I have utilized CTL on the STM32E407 equipped with FreeRTOS; nevertheless, each instance whereby the deque is employed results in program paralysis. Through experimentation, I discerned an anomalous memory consumption when the deque is in use. This discrepancy causes no issues when operating on a desktop computer; however, it leads to freezing when platforms such as microcontrollers are employed.
It's feasible to draw a comparison between the memory consumption of the deque container and that of the vector container.
printf("before init vec the min free stack size is %d \r\n",(int32_t)uxTaskGetStackHighWaterMark(NULL));
printf("before init vec the min free heap size is %d \r\n",(int32_t)xPortGetFreeHeapSize());
vec_double temp =vec_double_init();
printf("after init vec the the min free stack size is %d \r\n",(int32_t)uxTaskGetStackHighWaterMark(NULL));
printf("after init vec the the min free heap size is %d \r\n",(int32_t)xPortGetFreeHeapSize());
vec_double_resize(&temp,3,0.0);
printf("after push back to vec the min free stack size is %d \r\n",(int32_t)uxTaskGetStackHighWaterMark(NULL));
printf("after push back to vec the min free heap size is %d \r\n",(int32_t)xPortGetFreeHeapSize());
printf("before init deq the min free stack size is %d \r\n",(int32_t)uxTaskGetStackHighWaterMark(NULL));
printf("before init deq the min free heap size is %d \r\n",(int32_t)xPortGetFreeHeapSize());
deq_double temp2 =deq_double_init();
printf("after init deq the the min free stack size is %d \r\n",(int32_t)uxTaskGetStackHighWaterMark(NULL));
printf("after init deq the the min free heap size is %d \r\n",(int32_t)xPortGetFreeHeapSize());
deq_double_resize(&temp2,3,0.0);
printf("after push back to deq the min free stack size is %d \r\n",(int32_t)uxTaskGetStackHighWaterMark(NULL));
printf("after push back to deq the min free heap size is %d \r\n",(int32_t)xPortGetFreeHeapSize()); Output is:
[12:05:59:396] before init vec the min free stack size is 8866 ␍␊
[12:05:59:399] before init vec the min free heap size is 15880 ␍␊
[12:05:59:409] after init vec the the min free stack size is 8776 ␍␊
[12:05:59:413] after init vec the the min free heap size is 15880 ␍␊
[12:05:59:423] after push back to vec the min free stack size is 8776 ␍␊
[12:05:59:426] after push back to vec the min free heap size is 15848 ␍␊
[12:05:59:437] before init deq the min free stack size is 8758 ␍␊
[12:05:59:440] before init deq the min free heap size is 15848 ␍␊
[12:05:59:450] after init deq the the min free stack size is 8758 ␍␊
[12:05:59:453] after init deq the the min free heap size is 15848 ␍␊
[12:05:59:464] after push back to deq the min free stack size is 8758 ␍␊
[12:05:59:467] after push back to deq the min free heap size is 11720 ␍
Within the deque, storing a mere three double variables surprisingly consumed almost 4000 bytes of memory, a figure I perceive as incorrect. I have initiated the utilization of a vector container as a substitute for the deque, however, does a superior solution exists to remedy this issue? I avidly anticipate your response.
I have utilized CTL on the STM32E407 equipped with FreeRTOS; nevertheless, each instance whereby the deque is employed results in program paralysis. Through experimentation, I discerned an anomalous memory consumption when the deque is in use. This discrepancy causes no issues when operating on a desktop computer; however, it leads to freezing when platforms such as microcontrollers are employed.
It's feasible to draw a comparison between the memory consumption of the deque container and that of the vector container.
Output is:
Within the deque, storing a mere three double variables surprisingly consumed almost 4000 bytes of memory, a figure I perceive as incorrect. I have initiated the utilization of a vector container as a substitute for the deque, however, does a superior solution exists to remedy this issue? I avidly anticipate your response.