Allows boards to supply their own gnu-elf linker script, used when building Nuttx applications and CONFIG_BUILD_KERNEL.
This is useful when building fully-linked applications (see #9395).
This patch add an example and pyhton base modules:
1.add memdump.py is an example to analyze memory
usage by python script.
2.add The most basic data structure analysis, like list, etc.
future ideas:
Maybe we can add modules related to "sched, drivers, arch, fs.." to
automatically analyze scripts to debug some problems
References:
linux kernel (https://github.com/torvalds/linux/tree/master/scripts/gdb)
The official manual of gdb (https://sourceware.org/gdb/onlinedocs/gdb/Python-API.html)
Change-Id: Ib9025a0a141cb89f3813526f7c55dcb28de31ed9
Signed-off-by: anjiahao <anjiahao@xiaomi.com>
This PR is a modification that optimizes priority inheritance
for only one holder. After the above modifications are completed,
the mutex lock->unlock process that supports priority inheritance
can be optimized by 200 cycles.
Before modify: 2000 cycle
After modify: 1742 cycle
Signed-off-by: zhangyuan21 <zhangyuan21@xiaomi.com>
This commit aims to provide a way to get the current applied defconfig on runtime by providing CONFIG_BASE_DEFCONFIG that is updated when the configure and build scripts are run.
The "CONFIG_APPS_DIR" generated in .config has double quotes and is not recognized as a path by most make systems. This commits removes these double quotes to make the compile successfully.
The offset of the relevant registers in xcp will change after
enabling the FPU, this PR will add fpu offset correct the register offset
Signed-off-by: chao an <anchao@xiaomi.com>
Currently when we modify some Kconfig related to some chip or
some architecture we need to run refresh.sh to all boards that
is too slow (more than one hour to finish).
This patch modify the script to update only the boards configs
of an specific chip or only the boards of an specific architecture.
Examples:
$ ./tools/refresh.sh --silent --defaults arch:renesas
Normalize us7032evb1/ostest
Normalize us7032evb1/nsh
Normalize skp16c26/ostest
Normalize rx65n-rsk2mb/netnsh
Normalize rx65n-rsk2mb/ipv6
Normalize rx65n-rsk2mb/nsh
Normalize rx65n-rsk1mb/netnsh
Normalize rx65n-rsk1mb/nsh
Normalize rx65n-grrose/netnsh
Normalize rx65n-grrose/ipv6
Normalize rx65n-grrose/nsh
Normalize rx65n/nsh
It updates the boards from all 'renesas' architecture (m9c, rx65n, sh1)
$ ./tools/refresh.sh --silent --defaults chip:rx65n
Normalize rx65n-rsk2mb/netnsh
Normalize rx65n-rsk2mb/ipv6
Normalize rx65n-rsk2mb/nsh
Normalize rx65n-rsk1mb/netnsh
Normalize rx65n-rsk1mb/nsh
Normalize rx65n-grrose/netnsh
Normalize rx65n-grrose/ipv6
Normalize rx65n-grrose/nsh
Normalize rx65n/nsh
It updates all the boards from renesas chip 'rx65n' family
The syntax check of kconfiglib is stricter than kconfig-frontends,
but the warnings of kconfiglib are not catched by Makefile. Since kconfiglib
is implemented by python, the return value of the shell is always zero($?),
In this PR, I redirected the relevant standard errors to the file to check
whether the kconfig warning is occurred
Signed-off-by: chao an <anchao@xiaomi.com>
The order of the registers was changed by the following commit:
---------------------------------
| armv7-a/r: modify regs index to improve exception performance
|
| Signed-off-by: zhangyuan21 <zhangyuan21@xiaomi.com>
Signed-off-by: chao an <anchao@xiaomi.com>
Some time nuttx will receive SIGSEGV while executing the function:
| Thread 1 "nuttx" received signal SIGSEGV, Segmentation fault.
| 0xf7bab4a0 in __sanitizer::common_flags_dont_use () from /lib/i386-linux-gnu/libasan.so.6
| The program being debugged was signaled while in a function called from GDB.
| GDB has restored the context to what it was before the call.
| Evaluation of the expression containing the function
| (up_check_tcbstack) will be abandoned.
