liangxiaowz

thanks for reply. it is realy concern me for this "quicand-dirty way". So I use inline assembly instead at my project. Anyway, I apprciate to your work on SDCC, and sorry for reply too later. I just note that a imporvement was applied. refer to thread https://sourceforge.net/p/sdcc/bugs/2642/ Thanks

thanks for reply. it is realy concern me for this "quicand-dirty way". So I use inline assembly instead at my project. and totally forgot this thread. Anyway I realy appreciate to your work on SDCC I just note that a imporvement was applied. refer to thread https://sourceforge.net/p/sdcc/bugs/2642/ Thanks

thanks for reply. it is realy concern me for this "quicand-dirty way". So I use inline assembly instead at my project. and totally forgot this thread. Anyway I realy appreciate to your work on SDCC Thanks

I try this code #include <stdint.h> typedef struct { volatile uint8_t ODR; volatile uint8_t IDR; volatile uint8_t DDR; volatile uint8_t CR1; volatile uint8_t CR2; } PORT_TypeDef; #define PERIPH_BASE ((uint16_t)0x5000) #define PA ((PORT_TypeDef*)PERIPH_BASE) #define BV(b) (1<<(b)) void main(void){ while(1){ PA->ODR ^= BV(0); PA->ODR ^= BV(1); PA->ODR ^= BV(2); PA->ODR ^= BV(3); PA->ODR ^= BV(4); PA->ODR ^= BV(5); PA->ODR ^= BV(6); PA->ODR ^= BV(7); PA->ODR |= BV(0); PA->ODR |= BV(1); PA->ODR |= BV(2);...

I see, but it's still a bug, right? and I post a ticket and it has been fixed at...

I see, but it's still a bug, right? and I post a ticket and it has been fixed at...

keyword switch's behavior

I try sdcc-snapshot-x86_64-w64-mingw32-20160615-9621. #include <stdint.h> #include...