Simon Glass [Wed, 16 May 2018 15:42:26 +0000 (09:42 -0600)]
Add a comment for board_quiesce_devices()
This exported function should have a comment describing what it does. Also
it should really be removed in favour of device_remove(), which handles
this sort of thing now. Add a comment with a TODO.
Signed-off-by: Simon Glass <sjg@chromium.org> Signed-off-by: Alexander Graf <agraf@suse.de>
Simon Glass [Wed, 16 May 2018 15:42:25 +0000 (09:42 -0600)]
Define board_quiesce_devices() in a shared location
This undocumented function relies on arch-specific code to declare a nop
weak version. Add the weak function in common code instead to avoid having
to duplicate the same function in each arch.
Signed-off-by: Simon Glass <sjg@chromium.org> Signed-off-by: Alexander Graf <agraf@suse.de>
Simon Glass [Wed, 16 May 2018 15:42:22 +0000 (09:42 -0600)]
sandbox: Add a setjmp() implementation
Add an implementation of setjmp() and longjmp() which rely on the
underlying host C library. Since we cannot know how large the jump buffer
needs to be, pick something that should be suitable and check it at
runtime. At present we need access to the underlying struct as well.
Signed-off-by: Simon Glass <sjg@chromium.org> Signed-off-by: Alexander Graf <agraf@suse.de>
The UEFI spec mandates that unaligned memory access should be enabled if
supported by the CPU architecture.
This patch implements the function unaligned_access() to reset the aligned
access flag in the system control register (SCTLR). It is called when the
bootefi command is invoked.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
[agraf: fix SPDX identifier] Signed-off-by: Alexander Graf <agraf@suse.de>
Emmanuel Vadot [Fri, 4 May 2018 08:26:55 +0000 (10:26 +0200)]
video: sunxi: de2: Reserve the fb region in the EFI memory map
If compile with support for the efi loader we need to mark the pages
allocated for the framebuffer as reserved so the kernel won't attempt
to use them for other uses.
Signed-off-by: Emmanuel Vadot <manu@freebsd.org> Acked-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Signed-off-by: Alexander Graf <agraf@suse.de>
Ivan Gorinov [Fri, 11 May 2018 20:18:25 +0000 (13:18 -0700)]
efi_loader: fix off-by-one bug in efi_get_variable
efi_get_variable() always stores an extra zero byte after the output data.
When the returned data size matches the output buffer size, the extra zero
byte is stored past the end of the output buffer.
Signed-off-by: Ivan Gorinov <ivan.gorinov@intel.com> Reviewed-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Signed-off-by: Alexander Graf <agraf@suse.de>
efi_loader: avoid using unitialized values as console size
If a request for the console size would be answered with a response
with less then three values, uninitialized stack memory would be
copied to the number of rows and columns of the terminal.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Reviewed-by: Simon Glass <sjg@chromium.org> Signed-off-by: Alexander Graf <agraf@suse.de>
efi_loader: adjust definitions of variable services
The definitons of the variable services are adjusted:
- use efi_uintn_t instead of unsigned long
- use u16 * instead of s16 * for Unicode strings
- correct definition of QueryVariableInfo
- rename efi_get_next_variable to efi_get_next_variable_name
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Signed-off-by: Alexander Graf <agraf@suse.de>
efi_loader: always check parameters in efi_cout_query_mode()
If we cannot determine the size of the serial terminal we still have
to check the parameters of efi_cout_query_mode().
Querying the size of the serial terminal drains the keyboard buffer.
So make sure we do this during the initialization and not in the midst
of an EFI application.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Signed-off-by: Alexander Graf <agraf@suse.de>
The current coding advances the column by the number of UTF-8 bytes.
The column should be increased by one per unicode character.
The UEFI spec provides advance rules for U+0000, U+0008, U+000A,
and U000D. All other characters, including control characters
U+0007 (bel) and U+0009 (tab), have to increase the column by one.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Signed-off-by: Alexander Graf <agraf@suse.de>
efi_loader: if EFI_LOADER is enabled, enable CFB_CONSOLE_ANSI
EFI applications like GRUB issue ANSI sequences for setting colors
and for positioning. So if the EFI subsystem is enabled, we should
also enable ANSI control sequences by default.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Signed-off-by: Alexander Graf <agraf@suse.de>
Ivan Gorinov [Wed, 2 May 2018 23:36:02 +0000 (16:36 -0700)]
efi_loader: correctly apply relocations from the .reloc section
Instead of difference between preferred and actual image base, the
actual base is added to the fields specified in the .reloc section.