Signed-off-by: chao an <anchao@xiaomi.com>
Fix:
```
mksymtab.c: In function ‘main’:
mksymtab.c:280:15: warning: the comparison will always evaluate as ‘true’ for the address of ‘g_parm’ will never be NULL [-Waddress]
280 | cond = (g_parm[COND_INDEX] && strlen(g_parm[COND_INDEX]) > 0);
```
Signed-off-by: Huang Qi <huangqi3@xiaomi.com>
Add header keywords into whitelist, skip the warning if the
particular headers will contain the functions with mixed case
Signed-off-by: chao an <anchao@xiaomi.com>
Current Toolchain.defs set the compile flags directly, it's OK for
target specified gcc toolchain.
But some LLVM based toolchains (Rust/Zig etc) use single toolchain to handle all supported paltform.
In this patch, arch level Toolchain.defs export standard LLVM style arch flags, and let <Lang>.defs to map them into internal style,
This will simplify the intergration of non-c language.
Signed-off-by: Huang Qi <huangqi3@xiaomi.com>
Some targets, such as qemu-rv support multiple architectures. The required flags need to passed to the linker when built "out-of-tree" in the application directory.
Initially supporting ESP32-C3 chip, to be followed by other RISC-V-based
chips from Espressif.
Signed-off-by: Gustavo Henrique Nihei <gustavo.nihei@espressif.com>
When using stm32, the starting address of the function parsed by mkallsyms.py is an odd number, one large than the actual address
Signed-off-by: yinshengkai <yinshengkai@xiaomi.com>
Purge warning:
Kconfig:249: warning: the 'modules' option is not supported.
Let me know if this is a problem for you, as it wouldn't be that
hard to implement. Note that modules are supported -- Kconfiglib
just assumes the symbol name MODULES, like older versions of the
C implementation did when 'option modules' wasn't used.
Signed-off-by: chao an <anchao@xiaomi.com>
First decoupling changes related to CMAKE
BINDIR/APPSBINDIR:
Output path of Kconfig which dynamically generated by NuttX Kernel/Apps
This option is consistent with the TOPDIR/APPSDIR by default, and will
be changed when out-of-tree compilation is supported
Signed-off-by: chao an <anchao@xiaomi.com>
Summary:
- I noticed that nuttx crashes if DEBUG_ASSERTIONS=y
- This commit fixes this issue by changing QEMU options
Impact:
- None
Testing:
- Tested with QEMU
Signed-off-by: Masayuki Ishikawa <Masayuki.Ishikawa@jp.sony.com>
Summary:
- I noticed that the following build error happened on my machine.
make -C sched libsched.a EXTRAFLAGS="-D__KERNEL__ "
make[1]: Entering directory '/mnt/m2ssd/opensource/RTOS/tmp/nuttx/sched'
/bin/sh: 1: echoCC: clock/clock_initialize.c : not found
- This commit fixes this issue
Impact:
- Should be none
Testing:
- Build on ubuntu 18.04 x86_64
Signed-off-by: Masayuki Ishikawa <Masayuki.Ishikawa@jp.sony.com>
Using the full path will make __FILE__ longer, which will lead to a larger flash footprint
revert 9c6c9fb0cd && e14309a92d
Signed-off-by: yinshengkai <yinshengkai@xiaomi.com>
In order to make compilation warnings and errors easier to be found out,
this commit will disable the printing of the compilation process as much
as possible, and also if you want to restore the log information of the
compilation process, please enable verbose build on command line:
$ make V=0
OR
$ make V=1
| V=0: Exit silent mode
| V=1,2: Enable echo of commands
| V=2: Enable bug/verbose options in tools and scripts
Signed-off-by: chao an <anchao@xiaomi.com>
1. Change IP address format to addr/prefix, to be compatible with both IPv4/IPv6.
- When adding address in CIDR type, netmask/route will be automatically added.
2. Since route of whole subnet is added automatically, not specifying NuttX's IP any more.
- Multiple NuttX simulators (with IP 10.0.1.x) attached to same bridge can surf the net at same time.
3. NAT66 is used to make sure it works even if host has only one IPv6 address.
Signed-off-by: Zhe Weng <wengzhe@xiaomi.com>
We found that some people don't know / forget to enable forward in linux, which keeps nat not working.
Because we don't know the previous state before 'on', and forward seldom has side effect, we don't disable it in 'off'.
