by: Marek Behún <marek.behun@nic.cz>
Reviewed-by: Ramon Fried <rfried.dev@gmail.com>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
4 years agonet: introduce helpers to get PHY ofnode from MAC
Marek Behún [Wed, 6 Apr 2022 22:32:57 +0000 (00:32 +0200)] 
net: introduce helpers to get PHY ofnode from MAC

Add helpers ofnode_get_phy_node() and dev_get_phy_node() and use it in
net/mdio-uclass.c function dm_eth_connect_phy_handle(). Also add
corresponding UT test.

This is useful because other part's of U-Boot may want to get PHY ofnode
without connecting a PHY.

Signed-off-by: Marek Behún <marek.behun@nic.cz>
Reviewed-by: Ramon Fried <rfried.dev@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
4 years agonet: mdio-uclass: use ARRAY_SIZE()
Marek Behún [Wed, 6 Apr 2022 22:32:56 +0000 (00:32 +0200)] 
net: mdio-uclass: use ARRAY_SIZE()

Use the ARRAY_SIZE() macro instead of hardcoding sizes of arrays in
macros.

Signed-off-by: Marek Behún <marek.behun@nic.cz>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
4 years agonet: mdio-uclass: fix type for phy_mode_str and phy_handle_str
Marek Behún [Wed, 6 Apr 2022 22:32:55 +0000 (00:32 +0200)] 
net: mdio-uclass: fix type for phy_mode_str and phy_handle_str

These global variables should both have type
  static const char * const

Signed-off-by: Marek Behún <marek.behun@nic.cz>
Reviewed-by: Ramon Fried <rfried.dev@gmail.com>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
4 years agonet: phy: dp83867: Fix a never true comparison
Haolin Li [Sat, 19 Mar 2022 14:02:42 +0000 (07:02 -0700)] 
net: phy: dp83867: Fix a never true comparison

The type of the return value of phy_read() and phy_read_mmd() is int.
Change the variable to not be unsigned so that we not get into an
unsigned compared against 0.

Signed-off-by: Haolin Li <li.haolin@qq.com>
Reviewed-by: Ramon Fried <rfried.dev@gmail.com>
4 years agonet: phy: atheros: avoid error in ar803x_of_init() when PHY has no OF node
Vladimir Oltean [Wed, 23 Feb 2022 13:20:56 +0000 (15:20 +0200)] 
net: phy: atheros: avoid error in ar803x_of_init() when PHY has no OF node

A DM_ETH driver may use phy_connect() towards a PHY address on an MDIO
bus which is not specified in the device tree, as evidenced by:

pfe_eth_probe
-> pfe_phy_configure
   -> phy_connect

When this happens, the PHY will have an invalid OF node.

When ar803x_config() runs, it silently fails at ar803x_of_init(), and
therefore, fails to run the rest of the initialization.

This makes MII_BMCR contain what it had after BMCR_RESET (0x8000) has
been written into it by phy_reset(). Since BMCR_RESET is volatile and
self-clearing, the MII_BMCR ends up having a value of 0x0. The further
configuration of this register, which is supposed to be handled by
genphy_config_aneg() lower in ar803x_config(), never gets a chance to
run due to this early error from ar803x_of_init().

As a result of having MII_BMCR as 0, the following symptom appears:

=> setenv ethact pfe_eth0
=> setenv ipaddr 10.0.0.1
=> ping 10.0.0.2
pfe_eth0 Waiting for PHY auto negotiation to complete......... TIMEOUT !
Could not initialize PHY pfe_eth0

Manually writing 0x1140 into register 0 of the PHY makes the connection
work, but it is rather desirable that the port works without any manual
intervention.

Fixes: fe6293a80959 ("phy: atheros: add device tree bindings and config")
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Ramon Fried <rfried.dev@gmail.com>
4 years agonet: phy: dp83867: avoid error in dp83867_of_init() when PHY has no OF node
Vladimir Oltean [Wed, 23 Feb 2022 13:20:55 +0000 (15:20 +0200)] 
net: phy: dp83867: avoid error in dp83867_of_init() when PHY has no OF node

A DM_ETH driver may use phy_connect() towards a PHY address on an MDIO
bus which is not specified in the device tree, as evidenced by:

pfe_eth_probe
-> pfe_phy_configure
   -> phy_connect

When this happens, the PHY will have an invalid OF node.

The dp83867_config() method has extra initialization steps which are
bypassed when the PHY lacks an OF node, which is undesirable because it
will lead to broken networking. Allow the rest of the code to run.

Fixes: 085445ca4104 ("net: phy: ti: Allow the driver to be more configurable")
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Ramon Fried <rfried.dev@gmail.com>
4 years agodm: add tag support
AKASHI Takahiro [Tue, 8 Mar 2022 11:36:46 +0000 (20:36 +0900)] 
dm: add tag support

With dm-tag feature, any U-Boot subsystem is allowed to associate
arbitrary number of data with a particular udevice. This can been
see as expanding "struct udevice" without modifying the definition.

