- Convert SYS_[DI]CACHE_OFF to Kconfig, introduce SPL variant.
- Various bcm96* fixes.
- Import include/android_bootloader_message.h from AOSP
- Assorted other small fixes.
NVMe was turned on in qemu-x86 but somehow we missed it for 64-bit.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
When dma_addr_t is u32 in 64-bit, there are some warnings when
building NVME driver. Fix it by doing an additional (long) cast.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
coreboot makes it possible to add own entries into coreboot's
table at a per mainboard basis. As there might be some custom
ones it makes sense to provide a way to process them.
Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Do not write to address indicated by NULL pointer.
UEFI SCT II 2.6 (2017), 3.6.5 GetNextMonotonicCount(), 5.1.5.5.1
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Not checking the parameters may lead reading or writing from NULL.
Implement the parameter checks prescribed in the UEFI spec.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
According to the UEFI spec InstallMultipleProtocolInterfaces() must check
if a device path has already been installed. In this case it must return
EFI_ALREADY_STARTED.
Cf. UEFI SCT II 2.6 A (2017),
3.3.16 InstallMultipleProtocolInterfaces(), 5.1.3.16.1.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
In EFI 1.10 a version of the Unicode collation protocol using ISO 639-2
language codes existed. This protocol is not part of the UEFI specification
any longer. Unfortunately it is required to run the UEFI Self Certification
Test (SCT) II, version 2.6, 2017. So we implement it here for the sole
purpose of running the SCT. It can be removed once a compliant SCT is
available.
The configuration option defaults to no.
Signed-off-by: Rob Clark <robdclark@gmail.com>
Most of Rob's original patch is already merged. Only the deprecated
protocol is missing. Rebase it and make it configurable.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Rename variables to make it clear they refer to the Unicode collation
protocol identified by the EFI_UNICODE_PROTOCOL2_GUID.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Check buffer pointers are not NULL as required by the UEFI 2.7 spec.
Return EFI_UNSUPPORTED instead of EFI_INVALID_PARAMETER when trying to
transmit with non-zero header_size.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
UEFI spec 2.7 erratum A leaves it undefined if Attributes should be set if
GetVariable() returns EFI_BUFFER_TOO_SMALL.
UEFI spec 2.8 defines that Attributes should be set if the return value is
either EFI_SUCCESS or EFI_BUFFER_TOO_SMALL.
Set Attributes if the return value is EFI_BUFFER_TOO_SMALL.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
The current prompt text for FIT external offset is identical to
SYS_TEXT_BASE which might confuse the users. Provided more accurate
description for the prompt text.
Signed-off-by: Ibai Erkiaga <ibai.erkiaga-elorza@xilinx.com>
Reviewed-by: Marek Vasut <marex@denx.de>
This takes the latest changes from AOSP from the file
bootloader_message/include/bootloader_message/bootloader_message.h
in the repository
https://android.googlesource.com/platform/bootable/recovery
and re-licensed them to BSD-3 for U-Boot.
Minimum local changes have been applied (convert C++ to C comments and
adding #ifndef __UBOOT__ block to skip all the function declarations).
Signed-off-by: Alex Deymo <deymo@google.com>
Reviewed-by: Sam Protsenko <semen.protsenko@linaro.org>
Add an option to open files in read-only mode in mmap_fdt so
that fit_check_sign can be used to inspect files on read-only
filesystems.
For example, this is useful when a key is shipped in a read-only
rootfs or squashfs.
Signed-off-by: Luca Boccassi <luca.boccassi@microsoft.com>
Configuration option CONFIG_CMD_BOOTEFI_SELFTEST is useful for the
development of the UEFI sub-system. For production it is not needed.
Remove CONFIG_CMD_BOOTEFI_SELFTEST from bcm968580xref_ram_defconfig.
Suggested-by: Tom Rini <trini@konsulko.com>
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Configuration option CONFIG_CMD_BOOTEFI_SELFTEST is useful for the
development of the UEFI sub-system. For production it is not needed.
Remove CONFIG_CMD_BOOTEFI_SELFTEST from bcm963158_ram_defconfig.
Suggested-by: Tom Rini <trini@konsulko.com>
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Set the spl_image->fdt_addr pointer both for simple fitImage configuration
as well as full fitImage configuration, to let spl_perform_fixups() access
the DT and perform modifications to it if necessary.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Tom Rini <trini@konsulko.com>
This board define the flag CONFIG_SYS_NAND_DRIVER_ECC_LAYOUT
but it's a mistake. It's a workaround for an issue in nand core.