Signed-off-by: Zhe Weng <wengzhe@xiaomi.com>
Config which enables nxcamera app and video framework.
Exclude from CI macOS test because macOS does not have V4L2.
(Maybe consider adding it back after porting with avfoundation)
Signed-off-by: Peter Bee <bijunda1@xiaomi.com>
to avoid the below error:
destination path '/Users/runner/work/nuttx/nuttx/sources/tools/bloaty-src' already exists and is not an empty directory.
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
NuttX has graduated the incubator phase and the releases should be renamed
to match the current status
Signed-off-by: Alin Jerpelea <alin.jerpelea@sony.com>
to fix the below error:
x86_64-elf-gcc --version
Copyright (C) 2002-2007 Andrew Tridgell
ccache: error: Could not find compiler "x86_64-elf-gcc" in PATH
Copyright (C) 2009-2022 Joel Rosdahl and other contributors
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
fix the error message when executing make -C nuttx/tools -f Makefile.host clean
/bin/sh: 1: nuttx/tools/../tools/incdir: not found
Signed-off-by: yinshengkai <yinshengkai@xiaomi.com>
In the past, header file paths were generated by the incdir command
Now they are generated by concatenating environment variables
In this way, when executing makefile, no shell command will be executed,
it will improve the speed of executing makfile
Signed-off-by: yinshengkai <yinshengkai@xiaomi.com>
In the past, predefined macros were generated by define.sh scripts
Now they are generated by concatenating environment variables
In this way, when executing makefile, no shell command will be executed,
it will improve the speed of executing makfile
Signed-off-by: yinshengkai <yinshengkai@xiaomi.com>
Clang relies on clang-extdef-mapping for ctu checking. But clang-arm doesn't provide this file
A tweaking way to support this feature is copy a `clang-extdef-mapping` from other clang version.
Signed-off-by: xinbingnan <xinbingnan@xiaomi.com>
fix the following error:
/tools/clang-arm-none-eabi/bin/clang: error while loading shared libraries: libtinfo.so.5: cannot open shared object file: No such file or directory
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
fix the following warning:
clang: error while loading shared libraries: libtinfo.so.5: cannot open shared object file: No such file or directory
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
This change adds the following:
- Rename the board configuration name from qemu-a53 to qemu-v8a.
- Add the configurations for Cortex-A57 and Cortex-A72.
Signed-off-by: Hidenori Matsubayashi <hidenori.matsubayashi@gmail.com>
.github/workflows/build.yml: added CodeChecker support for GitHub Workflow
tools/testbuild.sh: added support for CodeChecker checks
- Added support for CodeChecker checks.
- Generate inspection reports and summaries.
- After the task is executed, the logs are compressed and the database is packaged.
Signed-off-by: xinbingnan <xinbingnan@xiaomi.com>
When compiling my project, the time to generate allsyms decreased from tens of seconds to 2 seconds
Signed-off-by: yinshengkai <yinshengkai@xiaomi.com>
* tools/configure.sh:
(USAGE): Explain both styles of usage:
- <board-name>:<config-name> for in-tree boards
- path to custom out-of-tree boards (relative to TOPDIR
or absolute)
Suggested by TimH in mailing list discussion "build board from custom
directory" started 15 Sept 2022, archived:
https://lists.apache.org/thread/7t8k79mm4kxy9cbo7vmybd36nzh94qtd
In order to better test the lcdddev driver and framebuffer, newer
board configs (for the TTGO T-Display and for the simulator) were
added.
Adjusted references of the sim:lvgl_(fb/lcd) config.
It is not really needed; g_hart_stacks is only used during SBI init as
a temporary stack area. We can use the scratch area buffers for this, as
the scratch areas define almost 4K of extra space, which is used for
exception stacks anyway.
Summary:
- This commit adds exportandimport func for CONFIG_BUILD_KERNEL=y
Impact:
- ci with CONFIG_BUILD_KERNEL=y only
Testing:
- Tested with sabre-6quad:netknsh (will be updated later)
Signed-off-by: Masayuki Ishikawa <Masayuki.Ishikawa@jp.sony.com>
Summary:
- This commit generates .tar.gz file with make export
Impact:
- make export file format
Testing:
- Tested with sabre-6quad:netknsh
Signed-off-by: Masayuki Ishikawa <Masayuki.Ishikawa@jp.sony.com>
N/A
Summary:
Arm64 support for NuttX, Features supported:
1. Cortex-a53 single core and SMP support: it's can run into nsh shell at
qemu virt machine.
