??? 06/29/05 06:27 Read: times |
#96154 - Thank you Responding to: ???'s previous message |
Hi all,
Thank you for your feedback. I am aware of the logic behind Keil's handling of the issue. I am not disputing whether it is correct or wrong. My focus of the query is not that. I have a code which compiles correctly in IAR. The definitions are handled as indended. I find this feature useful in writing crystal depended definitions. Hence my query was, is it possible to do in the same way in Keil and how. After some reasearch and as well as from your feedback, I realise Keil needs some modifications. To illustrate, #define Xtal 22.1184 #define tic12 (12/Xtal) #define _10msec (65536 - ((1000 * 10) / tic12)) Before modification { . TL1 = _10msec; TH1 = _10msec >> 8 ; . } After modification { uint temp = _10msec; . TL1 = temp ; TH1 = temp >> 8 ; . } In the former case Keil produces a "Bad operand >> error", whereas the second case it produced the indended code. May be I am still ignorant and missing something. regards, |