Tom Rini [Thu, 23 Jun 2022 18:30:27 +0000 (14:30 -0400)]
Merge branch '2022-06-23-scmi-optee-and-smccc-updates' into next
This consists of two slightly related series. For the first, to quote
the author:
This series implements 2 features in driver/firmware/scmi.
First, a single change adds support for SCMI OP-TEE transport to
use OP-TEE native shared memory. See the 1st patch in this series:
"firmware: scmi: optee: use TEE shared memory for SCMI messages".
Then come changes for supporting multi-channel in the SCMI drivers.
I've split the implementation in 11 several small incremental changes
in the hope it helps the review. Few minor fixup commits are also
inserted in the series.
And the second series implements some smccc improvements.
Tom Rini [Thu, 23 Jun 2022 18:24:24 +0000 (14:24 -0400)]
Merge branch '2022-06-23-fuzzing-and-asan-for-sandbox' into next
To quote the author:
This series introduces ASAN and a basic fuzzing infrastructure that
works with sandbox. The example fuzz test towards the end of the series
will find something pretty quickly. That something is fixed by the
series "virtio: Harden and test vring" that needs to be applied for the
final patch in this series.
There is some refactoring to stop using '.' prefixed sections. ELF
defines sections with names that contain anything that isn't
alphanumeric or an underscore as being for system use which means
clang's ASAN instrumentation happily add redzones between the contained
objects. That's not what we want for things like linker lists where the
linker script has carefully placed the sections contiguously. By
renaming the sections, clang sees them as user sections and doesn't add
instrumentation.
ASAN is left disabled by default as there are still some tests that it
triggers on and will need some more investigation to fix. It can be
enabled with CONFIG_ASAN or passing `-a ASAN` to buildman.
Adds random number generator driver using Arm SMCCC TRNG interface to
get entropy bytes from secure monitor. The driver registers as an
Arm SMCCC feature driver to allow PSCI driver to bind a device for
when secure monitor exposes RNG support from Arm SMCCC TRNG interface.
firmware: psci: bind arm smccc features when discovered
Use PSCI device to query Arm SMCCC v1.1 support from secure monitor
and if so, bind drivers for the SMCCC features that monitor supports.
Drivers willing to be bound from Arm SMCCC features discovery can use
macro ARM_SMCCC_FEATURE_DRIVER() to register to smccc feature discovery,
providing target driver name and a callback function that returns
whether or not the SMCCC feature is supported by the system.
Etienne Carriere [Tue, 31 May 2022 16:09:29 +0000 (18:09 +0200)]
firmware: scmi: use multi channel in mailbox, optee and smccc agents
Updates .process_msg operators of the SCMI transport drivers that
supports multi-channel to use it now that drivers do provide
the reference through channel argument. These are the mailbox
agent, the optee agent and the smccc agent.
Etienne Carriere [Tue, 31 May 2022 16:09:27 +0000 (18:09 +0200)]
power: regulator: scmi: support SCMI multi-channel
Update SCMI regulator controller driver to get its assigned SCMI channel
during initialization. This change allows SCMI voltage domain protocol
to use a dedicated channel when defined in the DT. The reference is
saved in SCMI regulator controller driver private data.
Etienne Carriere [Tue, 31 May 2022 16:09:26 +0000 (18:09 +0200)]
reset: scmi: support SCMI multi-channel
Update SCMI reset controller driver to get its assigned SCMI channel
during initialization. This change allows SCMI reset domain protocol
to use a dedicated channel when defined in the DT. The reference is
saved in SCMI reset controller driver private data.
Etienne Carriere [Tue, 31 May 2022 16:09:25 +0000 (18:09 +0200)]
clk: scmi: support SCMI multi-channel
Update SCMI clock driver to get its assigned SCMI channel during
initialization. This change allows SCMI clock protocol to use a
dedicated channel when defined in the DT. The reference is saved
in SCMI clock driver private data.
