Post by thread:CCS compiler

Hi, You could try inline assembly. But I think having the code: long var; var=var/32; will generate the shifts and not using the div function. The function shift_right() will shift an array or structure one bit at a time and may not generate a tight code. regards, Reggie -- e-mail: spam_OUTrberdinTakeThisOuTbigfoot.com ICQ#: 31651436 URL: http://www.bigfoot.com/~rberdin Lawrence Lile wrote: > > I'm trying to divide a 16 bit variable by 32. Instead of invoking a > cumbersome divide routine I'd like to just shift right 5X. CCS seems to > have a couple of functions that do something like this but I can't seem to > figure out how they work. Anybody know a good way to do this?

Hi Lawrence, long data; byte x; for (x = 0; x < 5; x++) { shift_right(&data,2,0); } or the normal C method. long data; data = data >>5; is the same as long data; data >>= 5; regards, Peter. --- Peter Homann Email: .....peterhKILLspam@spam@adacel.com.au Adacel Technologies Ltd _/_/_/ _/_/_/ _/ Phone: +61 3 9596 2991 250 Bay St, Brighton _/ _/ _/ _/ Fax: +61 3 9596 2960 Victoria 3186 _/_/_/ _/ _/ Home: +61 3 9555 5603 AUSTRALIA _/ _/ _/ _/_/_/Mobile: 0414-494578 ------------------------------------------------------------------------ |{Original Message removed}

Hi Lawrence, I tried to compile the following code in CCS: i = i/32; i >>= 5; and got: .................... i=i/32; RRF 26,F RRF 25,F RRF 26,F RRF 25,F RRF 26,F RRF 25,F RRF 26,F RRF 25,F RRF 26,F RRF 25,F MOVLW 07 ANDWF 26,F .................... i >>= 5; RRF 26,F RRF 25,F RRF 26,F RRF 25,F RRF 26,F RRF 25,F RRF 26,F RRF 25,F RRF 26,F RRF 25,F MOVLW 07 ANDWF 26,F which already are very efficient. regards, Reggie -- e-mail: rberdinKILLspambigfoot.com ICQ#: 31651436 URL: http://www.bigfoot.com/~rberdin Peter Homann wrote: {Quote hidden}> > Hi Lawrence, > long data; > byte x; > > for (x = 0; x < 5; x++) > { > shift_right(&data,2,0); > } > > or the normal C method. > > long data; > > data = data >>5; > > is the same as > > long data; > > data >>= 5; > > regards, > > Peter. > ---

Igor, I will start my reply by giving your several resources to answer questions about the CCS compiler. I am just a user of the compiler, starting about a year ago, and I am no expert. 1. You can get technical support directly from CCS at .....ccsKILLspam.....ccsinfo.com . They are very helpful. 2. You can get technical help on CCS and anything related to a PIC microcontroller by subscribing to the PIClist. It is a nonmoderated mailing list devoted to PIC microcontrollers and anything remotely related (Such as cows, jokes about Bill Gates, Optical components, homemade nuclear weapons, and general electronics<G>.) Joking aside, it is the BEST resource for an electronics guy. It also has many Spanish speaking members, and has even spawned a Spanish list ( I don't know the contacts for this list right now.) To subsribe send a message to EraseMELISTSERVspam_OUTTakeThisOuTMITVMA.MIT.EDU with the body of the message stating SUBSCRIBE PICLIST on a line by itself with a carriage return at the end of the line. You will get a detailed email in return telling you how the list operates. SAVE THIS MESSAGE because it tells you how to get OFF the list if you so desire. 3. Back to your question. I have trouble with the OUTPUT_HIGH(pin_xx) command as well. I think it has to do with the old read-write-modify bug (MCHIP says it is a feature???!?!?!!!???) of the PIC microcontroller. CCS translates OUTPUT_HIGH(pin_xx) into BSF PORT, PIN in assembler, which is generally NOT the correct way to change a pin in a PIC. If there is any capacitance on the pin it will NOT go high. Instead, try this: Keep a variable called MIRROR with your pin settings in it. If you want to raise a pin, do this: #ASM BSF MIRROR, PIN MOVF MIRROR, W MOVWF PORT #ENDASM Or in C, bit_set(mirror, pin); port = mirror; 4. BTW - how did you find out about me? -- Lawrwence Lile {Original Message removed}