This issue was fixed by the commit 5f626e7849 ("mtd: nand: raw:
Fix CONFIG_SYS_NAND_DRIVER_ECC_LAYOUT behavior"). Now, this flag
break the nand on this board, so we simply remove it.
Signed-off-by: Philippe Reynes <philippe.reynes@softathome.com>
This board define the flag CONFIG_SYS_NAND_DRIVER_ECC_LAYOUT
but it's a mistake. It's a workaround for an issue in nand core.
This issue was fixed by the commit 5f626e7849 ("mtd: nand: raw:
Fix CONFIG_SYS_NAND_DRIVER_ECC_LAYOUT behavior"). Now, this flag
break the nand on this board, so we simply remove it.
Signed-off-by: Philippe Reynes <philippe.reynes@softathome.com>
This board define the flag CONFIG_SYS_NAND_DRIVER_ECC_LAYOUT
but it's a mistake. It's a workaround for an issue in nand core.
This issue was fixed by the commit 5f626e7849 ("mtd: nand: raw:
Fix CONFIG_SYS_NAND_DRIVER_ECC_LAYOUT behavior"). Now, this flag
break the nand on this board, so we simply remove it.
Signed-off-by: Philippe Reynes <philippe.reynes@softathome.com>
The watchdog should use a clock at 50 Mhz, so
instead of using the clock osc (200 Mhz), we
define a reference clock at 50Mhz and use it
for both watchdog.
Signed-off-by: Philippe Reynes <philippe.reynes@softathome.com>
Reviewed-by: Stefan Roese <sr@denx.de>
The watchdog should use a clock at 50 Mhz, so
instead of using the clock osc (200 Mhz), we
define a reference clock at 50Mhz and use it
for both watchdog.
Signed-off-by: Philippe Reynes <philippe.reynes@softathome.com>
Reviewed-by: Stefan Roese <sr@denx.de>
The function bcm6345_wdt_start use the argument timeout
as tick but it should be used as milliseconds.
A clock is added as requirement for this driver.
The frequency of the clock is then used to convert the
millisecond to ticks in the function bcm6345_wdt_start.
Signed-off-by: Philippe Reynes <philippe.reynes@softathome.com>
Reviewed-by: Stefan Roese <sr@denx.de>
Prioritize external dtb if its passed via EXT_DTB
than the dtb that was built in the tree. With this
patch it appends the specified external dtb to
the u-boot image.
Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Signed-off-by: Siva Durga Prasad Paladugu <siva.durga.paladugu@xilinx.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
While converting CONFIG_SYS_[DI]CACHE_OFF to Kconfig, there are instances
where these configuration items are conditional on SPL. This commit adds SPL
variants of these configuration items, uses CONFIG_IS_ENABLED(), and updates
the configurations as required.
Acked-by: Alexey Brodkin <abrodkin@synopsys.com>
Signed-off-by: Trevor Woerner <trevor@toganlabs.com>
[trini: Make the default depend on the setting for full U-Boot, update
more zynq hardware]
Signed-off-by: Tom Rini <trini@konsulko.com>
CONFIG_SYS_[DI]CACHE_OFF had been partially converted to Kconfig
parameters; only for the ARC architecture. This patch turns these two
parameters into Kconfig items everywhere else they are found.
All of the include/configs/* and defconfig changes in this patch are
for arm machines only. The Kconfig changes for arc, nds32, riscv,
and xtensa have been included since these symbols are found in code
under arch/{arc,nds32,riscv,xtensa}, however, no currently-defined
include/configs/* or defconfigs for these architectures exist which
include these symbols.
These results have been confirmed with tools/moveconfig.py.
Acked-by: Alexey Brodkin <abrodkin@snopsys.com>
Signed-off-by: Trevor Woerner <trevor@toganlabs.com>
[trini: Re-migrate for a few more boards]
Signed-off-by: Tom Rini <trini@konsulko.com>
Eventually these configuration items will be converted to Kconfig,
therefore there's little point in leaving commented-out versions of
them in include/configs.
Signed-off-by: Trevor Woerner <trevor@toganlabs.com>
This config is the only config that uses:
#define CONFIG_SYS_DCACHE_OFF 1
in its #define.
Remove the superfluous "1" so this cache #define is like all the others.