Use ImageBase from PE optional header to compute the delta,
exit early if the image is loaded at the preferred address.
Signed-off-by: Ivan Gorinov <ivan.gorinov@intel.com> Reviewed-by: Heinrich Schuchardt <xypron.glpk@gmx.de> Signed-off-by: Alexander Graf <agraf@suse.de>
This patch adds support to load secure bitstreams(authenticated or
encrypted or both). As of now, this feature is added and tested only
for xilinx bitstreams and the secure bitstream was generated using
xilinx bootgen tool, but the command is defined in more generic way.
Command example to load authenticated and device key
encrypted bitstream is as follows
"fpga loads 0 100000 2000000 0 1"
Signed-off-by: Siva Durga Prasad Paladugu <siva.durga.paladugu@xilinx.com> Signed-off-by: Michal Simek <michal.simek@xilinx.com>
This patch modifies the arguments parsing code by parsing
based on requested operation for fpga loadfs and then
parses the most common/basic args for other fpga load
commands. This makes it easy for new command extensions
or additions especially the commands with more args.
Signed-off-by: Siva Durga Prasad Paladugu <siva.durga.paladugu@xilinx.com> Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Marek Vasut [Thu, 31 May 2018 16:30:17 +0000 (18:30 +0200)]
ARM: dts: rmobile: Sync R8A77990 Ebisu DTS with Linux
Import the R8A77990 and Ebisu DTS from linux-next to get the latest
version. This makes AVB ethernet work in U-Boot since the ethernet
node is now present in DT, as well as GPIOs.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
ARM: dts: rmobile: Add Renesas R8A77990 SoC support
This patch adds basic support for the Renesas R-Car E3 (R8A77990) SoC:
- PSCI
- CPU (single)
- Cache controller
- Main clocks and controller
- Interrupt controller
- Timer
- PMU
- Reset controller
- Product register
- System controller
- UART for console
Marek Vasut [Thu, 31 May 2018 17:06:02 +0000 (19:06 +0200)]
clk: renesas: Pass clock rate around as 64bit number internally
The PLL rate could be in the GHz range, which could overflow a 32bit
data type. Since the hardware is 64bit anyway, pass the clock rates
as 64bit number internally to avoid this.
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
Tom Rini [Thu, 31 May 2018 18:17:36 +0000 (14:17 -0400)]
Merge tag 'arc-updates-for-2018.07-rc1' of git://git.denx.de/u-boot-arc
Here we do a couple of minor fixes like:
- Move .ivt section to the very beginning of the image
by default which allows us to use that image put right
at reset vector (usually 0x0)
- Improve relocation fix-up which became required once
we moved .ivt and understood a problem with existing implementation
where we relied on a particular placement of sections.
Now we don't care about placement because we just explicitly
check for .text and in case of ARCompact .ivt sections
- Re-implemnt do_reset() such that it calls reset_cpu() which
could implmented for a particular board
And hte most important part we introduce support for yet another
devboard from Synopsys - EMDK.
Alexey Brodkin [Mon, 28 May 2018 12:27:43 +0000 (15:27 +0300)]
ARC: Add support for EM Development Kit board
Synopsys DesignWare ARC EM Development Kit (ARC EMDK) is
an FPGA-based development platform from Synopsys aimed to speed-up
development of software for ARC EM cores and entire subsystems based on
ARC EM like Data Fusion, Secure and Sensor & Control subsystems.
U-Boot is supposed to be used as a primary bootloader on EMDK allowing
users to easily load and start their application from micro-SD card.
Alexey Brodkin [Tue, 29 May 2018 15:09:55 +0000 (18:09 +0300)]
ARC: Improve relocation fix-ups
We used to have the one and only linker script for all ARC boards
and so we relied on a particular order of symbols there.
Because of that we used __ivt_end as the marker of the end of all the
code which won't be true any longer if we move .ivt section to any other
place. That said we'd better check for each section separately.
A couple of other improvements:
1. There's no point to include the marker of section end in interested
range because its address is beyond the section, i.e. we should
compare with "<" but not "<=".
2. .ivt section for ARCv2 cores is just an array of 32-bit ints and
they are not swapped even on little-endia cores while in case of
ARCompact cores .ivt contains valid code so swapping is required.
