In commit a1274cc94a ("ARM: Page align secure section only when it is
executed in situ"), we used output section attributes (the "ALIGN"
keyword after the colon) to specify the alignment requirements. Using
the constant "COMMONPAGE" there was recently broken in binutils 2.31 [1].
Binutils maintainer Alan Modra suggested the former method would still
work. Since both methods achieve the same result, this patch does just
that. This fixes the "reboot after bootm" issue we've been seeing on
sunxi when booting non-secure.
[1] https://sourceware.org/bugzilla/show_bug.cgi?id=23571
Suggested-by: Alan Modra <amodra@gmail.com>
Signed-off-by: Chen-Yu Tsai <wens@csie.org>
Reviewed-by: Tom Rini <trini@konsulko.com>
Acked-by: Jagan Teki <jagan@openedev.com>
When re-syncing the DTS files from the kernel, something caused
the MMC driver to no longer detect the MMC card. Undoing the
CD-invert appears to fix the issue.
Fixes: e6ea2390cd ("ARM: DTS: Resync LogicPD-Torpedo-37xx-devkit
with Linux 4.18-RC4")
Signed-off-by: Adam Ford <aford173@gmail.com>
CPU (and hence cpu timers) on HSDK board runs at 500MHz after
preloader so fix wrong CPU frequency value in hsdk.dts
Signed-off-by: Eugeniy Paltsev <Eugeniy.Paltsev@synopsys.com>
This patch adds system suspend/resume support,
when linux kernel enters deep sleep mode, SoC will go
into below mode:
- CA7 platform goes into STOP mode;
- SoC goes into DSM mode;
- DDR goes into self-refresh mode;
- CPU0/SCU will be powered down.
When wake up event arrives:
- SoC DSM mdoe exits;
- CA7 platform exit STOP mode, SCU/CPU0 power up;
- Invalidate L1 cache;
- DDR exit self-refresh mode;
- Do secure monitor mode related initialization;
- Jump to linux kernel resume entry.
Belwo is the log of 1 iteration of system suspend/resume:
[ 338.824862] PM: suspend entry (deep)
[ 338.828853] PM: Syncing filesystems ... done.
[ 338.834433] Freezing user space processes ... (elapsed 0.001 seconds) done.
[ 338.842939] OOM killer disabled.
[ 338.846182] Freezing remaining freezable tasks ... (elapsed 0.001 seconds) done.
[ 338.869717] PM: suspend devices took 0.010 seconds
[ 338.877846] Disabling non-boot CPUs ...
[ 338.960301] Retrying again to check for CPU kill
[ 338.964953] CPU1 killed.
[ 338.968104] Enabling non-boot CPUs ...
[ 338.973598] CPU1 is up
[ 339.267155] mmc1: queuing unknown CIS tuple 0x80 (2 bytes)
[ 339.275833] mmc1: queuing unknown CIS tuple 0x80 (7 bytes)
[ 339.284158] mmc1: queuing unknown CIS tuple 0x80 (6 bytes)
[ 339.385065] PM: resume devices took 0.400 seconds
[ 339.389836] OOM killer enabled.
[ 339.392986] Restarting tasks ... done.
[ 339.398990] PM: suspend exit
The resume entry function has to initialize stack pointer before calling
C code, otherwise there will be an external abort occur, in additional,
invalidate L1 cache must be done in secure section as well, so this
patch also adds assembly code back and keep it as simple as possible.
Signed-off-by: Anson Huang <Anson.Huang@nxp.com>
Acked-by: Stefan Agner <stefan@agner.ch>
Tested-by: Stefan Agner <stefan@agner.ch>
Add i.MX7D GPC initialization for low power mode
support like system suspend/resume from linux kernel:
- Pending IOMUXC IRQ to workaround GPC state machine issue;
- Mask all GPC interrupts for M4/C0/C1;
- Configure SCU timing;
- Configure time slot ack;
- Configure C0/C1 power up/down timing;
- Configure wakeup source mechanism;
- Disable DSM/RBC related settings.
Signed-off-by: Anson Huang <Anson.Huang@nxp.com>
This patch improves cpu hotplug, previous cpu_off
implementation is NOT safe, a CPU can NOT power down
itself in runtime, it will cause system bus hang due
to pending transaction. So need to use other online
CPU to kill it when it is ready for killed.