Signed-off-by: Trevor Woerner <trevor@toganlabs.com>
According to De Morgan's Law[1]:
!(A && B) = !A || !B
!(A || B) = !A && !B
There are 5 places in the code where we find:
#if !(defined(CONFIG_SYS_ICACHE_OFF) && defined(CONFIG_SYS_DCACHE_OFF))
and 4 places in the code where we find:
#if (!defined(CONFIG_SYS_ICACHE_OFF) || !defined(CONFIG_SYS_DCACHE_OFF))
In words, the construct:
!defined(CONFIG_SYS_[DI]CACHE_OFF)
means:
"is the [DI]CACHE on?"
and the construct:
defined(CONFIG_SYS_[DI]CACHE_OFF)
means:
"is the [DI]CACHE off?"
Therefore
!(defined(CONFIG_SYS_ICACHE_OFF) && defined(CONFIG_SYS_DCACHE_OFF))
means:
"the opposite of 'are they both off?'"
in other words:
"are either or both on?"
and:
(!defined(CONFIG_SYS_ICACHE_OFF) || !defined(CONFIG_SYS_DCACHE_OFF)
means:
"are either or both on?"
As a result, I've converted the 4 instances of '(!A || !B)' to '!(A && B)' for
consistency.
[1] https://en.wikipedia.org/wiki/De_Morgan%27s_laws
Signed-off-by: Trevor Woerner <trevor@toganlabs.com>
These options only apply when not using DM_I2C. When using device
trees, the dt will enable and control the speeds of the I2C
controller(s) and these configuration options have no effect.
So disable them in DM_I2C mode. Otherwise they show up as decoys, and
make it look like one is enabling I2C controllers and setting the speed
when really it's doing nothing.
However, a system using a SPL build will not use DM_I2C in the SPL, even
if DM_I2C is enabled for the main u-boot. And so the SPL might use the
kconfig based I2C speed controls while the main u-boot does not.
Cc: Sriram Dash <sriram.dash@nxp.com>
Cc: Priyanka Jain <priyanka.jain@nxp.com>
Cc: Heiko Schocher <hs@denx.de>
Signed-off-by: Trent Piepho <tpiepho@impinj.com>
When using DM_I2C the speed value supplied to setup_i2c() is not used,
so this code required CONFIG_SYS_MXC_I2C[12]_SPEED to be defined to
compile, but did not actually use them.
Change this so we no longer need to define an unused macro to compile in
DM_I2C mode. Also make it more clear that they do not control the bus
speed. Otherwise it is quite easy to mistakenly believe they are used
to set the bus speed.
Cc: Heiko Schocher <hs@denx.de>
Cc: Anatolij Gustschin <agust@denx.de>
Signed-off-by: Trent Piepho <tpiepho@impinj.com>
This is an old driver that supports both device mapped and non-mapped
mode, and covers a wide range of hardware. It's hard to change without
risking breaking something. I have to tried to be exceedingly detailed
in this patch, so please excuse the length of the commit essay that
follows.
In device mapped mode the I2C xfer function does not handle plain read,
and some other, transfers correctly.
What it can't handle are transactions that:
Start with a read, or,
Have a write followed by a read, or,
Have more than one read in a row.
The common I2C/SMBUS read register and write register transactions
always start with a write, followed by a write or a read, and then end.
These work, so the bug is not apparent for most I2C slaves that only use
these common xfer forms.
The existing xfer loop initializes by sending the chip address in write
mode after it deals with bus arbitration and master setup. When
processing each message, if the next message will be a read, it sends a
repeated start followed by the chip address in read mode after the
current message.
Obviously, this does not work if the first message is a read, as the
chip is always addressed in write mode initially by i2c_init_transfer().
A write following a read does not work because the repeated start is
only sent when the next message is a read. There is no logic to send it
when the current message is a read and next is write. It should be sent
every time the bus changes direction.
The ability to use a plain read was added to this driver in
commit 2feec4eafd ("imx: mxc_i2c: tweak the i2c transfer method"),
but this applied only the non-DM code path.
This patch fixes the DM code path. The xfer function will call
i2c_init_transfer() with an alen of -1 to avoid sending the chip
address. The same way the non-DM code achieves this. The xfer
function's message loop will send the address and mode before each
message if the bus changes direction, and on the first message.