As a first user, UEFI subsystem makes use of tags to associate
an efi_disk object with a block device.

Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
4 years agovirtio: call device_probe() in scanning
AKASHI Takahiro [Tue, 8 Mar 2022 11:36:45 +0000 (20:36 +0900)] 
virtio: call device_probe() in scanning

virtio_init() enumerates all the peripherals that are to be materialised
with udevices(UCLASS_VIRIO) and creates particular device instances
(UCLASS_BlK or whatever else) as children.
On the other hand, device_probe() won't be invoked against those resultant
udevices unlike other ordinary device drivers do in the driver model.

This is particularly inconvenient when we want to add "event notification"
callback so that we will be able to automatically create all efi_disk
objects in a later patch.

With this patch applied, "virtio scan" will work in a similar way
to "scsi rescan", "usb start" or others in term of 'probe' semantics.

I didn't add this change to virtio_init() itself because this function
may be called in board_init_r() (indirectly in board_late_init())
before UEFI subsustem is initialized.

Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
4 years agoblock: ide: call device_probe() after scanning
AKASHI Takahiro [Tue, 8 Mar 2022 11:36:44 +0000 (20:36 +0900)] 
block: ide: call device_probe() after scanning

Every time an ide bus/port is scanned and a new device is detected,
we want to call device_probe() as it will give us a chance to run
additional post-processings for some purposes.

In particular, support for creating partitions on a device will be added.

Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
4 years agosata: call device_probe() after scanning
AKASHI Takahiro [Tue, 8 Mar 2022 11:36:43 +0000 (20:36 +0900)] 
sata: call device_probe() after scanning

Every time a sata bus/port is scanned and a new device is detected,
we want to call device_probe() as it will give us a chance to run
additional post-processings for some purposes.

In particular, support for creating partitions on a device will be added.

Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
4 years agonvme: call device_probe() after scanning
AKASHI Takahiro [Tue, 8 Mar 2022 11:36:42 +0000 (20:36 +0900)] 
nvme: call device_probe() after scanning

Every time a nvme bus/port is scanned and a new device is detected,
we want to call device_probe() as it will give us a chance to run
additional post-processings for some purposes.

In particular, support for creating partitions on a device will be added.

Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
4 years agommc: call device_probe() after scanning
AKASHI Takahiro [Tue, 8 Mar 2022 11:36:41 +0000 (20:36 +0900)] 
mmc: call device_probe() after scanning

Every time a mmc bus/port is scanned and a new device is detected,
we want to call device_probe() as it will give us a chance to run
additional post-processings for some purposes.

In particular, support for creating partitions on a device will be added.

Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
Reviewed-by: Jaehoon Chung <jh80.chung@samsung.com>
4 years agousb: storage: call device_probe() after scanning
AKASHI Takahiro [Tue, 8 Mar 2022 11:36:40 +0000 (20:36 +0900)] 
usb: storage: call device_probe() after scanning

Every time a usb bus/port is scanned and a new device is detected,
we want to call device_probe() as it will give us a chance to run
additional post-processings for some purposes.

In particular, support for creating partitions on a device will be added.

Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
4 years agoscsi: call device_probe() after scanning
AKASHI Takahiro [Tue, 8 Mar 2022 11:36:39 +0000 (20:36 +0900)] 
scsi: call device_probe() after scanning

Every time a scsi bus/port is scanned and a new block device is detected,
we want to call device_probe() as it will give us a chance to run
additional post-processings for some purposes.

In particular, support for creating partitions on a device will be added.

Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
4 years agoefi_loader: Use sysreset instead of reset command
Masami Hiramatsu [Mon, 21 Mar 2022 13:37:56 +0000 (22:37 +0900)] 
efi_loader: Use sysreset instead of reset command

Use sysreset_walk_halt() directly from reset-after-capsule-on-disk
feature to reboot (cold reset) machine instead of using reset command
interface, since this is not a command.
Note that this will make CONFIG_EFI_CAPSULE_ON_DISK depending on
the CONFIG_SYSRESET.

Signed-off-by: Masami Hiramatsu <masami.hiramatsu@linaro.org>
Reviewed-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
4 years agoefi_loader: Make efi_load_capsule_drivers() available even if EFI_CAPSULE_ON_DISK=n
Masami Hiramatsu [Mon, 21 Mar 2022 13:37:45 +0000 (22:37 +0900)] 
efi_loader: Make efi_load_capsule_drivers() available even if EFI_CAPSULE_ON_DISK=n

Make efi_load_capsule_drivers() available even if EFI_CAPSULE_ON_DISK
is disabled because the caller (efi_init_obj_list()) expects it only
relays on EFI_HAVE_CAPSULE_SUPPORT.

Suggested-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Masami Hiramatsu <masami.hiramatsu@linaro.org>
4 years agocmd: efidebug: Disable 'capsule disk-update' when CONFIG_EFI_CAPSULE_ON_DISK=n
Masami Hiramatsu [Mon, 21 Mar 2022 13:37:35 +0000 (22:37 +0900)] 
cmd: efidebug: Disable 'capsule disk-update' when CONFIG_EFI_CAPSULE_ON_DISK=n

Disable 'capsule disk-update' option for the efidebug command
when CONFIC_EFI_CAPSULE_ON_DISK is disabled, because this option
is available only when the EFI capsule update on disk is enabled.

