dimonomid / tneo Goto Github PK

Error accessing repository with tests.

Error: HTTP Error 503: Service Unavailable

Can this RTOS also support the Cortex-M7? Are there any plans?

Hey

As the dsPIC33 has the same Architecture except the additional dsp register and as you wrote your kernel is useable for dspic devices why are you not saving the dsp Registers during a context switch?

It should be inserted in File tneo/src/arch/pic24_dspic/tn_arch_pic24.S and during stack initialisation.

So that an working dsp operation with Registers ACCA*/ACCB*/DO is context save.

Hi, Dmitry.

The links on docs haven't worked. It opens the bitbucket.org main face instead.

Hello,
I've tried to use TNeo with my ATSAMC21E (Cortex-M0+) and GCC compiler, have configured everything (.S file too) but without success. I can start the RTOS and create tasks, but it stays always in the task with highest priority and doesn't switch to another tasks after tn_task_sleep(1). Even idle task and SysTick interrupt are not active anymore. It seems to be not usable with such configuration (Cortex-M0+ and GCC).
Very interesting project but not really functional with all CPUs from the list.

Architecture: ARM Cortex M4F
Compiler: Linaro gcc-arm-none-eabi-7-2017-q4-major (bundled with TI Code Composer Studio v8.0.0)

During the build this warning is treated as error, so the build fails:

src/arch/cortex_m/tn_arch_cortex_m.S:303:22: error: this use of "defined" may not be portable [-Werror=expansion-to-defined] #if _TN_NEED_SAVE_LR() ^ src/arch/cortex_m/tn_arch_cortex_m.S:400:22: error: this use of "defined" may not be portable [-Werror=expansion-to-defined] #if _TN_NEED_SAVE_LR() ^ src/arch/cortex_m/tn_arch_cortex_m.S:489:22: error: this use of "defined" may not be portable [-Werror=expansion-to-defined] #if _TN_NEED_SAVE_LR() ^ cc1: all warnings being treated as errors make[1]: *** [_obj/cortex_m4f/arm-none-eabi-gcc/tn_arch_cortex_m.o] Ошибка 1 make: *** [all] Ошибка 2

I'm trying to use tneo in my samd21 (m0+) project. Since I have my own build engine based on python, I'm doing the "build manually" route. One of the options I compile with is -Wcast-align. However, doing so is producing the following error:

/Users/travisg/Projects/twig_lora_tneo/TNEO/core/tn_fmem.c: In function 'tn_fmem_create':
/Users/travisg/Projects/twig_lora_tneo/TNEO/core/tn_fmem.c:333:20: error: cast increases required alignment of target type [-Werror=cast-align]
          p_tmp   = (void **)p_block;

I'm not sure what the best way to get around this is. I can remove that option, but once upon a time, I had convinced myself that was a good thing to check for.

Hi. I have some questions.

I would use a low-power mode. I didn't find any information in docs about that.

1. Is it a possible with TNeo?
2. Where is the best place to move MCU a sleep mode?

Hi.

1. When I'm using the 'Pedantic' mode I got warnings like this:

../wm-core/src/tneo/src/core/internal/_tn_timer.h: In function '_tn_timer_callback_call':
../wm-core/src/tneo/src/core/internal/_tn_timer.h:182:20: warning: ISO C99 requires at least one argument for the "..." in a variadic macro
    TN_INT_IRESTORE();

../wm-core/src/tneo/src/core/tn_tasks.c: In function '_task_terminate':
../wm-core/src/tneo/src/core/tn_tasks.c:356:25: warning: ISO C99 requires at least one argument for the "..." in a variadic macro
       _TN_FATAL_ERROR("");

2. I have seen strange macros in file tn_arch_cortex_m.h. This file is part of arm-cortex arch and PIC32 prefix surprised me.

#define TN_INT_RESTORE()    _TN_PIC32_INTSAVE_CHECK();                      \
                            tn_arch_sr_restore(TN_INTSAVE_VAR)

I got this error:

../tn_tasks.c: In function '_tn_task_set_dormant':
../tn_tasks.c:1117:39: error: 'struct TN_Task' has no member named 'mutex_queue'
    } else if (!_tn_list_is_empty(&task->mutex_queue)){

It looks like a great project!

In the readme, all the links to the documentation return a 404.
They all link to some special bitbucket repo, that doesn't seem to exist.
I'm curious about reading the documentation.

Hello dimonomid,

I see that the original TNKernel and the andersm/TNKernel-PIC32 are under the BSD license. Could you please add a LICENSE or COPYING file or section in the README with the license you choose?

Thanks.