3. Just in case add check for ARC600 which is also ARCompact
and its .ivt is normal code.
Alexey Brodkin [Tue, 29 May 2018 17:29:16 +0000 (20:29 +0300)]
ARC: Make sure .ivt section is not purged by garbage collector
Fixes commit fac4790491f6 ("arc: Eliminate unused code and data with GCC's garbage collector"),
see [1].
Since in case of ARCv2 .ivt only contains addrresses of
interrupt/exception handlers linker doesn't understand that this section
actually make a lot of sense and decides to get rid of it if we use
"--gc-sections".
Alexey Brodkin [Fri, 25 May 2018 17:22:23 +0000 (20:22 +0300)]
ARC: Cache: Don't compare I$ and D$ line lengths
We don't care much about I$ line length really as there're
no per-line ops on I$ instead we only do full invalidation of it
on occasion of relocation and right before jumping to the OS.
Also as compared to Linux kernel where we don't support different
lengths of I$ and D$ lines in U-Boot we have to deal with such an
exotic configs if the target board is not supposed to run Linux kernel.
Tom Rini [Fri, 25 May 2018 12:28:45 +0000 (08:28 -0400)]
tests: Make test_log.py tests depend on cmd_log not log
While the tests in this testcase are for the log subsystem they are only
able to be run if CONFIG_CMD_LOG is enabled as well as CONFIG_LOG, so
update the buildconfigspec requirement.
Eugeniu Rosca [Sat, 19 May 2018 12:13:55 +0000 (14:13 +0200)]
test: dm: Fix wrong aliases property names
After importing v4.17-rc1 Linux commit 9130ba884640 ("scripts/dtc:
Update to upstream version v1.4.6-9-gaadd0b65c987"), sandbox build
reports below warnings:
arch/sandbox/dts/test.dtb: Warning (alias_paths): /aliases: aliases property name must include only lowercase and '-'
arch/sandbox/dts/test.dtb: Warning (alias_paths): /aliases: aliases property name must include only lowercase and '-'
arch/sandbox/dts/test.dtb: Warning (alias_paths): /aliases: aliases property name must include only lowercase and '-'
arch/sandbox/dts/test.dtb: Warning (alias_paths): /aliases: aliases property name must include only lowercase and '-'
Silent them by applying the 's/_/-/' substitution in the names of the
'fdt_dummy0', 'fdt_dummy1', 'fdt_dummy2', 'fdt_dummy3' properties.
Similar DTC warnings have been recently fixed in Linux kernel, e.g. via
v4.17-rc1 commit d366c30d19f4 ("ARM: dts: STi: Fix aliases property name
for STi boards").
If done alone, the DTS update generates a failure of the
`ut dm fdt_translation` unit test in sandbox environment as seen below:
Fix this issue in place, by updating the "name" string in the
UCLASS_DRIVER(fdt_dummy) definition, so that it matches the newly
updated aliases properties. After that, the test passes:
Eugeniu Rosca [Sat, 19 May 2018 12:13:54 +0000 (14:13 +0200)]
scripts/dtc: Re-sync with Linux 4.17-rc4
To achieve a DTC state (more or less) equivalent to Linux 4.17-rc4,
backport the DTC-specific part from below Linux commits:
b23d1a241f4eb4 ("kbuild: add %.lex.c and %.tab.[ch] to 'targets' automatically") 9a8dfb394c0467 ("kbuild: clean up *.lex.c and *.tab.[ch] patterns from top-level Makefile") 59889300274569 (".gitignore: move *.lex.c *.tab.[ch] patterns to the top-level .gitignore") e039139be8c251 ("scripts/dtc: generate lexer and parser during build instead of shipping") b24413180f5600 ("License cleanup: add SPDX GPL-2.0 license identifier to files with no license")
With this commit, the diff between Linux and U-boot common DTC files is:
$ git diff --stat --diff-filter=M v4.17-rc4 <this-commit> -- scripts/dtc
scripts/dtc/.gitignore | 2 +-
scripts/dtc/Makefile | 3 +++
The delta is coming from U-boot commits:
* v2017.11 de163ecedb8a ("scripts: dtc: Add .gitignore")
* v2018.01 15b97f5c5e6d ("pylibfdt: move pylibfdt to scripts/dtc/pylibfdt and refactor makefile")