Here use SRC parameter register and a magic number
of ~0 as handshake for killing a offline CPU,
when the online CPU checks the psci_affinity_info,
it will help kill the offline CPU according to
the magic number stored in SRC parameter register.
Signed-off-by: Anson Huang <Anson.Huang@nxp.com>
The existing imx_watchdog driver is compatible with mx25 chipsets.
Add a WDOG1_BASE_ADDR define for the base address and enable the driver
in watchdog's Makefile.
To use the driver, a board must define CONFIG_IMX_WATCHDOG and
CONFIG_HW_WATCHDOG.
This fixes an issue when booting an mx25 chip via usb/serial. In this
case, the boot rom will always enable the watchdog. If u-boot is running
in interactive mode and the watchdog is not serviced, the system is
rebooted when the watchdog expires.
Signed-off-by: Martin Kaiser <martin@kaiser.cx>
Reviewed-by: Fabio Estevam <fabio.estevam@nxp.com>
Add the board DTS files for 14x14 EVK and 9x9 EVK. They are necessary
for converting to use u-boot DM driver.
Two -u-boot.dtsi are added to modify compatible string of SPI flash
device to "spi-flash".
Signed-off-by: Ye Li <ye.li@nxp.com>
On iMX7D SabreSD board, the QSPI has pins conflict with EPDC (default).
To use QSPI, users have to rework the board (de-populate R388-R391, R396-R399
populate R392-R395, R299, R300). So we add new DTS file and new defconfig
dedicated for QSPI.
Other changes to support the DM QSPI:
- Add QSPI node and alias spi0.
- Modify spi4 (spi-gpio) node and add alias spi5 for it to avoid req
conflict
- Add EPDC node in imx7d.dtsi and disable it in imx7d-sdb-qspi.dts to
align with kernel and also present the conflict.
- Add -u-boot.dtsi to modify compatible string of mx25l51245g@0 to
"spi-flash"
- Remove iomux settings of qspi in board codes which is not needed
for DM driver.
Signed-off-by: Ye Li <ye.li@nxp.com>
To support DM QSPI driver:
- Add -u-boot.dtsi to modify n25q256a@0 and n25q256a@1 compatible string
to "spi-flash" and add "num-cs" property.
- Enable DM SPI and DM SPI FLASH configurations
- Remove iomux settings of qspi1 in board codes which is not needed
for DM driver.
Signed-off-by: Ye Li <ye.li@nxp.com>
To support DM QSPI driver
- Add spi0 and spi1 alias for qspi1 and qspi2.
- Add -u-boot.dtsi to modify n25q256a@0 and n25q256a@1 compatible string
to "spi-flash" and add "num-cs" property.
- Enable DM SPI/QSPI relavent configurations
- Remove iomux settings of qspi2 in board codes which is not needed
for DM driver.
- Add sf default settings. So running "sf probe" can detect the flash
Signed-off-by: Ye Li <ye.li@nxp.com>
As of today, the proposal of adding "acpi_rsdp_addr" to the kernel
boot protocol does not make its way to the kernel mainline. This
creates some confusion if we leave it in the U-Boot code base.
Remove it for now until we have a clear picture with kernel upstream.
Note this eventually does a partial revert to commit 3469bf4274
("x86: zImage: Propagate acpi_rsdp_addr to kernel via boot parameters")
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Acked-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
At present Linux kernel loaded from U-Boot as an EFI payload does
not boot. This fills in kernel's boot params structure with the
required critical EFI information like system table address and
memory map stuff so that kernel can obtain essential data like
runtime services and ACPI table to boot.
With this patch, now U-Boot as an EFI payload becomes much more
practical: it is another option of kernel bootloader, ie, can be
a replacement for grub.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
This implements payload-specific install_e820_map() to get E820 map
from the EFI memory map descriptors.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Fix warning below:
unnecessary #address-cells/#size-cells without "ranges" or child "reg"
property
Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com>
It is not preferred to put SUNXI-specific code in the common place.
Change it to 'imply' property of ARCH_SUNXI.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
In an earlier proposed patch, bcmstb.c implemented timer_read_counter,
but it was updated to implement get_ticks instead. This patch updates
the declaration in timer.h accordingly.
Signed-off-by: Thomas Fitzsimmons <fitzsim@fitzsim.org>
The NAND reset is missing from DT, so the reset manager cannot unreset the NAND.
