Post by thread:Bit test commands and ports

Hi Piclisters, I have a device that I wish to connect to one of the I/O ports o f a PIC that can be read and written to. What I wish to know is if the PIC bit test instuctions (BTFSS and BTFSC) are read-modify-write instructions, that is if I perfom a bit test on the port whilst it is in input mode does the PIC read the port and then place the value read into the ports data latch, so that if I then change the port into an output will it then output the value read by the bit test command. For example the port has B'11110000' on it when configured as an input,carry out a bit test,change the port to output, will it output B'11110000'? or will it output the value last written to it when it was configured configured as an output? Regards Geoff :) ================== spam_OUTgeoffTakeThisOuTtechie.com

Geoff Thornton wrote: > > Hi Piclisters, > I have a device that I wish to connect to one of the I/O ports of a PIC > that can be read and written to. What I wish to know is if the PIC bit test > instuctions (BTFSS and BTFSC) are read-modify-write instructions, that is if > I perfom a bit test on the port whilst it is in input mode does the PIC read Hello Geoff. Don't worry about that. BTFSx is read only commands. Probably you wrongly thinking BSF or BCF command ... Make you sure that required i/o pins in input mode and all'll be ok. WBR Dmitry.

Dmitry, Thanks. But that is not really what I'm after you see, I want the port t o set the output to what was read after the last read of the port when it was an input and I was hoping the bit test commands would do this, can you please advise. > Hello Geoff. Don't worry about that. BTFSx is read only commands. > Probably you wrongly thinking BSF or BCF command ... Make you sure > that required i/o pins in input mode and all'll be ok. > > WBR Dmitry. >

Geoff Thornton wrote: > [...] not really what I'm after you see, I want the port to set the > output to what was read after the last read of the port when it was > an input and I was hoping the bit test commands would do this, can you > please advise. You hope vainly. Bit test instructions (BTFSC, BTFSS) do not change port bits. They change PC only, if at all. Use BCF, BSF, MOVF, et. al. instead. -- James Cameron (.....cameronKILLspam@spam@stl.dec.com) OpenVMS, Linux, Firewalls, Software Engineering, CGI, HTTP, X, C, FORTH, COBOL, BASIC, DCL, csh, bash, ksh, sh, Electronics, Microcontrollers, Disability Engineering, Netrek, Bicycles, Pedant, Farming, Home Control, Remote Area Power, Greek Scholar, Tenor Vocalist, Church Sound, Husband. "Specialisation is for insects." -- Robert Heinlein.

On Thu, 31 Dec 1998 09:10:29 +1000 Geoff Thornton <geoffKILLspamTECHIE.COM> writes: >Dmitry, > Thanks. But that is not really what I'm after you see, I want >the port to >set the output to what was read after the last read of the port when >it was >an input and I was hoping the bit test commands would do this, can you >please advise. The bit test commands don't change the port's output latch. Only instructions with a destination of "f" (including bsf and bcf) do. The easiest way to do what you are talking about would be movf PORTB,f This reads all the pins of the port and sets the output latch to the value read. Changing the port pins from input to output or vice versa by writng the TRIS register doesn't change the output latch. So the pins that are newly made outputs will immediately output the level set by the output latch. I'm fairly sure that nothing changes the output latch value other than software writes to the port register or turning the power off. ___________________________________________________________________ You don't need to buy Internet access to use free Internet e-mail. Get completely free e-mail from Juno at http://www.juno.com/getjuno.html or call Juno at (800) 654-JUNO [654-5866]

> This reads all the pins of the port and sets the output latch to the > value read. <snip> > I'm fairly sure that nothing changes the output latch > value other than software writes to the port register or turning the > power off. Your second statement is correct if the word "all" is noticed in the first statement. Doing a BSF on one bit can change the output latch value on another bit because _all_ bits of the byte are read and rewritten, even though the instruction is specific for a single bit. Steve. ====================================================== Steve Baldwin Electronic Product Design TLA Microsystems Ltd Microcontroller Specialists PO Box 15-680, New Lynn http://www.tla.co.nz Auckland, New Zealand ph +64 9 820-2221 email: EraseMEstevebspam_OUTTakeThisOuTtla.co.nz fax +64 9 820-1929 ======================================================

Geoff Thornton wrote: > It looks like I will have to do a MOVF port,W followed by a BTFSS or > BTFSC to achieve what I'm after, thanks for the help, and thank you too > Dmitry. Actually, based on what you have said before, I think what you will end up with is something like this ... MOVF port,W MOVWF port MOVWF file BTFSS file,bit Derivation follows ... You said you wanted to cause the port output latches to adopt their input states, and also to test a bit. Using "MOVF port,F" will do the first step, and "BTFSS" the second step, so the initial version would be; MOVF port,F BTFSS port,bit But the problem with this is that the port is read twice, once in each instruction, and so it is possible for the state of bits to change between each instruction. This may not be a problem for you in your application. However, to avoid the double read you'd need to buffer it in W or a file register. MOVF port,W MOVWF port BTFSS W,bit ; (sic) But you can't test a bit in W. The BTFSS works only with a port or file register addresses. So you must test a bit in either of those. But you can't test twice, so you must use a file register as a buffer. MOVF port,W ; read the port bits into W MOVWF port ; copy them back to the port MOVWF file ; save in a temporary file register BTFSS file,bit ; test the bit And when I look at code like that I wonder if I could simplify my external circuitry so that I don't have to code like that! -- James Cameron (cameronspam_OUTstl.dec.com) OpenVMS, Linux, Firewalls, Software Engineering, CGI, HTTP, X, C, FORTH, COBOL, BASIC, DCL, csh, bash, ksh, sh, Electronics, Microcontrollers, Disability Engineering, Netrek, Bicycles, Pedant, Farming, Home Control, Remote Area Power, Greek Scholar, Tenor Vocalist, Church Sound, Husband. "Specialisation is for insects." -- Robert Heinlein.