aadd0b65c987 checks: centralize printing of property names in failure messages 88960e398907 checks: centralize printing of node path in check_msg f1879e1a50eb Add limited read-only support for older (V2 and V3) device tree to libfdt. 37dea76e9700 srcpos: drop special handling of tab 65893da4aee0 libfdt: overlay: Add missing license 962a45ca034d Avoid installing pylibfdt when dependencies are missing cd6ea1b2bea6 Makefile: Split INSTALL out into INSTALL_{PROGRAM,LIB,DATA,SCRIPT} 51b3a16338df Makefile.tests: Add LIBDL make(1) variable for portability sake 333d533a8f4d Attempt to auto-detect stat(1) being used if not given proper invocation e54388015af1 dtc: Bump version to v1.4.6 a1fe86f380cb fdtoverlay: Switch from using alloca to malloc c8d5472de3ff tests: Improve compatibility with other platforms c81d389a10cc checks: add chosen node checks e671852042a7 checks: add aliases node checks d0c44ebe3f42 checks: check for #{size,address}-cells without child nodes 18a3d84bb802 checks: add string list check for *-names properties 8fe94fd6f19f checks: add string list check 6c5730819604 checks: add a string check for 'label' property a384191eba09 checks: fix sound-dai phandle with arg property check b260c4f610c0 Fix ambiguous grammar for devicetree rule fe667e382bac tests: Add some basic tests for the pci_bridge checks 7975f6422260 Fix widespread incorrect use of strneq(), replace with new strprefixeq() fca296445eab Add strstarts() helper function cc392f089007 tests: Check non-matching cases for fdt_node_check_compatible() bba26a5291c8 livetree: avoid assertion of orphan phandles with overlays c8f8194d76cc implement strnlen for systems that need it c8b38f65fdec libfdt: Remove leading underscores from identifiers 3b62fdaebfe5 Remove leading underscores from identifiers 2d45d1c5c65e Replace FDT_VERSION() with stringify() 2e6fe5a107b5 Fix some errors in comments b0ae9e4b0ceb tests: Correct warning in sw_tree1.c
Commit c8b38f65fdec upstream ("libfdt: Remove leading underscores from
identifiers") changed the multiple inclusion define protection, so the
kernel's libfdt_env.h needs the corresponding update.
Signed-off-by: Rob Herring <robh@kernel.org>
[ Linux commit: 9130ba884640328bb78aaa4840e5ddf06ccafb1c ]
[erosca: - Fixup conflicts in include/linux/libfdt_env.h caused by v2018.03-rc4
commit b08c8c487083 ("libfdt: move headers to <linux/libfdt.h>
and <linux/libfdt_env.h>")
- Fix build errors in lib/libfdt/fdt_ro.c, tools/libfdt/fdt_rw.c by:
- s/_fdt_mem_rsv/fdt_mem_rsv_/
- s/_fdt_offset_ptr/fdt_offset_ptr_/
- s/_fdt_check_node_offset/fdt_check_node_offset_/
- s/_fdt_check_prop_offset/fdt_check_prop_offset_/
- s/_fdt_find_add_string/fdt_find_add_string_/] Signed-off-by: Eugeniu Rosca <erosca@de.adit-jv.com> Reviewed-by: Simon Glass <sjg@chromium.org> Reviewed-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Eugeniu Rosca [Sat, 19 May 2018 12:13:51 +0000 (14:13 +0200)]
board: eets: pdu001: Fix wrong default value in Kconfig
After importing linux v4.16-rc1 commit 2c37e08464a8 ("kconfig: Warn if
choice default is not in choice"), Kconfig complains:
scripts/kconfig/conf --syncconfig Kconfig
board/eets/pdu001/Kconfig:22:warning: \
choice default symbol 'PDU001_RUN_LED_RED' \
is not contained in the choice
This looks to be caused by a typo. Fix it.
Fixes: 85ab0452fefc ("arm: add support for PDU001") Signed-off-by: Eugeniu Rosca <erosca@de.adit-jv.com> Reviewed-by: Masahiro Yamada <yamada.masahiro@socionext.com> Reviewed-by: Felix Brack <fb@ltec.ch> Tested-by: Felix Brack <fb@ltec.ch>
Eugeniu Rosca [Sat, 19 May 2018 12:13:50 +0000 (14:13 +0200)]
kconfig: re-sync with Linux 4.17-rc4
Align Kconfig to Linux 4.17-rc4 with minimal impact on non-kconfig files.