Cc: Lukasz Majewski <lukma@denx.de> Cc: Sean Anderson <seanga2@gmail.com> Signed-off-by: Etienne Carriere <etienne.carriere@linaro.org>
Etienne Carriere [Tue, 31 May 2022 16:09:21 +0000 (18:09 +0200)]
firmware: scmi: add multi-channel support
Adds resources for SCMI protocols to possibly use a dedicated SCMI
channel instead of the default channel allocated by the SCMI agent
during initialization. As per DT binding documentation, some SCMI
transports can define a specific SCMI communication channel for
given SCMI protocols. It allows SCMI protocols to pass messages
concurrently each other.
This change introduces new scmi agent uclass API function
devm_scmi_of_get_channel() for SCMI drivers probe sequences to get
a reference to the SCMI channel assigned to its related SCMI protocol.
The function queries the channel reference to its SCMI transport driver
through new scmi agent uclass operator .of_get_channel that uses Device
Tree information from related SCMI agent node.
Operator .of_get_channel returns a reference to the SCMI channel
assigned to SCMI protocol used by the caller device. SCMI transport
drivers that do not support multi-channel are not mandated to register
this operator. When so, API function devm_scmi_of_get_channel() returns
NULL and SCMI transport driver are expected to retrieve by their own
means the reference to the unique SCMI channel, for example using
platform data as these drivers currently do in U-Boot source tree.
Etienne Carriere [Tue, 31 May 2022 16:09:18 +0000 (18:09 +0200)]
firmware: scmi: prepare scmi uclass API to multi-channel
Changes SCMI driver API function devm_scmi_process_msg() to add
an SCMI channel reference argument for when SCMI agent supports
SCMI protocol specific channels. First argument of devm_scmi_process_msg()
is also change to point to the caller SCMI protocol device rather
than its parent device (the SCMI agent device).
The argument is a pointer to opaque struct scmi_channel known from
the SCMI transport drivers. It is currently unused and caller a pass
NULL value. A later change will enable such support once SCMI protocol
drivers have means to get the channel reference during initialization.
Cc: Lukasz Majewski <lukma@denx.de> Cc: Sean Anderson <seanga2@gmail.com> Cc: Jaehoon Chung <jh80.chung@samsung.com> Signed-off-by: Etienne Carriere <etienne.carriere@linaro.org>
Etienne Carriere [Tue, 31 May 2022 16:09:16 +0000 (18:09 +0200)]
firmware: scmi: optee: use TEE shared memory for SCMI messages
Changes implementation when using TEE dynamically allocated shared
memory to synchronize with the Linux implementation where the legacy
SMT protocol cannot be used with such memory since it is expected from
device mapped memory whereas OP-TEE shared memory is cached and
hence should not be accessed using memcpy_toio()/memcpy_fromio().
This change implements the MSG shared memory protocol introduced
in Linux [1]. The protocol uses a simplified SMT header of 32bit
named MSG_SMT to carry SCMI protocol information and uses side channel
means to carry exchanged buffer size information, as TEE invocation API
parameters when used in the SCMI OP-TEE transport.
Andrew Scull [Mon, 30 May 2022 10:00:11 +0000 (10:00 +0000)]
sandbox: Add libfuzzer integration
Add an implementation of LLVMFuzzerTestOneInput() that starts the
sandbox on a secondary thread and exposes a function to synchronize the
generation of fuzzing inputs with their consumption by the sandbox.
Signed-off-by: Andrew Scull <ascull@google.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Andrew Scull [Mon, 30 May 2022 10:00:07 +0000 (10:00 +0000)]
CI: Azure: Build with ASAN enabled
In order to prevent build regressions with ASAN, add the builds to CI.
The longer term objective will be to enabled test targets with ASAN
enabled, but there are too many at the moment.
Andrew Scull [Mon, 30 May 2022 10:00:06 +0000 (10:00 +0000)]
test/py: test_stackprotector: Disable for ASAN
The stack protector test intentionally overflows a buffer in order to
corrupt the stack canary so that it can test that the corruption is
detected as expected. However, this is incompatible with ASAN, which
detects the buffer overflow and interrupts the test, so disable the test
for such configurations.
Andrew Scull [Mon, 30 May 2022 10:00:05 +0000 (10:00 +0000)]
sandbox: Add support for Address Sanitizer
Add CONFIG_ASAN to build with the Address Sanitizer. This only works
with the sandbox so the config is likewise dependent. The resulting
executable will have ASAN instrumentation, including the leak detector
that can be disabled with the ASAN_OPTIONS environment variable:
ASAN_OPTIONS=detect_leaks=0 ./u-boot
Since u-boot uses its own dlmalloc, dynamic allocations aren't
automatically instrumented, but stack variables and globals are.
