Now that those common Allwinner config symbols are defined automatically
for all boards in their Kconfig files, we can remove the now redundant
definitions from the boards' _defconfig files.
Some boards had a differing definiton for some of those symbols, it
looks like mostly to "merge races" when the symbol was introduced (new
board *_defconfig file missed the "add symbol to all files" patch).
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Acked-by: Maxime Ripard <mripard@kernel.org>
Tested-by: Jagan Teki <jagan@amarulasolutions.com> # Amarula A64-Relic
Reviewed-by: Jagan Teki <jagan@amarulasolutions.com>
Some config symbols are found in *almost* every _defconfig file for
Allwinner boards, because those options are actually a platform choice,
and not a per-board decision.
Some of these options are older, some have recently been added.
Move those options to be set for all Allwinner boards in their
respective Kconfig files.
The rationales are as follows:
- NR_DRAM_BANKS: All Allwinner SoC map DRAM at one contiguous region of
address space only, starting at 1 GB. So it's always one bank.
- SPL_{DOS,EFI}_PARTITION: The Allwinner SPL does only support raw MMC
accesses, we don't care about filesystems or partitions in there, so
there is no need to define those symbols at all.
- USE_PREBOOT: We start USB early when a keyboard is configured, using the
preboot env variable, so we need to set this variable.
- SYS_RELOC_GD_ENV_ADDR: We don't specify any ENV_ADDR, so need this
symbol to be set (according to 8d8ee47e03).
- SYS_USB_EVENT_POLL_VIA_INT_QUEUE: According to commit eab9433aa5,
specifying this reduces the latency of the USB keyboard handling, so
this was formerly enabled in config headers for all Allwinner boards.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Acked-by: Maxime Ripard <mripard@kernel.org>
Tested-by: Jagan Teki <jagan@amarulasolutions.com> # Amarula A64-Relic
Reviewed-by: Jagan Teki <jagan@amarulasolutions.com>
The Pine H64 board comes with some onboard SPI flash chip, which is
perfect for loading SPL, ATF and U-Boot proper from there.
Enable the functionality in the defconfig, so that we use sunxi-fel to
transfer a bootable image to the NOR flash.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Reviewed-by: Jagan Teki <jagan@amarulasolutions.com>
The Allwinner H6 SoC uses a quite different memory map, also changes the
clocks quite a bit. This requires some changes to the SPL SPI routine,
which hardcodes these values so far.
Using the just introduced helper functions to determine base address
and SPI controller generation, we can cover some of these differences
easily.
The clock setup is different, so requires some explicit code changes
there (reset and clock gate in one register at a different address).
Also we need to change the pinmux function to use a different set of
pins that the H6 uses for SPI0.
Eventually we can enable the H6 to use SPI booting in Kconfig.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Reviewed-by: Jagan Teki <jagan@amarulasolutions.com>
Now that we can easily select an SoC specific SPI0 base address, adding
support for the Allwinner R40 is fairly trivial:
We set the base address, add this SoC to the ones that use PC23 and
enable it in Kconfig.
This allows booting from SPI flash on R40 boards.
Tested on a Bananapi M2 Berry with SPI flash connected to the header pins.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Reviewed-by: Jagan Teki <jagan@amarulasolutions.com>
So far we were using the CONFIG_SUNXI_GEN_SUN6I symbol to select between
the two SPI controller generations used on Allwinner SoCs. This is a
convenience symbol to roughly differentiate between "older" and "newer"
generation of SoCs.
The H6 SoCs is the newest SoC so far, but is sufficiently different to
not define this symbol. However it is using a SPI controller compatible
to the "new gen" SoCs.
To prepare for H6 support, we replace the check for this single symbol
with an explicit function, which can later be extended.