Previous Kconfig sync was done by commit bf7ab1e70fd762 ("kconfig:
re-sync with Linux 4.10") and it achieved almost perfect alignment with
a few (intended) exceptions, caused by below U-boot commits:
Here is the list of Kconfig commits which followed the v4.10 alignment:
[F] v2018.01 0931ed3c0d0f ("kconfig/symbol.c: use correct pointer type argument for sizeof")
[G] v2018.03 1414e09b4f25 ("kconfig: revert change that was not needed for -Wformat-security")
[H] v2018.05 83d290c56fab ("SPDX: Convert all of our single license tags to Linux Kernel style")
Commit [F] was subsequently applied to Linux kernel as commit [I]
with the same patch id, so it won't contribute to further misalignment.
[I] v4.15-rc1 88127dae6ed9 ("kconfig/symbol.c: use correct pointer type argument for sizeof")
Commit [G] is a Kconfig-specific revert of commit [E].
Commit [H] relocated and reformatted the license doing no functional change.
In summary, the only functional change that makes U-boot Kconfig
diverge from Linux Kconfig is commit [B]. After a brief analysis,
the purpose of [B] seems to be placing "\n" literals in string symbols
like CONFIG_AUTOBOOT_PROMPT="autoboot in %d seconds\n" in order to pass
them directly to printf and expect correct output at runtime. Currently,
Linux doesn't seem to have this requirement, so for the moment [B] looks
like a U-boot specific feature/fix. From point of view of further Kconfig
alignment and backporting efforts, it is highly desired that commits
like [B] are propagated to Linux and any Kconfig fixes/features are
contributed to Linux kernel first. This specific Kconfig re-sync just
keeps [B] in place.
Contrary to 4.10 Kconfig re-sync (which achieves zero non-kconfig
changes), 4.17-rc4 re-sync does some amount of updates in Kbuild
(striving to keep them at minimum), due to a number of reasons:
* Kbuild is affected by the removal of Kconfig "*shipped" files and now
requires flex and bison pre-installed on the host.
* PYTHON{2,3} variables are defined in top-level Makefile as
prerequisite for running the newly developed Kconfig unit tests.
* silentoldconfig becomes an "internal implementation detail" deprecated
for external use, being renamed to syncconfig.
The exact non-kconfig files touched by this commit are:
The imported Linux commits touching the above files are:
c054be10ffdbd5 ("remove gperf left-overs from build system") 73a4f6dbe70a1b ("kbuild: add LEX and YACC variables") 033dba2ec06c47 ("kbuild: prepare to remove C files pre-generated by flex and bison") eea199b445f64c ("kbuild: remove unnecessary LEX_PREFIX and YACC_PREFIX") e71de5ee08dcb0 ("kbuild: remove remaining use of undefined YACC_PREFIX") d59fbbd09d5d6b ("kbuild: replace hardcoded bison in cmd_bison_h with $(YACC)") 911a91c39cabcb H ("kconfig: rename silentoldconfig to syncconfig") 59889300274569 (".gitignore: move *.lex.c *.tab.[ch] patterns to the top-level .gitignore") 9a8dfb394c0467 ("kbuild: clean up *.lex.c and *.tab.[ch] patterns from top-level Makefile") 833e622459432e H ("genksyms: generate lexer and parser during build instead of shipping") b23d1a241f4eb4 H ("kbuild: add %.lex.c and %.tab.[ch] to 'targets' automatically") e9781b52d4e0e3 H ("kbuild: add PYTHON2 and PYTHON3 variables")
The commits marked with 'H' are assessed as "hard" (build will fail)
prerequisites and the rest of them are assessed as "soft" prerequisites
for the re-sync. In spite of relatively high number of non-H commits,
they belong to this Kconfig update topic-wise and decrease the number of
cherry pick conflicts for many commits in this series. Additional effort
can be put in eliminating the soft prerequisites, if really needed.
The commits which contributed to this Kconfig re-sync are listed below.