Signed-off-by: Masami Hiramatsu <masami.hiramatsu@linaro.org>
Suggested-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
4 years agoefi_loader: optional persistence of variables
Tom Saeger [Tue, 22 Mar 2022 21:21:10 +0000 (15:21 -0600)] 
efi_loader: optional persistence of variables

Since be66b89da306 ("efi_loader: configuration of variables store")
the choice of EFI_VARIABLE_FILE_STORE or EFI_MM_COMM_TEE
is mutually-exclusive, however efi_var_to_file also allows
for "neither". Provide an additional Kconfig option.

Signed-off-by: Tom Saeger <tom.saeger@oracle.com>
Reviewed-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
4 years agodoc: distro: Update list of all support boot types
Pali Rohár [Wed, 6 Apr 2022 09:35:46 +0000 (11:35 +0200)] 
doc: distro: Update list of all support boot types

Add HOST, UBIFS.

Signed-off-by: Pali Rohár <pali@kernel.org>
Reviewed-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
4 years agodoc: debug UART for RISC-V QEMU virt machine
Heinrich Schuchardt [Tue, 5 Apr 2022 14:14:27 +0000 (16:14 +0200)] 
doc: debug UART for RISC-V QEMU virt machine

Provide settings for enabling the debug UART of the virt machine on
RISC-V QEMU.

Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
4 years agodoc: correct bootefi.rst
Heinrich Schuchardt [Sat, 2 Apr 2022 05:16:56 +0000 (07:16 +0200)] 
doc: correct bootefi.rst

* add link to booti man-page
* correct link description for efidebug command
* correct punctuation

Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
4 years agodm: Add docs to explain how to enable DM_SERIAL for a board
Simon Glass [Wed, 16 Mar 2022 03:03:20 +0000 (21:03 -0600)] 
dm: Add docs to explain how to enable DM_SERIAL for a board

This is an attempt to cover the common cases found when enabling driver
model for serial on a new board.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Fabio Estevam <festevam@denx.de>
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
4 years agogpio: pca953x_gpio: add PCA957X support
Luca Ellero [Tue, 22 Mar 2022 14:38:36 +0000 (15:38 +0100)] 
gpio: pca953x_gpio: add PCA957X support

Take as reference Linux kernel code:
drivers/gpio/gpio-pca953x.c

Signed-off-by: Luca Ellero <l.ellero@asem.it>
Reviewed-by: Simon Glass <sjg@chromium.org>
4 years agogpio: Enable hogging support in SPL
Eddie James [Mon, 7 Feb 2022 23:09:01 +0000 (17:09 -0600)] 
gpio: Enable hogging support in SPL

Use the CONFIG macros to conditionally build the GPIO hogging support in
either the SPL or U-Boot, or both, depending on the configuration. Also
call the GPIO hog probe function in the common SPL board initialization
as an equivalent to adding it to the U-Boot init sequence functions.

Signed-off-by: Eddie James <eajames@linux.ibm.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
4 years agogpio: Add gpio_request_by_line_name()
Andrew Jeffery [Mon, 31 Jan 2022 03:24:05 +0000 (13:54 +1030)] 
gpio: Add gpio_request_by_line_name()

Add support for the upstream gpio-line-names property already described
in the common GPIO binding document[1]. The ability to search for a line
name allows boards to lift the implementation of common GPIO behaviours
away from specific line indexes on a GPIO controller.

[1] https://github.com/devicetree-org/dt-schema/blob/3c35bfee83c2e38e2ae7af5f83eb89ca94a521e8/dtschema/schemas/gpio/gpio.yaml#L17

Signed-off-by: Andrew Jeffery <andrew@aj.id.au>
4 years agotest/py:Update python tests for ‘gpio’ cmd
Adarsh Babu Kalepalli [Mon, 31 May 2021 10:53:51 +0000 (16:23 +0530)] 
test/py:Update python tests for ‘gpio’ cmd

Generic Python Test cases are developed to verfiy 'gpio' command.

Signed-off-by: Adarsh Babu Kalepalli <opensource.kab@gmail.com>
4 years agoconfigs: Resync with savedefconfig
Tom Rini [Fri, 8 Apr 2022 17:36:51 +0000 (13:36 -0400)] 
configs: Resync with savedefconfig

Rsync all defconfig files using moveconfig.py

Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agoMerge branch '2022-04-08-Kconfig-updates-and-dead-code-removal'
Tom Rini [Fri, 8 Apr 2022 17:33:14 +0000 (13:33 -0400)] 
Merge branch '2022-04-08-Kconfig-updates-and-dead-code-removal'

- Migration of more symbols from board config.h files to defconfig
- Remove unused network drivers

4 years agoarm: Remove unused ep93xx code
Tom Rini [Thu, 31 Mar 2022 17:46:53 +0000 (13:46 -0400)] 
arm: Remove unused ep93xx code

There are no platforms for this architecture anymore, remove unused
code.

Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agonet: Remove uli526x driver
Tom Rini [Thu, 31 Mar 2022 17:46:52 +0000 (13:46 -0400)] 
net: Remove uli526x driver

This driver is not enabled by any board and not converted to DM_ETH.
Remove.

Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agonet: Remove ns8382x driver
Tom Rini [Thu, 31 Mar 2022 17:46:51 +0000 (13:46 -0400)] 
net: Remove ns8382x driver

This driver is not enabled by any board and not converted to DM_ETH.
Remove.

Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agonet: Remove natsemi driver
Tom Rini [Thu, 31 Mar 2022 17:46:50 +0000 (13:46 -0400)] 
net: Remove natsemi driver

This driver is not enabled by any board and not converted to DM_ETH.
Remove.

Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agonet: Remove lan91c96 driver
Tom Rini [Thu, 31 Mar 2022 17:46:49 +0000 (13:46 -0400)] 
net: Remove lan91c96 driver

This driver is not enabled by any board and not converted to DM_ETH.
Remove.

Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agonet: Remove ftmac110 driver
Tom Rini [Thu, 31 Mar 2022 17:46:48 +0000 (13:46 -0400)] 
net: Remove ftmac110 driver

This driver is not enabled by any board and not converted to DM_ETH.
Remove.

Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agonet: Remove dnet driver
Tom Rini [Thu, 31 Mar 2022 17:46:47 +0000 (13:46 -0400)] 
net: Remove dnet driver

This driver is not enabled by any board and not converted to DM_ETH.
Remove.

Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agonet: Remove cs8900 driver
Tom Rini [Thu, 31 Mar 2022 17:46:46 +0000 (13:46 -0400)] 
net: Remove cs8900 driver

This driver is not enabled by any board and not converted to DM_ETH.
Remove.

Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agonet: Remove ax88180 driver
Tom Rini [Thu, 31 Mar 2022 17:46:45 +0000 (13:46 -0400)] 
net: Remove ax88180 driver

This driver is not enabled by any board and not converted to DM_ETH.
Remove.

Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agonet: Remove armada100_fec driver
Tom Rini [Thu, 31 Mar 2022 17:46:44 +0000 (13:46 -0400)] 
net: Remove armada100_fec driver

This driver is not enabled by any board and not converted to DM_ETH.
Remove.

Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agoarm: fsl-layerscape: Migrate more DP-DDR options to Kconfig
Tom Rini [Wed, 30 Mar 2022 22:07:35 +0000 (18:07 -0400)] 
arm: fsl-layerscape: Migrate more DP-DDR options to Kconfig

Based on current usage, migrate a number of DP-DDR related options to
Kconfig.

Cc: Priyanka Jain <priyanka.jain@nxp.com>
Cc: Rajesh Bhagat <rajesh.bhagat@nxp.com>
Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agosiemens-am33x-common: Drop CONFIG_DMA_COHERENT*
Tom Rini [Wed, 30 Mar 2022 22:07:34 +0000 (18:07 -0400)] 
siemens-am33x-common: Drop CONFIG_DMA_COHERENT*

These values are not used in the code, remove them.

Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agom53menlo: Drop CONFIG_DISCOVER_PHY
Tom Rini [Wed, 30 Mar 2022 22:07:33 +0000 (18:07 -0400)] 
m53menlo: Drop CONFIG_DISCOVER_PHY

This is not used in code, drop.

Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agoConvert CONFIG_FSL_QIXIS et al to Kconfig
Tom Rini [Wed, 30 Mar 2022 22:07:32 +0000 (18:07 -0400)] 
Convert CONFIG_FSL_QIXIS et al to Kconfig

This converts the following to Kconfig:
   CONFIG_FSL_QIXIS
   CONFIG_QIXIS_I2C_ACCESS

Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agoConvert CONFIG_DIMM_SLOTS_PER_CTLR to Kconfig
Tom Rini [Wed, 30 Mar 2022 22:07:31 +0000 (18:07 -0400)] 
Convert CONFIG_DIMM_SLOTS_PER_CTLR to Kconfig

This converts the following to Kconfig:
   CONFIG_DIMM_SLOTS_PER_CTLR

Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agogalileo: Remove CONFIG_DESIGNWARE_ETH reference
Tom Rini [Wed, 30 Mar 2022 22:07:30 +0000 (18:07 -0400)] 
galileo: Remove CONFIG_DESIGNWARE_ETH reference

This platform has CONFIG_ETH_DESIGNWARE set already, which is the
correct value for the driver in question.  Remove this incorrect line.

Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agopowerpc: mpc83xx: Migrate DEFAULT_IMMR to Kconfig
Tom Rini [Wed, 30 Mar 2022 22:07:29 +0000 (18:07 -0400)] 
powerpc: mpc83xx: Migrate DEFAULT_IMMR to Kconfig

As no platforms override this value, set it for all mpc83xx platforms.