For now we just return CONFIG_SUNXI_GEN_SUN6I in there, so this does not
create a functional change.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Reviewed-by: Jagan Teki <jagan@amarulasolutions.com>
So far on all supported Allwinner SoCs we find the old generation SPI
controller always at address 0x1c05000, and the new generation one at
0x1c68000. However the Allwinner R40 SoC has a new generation SPI at
the old address, and the H6 uses a completely different address.
So split off the base address from the respective SPI registers, by
changing the #defines to just contain offsets.
The base address is provided by a function, so it can easily be extended
later when support for those SoCs materialises.
This does not change the code size (since the toolchain is clever enough
to optimise this properly), also does not bring any functional change at
this point.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Reviewed-by: Jagan Teki <jagan@amarulasolutions.com>
This series fixes several bugs:
* consider the /reserved-memory node in the device tree
* consider memory reservations created in ft_board_setup()
* correct output of 'efidebug memmap' on the sandbox
* correct the definition of efi_capsule_header()
Furthermore some definitions needed for future patches are added to
header files.
-----BEGIN PGP SIGNATURE-----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=zB6g
-----END PGP SIGNATURE-----
Merge tag 'efi-2020-04-rc4-4' of https://gitlab.denx.de/u-boot/custodians/u-boot-efi
Pull request for UEFI sub-system for efi-2020-04-rc4 (4)
This series fixes several bugs:
* consider the /reserved-memory node in the device tree
* consider memory reservations created in ft_board_setup()
* correct output of 'efidebug memmap' on the sandbox
* correct the definition of efi_capsule_header()
Furthermore some definitions needed for future patches are added to
header files.
Using memcpy() for some MMIO access is generally frowned upon and might
break things on some platforms. Allwinner H3, which fails to boot, being
an example here.
Moreover, fields being accessed are naturally aligned and warnings
produced by GCC have been quiesced for some time already by:
53dc8ae ("gcc-9: silence 'address-of-packed-member' warning")
That said, it should be okay to revert this commit.
This reverts commit 9bd34a69a4.
Cc: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Tomasz Duszynski <tduszyns@gmail.com>
Acked-by: Jagan Teki <jagan@amarulasolutions.com>
K3 J721E:
* OSPI boot support
* Support for loading remote cores in R5 SPL
* PMIC ESM Support
* Minor fixes for R5F and C7x remoteproc drivers
K3 AM654:
* Update AVS class 0 voltages.
* Add I2C nodes
DRA7xx/AM57xx:
* Fixed Android boot on AM57xx
AM33/AM43/Davinci:
* switch to driver model for the net and mdio driver for baltos
* Add DM/DTS support for omap video driver
* Enable fastboot on am335x-evm
i2c: for next
- i2c-gpio: make it possible to run deblock sequence on driver probe
- i2c-gpio: add clock stretching support
- updates the Designware I2C driver for high speed mode,
fix a bug and add some improvements.
- add DM support for memory based bootcounter driver
See UEFI specification, section 8.5.3.
In addition, the structure, efi_capsule_header, should be "packed"
as it is a serialized binary format in a capsule file.
Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
This is a preparatory patch.
Those functions will be used in an implementation of UEFI firmware
management protocol as part of my capsule update patch.
Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
Some of those values will be used in an implementation of UEFI firmware
management protocol as part of my capsule update patch.
Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
These flags are expected to be set in OsIndicationsSupported variable
if corresponding features are supported. See UEFI specification,
section 8.5.4.
In particular, capsule-related flags will be used in my capsule
update patch.
Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
Fix misspelled EFI_OS_INDICATIONS_BOOT_TO_FW_UI.
Reviewed-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Currently, bootefi only parses memory reservation block to setup
EFI reserved memory mappings. However, it doesn't parse the
reserved-memory[1] device tree node that also can contain the
reserved memory regions.
Add capability to parse reserved-memory node and update the EFI memory
mappings accordingly.
1. <U-Boot source>/doc/device-tree-bindings/reserved-memory/reserved-memory.txt]
Signed-off-by: Atish Patra <atish.patra@wdc.com>
Fix an endless loop.