Instrumentation could be added to dlmalloc to poison and unpoison memory
as it is allocated and deallocated, and to introduce redzones between
allocations. Alternatively, the sandbox may be able to play games with
the system allocator and somehow still keep the required memory
abstraction. No effort to address dynamic allocation is made by this
patch.
The config is not yet enabled for any targets by default.
Signed-off-by: Andrew Scull <ascull@google.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Andrew Scull [Mon, 30 May 2022 10:00:04 +0000 (10:00 +0000)]
linker_lists: Rename sections to remove . prefix
Rename the sections used to implement linker lists so they begin with
'__u_boot_list' rather than '.u_boot_list'. The double underscore at the
start is still distinct from the single underscore used by the symbol
names.
Having a '.' in the section names conflicts with clang's ASAN
instrumentation which tries to add redzones between the linker list
elements, causing expected accesses to fail. However, clang doesn't try
to add redzones to user sections, which are names with all alphanumeric
and underscore characters.
Signed-off-by: Andrew Scull <ascull@google.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Andrew Scull [Mon, 30 May 2022 10:00:03 +0000 (10:00 +0000)]
sandbox: Rename getopt sections
Rename the sections used for defining sandbox command line options so
that they don't start with a '.'. ELF says that sections starting with a
'.' are reserved for system use, but the sandbox runs as a normal user
process so should be using user sections instead.
Clang's ASAN adds redzones to non-user sections and the extra padding
meant that the list of options was being corrupted. Naming the sections
as user sections avoids this issue as clang handles them as we intended.
Andrew Scull [Mon, 30 May 2022 10:00:02 +0000 (10:00 +0000)]
sandbox: Rename EFI runtime sections
Rename the sections used for placing the EFI runtime so that they don't
start with a '.'. ELF says that sections starting with a '.' are
reserved for system use, but the sandbox runs as a normal user process
so should be using user sections instead.
Clang's ASAN adds redzones to non-user sections and the extra padding
meant that the list of options was being corrupted. Naming the sections
as user sections avoids this issue as clang handles them as we intended.
Andrew Scull [Mon, 30 May 2022 10:00:01 +0000 (10:00 +0000)]
serial: sandbox: Fix buffer underflow in puts
Fix the buffer underflow that would occur if puts is called with length
of zero.
Fixes: efa51f2bd64 ("serial: sandbox: Implement puts") Cc: Sean Anderson <sean.anderson@seco.com> Cc: Simon Glass <sjg@chromium.org> Reviewed-by: Sean Anderson <sean.anderson@seco.com>
Chia-Wei Wang [Wed, 1 Jun 2022 08:21:15 +0000 (16:21 +0800)]
configs: ast2600: Move SPL bss section to DRAM space
The commit b583348ca8c8 ("image: fit: Align hash output buffers") places
the hash output buffer at the .bss section. However, AST2600 by default
executes SPL in the NOR flash XIP way. This results in the hash output
cannot be written to the buffer as it is located at the R/X only region.
We need to move the .bss section out of the SPL body to the DRAM space,
where hash output can be written to. This patch includes:
- Define the .bss section base and size
- A new SPL linker script is added with a separate .bss region specified
- Enable CONFIG_SPL_SEPARATE_BSS kconfig option
Signed-off-by: Chia-Wei Wang <chiawei_wang@aspeedtech.com> Reviewed-by: Neal Liu <neal_liu@aspeedtech.com>
Nick Hawkins [Wed, 8 Jun 2022 21:21:39 +0000 (16:21 -0500)]
ARM: dts: Add device tree files for hpe gxp soc
The HPE SoC is new to linux. A basic device tree layout with minimum
required for linux to boot including a timer and watchdog support has
been created.
The dts file is empty at this point but will be updated in subsequent
updates as board specific features are enabled.