2. qemu-a53 board configuration support: it's only for evaluate propose
3. FPU support for armv8-a: FPU context switching at NEON/floating-point
TRAP is supported.
4. psci interface, armv8 cache operation(data cache) and smccc support.
5. fix mass code style issue, thank for @xiaoxiang781216, @hartmannathan @pkarashchenko
Please refer to boards/arm64/qemu/qemu-a53/README.txt for detail
Note:
1. GCC MACOS issue
The GCC 11.2 toolchain for MACOS may get crash while compiling
float operation function, the following link describe the issue
and give analyse at the issue:
https://bugs.linaro.org/show_bug.cgi?id=5825
it's seem GCC give a wrong instruction at certain machine which
without architecture features
the new toolchain is not available still, so just disable the MACOS
cibuild check at present
Signed-off-by: qinwei1 <qinwei1@xiaomi.com>
Targets build during the kernel phase did not have their dependencies
specified and thus they were not rebuilt after their dependencies have
changed, for example by changing options in menuconfig.
Signed-off-by: Abdelatif Guettouche <abdelatif.guettouche@espressif.com>
This reverts commit 848d1ef0b7.
It doesn't make sense because the generated file just have names of
symbols as strings. Nothing to worry about builtins.
Follow: http://glennastory.net/boot/sysinit.html
This is first script that init runs is rc.sysinit. This
script does serval initialization tasks about basic service.
The boot sequence currently provided to the board level is:
board_earlyinitialize->
board_lateinitialize(Peripherals driver, core driver, ...)->
run rcS script(mount fs, run service) ->
board_appinitialize->
After this patch:
The boot sequence currently provided to the board level is:
board_earlyinitialize->
board_lateinitialize(core driver,...)->
run rc.sysinit script(mount fs, run core service) ->
board_appinitialize(Peripherals driver)->
run rcS script(run other service)->
So, Peripheral drivers can do more with the file system and
core services.
Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
follow up the bellow chnage:
commit 0f2f48f8ba
Author: Xiang Xiao <xiaoxiang@xiaomi.com>
Date: Sun Mar 20 18:12:26 2022 +0800
sys/type.h: Change pid_t from int16_t to int
to fix the following warning:
include/unistd.h:302:9: error: incompatible redeclaration of library function 'vfork' [-Werror,-Wincompatible-library-redeclaration]
pid_t vfork(void);
^
include/unistd.h:302:9: note: 'vfork' is a builtin with type 'int (void)'
and change 32768 to INT_MAX to match the type change
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
This script will help you analyze memdump log files,
analyze the number of occurrences of backtrace,
and output stack information
memdump log files need this format:
pid size addr mem
Signed-off-by: anjiahao <anjiahao@xiaomi.com>
Signed-off-by: Jiuzhu Dong <dongjiuzhu1@xiaomi.com>
I plan to use this to run sim:ostest (and probably other tests later)
on the CI.
The script indirection might allow future non-sim usage as well.
(like running it with qemu, or even on the real hardware.)
I have no plan to do it by myself right now though.
## Summary
A lot of linker scripts were listed twice, once for unix, once for windows.
This PR cleans up the logic so they're only listed once.
## Impact
Any opportunity to use a single source of truth and reduce lines of code is a win!
## Testing
CI will test all build
If attaching to a target that is already running JLinkGDBServer calls
RTOS_GetNumThreads() without a prior call to RTOS_UpdateThreads(). So
do this within RTOS_GetNumThreads() if g_plugin_priv.ntcb has not yet
been initialized.
Note: If after attaching the debugger to the target, the target is
resumed and then stopped again, the RTOS_UpdateThreads is actually
called. Thus, we are not running on stale thread data in this case.
I also changed PLUGIN_VER to API_VER and its value to 101, as
RTOS_GetVersion() does not query the version of the plugin, but the API
version implemented by the plugin, which in our case is 1.1.
Signed-off-by: Michael Jung <mijung@gmx.net>
The packed-attribute on the tcb_info_s type was misplaced, which caused
incompatible memory layout between host and target. According to
current GCC documentation:
> You may specify type attributes in an enum, struct or union type
> declaration or definition by placing them immediately after the struct,
> union or enum keyword. You can also place them just past the closing
> curly brace of the definition, but this is less preferred because
> logically the type should be fully defined at the closing brace.
