cc-caixf issues

Results 4 issues of


                                            cc-caixf

generate_blobs.py: handle missing PrgData section

![image](https://user-images.githubusercontent.com/19278194/121473141-c9379b80-c9f4-11eb-8d64-e17ed140c6f8.png) ![image](https://user-images.githubusercontent.com/19278194/121473162-d0f74000-c9f4-11eb-8ec1-7472f945b3c6.png) ![image](https://user-images.githubusercontent.com/19278194/121473203-e0768900-c9f4-11eb-9c23-5e82269a8fed.png) ![image](https://user-images.githubusercontent.com/19278194/121473217-e5d3d380-c9f4-11eb-8060-fb6f0b9989fc.png) ![image](https://user-images.githubusercontent.com/19278194/121473241-f126ff00-c9f4-11eb-9def-8e740e2bda25.png) use gnu toolchain,no section name PrgData in output file *.elf

bug

component: scripts

建议增加expect rsp packet len.

可以用requst packet推出rsp packet len, 这样提高分包效率

stm32串口掉数据

场景: - 串口1 `115200bps` 以`中断`的方式接收/发送数据 - 串口2 `115200bps` 以`DMA`的方式接收/发送数据现象 - 串口2接收大批量数据时, 串口1就会出现接收的数据包中掉1个字节的情况分析 - `serial_v2.c` 的[这里](https://github.com/RT-Thread/rt-thread/blob/ed4d037d7b09cee8e53810affd215ae504c79fc7/components/drivers/serial/serial_v2.c#L432C16-L432C16)从ringbuf copy数据长时间的关闭中断, 导致系统响应中断不及时. 将这里的关中断去掉, 串口1就不会掉数据(话说, 这里的互斥不应该应用层来做吗?) 还发现的问题 - `drv_usart.c` 的stm32 实现, ORE在读SR再读DR就会被清掉, 所以[这种写法](https://github.com/RT-Thread/rt-thread/blob/ed4d037d7b09cee8e53810affd215ae504c79fc7/bsp/stm32/libraries/HAL_Drivers/drivers/drv_usart.c#L461C16-L461C16)根本不能发现ORE, 判断ORE应该在读DR之前

pyocd gdbserver crash

what happened to this, do I need to set something? Steps to Reproduce - vscode start debug Environment - Operating System: window10 - pyOCD Version: 0.36.0 - gdb version: GNU...