`puts("");` did not print a newline. The standard
behavior is to print a newline even if the string
is empty.
Signed-off-by: liamHowatt <liamjmh0@gmail.com>
reason:
1: spin_lock_init and spin_initialize have similar functionalities.
2: spin_lock and spin_unlock should be called in matching pairs.
Signed-off-by: hujun5 <hujun5@xiaomi.com>
In devices without storage media, you can export data to the
command line and then generate the corresponding gcda file
It can save the result output by calling __gcov_info_to_gcda
The usage is similar to:
https://gcc.gnu.org/onlinedocs/gcc/Freestanding-Environments.html#Profiling-and-Test-Coverage-in-Freestanding-Environments
Usage:
./tools/configure.sh qemu-armv7a:nsh
Modify the configuration
+CONFIG_COVERAGE_ALL=y
+CONFIG_COVERAGE_MINI=y
+CONFIG_SYSTEM_GCOV=y
Run:
qemu-system-arm -cpu cortex-a7 -nographic -smp 4 \
-machine virt,virtualization=off,gic-version=2 \
-net none -chardev stdio,id=con,mux=on -serial chardev:con \
-mon chardev=con,mode=readline -kernel ./nuttx/nuttx -semihosting -s | tee gcov.txt
./nuttx/tools/gcov_convert.py -i ./gcov.txt
./nuttx/tools/gcov.sh -t arm-none-eabi-gcov
Signed-off-by: yinshengkai <yinshengkai@xiaomi.com>
If the pointer is not set to zero, it will erroneously point to itself, resulting in an erroneous loop pointing
Signed-off-by: wangchen <wangchen41@xiaomi.com>
reason:
new implementation does not requires the use of enter_critical_section,
so the source code needs to be moved to user space
This reverts commit d189a86a35.
syslog_putc() have a lot of duplicate logic with syslog_write().
remove syslog_putc() and reuse syslog_write() to simplify syslog printing.
Signed-off-by: chao an <anchao@lixiang.com>
reason:
When entering an exception or interrupt, there are two sets of registers:
one is the "running regs", which we need to save,
and the other is the "ready to running regs", which we may soon use.
For consistency in logic, we can always store the "running regs" in the regs field of g_running_tasks,
otherwise it may lead to errors in the storage location of the "running regs."
When we need to access the "running regs," we should uniformly retrieve them from the regs field of g_running_tasks.
As the next step, we will rename the set_current_regs/up_current_regs functions
for each architecture to more appropriate names, solely for the purpose of identifying interrupts.
Signed-off-by: hujun5 <hujun5@xiaomi.com>
reason:
We decouple semcount from business logic by using an independent counting variable,
which allows us to remove critical sections in many cases.
Signed-off-by: hujun5 <hujun5@xiaomi.com>
We could call backtrace from mm module and access to kasan protected mm_node.backtrace field. Disable kasan check for backtrace_format.
Signed-off-by: xuxingliang <xuxingliang@xiaomi.com>
Some armv7-m-based SoCs do not work with atomic instructions,
even though armv7-m supports them.
To avoid using atomic instructions generated by gcc,
CONFIG_LIBC_ARCH_ATOMIC is newly introduced with which
arch_atomic.c is linked explicitly.
However, the function names need to be changed to avoid
build errors, since the functions described in stdatomic.h
are gcc built-in and inlined when the code is compiled.
About libcxx with CONFIG_LIBC_ARCH_ATOMIC, it still
does not work. It is also needed to call nx_atomic_ ver
instead of __atomic ver in
libcxx/include/__atomic/cxx_atomic_lmpl.h.
Signed-off-by: Takuya Miyasita <Takuya.Miyashita@sony.com>
Currently, there are four code coverage statistics implementations:
1. GCC native implementation
2. CLANG native implementation
3. GCC coverage nuttx mini version
4. CLANG coverage nuttx mini version
Coverage Support:
GCC native implementation CLANG native implementation GCC coverage nuttx mini version CLANG coverage nuttx mini version
Compiler version requirements ALL ALL GCC 13.2 and below CLANG 17.0 and below
Program coverage statistics support Main program √ √ √ √
Program coverage statistics support Interrupt program X √ √ √
Configuration options CONFIG_COVERAGE_TOOLCHAIN CONFIG_COVERAGE_COMPILER_RT CONFIG_COVERAGE_MINI CONFIG_COVERAGE_MINI
Platform Support ARM √ √ √ √
ARM64 √ √
(Coming soon, not yet merged)
Signed-off-by: wangmingrong1 <wangmingrong1@xiaomi.com>
1. Excerpted from: https://github.com/ARM-software/LLVM-embedded-toolchain-for-Arm/blob/main/samples/src/cpp-baremetal-semihosting-prof/proflib.c
2. Since llvm profile supports more than just gcov, and some features have not yet been explored, two clang gcov implementations are supported after this patch
3. Using this lib only supports the gcov compilation options of "-fprofile-instr-generate -fcoverage-mapping"
4. This file is heavily dependent on the compiler clang version, and is currently aligned with ci, supporting 17.0.1 and below. 18 and above are not supported by this library due to different internal implementations of the compiler
Signed-off-by: wangmingrong1 <wangmingrong1@xiaomi.com>
Some armv7-m-based SoCs do not work with atomic instructions,
even though armv7-m supports them.
To avoid using atomic instructions generated by gcc,
CONFIG_LIBC_ARCH_ATOMIC is newly introduced with which
arch_atomic.c is linked explicitly.
However, the function names need to be changed to avoid
build errors, since the functions described in stdatomic.h
are gcc built-in and inlined when the code is compiled.
Signed-off-by: Takuya Miyasita <Takuya.Miyashita@sony.com>
1. bin and kbin folders are dynamically generated
2. Context comes before depend, so the bin and kbin folders are created in the context construction phase. When the depend process comes, each library is downloaded, and a relative bin/xxx and kbin/xxx directory is created for each library to place files. This is to prevent duplicate file names.
Signed-off-by: wangmingrong1 <wangmingrong1@xiaomi.com>
1. Calling CLEAN on these folders did not have any effect
2. bin,kbin will be created during the context construction process
Signed-off-by: wangmingrong1 <wangmingrong1@xiaomi.com>