Add the missing DT reset entry.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Chin Liang See <chin.liang.see@intel.com>
Cc: Dinh Nguyen <dinguyen@kernel.org>
Cc: Ley Foon Tan <ley.foon.tan@intel.com>
The UART clock frequency can be obtained from the clock framework by the
ns16550 driver, so drop this redundant DT node.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Chin Liang See <chin.liang.see@intel.com>
Cc: Dinh Nguyen <dinguyen@kernel.org>
Cc: Ley Foon Tan <ley.foon.tan@intel.com>
Switch the Arria10 from ad-hoc hardcoded timer to timer framework
and the DW APB timer driver. This allows the A10 to extract timer
information, like timer rate, from clock framework and thus DT
instead of having it hardcoded in U-Boot configuration files.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Chin Liang See <chin.liang.see@intel.com>
Cc: Dinh Nguyen <dinguyen@kernel.org>
Cc: Ley Foon Tan <ley.foon.tan@intel.com>
Flag timer clock as DM pre-reloc, so that a timer driver can be used and
it can extract information about it's clock rate using the clock framework.
This patch also moves some of the pre-reloc flags into the core dtsi file,
this is because the timer is not board specific, but rather is used on all
boards.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Chin Liang See <chin.liang.see@intel.com>
Cc: Dinh Nguyen <dinguyen@kernel.org>
Cc: Ley Foon Tan <ley.foon.tan@intel.com>
The Arria10 SPL is a complete mess of calls to functions which are
called in the wrong context and it is surprise it works at all. This
patch tries to clean that mess up by shuffling the function calls
around and moving the calls into the correct context. Due to the
delicate nature of the reordering, this is done in one huge patch.
The following changes happen in this patch:
- Security policy init and NIC301 happens first in board_init_f()
- The clock init happens very early in board_init_f() in SPL only
- arch_early_init_r() only registers the FPGA, just like on Gen5
- arch_early_init_r() is never called from any _f() function
- Dedicated FPGA pins are inited in board_init_f() as on Gen5
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Chin Liang See <chin.liang.see@intel.com>
Cc: Dinh Nguyen <dinguyen@kernel.org>
Cc: Ley Foon Tan <ley.foon.tan@intel.com>
According to mailbox spec, software should send urgent command with
urgent register instead of COUT location. This patch write urgent
command index to urgent register.
Signed-off-by: Ley Foon Tan <ley.foon.tan@intel.com>
Code checking and setting EMAC use fpga is in
populate_sysmgr_fpgaintf_module(). So, call to sysmgr_pinmux_init()
instead of populate_sysmgr_pinmux().
In sysmgr_pinmux_init(), it will call to both populate_sysmgr_pinmux()
and populate_sysmgr_fpgaintf_module().
Signed-off-by: Ley Foon Tan <ley.foon.tan@intel.com>
This partially reverts commit 7e21fbca26.
That change broke sandbox EFI support for unknown reasons. It also changes
sandbox to use--gc-sections which we don't want.
For now I am just reverting the sandbox portion as presumably this change
is safe on other architectures.
Fixes: 7e21fbca26 (efi_loader: Rename sections to allow for implicit data)
Signed-off-by: Simon Glass <sjg@chromium.org>
Update README.falcon to use "none" for compression property for
ramdisk image to avoid being uncompressed upon loading.
Signed-off-by: York Sun <york.sun@nxp.com>
A few fixes for 2018.09. Most noticable are:
- unbreak x86 target (-fdata-section fallout)
- fix undefined behavior in a few corner cases
- make Jetson TX1 boot again
- RTS fixes
- implement reset for simple output
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2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=Zi4z
-----END PGP SIGNATURE-----
Merge tag 'signed-efi-2018.09' of git://github.com/agraf/u-boot
Patch queue for efi - 2018-08-21
A few fixes for 2018.09. Most noticable are:
- unbreak x86 target (-fdata-section fallout)
- fix undefined behavior in a few corner cases
- make Jetson TX1 boot again
- RTS fixes
- implement reset for simple output
A secure monitor that runs before U-Boot, and hence causes U-Boot to run
in non-secure world, must implement a few operations that U-Boot
otherwise implements when running in secure world. Fix U-Boot to skip
these operations when running in non-secure world. In particular:
- The secure monitor must provide the LP0 resume code and own LP0
configuration in order to maintain security, so must initialize all
the PMC scratch registers used by the boot ROM during LP0 resume.
Consequently, U-Boot should not attempt to clear those registers,
since the register accesses will fail or cause an error.
- The secure monitor owns system security, and so is responsible for
configuring security-related items such as the VPR.
