Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.arch.embedded > #24495

Re: Early 8051 variants with dual dptr (AMD?)

From Philipp Klaus Krause <pkk@spth.de>
Newsgroups comp.arch.embedded
Subject Re: Early 8051 variants with dual dptr (AMD?)
Date 2016-07-01 08:06 +0200
Organization solani.org
Message-ID <nl518i$n85$1@solani.org> (permalink)
References (1 earlier) <nl03qh$s7i$1@solani.org> <n029nb5r45t3n80p6hjahjdkhivh4k8itf@4ax.com> <td39nbhn82svvq4c2k3r77rkq27hp1mm5k@4ax.com> <nl26d6$3if$2@dont-email.me> <dtlf5hF90sdU1@mid.dfncis.de>

Show all headers | View raw


On 30.06.2016 22:52, Hans-Bernhard Bröker wrote:
> Am 30.06.2016 um 06:15 schrieb rickman:
> 
>> With multiple DPTR registers, wouldn't the instruction set be
>> modified making it no longer code compatible?
> 
> Of course not.  That would be silly.
> 
>> Or is the DPTR register
>> switched out by some mechanism that is essentially transparent to
>> existing code?
> 
> The 8051 has an entire dedicated memory space for "special function
> registers".  The obvious place to put the switch that flips the two
> available DPTRs is in there.
> 
> Normal compiled code won't get to use the additional DPTR, though.  It
> would be used mainly to speed up optimized or inlined versions of stuff
> like memcpy(), memset() etc.  It was great fun to deal with this in
> assembly, though.
> 

Yes. The only exception I know of are the ds390 and ds400 backends in
SDCC, which use the dual dptr for normal code. But one could argue that
the DS390 and DS400 with their 24-bit pointers are more than just normal
8051 drivatives.

Philipp

Back to comp.arch.embedded | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread


Thread

Early 8051 variants with dual dptr (AMD?) Philipp Klaus Krause <pkk@spth.de> - 2016-06-29 09:49 +0200
  Re: Early 8051 variants with dual dptr (AMD?) Philipp Klaus Krause <pkk@spth.de> - 2016-06-29 11:19 +0200
    Re: Early 8051 variants with dual dptr (AMD?) Martin Riddle <martin_ridd@verizon.net> - 2016-06-29 22:52 -0400
      Re: Early 8051 variants with dual dptr (AMD?) Martin Riddle <martin_ridd@verizon.net> - 2016-06-29 23:28 -0400
        Re: Early 8051 variants with dual dptr (AMD?) rickman <gnuarm@gmail.com> - 2016-06-30 00:15 -0400
          Re: Early 8051 variants with dual dptr (AMD?) Philipp Klaus Krause <pkk@spth.de> - 2016-06-30 09:37 +0200
          Re: Early 8051 variants with dual dptr (AMD?) Martin Riddle <martin_ridd@verizon.net> - 2016-06-30 08:26 -0400
          Re: Early 8051 variants with dual dptr (AMD?) Hans-Bernhard Bröker <HBBroeker@t-online.de> - 2016-06-30 22:52 +0200
            Re: Early 8051 variants with dual dptr (AMD?) Philipp Klaus Krause <pkk@spth.de> - 2016-07-01 08:06 +0200
        Re: Early 8051 variants with dual dptr (AMD?) Philipp Klaus Krause <pkk@spth.de> - 2016-07-07 01:19 +0200
      Re: Early 8051 variants with dual dptr (AMD?) Philipp Klaus Krause <pkk@spth.de> - 2016-06-30 09:28 +0200

csiph-web