Whenever a conflict resolution has been performed (mostly by hand, but
sometimes automatically by git), it is revealed by the '!' sign in the
second column, which means a patch id mismatch between Linux and U-boot
commits:
9be3213b14d44f ("gconfig: remove misleading parentheses around a condition") ff85a1a80e0034 ("kconfig: Check for libncurses before menuconfig") ad8181060788c8 ("kconfig: fix sparse warnings in nconfig") cb77f0d623ff33 ! ("scripts: Switch to more portable Perl shebang") bb3290d91695bb ! ("Remove gperf usage from toolchain") c054be10ffdbd5 ("remove gperf left-overs from build system") b24413180f5600 ! ("License cleanup: add SPDX GPL-2.0 license identifier to files with no license") 9059a3493efea6 ! ("kconfig: fix relational operators for bool and tristate symbols") 2c37e08464a850 ("kconfig: Warn if choice default is not in choice") 33ca1a24866373 ("kconfig: Document the 'menu' struct") 52aede4ba5efd1 ("kconfig: Document the 'symbol' struct") c873443430ebd1 ("kconfig: Sync zconf.y with zconf.tab.c_shipped") 9a826842ff2fbd ("kconfig: Rename menu_check_dep() to rewrite_m()") fa8cedaef814ce ("kconfig: Clarify expression rewriting") f77850d3fe0c96 ("kconfig: Clean up modules handling and fix crash") e3b03bf29d6b99 ("kconfig: display recursive dependency resolution hint just once") 73a4f6dbe70a1b ! ("kbuild: add LEX and YACC variables") 033dba2ec06c47 ! ("kbuild: prepare to remove C files pre-generated by flex and bison") 29c833061c1d8c ("kconfig: generate lexer and parser during build instead of shipping") 26e47a3c11a25c ("kconfig: Don't leak symbol names during parsing") 24161a6711c945 ("kconfig: Don't leak 'source' filenames during parsing") bc28fe1d5ede88 ("kconfig: Don't leak 'option' arguments during parsing") 0724a7c32a54e3 ("kconfig: Don't leak main menus during parsing") ae7440ef0c8013 ("kconfig: Fix automatic menu creation mem leak") 5b1374b3b3c2fc ("kconfig: Fix expr_free() E_NOT leak") 7cf33f88e29410 ("kconfig: Fix choice symbol expression leak") 05cccce580456d ("kconfig: Document automatic submenu creation code") 0735f7e5def2ab ("kconfig: Document important expression functions") df60f4b92d3d0b ("kconfig: Remove menu_end_entry()") b92d804a51796b ("kconfig: drop 'boolean' keyword") 6479f327dea60d ("kconfig: Warn if there is more than one help text") 52e58a3caeba5d ("kconfig: make input_mode static") 5a3dc717b3c785 ("kconfig: make xfgets() really static") 84dd95d4f87a0d ("kconfig: make conf_unsaved a local variable of conf_read()") 765f4cdef6f80d ("kconfig: use default 'yy' prefix for lexer and parser") eea199b445f64c ("kbuild: remove unnecessary LEX_PREFIX and YACC_PREFIX") e71de5ee08dcb0 ("kbuild: remove remaining use of undefined YACC_PREFIX") d59fbbd09d5d6b ! ("kbuild: replace hardcoded bison in cmd_bison_h with $(YACC)") 3e41ba05b6d60c ("kconfig: Document SYMBOL_OPTIONAL logic") d3465af60f4471 ("kconfig: Clarify choice dependency propagation") 9d1a9e8bc18bea ("kconfig: Document 'if' flattening logic") b53688014e3325 ("kconfig: Clarify menu and 'if' dependency propagation") d0fd0428ecf04b ("kconfig: fix make xconfig when gettext is missing") 312ee68752faaa ("kconfig: announce removal of oldnoconfig if used") 1ccb27143360bd ("kconfig: make "Selected by:" and "Implied by:" readable") cedd55d49dee94 ! ("kconfig: Remove silentoldconfig from help and docs; fix kconfig/conf's help") 1b9eda2e4892cb ("kconfig: Warn if help text is blank") cb67ab2cd2b8ab ("kconfig: do not write choice values when their dependency becomes n") 4f208f392103e8 ("kconfig: show '?' prompt even if no help text is available") cd58a91def2acc ("kconfig: remove 'config*' pattern from .gitignnore") d2a04648a5dbc3 ("kconfig: remove check_stdin()") f3ff6fb5db68bc ("kconfig: echo stdin to stdout if either is redirected") 9e3e10c725360b ("kconfig: send error messages to stderr") d717f24d8c6808 ("kconfig: add xrealloc() helper") 523ca58b7db2e3 ("kconfig: remove const qualifier from sym_expand_string_value()") cd81fc82b93fa4 ("kconfig: add xstrdup() helper") f4bc1eefc1608e ("kconfig: set SYMBOL_AUTO to the symbol marked with defconfig_list") bf0bbdcf100322 ("kconfig: Don't leak choice names during parsing") 1a90ce36c6eff6 ("kconfig: Update ncurses package names for menuconfig") 5ae6fcc4bb82bd ("kconfig: fix line number in recursive inclusion error message") 07a422bb213adb ! ("kbuild: restore autoksyms.h touch to the top Makefile") 9a47ceec543bfb ("kconfig: clean-up reverse dependency help implementation") d9119b5925a03b ("kconfig: Print reverse dependencies in groups") f467c5640c29ad ("kconfig: only write '# CONFIG_FOO is not set' for visible symbols") 59a80b5e892dde ("kconfig: do not call check_conf() for olddefconfig") 4bb3a5b085cd6f ("kconfig: remove unneeded input_mode test in conf()") 99f0b6578bab44 ("kconfig: remove redundant input_mode test for check_conf() loop") 2aad9b89621386 ("kconfig: hide irrelevant sub-menus for oldconfig") 81d2bc2273052e ("kconfig: invoke oldconfig instead of silentoldconfig from local*config") 911a91c39cabcb ! ("kconfig: rename silentoldconfig to syncconfig") 2a61625835c7c8 ! ("kconfig: remove redundant streamline_config.pl prerequisite") 022a4bf6b59dfd ("kconfig: tests: add framework for Kconfig unit testing") 1903c511905984 ("kconfig: tests: add basic choice tests") 49ac3c0c3aa3b7 ("kconfig: tests: test automatic submenu creation") b76960c0f6b25d ("kconfig: tests: test if new symbols in choice are asked") 930c429a656fdb ("kconfig: tests: check unneeded "is not set" with unmet dependency") ee236610653ede ("kconfig: tests: check visibility of tristate choice values in y choice") beaaddb625400e ("kconfig: tests: test defconfig when two choices interact") 3e4888c2e3d77d ("kconfig: tests: test randconfig for choice in choice") 29c434f367ea7b ("kconfig: tests: test if recursive dependencies are detected") e2c75e7667c737 ("kconfig: tests: test if recursive inclusion is detected") f622f827958162 ("kconfig: warn unmet direct dependency of tristate symbols selected by y") f8f69dc0b4e070 ("kconfig: make unmet dependency warnings readable") 26561514cc9def ("kconfig: do not include both curses.h and ncurses.h for nconfig") 32a94b8b0c3e5a ("kconfig: remove duplicated file name and lineno of recursive inclusion") 379a8eb8eb1a55 ("kconfig: detect recursive inclusion earlier") 18492685e479fd ("kconfig: use yylineno option instead of manual lineno increments") 59889300274569 ! (".gitignore: move *.lex.c *.tab.[ch] patterns to the top-level .gitignore") 9a8dfb394c0467 ! ("kbuild: clean up *.lex.c and *.tab.[ch] patterns from top-level Makefile") 833e622459432e ! ("genksyms: generate lexer and parser during build instead of shipping") b23d1a241f4eb4 ! ("kbuild: add %.lex.c and %.tab.[ch] to 'targets' automatically") 17baab68d337a0 ("kconfig: extend output of 'listnewconfig'") e9781b52d4e0e3 ! ("kbuild: add PYTHON2 and PYTHON3 variables")
This seems to happen because the Kbuild updates apparently didn't make
room for both "*shipped"-based builds and flex/bison-based builds. A
similar problem has been reported for genksyms parser in v4.17-rc1
commit 833e622459432e ("genksyms: generate lexer and parser during build
instead of shipping"). I have figured out empirically that the warnings
are healed after updating the in-tree U-boot DTC to upstream v1.4.6-9,
same as done by Linux v4.17-rc1 commit 9130ba88464032 ("scripts/dtc:
Update to upstream version v1.4.6-9-gaadd0b65c987"). Whether fixing the
DTC-related yacc warnings should be done together with the Kconfig
re-sync, I would like to hear from community.
My testing was limited to:
- make defconfig all
- make ARCH=arm CROSS_COMPILE=aarch64-linux-gnu- r8a7795_ulcb_defconfig all
- comparing .config before and after the re-sync
- running the newly imported Kconfig unit tests as seen below:
projects / u-boot / log