Cc: Mario Six <mario.six@gdsys.cc>
Cc: Wolfgang Denk <wd@denx.de>
Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agonds32: Migrate CONFIG_DEBUG_LED to Kconfig
Tom Rini [Wed, 30 Mar 2022 22:07:28 +0000 (18:07 -0400)] 
nds32: Migrate CONFIG_DEBUG_LED to Kconfig

Move this value to the board Kconfig file.

Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agols1021atwr: Use DEBUG and not CONFIG_DEBUG
Tom Rini [Wed, 30 Mar 2022 22:07:27 +0000 (18:07 -0400)] 
ls1021atwr: Use DEBUG and not CONFIG_DEBUG

We use 'DEBUG' and not 'CONFIG_DEBUG' tree-wide for debug code that is
left in, and not wrapped by some other regular debugging type print
macro.

Cc: Alison Wang <alison.wang@nxp.com>
Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agomicroblaze: Migrate DCACHE/ICACHE to Kconfig
Tom Rini [Wed, 30 Mar 2022 22:07:26 +0000 (18:07 -0400)] 
microblaze: Migrate DCACHE/ICACHE to Kconfig

Move these two options to the arch Kconfig file.

Cc: Michal Simek <michal.simek@xilinx.com>
Signed-off-by: Tom Rini <trini@konsulko.com>
Acked-by: Michal Simek <michal.simek@xilinx.com>
4 years agotegra: Drop CONFIG_CHROMEOS_EXTRA_ENV_SETTINGS
Tom Rini [Wed, 30 Mar 2022 22:07:25 +0000 (18:07 -0400)] 
tegra: Drop CONFIG_CHROMEOS_EXTRA_ENV_SETTINGS

This is not currently set by any boards, so drop.

Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agocontrolcenterdc: Migrate CUSTOMER_BOARD_SUPPORT to Kconfig
Tom Rini [Wed, 30 Mar 2022 22:07:24 +0000 (18:07 -0400)] 
controlcenterdc: Migrate CUSTOMER_BOARD_SUPPORT to Kconfig

This symbol is only possibly useful on some mvebu platforms, so move the
symbol there and select it for the only current user.

Cc: Mario Six <mario.six@gdsys.cc>
Cc: Stefan Roese <sr@denx.de>
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Stefan Roese <sr@denx.de>
4 years agospi: cadence_qspi: Migrate CONFIG_CQSPI_REF_CLK to Kconfig
Tom Rini [Wed, 30 Mar 2022 22:07:23 +0000 (18:07 -0400)] 
spi: cadence_qspi: Migrate CONFIG_CQSPI_REF_CLK to Kconfig

This is a little tricky since SoCFPGA has code to determine this as
runtime.  Introduce a guard variable for platforms to select if they
have a static value to use.  Then for ARCH_SOCFPGA, call
cm_get_qspi_controller_clk_hz() and otherwise continue the previous
behavior.

Cc: Jagan Teki <jagan@amarulasolutions.com>
Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agoConvert CONFIG_SH_SCIF_CLK_FREQ to Kconfig
Tom Rini [Wed, 30 Mar 2022 22:07:22 +0000 (18:07 -0400)] 
Convert CONFIG_SH_SCIF_CLK_FREQ to Kconfig

This converts the following to Kconfig:
   CONFIG_SH_SCIF_CLK_FREQ

Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agoConvert CONFIG_CONS_SCIF0 et al to Kconfig
Tom Rini [Wed, 30 Mar 2022 22:07:21 +0000 (18:07 -0400)] 
Convert CONFIG_CONS_SCIF0 et al to Kconfig

This converts the following to Kconfig:
   CONFIG_CONS_SCIF0
   CONFIG_CONS_SCIF1
   CONFIG_CONS_SCIF2
   CONFIG_CONS_SCIF4
   CONFIG_CONS_SCIFA0

Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agoat91: Remove unused LED code
Tom Rini [Wed, 30 Mar 2022 22:07:20 +0000 (18:07 -0400)] 
at91: Remove unused LED code

These LED files (and CONFIG values) are unused today, remove the code in
question.

Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agostm32f429-discovery: Migrate CONFIG_*_LED out of CONFIG namespace
Tom Rini [Wed, 30 Mar 2022 22:07:19 +0000 (18:07 -0400)] 
stm32f429-discovery: Migrate CONFIG_*_LED out of CONFIG namespace

These values are only used in one file, reference them more directly.

Cc: Kamil Lulko <kamil.lulko@gmail.com>
Cc: Patrick Delaunay <patrick.delaunay@foss.st.com>
Cc: Patrice Chotard <patrice.chotard@foss.st.com>
Cc: uboot-stm32@st-md-mailman.stormreply.com
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Patrice Chotard <patrice.chotard@foss.st.com>
4 years agocorvus: Migrate CONFIG_*_LED out of CONFIG namespace
Tom Rini [Wed, 30 Mar 2022 22:07:18 +0000 (18:07 -0400)] 
corvus: Migrate CONFIG_*_LED out of CONFIG namespace

This code is only used on the corvus platform, so migrate the LED on/off
code to this platform and remove it from the CONFIG namespace.  In
theory, this should likely be moved to the modern GPIO LED driver as a
further cleanup.