I also added jlink-nuttx.so to the .gitignore list and updated nxstyle
to ignore the camel case function names required by JLinkGDBServer.
Signed-off-by: Michael Jung <mijung@gmx.net>
This fixes error:
tools/Unix.mk:49: *** missing separator. Stop.
that can happen if version.sh is invoked on a git repository
that contains no tags.
See https://github.com/apache/incubator-nuttx/issues/5324
Signed-off-by: Juha Niskanen <juha.niskanen@haltian.com>
This PR renames `tools/Makefile.*` to have the `.mk` extension. This PR also updates `README.md` and other files that references the other files.
Note: Skipped Makefile.host for this PR since it caused failures in CI for the sim build
By using a standard extension for Makefiles (https://www.file-extension.info/format/mk), editors will auto-format files.
This change will also improve developer ergonomics when searching for specific files
Verified locally that build still works, CI will verify more!
`include/nuttx/version.h` is dependent on `.version`, which needs to run for every invocation of make, since git info can update at any time.
Currently the `.version` target is PHONY, thus making `include/nuttx/version.h` and `context` targets (and every target that depends on `context`) have to run every time.
This PR modifies `.version` so it's a real rule. Additionally, `tools/version.sh` is run every invocation of make. This maintains the current behavior while preventing some targets from needlessly running every time
Polarfire Icicle board has only (128K - 256) bytes for the bootloader
in the non-volatile eNVM. This space is barely enough for running NuttX.
If OpenSBI is selected, it will be placed in DDR. This all means the
nuttx.bin file grows into gigabyte size, filling the unused space (ddr -
envm) with zeroes.
The memory layout is as follows:
MEMORY
{
ddr (rx) : ORIGIN = 0x80000000, LENGTH = 4M
envm (rx) : ORIGIN = 0x20220100, LENGTH = 128K - 256
l2lim (rwx) : ORIGIN = 0x08000000, LENGTH = 1024k
l2zerodevice (rwx) : ORIGIN = 0x0A000000, LENGTH = 512k
}
OpenSBI library is used as a separate binary, which is stored into
eMMC or SD-card. It is then loaded into its precise location in DDR.
Thus, we separate OpenSBI from NuttX and end up with two images
by utilizing the objcopy options:
--only-section=sectionpattern (-j in short)
--remove-section=sectionpattern (-R in short)
This is only valid when CONFIG_MPFS_OPENSBI is set.
Signed-off-by: Eero Nurkkala <eero.nurkkala@offcode.fi>
This PR updates the dependencies for `dirlinks` so they're all real files. This allows `dirlinks` rule to not have to be rerun every time.
This PR also changes the name from `dirlinks` to `.dirlinks`, since a file named `.dirlinks` is created to denote that all symlinks have been created
Changes:
* tools/link.sh
* link.sh now detects broken symlinks. Previously, it would return `0` if the symlink existed, but didn't point to anything.
* tools/Makefile.unix
* Added dependencies to symlinks as order-only prerequisites. See https://www.gnu.org/software/make/manual/html_node/Prerequisite-Types.html
* Removed `touch` from symlink recipes by specifying them as order only prerequisites.
* Check Kconfig variables before adding directories as symlinks
* Added rule for `$(TOPDIR)/arch/dummy/Kconfig`
* Added rule for `$(ARCH_SRC)/board/board`
* Added pattern rule (similar to `CONTEXTDIRS_DEPS`) for external folder dirlink dependencies
* Use $(APPDIR) instead of $(CONFIG_APPS_DIR), since on line 64 $(APPDIR) is validated and `realpath` is called on the path
* Added a rule `clean_dirlinks` to cleanup the symlinks in the correct order
* .gitignore
* Added ignore rule for `.dirlinks`
Testing
Step 1: configure nuttx:
```
$ (cd tools && ./configure.sh -a ../incubator-nuttx-apps stm32f3discovery:nsh)
```
part of the configure step ends up calling `.dirlinks`.
Step 2: We can confirm that `.dirlinks` doesn't need to be run again by running with the question flag: `--question`
```
$ make .dirlinks --question
$ echo $?
0
```
Step 3: confirm `make` succeeds.
```
make
```