The /reserved-memory node may have children without reg property. Remove
a superfluous debug statement.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Addresses in the sandbox's device tree are in the sandbox's virtual address
space. If we want to compare memory reservations in the device-tree with
the output of 'efidebug memmap', we need to convert back to this address
space.
Adjust the output of the 'efidebug memmap' command.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Some memory reservations are made in ft_board_setup(). Ensure that we
create reserved memory map entries after ft_board_setup().
The downside of this patch is that if bootefi is called multiple times with
an devicetree argument superfluous reservations for the old copies of the
device tree will exist. But that is still better than missing a reservation.
Deleting the superfluous reservations is not possible because reservations
in the memory map are rounded to page size and may be coallesced.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Few v0.1 SBI calls are being replaced by new SBI calls that follows
v0.2 calling convention.
Implement the replacement extensions and few additional new SBI
function calls that makes way for a better SBI interface in future.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Pragnesh Patel <pragnesh.patel@sifive.com>
We now have SBI v0.2 which is more scalable and extendable to handle
future needs for RISC-V supervisor interfaces.
Introduce a new config and move all SBI v0.1 code under that config.
This allows to implement the new replacement SBI extensions cleanly
and remove v0.1 extensions easily in future. Currently, the config
is enabled by default. Once all M-mode software, with v0.1, is no
longer in use, this config option and all relevant code can be easily
removed.
This commit is inspired from Linux kernel patch:
https://patchwork.kernel.org/patch/11407361/
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Pragnesh Patel <pragnesh.patel@sifive.com>
Few v0.1 SBI calls are being replaced by new SBI calls that follows
v0.2 calling convention.
This patch just defines these new extensions.
This commit is inspired from Linux kernel patch:
https://patchwork.kernel.org/patch/11407359/
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Pragnesh Patel <pragnesh.patel@sifive.com>
The SBI v0.2 introduces a base extension which is backward compatible
with v0.1. Implement all helper functions and minimum required SBI
calls from v0.2 for now. All other base extension function will be
added later as per need.
As v0.2 calling convention is backward compatible with v0.1, remove
the v0.1 helper functions and just use v0.2 calling convention.
Add a new Kconfig options CONFIG_SBI for the new SBI v0.2 codes, and
let CONFIG_SBI_IPI depend on it.
This commit is inspired from Linux kernel patch:
https://patchwork.kernel.org/patch/11407363/
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Pragnesh Patel <pragnesh.patel@sifive.com>
As per the new SBI specification, current SBI implementation version
is defined as 0.1 and will be removed/replaced in future. Each of the
function call in 0.1 is defined as a separate extension which makes
easier to replace them one at a time.
Rename existing implementation to reflect that. This patch is just
a preparatory patch for SBI v0.2 and doesn't introduce any functional
changes.
This commit is inspired from Linux kernel patch:
https://patchwork.kernel.org/patch/11407355/
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Pragnesh Patel <pragnesh.patel@sifive.com>
There is no need for S-mode U-Boot to call sbi_clear_ipi() as it
can be cleared directly from S-mode. This saves some cycles.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Lukas Auer <lukas@auer.io>
Currently sbi_remote_sfence_vma{,_asid} does not pass their arguments
to SBI at all, which is semantically incorrect.
This keeps in sync with Linux kernel commit:
a21344dfc6ad: fix sbi_remote_sfence_vma{,_asid}
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Lukas Auer <lukas@auer.io>
In a few boards we had overridden, intentionally, the value used for
CONFIG_SPL_LDSCRIPT. However, rather than using the ARCH value (arm)
they used the $(ARCH) variable in make. This doesn't help really, so
switch to a hard-coded value.
Signed-off-by: Tom Rini <trini@konsulko.com>
Align Kconfig and Kbuild logic to Linux 4.18 release with minimal impact
on files outside of this scope.
Our previous Kconfig sync was done by commit e91610da7c ("kconfig:
re-sync with Linux 4.17-rc4").