Signed-off-by: Nick Hawkins <nick.hawkins@hpe.com>
Nick Hawkins [Wed, 8 Jun 2022 21:21:36 +0000 (16:21 -0500)]
spi: gxp_spi: Add GXP SPI controller driver
The GXP supports 3 separate SPI interfaces to accommodate the system
flash, core flash, and other functions. The SPI engine supports variable
clock frequency, selectable 3-byte or 4-byte addressing and a
configurable x1, x2, and x4 command/address/data modes. The memory
buffer for reading and writing ranges between 256 bytes and 8KB. This
driver supports access to the core flash.
Signed-off-by: Nick Hawkins <nick.hawkins@hpe.com>
Nick Hawkins [Wed, 8 Jun 2022 21:21:35 +0000 (16:21 -0500)]
timer: gxp: Add HPE GXP timer support
Add support for the HPE GXP SOC timer. The GXP supports several different
kinds of timers but for the purpose of this driver there is only support
for the General Timer. The timer has a 1us resolution and is 56 bits.
Signed-off-by: Nick Hawkins <nick.hawkins@hpe.com>
Nick Hawkins [Wed, 8 Jun 2022 21:21:34 +0000 (16:21 -0500)]
ARM: hpe: gxp: add core support
The GXP is the HPE BMC SoC that is used in the majority
of current generation HPE servers. Traditionally the asic will
last multiple generations of server before being replaced.
Info about SoC:
HPE GXP is the name of the HPE Soc. This SoC is used to implement many BMC
features at HPE. It supports ARMv7 architecture based on the Cortex A9
core. It is capable of using an AXI bus to whicha memory controller is
attached. It has multiple SPI interfaces to connect boot flash and BIOS
flash. It uses a 10/100/1000 MAC for network connectivity. It has multiple
i2c engines to drive connectivity with a host infrastructure. There
currently are no public specifications but this process is being worked.
Signed-off-by: Nick Hawkins <nick.hawkins@hpe.com>
Tom Rini [Wed, 22 Jun 2022 20:08:56 +0000 (16:08 -0400)]
cmd/misc: Stop using a function pointer
Currently, enabling CMD_MISC gives:
cmd/misc.c:67:25: warning: assignment to 'int (*)(struct udevice *, int, void *, int)' from incompatible pointer type 'int (*)(struct udevice *, int, const void *, int)' [-Wincompatible-pointer-types]
Because 'misc_read' takes a void * and 'misc_write' takes a const void
*, both of which make sense for their operation. Given there's one
place we make use of the function pointer, just call read or write
directly for the operation we're called with.
Reviewed-by: Bin Meng <bmeng.cn@gmail.com> Reviewed-by: Sean Anderson <seanga2@gmail.com> Signed-off-by: Tom Rini <trini@konsulko.com>
Rui Miguel Silva [Wed, 11 May 2022 09:55:41 +0000 (10:55 +0100)]
arm: add support to corstone1000 platform
Corstone1000 is a platform from arm, which includes pre
verified Corstone SSE710 sub-system that combines Cortex-A and
Cortex-M processors [0].
This code adds the support for the Cortex-A35 implementation
at host side, it contains also the necessary bits to support
the Corstone 1000 FVP (Fixed Virtual Platform) [1] and also the
FPGA MPS3 board implementation of this platform. [2]
Rui Miguel Silva [Wed, 11 May 2022 09:55:40 +0000 (10:55 +0100)]
cmd: load: add load command for memory mapped
cp.b is used a lot as a way to load binaries to memory and execute
them, however we may need to integrate this with the efi subsystem to
set it up as a bootdev.
So, introduce a loadm command that will be consistent with the other
loadX commands and will call the efi API's.
Tom Rini [Mon, 20 Jun 2022 12:09:24 +0000 (08:09 -0400)]
Merge tag 'u-boot-stm32-20220620' of https://source.denx.de/u-boot/custodians/u-boot-stm into next
- Add STM32MP13 SoCs support with associated board STM32M135F-DK
- Correct livetree support in stm32mp1 boards
- Activate livetree for stm32mp15 DHSOM boards
Michael Walle [Mon, 30 May 2022 21:02:09 +0000 (23:02 +0200)]
board: sl28: support 8 GiB memory
The board supports up to 8 GiB memory. The memory is soldered on the
board but the configuration is equivalent to a dual chip select, dual
rank DIMM module.
