Email: Password: Remember Me | Create Account (Free)

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
???
10/20/06 19:22
Read: times


 
#126829 - map registers
Responding to: ???'s previous message
if you have a set of map registers, one of which is dedicated to moving the bit-mapped registers then you can move it about by adding the map-register content to the current address of the bitmapped region. That happens transparently to the instructions. You do have to program the thing, though. If you look carefully at the instruction binaries, you'll see how the bit-addressable registers' bits are addressed. Extend that same concept to addressing the SFR's, with an offset from the page registers, i.e. transparently to the user/code, but imposed via the hardware, a different part of the SFR-space is now bit-addressable.

The page registers can be accessed via two locations in SFR-space, one for the address and another for the content, simiarly to the way in which the innards of the HD44780 is addressed. The concept can be implemented in a way that's infinitely extensible, so anything's possible with this approach. There can be an offset register for each bit, just as there can be an offset register for each page of memory, either code or data. It's all up to the designer/implementor.

It would be up to the programmer to keep track of the effect of his modification of the page table(s).

That way, the entire mess can be kept instruction-set compatible.

RE


List of 59 messages in thread
TopicAuthorDate
How much flash should an 8051 have?            01/01/70 00:00      
   enough            01/01/70 00:00      
      Code Flash VS Data Flash            01/01/70 00:00      
         hints            01/01/70 00:00      
            Max Clock Speed, Etc.            01/01/70 00:00      
               that\'s a cop-out            01/01/70 00:00      
                  Not a cop-out            01/01/70 00:00      
                     can you sell it for less than $1.41            01/01/70 00:00      
                        Replacement costs            01/01/70 00:00      
                           then please stop arguing one point from the perspe            01/01/70 00:00      
                              The replacement market            01/01/70 00:00      
                                 How will that help?            01/01/70 00:00      
                                    Flash Reliability            01/01/70 00:00      
                                       not really - look out            01/01/70 00:00      
                                          More on Flash Reliability            01/01/70 00:00      
                                             XRAM replacement            01/01/70 00:00      
                                                Write-While-Fetch            01/01/70 00:00      
                                    That does not appear to be the target.            01/01/70 00:00      
                                       6/12 Clock Mode and XDATA support            01/01/70 00:00      
                     Have you looked at the one-clockers?            01/01/70 00:00      
                        How do you propose to fit that in the instruction            01/01/70 00:00      
                           map registers            01/01/70 00:00      
               clock speed            01/01/70 00:00      
                  Not a bad idea ... that's what Dallas did!            01/01/70 00:00      
                     Clock Speed Reduction            01/01/70 00:00      
                        if you knew the chip            01/01/70 00:00      
                           actually, it does either speed up or slow-down            01/01/70 00:00      
                              Variable clock speed            01/01/70 00:00      
                                 clock vs power            01/01/70 00:00      
                                    SIngle clock power            01/01/70 00:00      
               when i did this            01/01/70 00:00      
                  0.35u process & 64K            01/01/70 00:00      
         Big Flash - coda and data!            01/01/70 00:00      
            Does Zylogic = Triscend?            01/01/70 00:00      
               Zylogic; Triscend            01/01/70 00:00      
            that should make it clear            01/01/70 00:00      
         Data Apps            01/01/70 00:00      
   bank switching mechanism (RFC)            01/01/70 00:00      
      Bank Switching            01/01/70 00:00      
         that's how it's done traditionally ...            01/01/70 00:00      
         Keil support            01/01/70 00:00      
      bank switching no0t needed            01/01/70 00:00      
         "bank switching no0t needed"            01/01/70 00:00      
            if you read it all            01/01/70 00:00      
               I'd even tend to believe...            01/01/70 00:00      
      No lock IRQ            01/01/70 00:00      
   Parkinson's Law...            01/01/70 00:00      
   The trouble with PLLs            01/01/70 00:00      
      nearly every ARM has a PLL            01/01/70 00:00      
   yeah i understand that            01/01/70 00:00      
      ethernet mac            01/01/70 00:00      
   Dougal was..            01/01/70 00:00      
      Sorry to hear about your misfortune            01/01/70 00:00      
   no not seriously            01/01/70 00:00      
      I don't think he hates you, Jez            01/01/70 00:00      
      That must be the most painful            01/01/70 00:00      
         it's not a killer, but you do notice it            01/01/70 00:00      
   well jasmine can ride him all day            01/01/70 00:00      
   Gives a new meaning to that old interview question            01/01/70 00:00      

Back to Subject List