Cc: Heiko Schocher <hs@denx.de>
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Heiko Schocher <hs@denx.de>
4 years agoMPC837XERDB: Stop using CONFIG_RAMDISKFILE
Tom Rini [Wed, 30 Mar 2022 22:07:17 +0000 (18:07 -0400)] 
MPC837XERDB: Stop using CONFIG_RAMDISKFILE

We don't really configure this, just set it directly in the environment
section.

Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agos5p_goni, smdkc100: Move some environment settings out of CONFIG
Tom Rini [Wed, 30 Mar 2022 22:07:16 +0000 (18:07 -0400)] 
s5p_goni, smdkc100: Move some environment settings out of CONFIG

A number of CONFIG options are used on these platforms as part of the
default environment.  Set some of these more directly and in other
cases, just reference them directly.

Cc: Jaehoon Chung <jh80.chung@samsung.com>
Cc: Minkyu Kang <mk7.kang@samsung.com>
Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agoarm: integrator: Migrate platform-specific options and cleanup armcoremodule.h
Tom Rini [Wed, 30 Mar 2022 22:07:15 +0000 (18:07 -0400)] 
arm: integrator: Migrate platform-specific options and cleanup armcoremodule.h

This converts the following to Kconfig:
   CONFIG_CM_INIT
   CONFIG_CM_REMAP
   CONFIG_CM_SPD_DETECT
   CONFIG_CM_MULTIPLE_SSRAM
   CONFIG_CM_TCRAM

We make the first three of these options be always enabled, as that
matches usage.  We select the last two based on how they were defined in
armcoremodule.h.  This also allows us to remove some unused code in
board/armltd/integrator/lowlevel_init.S

Cc: Linus Walleij <linus.walleij@linaro.org>
Cc: Andre Przywara <andre.przywara@arm.com>
Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agoConvert CONFIG_CMDLINE_PS_SUPPORT to Kconfig
Tom Rini [Wed, 30 Mar 2022 22:07:14 +0000 (18:07 -0400)] 
Convert CONFIG_CMDLINE_PS_SUPPORT to Kconfig

This converts the following to Kconfig:
   CONFIG_CMDLINE_PS_SUPPORT

Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agotegra: Migrate CI_UDC_HAS_HOSTPC to Kconfig
Tom Rini [Wed, 30 Mar 2022 22:07:13 +0000 (18:07 -0400)] 
tegra: Migrate CI_UDC_HAS_HOSTPC to Kconfig

This option is only enabled for CI_UDC and !TEGRA20, so implement it as
such in Kconfig directly.

Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Peter Robinson <pbrobinson@gmail.com>
Tested-by: Peter Robinson <pbrobinson@gmail.com>
4 years agomvebu: Move BOOTROM_ERR_REG out of CONFIG namespace
Tom Rini [Wed, 30 Mar 2022 22:07:12 +0000 (18:07 -0400)] 
mvebu: Move BOOTROM_ERR_REG out of CONFIG namespace

This register is referenced in one location and does not seem
configurable, so remove it from CONFIG namespace.

Cc: Stefan Roese <sr@denx.de>
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Stefan Roese <sr@denx.de>
4 years agoConvert CONFIG_BOARD_SIZE_LIMIT to Kconfig
Tom Rini [Wed, 30 Mar 2022 22:07:11 +0000 (18:07 -0400)] 
Convert CONFIG_BOARD_SIZE_LIMIT to Kconfig

This converts the following to Kconfig:
   CONFIG_BOARD_SIZE_LIMIT

To do this, introduce CONFIG_HAS_BOARD_SIZE_LIMIT.

Signed-off-by: Tom Rini <trini@konsulko.com>
4 years agoMerge branch '2022-04-08-env-updates'
Tom Rini [Fri, 8 Apr 2022 12:27:50 +0000 (08:27 -0400)] 
Merge branch '2022-04-08-env-updates'

- Assorted env tooling updates
- Bug fix around multiple possible env locations and ENV_IS_NOWHERE
- Add 'indirect' sub-command to env
- Allow for FAT env to have the location overridden by the board code.

4 years agoenv: Allow text-env tests to run with awk
Simon Glass [Sun, 13 Mar 2022 05:47:49 +0000 (22:47 -0700)] 
env: Allow text-env tests to run with awk

At present the tests assume that gawk is being used. Adjust the tests so
that the names are inserted in alphabetical order, so that awk is happy.

Also use PROCINFO to make gawk output in alphabetical order. This is not
ideal, since it changes the env-car ordering from what the user provided,
but it may be acceptable.

Signed-off-by: Simon Glass <sjg@chromium.org>
Reported-by: Patrick Delaunay <patrick.delaunay@foss.st.com>
Fixes: https://source.denx.de/u-boot/u-boot/-/issues/10
4 years agoenv: Drop the unncessary protocol mention in autoload
Simon Glass [Fri, 11 Mar 2022 23:22:39 +0000 (16:22 -0700)] 
env: Drop the unncessary protocol mention in autoload

Drop this text at the end since it already mentions bootp and dhcp.