Michael Walle [Mon, 30 May 2022 21:02:07 +0000 (23:02 +0200)]
board: sl28: set CPO value
With a 8GiB memory board, it seems that the "very unlikely event" of a
DDR initialization with non-optimal values are not really that unlikely.
It happens in about every other reboot. As described in erratum
A-009942, preset the DEBUG_28 register with an optimal value. The value
iself depends on the memory configuration of the board, but the used
value seems to work well for all variants.
Michael Walle [Mon, 30 May 2022 21:02:05 +0000 (23:02 +0200)]
armv8: layerscape: add missing RCW source defines
A board might need to get the source of the RCW word, which is also the
boot source in most cases.
These defines are taken from the LS1028A and I expect they are the same
across the SoCs with the same chassis, after all, there was already a
reset source for NOR flash.
Pali Rohár [Thu, 26 May 2022 08:52:27 +0000 (10:52 +0200)]
board: freescale: p1_p2_rdb_pc: Move boot reset macros to p1_p2_bootsrc.h
Code for changing boot source is platform generic and can be used by any
P1* and P2* compatible RDB board. Not only by boards which use config
header file p1_p2_rdb_pc.h.
So move this code from p1_p2_rdb_pc.h to p1_p2_bootsrc.h and cleanup macros
for generating boot source env variables in CONFIG_EXTRA_ENV_SETTINGS.
This allows to use code for resetting board and rebooting to other boot
source also by other boards in future.
Pali Rohár [Wed, 11 May 2022 18:27:13 +0000 (20:27 +0200)]
mmc: fsl_esdhc: Add new config option for default fallback mode
Currently default fallback SDHC mode is 1-bit. Add new config option
CONFIG_SYS_FSL_ESDHC_DEFAULT_BUS_WIDTH to allow specifying default fallback
mode. This is useful e.g. for SPL builds which loads other parts from SD
card during boot process.
Pali Rohár [Wed, 11 May 2022 18:27:12 +0000 (20:27 +0200)]
mmc: fsl_esdhc: Set fallback mode to 1-bit
8-bit mode is not supported by SD cards and on P2020 are four SDHC pins
shared with SPI (so if P2020 board have also SPI then only 4-bit SDHC mode
is provided). So 8-bit SDHC mode is really bad default.
When max bus width is not provided then set mode to 1-bit. This mode is
supported by all cards, so it is the best option for fallback mode.
Also P2020 bootrom sets mode to 1-bit when booting from SD/MMC card.
Alison Wang [Tue, 10 May 2022 10:29:10 +0000 (18:29 +0800)]
ls1028a: hdp: Add config support for HDP firmware loading
This patch adds config support for HDP firmware loading on LS1028A.
Signed-off-by: Oliver Brown <oliver.brown@nxp.com> Signed-off-by: Alison Wang <alison.wang@nxp.com> Signed-off-by: Ye Li <ye.li@nxp.com> Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
Pali Rohár [Sun, 1 May 2022 15:45:58 +0000 (17:45 +0200)]
powerpc: fsl_law: Add definition for first PCIe target interface
Header file asm/fsl_law.h already provides correct definition for second
and third PCIe controller (LAW_TRGT_IF_PCIE_2 and LAW_TRGT_IF_PCIE_3). But
is missing definition for the first PCIe controller (LAW_TRGT_IF_PCIE_1).
Note that existing definition for LAW_TRGT_IF_PCIE_2 and LAW_TRGT_IF_PCIE_3
are slightly complicated, but are really correct for P2020 platform.
When CONFIG_WDT is enabled then non-DM watchdog code cannot be used due to
conflicting functions like watchdog_reset(). So disable compilation of
mpc85xx watchdog_reset() function when CONFIG_WDT is enabled.
Sean Anderson [Fri, 22 Apr 2022 18:34:19 +0000 (14:34 -0400)]
ARM: dts: ls1021a: update the clockgen node
QorIQ platforms now use different clock bindings. Although we don't use
the device tree for clocks on this platform, it is helpful to sync it
because then the bindings will more closely match Linux. Additionally,
it allows for using more clock fractions (such as platform/4).
This corresponds to Linux commit b6f5e7019391 ("ARM: dts: ls1021a:
update the clockgen node").