A very small number of changes upstream since our sync with v4.17-rc4
that exist in the v4.18 release have already been applied here and have
been omitted from the list in this commit (and are readily available in
our own git history).
The imported Linux commits are:
[From prior to v4.17-rc4]
39a33ff80a25 kbuild: remove cc-option-align
db547ef19064 Kbuild: don't add obj tree in additional includes
b999596b963a Kbuild: don't add ../../ to include path
[From v4.17 to v4.18]
b3aa58d2e85d fixdep: suppress consecutive / from file paths in dependency list files
74656b682902 kbuild: disable new dtc graph and unit-address warnings
74d931716151 genksyms: remove symbol prefix support
e6ecfb45072c kbuild: do not display CHK for filechk
0b669a5076fd kconfig: refactor Qt package checks for building qconf
b464ef583dc7 kconfig: refactor GTK+ package checks for building gconf
1c5af5cf9308 kconfig: refactor ncurses package checks for building mconf and nconf
694c49a7c01c kconfig: drop localization support
96f60dfa5819 trace: Use -mcount-record for dynamic ftrace
bb222ceeb327 kconfig: remove string expansion in file_lookup()
96d8e48da55a kconfig: remove string expansion for mainmenu after yyparse()
5b31a9746756 kconfig: remove sym_expand_string_value()
137c0118a900 kconfig: make default prompt of mainmenu less specific
e298f3b49def kconfig: add built-in function support
2fd5b09c201e kconfig: add 'shell' built-in function
9de071536c87 kconfig: begin PARAM state only when seeing a command keyword
9ced3bddec08 kconfig: support user-defined function and recursively expanded variable
1175c02506ff kconfig: support simply expanded variable
ed2a22f277c6 kconfig: support append assignment operator
82bc8bd82e5c kconfig: expand lefthand side of assignment statement
1d6272e6fe43 kconfig: add 'info', 'warning-if', and 'error-if' built-in functions
a702a6176e2f kconfig: add 'filename' and 'lineno' built-in variables
915f64901eb3 kconfig: error out if a recursive variable references itself
2bece88f89fa kconfig: test: add Kconfig macro language tests
21c54b774744 kconfig: show compiler version text in the top comment
59f7b5847b0c kbuild: $(CHECK) doesnt need NOSTDINC_FLAGS twice
145167650b96 kbuild: add endianness flag to CHEKCFLAGS
1f2f01b122d7 kbuild: add machine size to CHECKFLAGS
d6a0c8a1326b kconfig: Add testconfig into make help output
bb6d83dde191 kbuild: Move last word of nconfig help to the previous line
8593080c0fcf kconfig: fix localmodconfig
ed7d40bc67b8 tracing: Fix SKIP_STACK_VALIDATION=1 build due to bad merge with -mrecord-mcount
b2d00d7c61c8 kconfig: fix line numbers for if-entries in menu tree
ecd53ac2f2c6 kconfig: handle P_SYMBOL in print_symbol()
73d1c580f92b kconfig: loop boundary condition fix
48f6e3cf5bc6 kbuild: do not drop -I without parameter
bd412d81b7ea kbuild: .PHONY is not a variable, but PHONY is
6916162c7308 kbuild: remove duplicated comments about PHONY
Cc: Masahiro Yamada <masahiroy@kernel.org>
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Masahiro Yamada <masahiroy@kernel.org>
This adds the following commits from upstream:
84e414b0b5bc tests: Add a test case for the omit-if-no-ref keyword
4038fd90056e dtc: add ability to make nodes conditional on them being referenced
e1f139ea4900 checks: drop warning for missing PCI bridge bus-range
f4eba68d89ee checks: Print duplicate node name instead of parent name
46df1fb1b211 .travis.yml: Run valgrind checks via Travis
14a3002a1aee tests: Update valgrind suppressions for sw_tree1
02c5fe9debc0 tests: Remove valgrind error from tests/get_path
df536831d02c checks: add graph binding checks
2347c96edcbe checks: add a check for duplicate unit-addresses of child nodes
8f1b35f88395 Correct overlay syntactic sugar for generating target-path fragments
afbddcd418fb Suppress warnings on overlay fragments
119e27300359 Improve tests for dtc overlay generation
[From Linux Kernel commit 50aafd60898a8b3edf2f60e014a8288da3b2e5e3]
Signed-off-by: Rob Herring <robh@kernel.org>
[For applying to U-Boot]
Signed-off-by: Tom Rini <trini@konsulko.com>
The Kconfig language provides a unit test that can be run. As these
require pytest to be installed and run very quickly, bundle them in to
an existing CI job.