Signed-off-by: Simon Glass <sjg@chromium.org>
Suggested-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
4 years agoenv: Move the doc comment to the code
Simon Glass [Fri, 11 Mar 2022 23:22:38 +0000 (16:22 -0700)] 
env: Move the doc comment to the code

This doesn't really make much sense in the documentation. Add a code
comment instead.

Suggested-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Simon Glass <sjg@chromium.org>
4 years agoMakefile: Sort u-boot-initial-env output
Christoph Niedermaier [Tue, 1 Mar 2022 08:38:51 +0000 (09:38 +0100)] 
Makefile: Sort u-boot-initial-env output

This will sort the default environment file generated by
"make u-boot-initial-env", but won't sort the default
environment in the compiled u-boot binary. The file
u-boot-initial-env is considered to use for the userpace
environment access tools [1] in case of that the
environments is written the first time into its location.
This is done on the one hand for a better overview and
comparison of the generated environment file. On the other
hand it is to synchronize the output with the script
get_default_env.sh, which generated a sorted default
environment file. The sorting preserves the order of equal
variable names by sorting only the variable name, and
disable the last-resort comparison. After sorting,
unnoticed blank lines at the end move to the top. Avoid
that by removing it before sorting.

[1] https://github.com/sbabic/libubootenv

Signed-off-by: Christoph Niedermaier <cniedermaier@dh-electronics.com>
Reviewed-by: Stefano Babic <sbabic@denx.de>
Tested-by: Stefano Babic <sbabic@denx.de>
Cc: Stefano Babic <sbabic@denx.de>
Cc: Simon Glass <sjg@chromium.org>
Cc: Marek Behún <marek.behun@nic.cz>
To: u-boot@lists.denx.de
Reviewed-by: Simon Glass <sjg@chromium.org>
4 years agoscripts/get_default_envs.sh: Remove blank lines
Christoph Niedermaier [Wed, 23 Feb 2022 09:33:36 +0000 (10:33 +0100)] 
scripts/get_default_envs.sh: Remove blank lines

After sorting, unnoticed blank lines at the end move to the top.
Avoid this by removing it before sorting.

Signed-off-by: Christoph Niedermaier <cniedermaier@dh-electronics.com>
Cc: Lukasz Majewski <lukma@denx.de>
Cc: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
Cc: Tom Rini <trini@konsulko.com>
To: u-boot@lists.denx.de

4 years agoenv: fat: Allow overriding interface, device and partition
He Yong [Thu, 17 Feb 2022 16:07:25 +0000 (00:07 +0800)] 
env: fat: Allow overriding interface, device and partition

For platform which can boot on different device, this allows
to override interface, device and partition from board code

Signed-off-by: He Yong <hyyoxhk@163.com>
4 years agocmd: env: Add `indirect` to indirectly set values
Samuel Dionne-Riel [Mon, 20 Dec 2021 23:31:56 +0000 (18:31 -0500)] 
cmd: env: Add `indirect` to indirectly set values

This allows an ergonomic-enough approximation of ${!variable} expansion.
This could be used the following way:

```
for target in ${boot_targets}; do
   env indirect target_name target_name_${target}
   # ...
done
```

Assuming `target_name_mmc0` and similar are set appropriately.

A default value can be optionally provided.

Note: this acts on environment variables, not hush variables.

Signed-off-by: Samuel Dionne-Riel <samuel@dionne-riel.com>
Cc: Simon Glass <sjg@chromium.org>
Cc: "Marek Behún" <marek.behun@nic.cz>
[trini: Don't enable by default]

4 years agoenv: Load env when ENV_IS_NOWHERE is only location selected
Felix.Vietmeyer@jila.colorado.edu [Wed, 21 Apr 2021 02:04:26 +0000 (20:04 -0600)] 
env: Load env when ENV_IS_NOWHERE is only location selected

This patch prevent u-boot from hanging on a UltraZed EG board (zynqmp).

Without the patch,
(drv = env_driver_lookup(ENVOP_INIT, prio))
evaluates to 0, causing prio = 0
Then, (!prio) is hit, returning -ENODEV causing a stall.

With the patch,
instead of returning -ENODEV and causing a stall, we
set gd->env_addr (is this really needed?)
and then
mark gd->env_valid = ENV_INVALID to use the default env.

4 years agoimx8/ls10xx: Use a sane SYS_MALLOC_F_LEN default
Fabio Estevam [Sat, 26 Mar 2022 14:47:40 +0000 (11:47 -0300)] 
imx8/ls10xx: Use a sane SYS_MALLOC_F_LEN default

When adding new features to imx8m boards, such as DM clock support,
the malloc area can be exhausted.

To avoid such issue, provide a reasonable default for the
SYS_MALLOC_F_LEN size.

Quoting Tom Rini:

"This seems to be an area where everyone is either:
- Kicking the value up a bit for themselves
- Having hard to figure out problems booting the platform because it's
  too small a value until they see someone else picked a larger value.