When reading data, the master hardware is one byte ahead of what we
receive. I.e., reading a byte from the data register returns a byte
*already received* by the master, and causes the master to start the RX
of the *next* byte. Therefor, before we read the final byte of a
message, we must tell the master what to do next. I add a "last" flag
to i2c_read_data() to tell it if the message is to be followed by a stop
or a repeated start. When last == true it acts exactly as before.
The non-DM code can only create an xfer where the read, if any, is the
final message of the xfer. And so the only callsite of i2c_read_data()
in the non-DM code has the "last" parameter as true. Therefore, this
change has no effect on the non-DM code. As all other changes are in
the DM xfer function, which is not even compiled in non-DM code, I am
confident that this patch has no effect on boards not using I2C_DM.
This greatly reduces the range of hardware that could be affected.
For DM boards, I have verified every transaction the "i2c" command can
create on a scope and they are all exactly as they are supposed to be.
I also tested write->read->write, which isn't possible with the i2c
command, and it works as well. I didn't fix multiple reads in a row, as
it's a lot more invasive and obviously no one has every wanted them
since they've never worked. It didn't seem like the extra complexity
was justified to support something no one uses.
Cc: Nandor Han <nandor.han@ge.com>
Cc: Heiko Schocher <hs@denx.de>
Cc: Stefano Babic <sbabic@denx.de>
Cc: Fabio Estevam <festevam@gmail.com>
Cc: Breno Matheus Lima <brenomatheus@gmail.com>
Signed-off-by: Trent Piepho <tpiepho@impinj.com>
It is not very clear how these work in relation to the exact I2C xfers
they produce. In paticular, the address length is somewhat overloaded
in the read method. Clearly document the existing behavior. Maybe this
will help the next person who needs to work on this driver and not break
non-DM boards.
Cc: Nandor Han <nandor.han@ge.com>
Cc: Heiko Schocher <hs@denx.de>
Cc: Stefano Babic <sbabic@denx.de>
Cc: Fabio Estevam <festevam@gmail.com>
Cc: Breno Matheus Lima <brenomatheus@gmail.com>
Signed-off-by: Trent Piepho <tpiepho@impinj.com>
Arndale board has an Asix AX88760 USB 2.0 Hub and Fast Ethernet combo.
The appropriate driver for it is USB_ETHER_ASIX.
The mistake probably came from misinterpretation of commit e9954b867c
("usb: eth: add ASIX AX88179 DRIVER") which was tested on RECS5250 COM
module. This module indeed has Exynos5250 and some similarities with
Arndale 5250 board but the USB/Ethernet chip used there is apparently
different.
Fixes: f58ad98a62 ("usb: net: migrate USB Ethernet adapters to Kconfig")
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
Reviewed-by: Lukasz Majewski <lukma@denx.de>
Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
The CONFIG_DM_I2C_COMPAT was introduced in
include/configs/exynos5-common.h in commit 189d80166b ("exynos5:
enable dm i2c") and then it propagated up to configs/arndale_defconfig.
However since beginning the Arndale board (Exynos5250) was not using
I2C.
In fact, the Arndale board is not configuring its PMIC (S5M8767) which
uses I2C bus. This setting can be thus safely removed to fix build
warning:
This board uses CONFIG_DM_I2C_COMPAT. Please remove
(possibly in a subsequent patch in your series)
before sending patches to the mailing list.
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
The CONFIG_POWER and CONFIG_POWER_I2C were introduced in
include/configs/exynos5-common.h in commit 19bd3aaa59 ("exynos5: fix
build break by adding CONFIG_POWER") and then it propagated up to
include/configs/arndale.h. However before that commit, there was no
build break at all on Arndale and SMDK5250 boards. It seems the commit
fixed nothing and just added unused defines.
In fact, the Arndale board is not configuring its PMIC (S5M8767) which
uses I2C bus.
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
In mt7629-rfb.dts, the phy-handle is a reference to the node phy0, not the
node itself:
phy-handle = <&phy0>;
phy0: ethernet-phy@0 {
reg = <0>;
}
However the driver used ofnode_find_subnode("phy-handle") to read the node.
It will always fail.
This patch replaces ofnode_find_subnode with dev_read_phandle_with_args to
make sure the node can be read correctly.
Cc: Joe Hershberger <joe.hershberger@ni.com>
Signed-off-by: Weijie Gao <weijie.gao@mediatek.com>
Acked-by: Joe Hershberger <joe.hershberger@ni.com>