Signed-off-by: Tom Rini <trini@konsulko.com>
As there is only one linker script to use in this case, rather than use
the BOARDDIR variable hard-code the path.
Cc: Lokesh Vutla <lokeshvutla@ti.com>
Suggested-by: Masahiro Yamada <masahiroy@kernel.org>
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Lokesh Vutla <lokeshvutla@ti.com>
Reviewed-by: Masahiro Yamada <masahiroy@kernel.org>
As there is only one machine under mach-orion5x, having a Kconfig entry
for SPL_LDSCRIPT is not helpful, move this to the defconfig file.
Suggested-by: Masahiro Yamada <masahiroy@kernel.org>
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Masahiro Yamada <masahiroy@kernel.org>
As there is only one mx31pdk config file and with upcoming updates to
the Kconfig parsing logic, rather than have an entry in
board/freescale/mx31pdk/Kconfig, move this single setting to the
defconfig file.
Cc: Magnus Lilja <lilja.magnus@gmail.com>
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Magnus Lilja <lilja.magnus@gmail.com>
In a few places we have Kconfig entries that set SPL_LDSCRIPT to what is
the default value anyways. Drop these.
Cc: Michal Simek <monstr@monstr.eu>
Cc: Rick Chen <rick@andestech.com>
Cc: Philippe Reynes <tremyfr@yahoo.fr>
Cc: Eric Jarrige <eric.jarrige@armadeus.org>
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Rick Chen <rick@andestech.com>
Reviewed-by: Michal Simek <michal.simek@xilinx.com> (for Microblaze)
Reviewed-by: Masahiro Yamada <masahiroy@kernel.org>
The default SPL / TPL linker script is in the $(ARCH) directory. The
way we use this today works but isn't ideal. With an update to Kconfig
to re-sync with the Linux Kernel, we need to escape the '$' here so that
it will end up being evaluated by make.
Cc: Masahiro Yamada <masahiroy@kernel.org>
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Masahiro Yamada <masahiroy@kernel.org>
This is U-Boot own code, and no longer used since commit 36dd5f1b8a
("dtc: Switch to building and using our own dtc unless provided").
Prior to that commit, U-Boot relied on an external dtc, so this script
was used to check the dtc version.
Now U-Boot bundles our own dtc in script/dtc/dtc like Linux kernel.
Users are still allowed to pass DTC= option from the command line,
but they are supposed to choose correct version of dtc in this case.
So, we do not check the dtc version any more.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
The static inline function spl_phase needs <asm/global_data.h>.
Some functions take pointers to struct blk_desc or image_header.
Add forward declarations.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Reviewed-by: Simon Glass <sjg@chromium.org>
'uint' is not a primitive type. You need to include <linux/types.h>
or otherwise change it to (unsigned int).
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
This header uses 'phys_addr_t' and 'ulong'. Include the definitions.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
[trini: Move include to below __ASSEMBLY__ test]
Signed-off-by: Tom Rini <trini@konsulko.com>
The compiler never knows what 'bd_t' is without including <asm/u-boot.h>.
By changing it to (struct bd_info), the compiler learns it is struct.
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
This exists in Linux Kernel with commit 70523a3ce5ff so put it in the
list of DTC_FLAGS that mirror Linux as we will catch up there.