So lets raise these a bit and get some acks, please."

Suggested-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Fabio Estevam <festevam@denx.de>
Tested-by: Adam Ford <aford173@gmail.com> #imx8mm-beacon, imx8mn-beacon
Tested-by: Heiko Thiery <heiko.thiery@gmail.com>
4 years agoMerge branch '2022-04-06-various-verified-boot-bugfixes'
Tom Rini [Thu, 7 Apr 2022 13:49:27 +0000 (09:49 -0400)] 
Merge branch '2022-04-06-various-verified-boot-bugfixes'

To quote the author:

This series collects together various misc patches that were needed when
building mainline U-Boot against Chromium OS verified boot. Most of them
fix minor bugs.

4 years agodm: blk: Expand iteration and add tests
Simon Glass [Mon, 28 Feb 2022 19:08:35 +0000 (12:08 -0700)] 
dm: blk: Expand iteration and add tests

Add some functions which support iteration before probing. Also add tests
for the functions.

Signed-off-by: Simon Glass <sjg@chromium.org>
4 years agodm: core: Tidy up comments in uclass headers
Simon Glass [Mon, 28 Feb 2022 19:08:34 +0000 (12:08 -0700)] 
dm: core: Tidy up comments in uclass headers

Improve some of the comments in these files, which don't follow the
correct style.

Signed-off-by: Simon Glass <sjg@chromium.org>
4 years agodm: core: Add a required struct declaration
Simon Glass [Mon, 28 Feb 2022 19:08:33 +0000 (12:08 -0700)] 
dm: core: Add a required struct declaration

This file uses struct driver, so declare it at the top in case the
header-inclusion order is not as expected.

Signed-off-by: Simon Glass <sjg@chromium.org>
4 years agostdint: Add a definition of UINT8_MAX
Simon Glass [Mon, 28 Feb 2022 19:08:32 +0000 (12:08 -0700)] 
stdint: Add a definition of UINT8_MAX

This is normally defined in stdint.h but is not used in U-Boot. When
libraries (such as Chromium OS vboot) are built against U-Boot they may
expect this value to be available. Add it to avoid build errors in this
case.

Signed-off-by: Simon Glass <sjg@chromium.org>
4 years agoMake ASYMMETRIC_KEY_TYPE depend on FIT_SIGNATURE
Simon Glass [Mon, 28 Feb 2022 19:08:31 +0000 (12:08 -0700)] 
Make ASYMMETRIC_KEY_TYPE depend on FIT_SIGNATURE

Add this dependency to avoid a build error if FIT_SIGNATURE is not
enabled.

Signed-off-by: Simon Glass <sjg@chromium.org>
4 years agoAdd a default for TPL_TEXT_BASE
Simon Glass [Mon, 28 Feb 2022 19:08:30 +0000 (12:08 -0700)] 
Add a default for TPL_TEXT_BASE

If this value is not provided it causes a hang in the build. Add a default
value to avoid this.

Signed-off-by: Simon Glass <sjg@chromium.org>
4 years agodisk: Correct the conditions for SPL
Simon Glass [Mon, 28 Feb 2022 19:08:29 +0000 (12:08 -0700)] 
disk: Correct the conditions for SPL

These filesystems may have different settings for SPL and TPL. Use the
correct Makefile variable to handle this.

Signed-off-by: Simon Glass <sjg@chromium.org>
4 years agospi: Avoid checking console in SPL
Simon Glass [Mon, 28 Feb 2022 19:08:28 +0000 (12:08 -0700)] 
spi: Avoid checking console in SPL

When SPI flash is used in SPL there is no console, so ctrlc() cannot be
called. Add a condition to fix this.

Signed-off-by: Simon Glass <sjg@chromium.org>
4 years agocros_ec: Complete the comment for cros_ec_read_batt_charge()
Simon Glass [Mon, 28 Feb 2022 19:08:27 +0000 (12:08 -0700)] 
cros_ec: Complete the comment for cros_ec_read_batt_charge()

Add the missing 'Return' information.

Signed-off-by: Simon Glass <sjg@chromium.org>
4 years agocbfs: Add some more definititions
Simon Glass [Mon, 28 Feb 2022 19:08:26 +0000 (12:08 -0700)] 
cbfs: Add some more definititions

Add definitions for a payload or 'self', which is like an unpacked ELF
file.

Signed-off-by: Simon Glass <sjg@chromium.org>
4 years agolzma: Tidy up the function prototype
Simon Glass [Mon, 28 Feb 2022 19:08:25 +0000 (12:08 -0700)] 
lzma: Tidy up the function prototype

This should use a const pointer for the input stream. Fix this and also
add a proper comment.

Signed-off-by: Simon Glass <sjg@chromium.org>
4 years agofdt: sandbox: Avoid looking for an appended device tree
Simon Glass [Mon, 28 Feb 2022 19:08:24 +0000 (12:08 -0700)] 
fdt: sandbox: Avoid looking for an appended device tree

We don't use an appended tree for sandbox and the required symbols are
not present. Add a condition to avoid a build error.

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