Forums before death by AOL, social media and spammers... "We can't have nice things"
|    comp.arch    |    Apparently more than just beeps & boops    |    131,241 messages    |
[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]
|    Message 130,314 of 131,241    |
|    Michael S to MitchAlsup    |
|    Re: Multi-precision addition and archite    |
|    17 Nov 25 23:35:37    |
   
   From: already5chosen@yahoo.com   
      
   On Mon, 17 Nov 2025 18:54:17 GMT   
   MitchAlsup wrote:   
      
   > anton@mips.complang.tuwien.ac.at (Anton Ertl) posted:   
   >   
   > > Robert Finch writes:   
   > > >Skimming through the SPARC architecture manual I am wondering how   
   > > >they handle register renaming with a windowed register file. If   
   > > >the register window file is deep there must be a ginormous number   
   > > >of registers for renaming. Would it need to keep track of the   
   > > >renames for all the registers? How does it dump the rename state   
   > > >to memory?   
   >   
   > I don't remember SPARC ever getting OoO. The windowed register file   
   > is but one cause.   
   >   
      
   The first production OoO SPARC was HAL SPARC64 manufactured for   
   Fujitsu on Fujitsu's own fabs back in 1995, so contemporary of PPro. It   
   was 4-die chipset.   
   HAL SPARC64-GP was first single-chip implementation in 1997.   
   https://en.wikipedia.org/wiki/HAL_SPARC64   
   The line was continued by Fujitsu:   
   https://en.wikipedia.org/wiki/SPARC64_V   
   Since then and up to 2017 there were many generations made by Fujitsu.   
      
   There were also few OoO SPARCs designed by Oracle, independently of   
   Fujitsu. I think that they all shared the same core uArch originally   
   introduced in SPARC T4 (2011).   
   https://en.wikipedia.org/wiki/SPARC_T4   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)   
|
[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]
(c) 1994, bbs@darkrealms.ca