Signed-off-by: Stephen Warren <swarren@nvidia.com>
Signed-off-by: Tom Warren <twarren@nvidia.com>
Since commit f2faffecb0, tegra: Convert to use binman
the dm-pre-reloc properties are removed.
This leads U-Boot not to enable the display on paz00
This patch restore the dm-pre-reloc properties allowing
the bootloader to output to the display panel
v4: - Spell project name as appropriate
v3: - Fix few typos
v2: - Add more characters to commit hash
Signed-off-by: Nicolas Chauvet <kwizart@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Tom Warren <twarren@nvidia.com>
In some cases it can be useful to be able to bind a device to a driver from
the command line.
The obvious example is for versatile devices such as USB gadget.
Another use case is when the devices are not yet ready at startup and
require some setup before the drivers are bound (ex: FPGA which bitsream is
fetched from a mass storage or ethernet)
usage example:
bind usb_dev_generic 0 usb_ether
unbind usb_dev_generic 0 usb_ether
or
unbind eth 1
bind /ocp/omap_dwc3@48380000/usb@48390000 usb_ether
unbind /ocp/omap_dwc3@48380000/usb@48390000
Signed-off-by: Jean-Jacques Hiblot <jjhiblot@ti.com>
Also add device_find_global_by_ofnode() that also find a device based on
the OF node, but doesn't probe the device.
Signed-off-by: Jean-Jacques Hiblot <jjhiblot@ti.com>
When we build with -fdata-sections we may end up with bss subsections. Our
linker script explicitly lists only a single consecutive bss section though.
Adapt the statement to also include subsections.
Signed-off-by: Alexander Graf <agraf@suse.de>
We left -fdata-sections disabled for x86_64 before because we encountered
random bugs that were at that time inexplicable.
Turns out this really was just side effects of missing .bss* statements
in the linker scripts. With those fixed, we can enable data sections for all
targets.
Signed-off-by: Alexander Graf <agraf@suse.de>
When we build with -fdata-sections we may end up with bss subsections. Our
linker script explicitly lists only a single consecutive bss section though.
Adapt the statement to also include subsections.
This fixes booting efi-x86_app_defconfig.
Signed-off-by: Alexander Graf <agraf@suse.de>
In the current H6 CPU memory space code, the SUNXI in the macro name of
the SID address base is wrongly spelled as SNUXI, which leads to SID
readout not working.
Fix this macro name.
Fixes: 55f6b1c351c9 ("sunxi: add basic memory map definitions of H6 SoC")
Signed-off-by: Icenowy Zheng <icenowy@aosc.io>
Reviewed-by: Jagan Teki <jagan@amarulasolutions.com>
Something went wrong when writing the sandbox linker scripts and so we
ended up with a .bss section marker right before the efi runtime sections.
That obviously is a terrible idea, as it may result in overwriting efi
runtime code and data. So let's move the .bss identifier behind the efi
sections.
Signed-off-by: Alexander Graf <agraf@suse.de>
It was observed sometimes U-Boot as the EFI payload fails to boot on
QEMU. This is because TSC calibration fails with no valid frequency.
This adds default TSC frequency in the device tree.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
It was observed sometimes U-Boot as the coreboot payload fails to
boot on QEMU. This is because TSC calibration fails with no valid
frequency. This adds default TSC frequency in the device tree.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
There is no need to keep a separate coreboot_fb.dtsi since now we
have a generic coreboot payload dts.
While we are here, this also remove the out-of-date description in
the documentation regarding to coreboot framebuffer driver with
U-Boot loaded as a payload from coreboot. As the testing result with
QEMU 2.5.0 shows, the driver just works like a charm.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Currently building U-Boot as the coreboot payload requires user
to change the build configuration for a specific board during
menuconfig process. This uses the board's native device tree
to configure the hardware. For example, the device tree provides
PCI address range for the PCI host controller and U-Boot will
re-program all PCI devices' BAR to be within this range. In order
to make sure we don't mess up the hardware, we should guarantee
the range matches what coreboot programs the chipset.
But we really should make the coreboot payload support easier.
Just like EFI payload, we can create a generic coreboot payload
for all x86 boards as well. The payload is configured to include
as many generic drivers as possible. All stuff that touches low
level initialization are not allowed as such is the coreboot's
responsibility. Platform specific drivers (like gpio, spi, etc)
are not included.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Chen-Yu Tsai