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Masahiro Yamada <masahiroy@kernel.org>
While we are working on correcting usage related to the pci_bridge and
pci_device_bus_num warnings, disable these flags for now.
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Masahiro Yamada <masahiroy@kernel.org>
The way that we have been handling additional DTC warning flags hasn't
matched the way the Linux Kernel does. Resync this logic with v4.17.
Signed-off-by: Tom Rini <trini@konsulko.com>
The previous kbuild resync of e91610da7c ("kconfig: re-sync with Linux
4.17-rc4") accidentally did not sync the fixdep program. This commit
brings fixdep in line with the rest of that previous resync.
This includes all of the following Linux kernel commits:
fbfa9be9904e kbuild: move include/config/ksym/* to include/ksym/*
5b8ad96d1a44 fixdep: remove some false CONFIG_ matches
14a596a7e6fd fixdep: remove stale references to uml-config.h
ab9ce9feed36 fixdep: use existing helper to check modular CONFIG options
87b95a81357d fixdep: refactor parse_dep_file()
5d1ef76f5a22 fixdep: move global variables to local variables of main()
ccfe78873c22 fixdep: remove unneeded memcpy() in parse_dep_file()
4003fd80cba9 fixdep: factor out common code for reading files
01b5cbe7012f fixdep: use malloc() and read() to load dep_file to buffer
41f92cffba19 fixdep: remove unnecessary <arpa/inet.h> inclusion
7c2ec43a2154 fixdep: exit with error code in error branches of do_config_file()
4e433fc4d1a9 fixdep: trivial: typo fix and correction
dee81e988674 fixdep: faster CONFIG_ search
c1a95fda2a40 kbuild: add fine grained build dependencies for exported symbols
d8329e35cc08 fixdep: accept extra dependencies on stdin
4c835b57b8de fixdep: constify strrcmp arguments
Of note is that when applying dee81e988674 above our logic in that area
required some careful consideration to continue to apply.
[Fold in bugfix to allow us to include 638e69cf2230 from upstream]
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
[Merge everything to U-Boot, rework dee81e988674]
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Masahiro Yamada <masahiroy@kernel.org>
Currently, fixdep skips parsing include/linux/kconfig.h, but if it
parsed it, it would translate the following code in kconfig.h
config_enabled(CONFIG_VAL(option##_MODULE)
into:
$(wildcard include/config/option##/module.h)
When Kbuild includes .*.cmd, it would emit the following error:
*** unterminated call to function 'wildcard': missing ')'. Stop.
This issue prevents us from importing the upstream Linux commit
638e69cf2230 ("fixdep: do not ignore kconfig.h").
Fix this by handling only alphanumerical characters and underscores.
This makes sense because they match to the valid character sets in
Kconfig symbols.
As a side-note, you can reproduce this issue only on GNU Make <= 4.2.1
For GNU Make <= 4.2.1, the '#' always means the start of a comment.
Hence, GNU Make thinks the closing ')' is missing.
The following commit in GNU Make changed how it handles '#' in
function invocations. So, this does not happen for GNU Make 4.3
| commit c6966b323811c37acedff05b576b907b06aea5f4
| Author: Paul Smith <psmith@gnu.org>
| Date: Thu Dec 22 18:47:26 2016 -0500
|
| [SV 20513] Un-escaped # are not comments in function invocations
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Reported-by: Tom Rini <trini@konsulko.com>
On some platforms/architectures the value from get_timer() can wrap.
This is particularly problematic when long-running code needs to measure
a time difference as is the case with watchdog_reset() which tries to
avoid tickling the watchdog too frequently.
Use time_after() from time.h instead of a plain > comparison to avoid
any issues with the time wrapping on a system that has been sitting in
u-boot for a long time.
Signed-off-by: Chris Packham <judge.packham@gmail.com>
Reviewed-by: Stefan Roese <sr@denx.de>
