ARM Cortex G＆＃x2B;＆＃x2B;我正在为Cortex M7＆＃x2B写一个EDMA串行驱动程序。我正在尝试使用c＆＃x2b;＆＃x2b;，想知道是否可能

发布于 2025-01-28 11:48:51 字数 505 浏览 2 评论 0原文

我正在编写Cortex M7+ EDMA串行驱动程序。我想在C ++中这样做，从制造商的示例中可以在C中工作。我遇到了一些问题之一。我知道如何从中断中调用静态C ++函数，这不是问题。问题在于他们的驱动程序要求EDMA转移的Ciruclar缓冲区位于不可缓存的内存中。

/* allocate ring buffer section. */
AT_NONCACHEABLE_SECTION_INIT(uint8_t _ringBuffer[RX_RING_BUFFER_SIZE]) = {0};
/* Allocate TCD memory poll with ring buffer used. */
AT_NONCACHEABLE_SECTION_ALIGN(static edma_tcd_t _tcdMemoryPoolPtr[1], 
sizeof(edma_tcd_t));

有没有办法使用类变量来执行此操作，或者当我实例化班级时，我需要将整个班级放置在该领域？还是我应该把那些可怕的东西留在c中？

谢谢

原文

I am writing a cortex M7+ edma serial driver. I want to do it in c++ I have it working in c from the examples from the manufacturer. One of the problems I am having a few problems. I know how to call a static C++ function from the interrupt that is not the problem. The problem is that their driver requires that the ciruclar buffer for the edma transfer be located in non cacheable memory.

/* allocate ring buffer section. */
AT_NONCACHEABLE_SECTION_INIT(uint8_t _ringBuffer[RX_RING_BUFFER_SIZE]) = {0};
/* Allocate TCD memory poll with ring buffer used. */
AT_NONCACHEABLE_SECTION_ALIGN(static edma_tcd_t _tcdMemoryPoolPtr[1], 
sizeof(edma_tcd_t));

Is there a way to do this with class variables, or when I instantiate my class do I need to place the whole class in this area? Or should I just leave the darn thing in C??

Thanks

分享到QQ

分享到微博