Signed-off-by: Sean Anderson <sean.anderson@seco.com>
Sean Anderson [Fri, 22 Apr 2022 18:34:18 +0000 (14:34 -0400)]
arm: layerscape: Add sfp driver
This adds a driver for the Security Fuse Processor (SFP) present on
LS1012A, LS1021A, LS1043A, and LS1046A processors. It holds the
Super-Root Key (SRK), One-Time-Programmable Master Key (OTPMK), and
other "security" related fuses. Similar devices (sharing the same name)
are present on other processors, but for the moment this just supports
the LS2 variants.
The mirror registers are loaded during power-on reset. All mirror
registers must be programmed or read at once. Because of this, `fuse
prog` will program all fuses, even though only one might be specified.
To prevent accidentally burning through all your fuse programming cycles
with something like `fuse prog 0 0 A B C D`, we limit ourselves to one
programming cycle per reset. Fuses are numbered based on their address.
The fuse at 0x1e80200 is 0, the fuse at 0x1e80204 is 1, etc.
The TA_PROG_SFP supply must be enabled when programming fuses, but must
be disabled when reading them. Typically this supply is enabled by
inserting a jumper or by setting a register in the board's FPGA. I've
also added support for using a regulator. This could be helpful for
automatically issuing the FPGA write, or for toggling a GPIO controlling
the supply.
I suggest using the following procedure for programming:
1. Override the fuses you wish to program
=> fuse override 0 2 A B C D
2. Inspect the values and ensure that they are what you expect
=> fuse sense 0 2 4
3. Enable TA_PROG_SFP
4. Issue a program command using OSPR0 as a dummy. Since it contains the
write-protect bit you will usually want to write it last anyway.
=> fuse prog 0 0 0
5. Disable TA_PROG_SFP
6. Read back the fuses and ensure they are correct
=> fuse read 0 2 4
Signed-off-by: Sean Anderson <sean.anderson@seco.com>
If CONFIG_VIDEO_DM=n we query the display size from the serial console.
Especially when using a remote console the response can be so late that
it interferes with autoboot.
Only query the console size when running an EFI binary.
Add debug output showing the determined console size.
Reported-by: Fabio Estevam <festevam@gmail.com> Fixes: a57ad20d07e8 ("efi_loader: split efi_init_obj_list() into two stages") Fixes: a9bf024b2933 ("efi_loader: disk: a helper function to create efi_disk objects from udevice") Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com> Tested-by: Fabio Estevam <festevam@denx.de> Tested-by: Heiko Thiery <heiko.thiery@gmail.com>
When the UEFI sub-system is initialized it sends an escape sequence to the
serial console to determine the terminal size. This stops the
run_command_list() function of the console emulation from recognizing the
U-Boot command line prompt.
Add a 'print -e' command as first command in the command list to work
around this issue.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Patrick Delaunay [Wed, 15 Jun 2022 17:41:48 +0000 (19:41 +0200)]
configs: stm32mp: cleanup the stm32mp15 file
Remove STM32_SYSRAM_END and clean the comments in stm32mp15_common.h file
after moving some CONFIG to Kconfig: CONFIG_SYS_CBSIZE,
CONFIG_SPL_MAX_FOOTPRINT, CONFIG_SYS_SPL_MALLOC_START and
CONFIG_SYS_SPL_MALLOC_SIZE.
Signed-off-by: Patrick Delaunay <patrick.delaunay@foss.st.com> Reviewed-by: Patrice Chotard <patrice.chotard@foss.st.com>
stm32mp: stpmic1: remove the debug unit request by debugger
Depending on backup register value, U-Boot SPL maintains the debug unit
powered-on for debugging purpose; only BUCK1 is required for powering
the debug unit, so revert the setting for all the other power lanes,
except BUCK3 that has to be always on.
To be functional this patch requires a modification in the debugger
,openocd for example, to update the STM32MP15 backup register when it is
required to debug SPL after reset. After deeper analysis this behavior
will be never supported in tools so the associated code, will be never
used and the associated code can be removed.
Signed-off-by: Patrick Delaunay <patrick.delaunay@foss.st.com> Reviewed-by: Patrice Chotard <patrice.chotard@foss.st.com>
projects / u-boot.git / log
