From: user5857@newsgrouper.org.invalid   
      
   Thomas Koenig  posted:   
      
   > EricP  schrieb:   
   > > Thomas Koenig wrote:   
   > >> https://old.chipsandcheese.com/2025/08/29/condors-cuzco-ris   
   -v-core-at-hot-chips-2025/   
   > >> has an interestig take on how to do OoO (quite patented,   
   > >> apparently).  Apparently, they predict how many cycles their   
   > >> instructions are going to take, and replay if that doesn't work   
   > >> (for example in case of an L1 cache miss).   
   > >>   
   > >> Sounds interesting, I wonder what people here think of it.   
   > >   
   > > I searched for "processor" "schedule" "time resource matrix" and got   
   > > a hit on a different company's patent for what looks like the same idea.   
   > >   
   > > Time-resource matrix for a microprocessor with time counter   
   > > for statically dispatching instructions   
   > > https://patents.google.com/patent/US11829762B2   
   >   
   > Maybe the same people/company.  Thang Minh Tran, the inventor   
   > of the patent, works for Simple (the owner of the patent), but   
   > previously worked for Andes, who held the presentation.  This   
   > might be a case of shared IP, or a licensing agreement.   
   >   
   > Mitch, from his bio on simplexmicro.com, it seems that he worked   
   > at AMD around the same time you did, maybe a little earlier.   
   > Do you know him?   
      
   I was at AMD from Oct 1999 to May 2006   
      
   I don't remember the name.   
      
   > > It basically puts all the schedule in one HW matrix of time_slots *   
   resources   
   > > and scans forward looking for empty slots to allocate to each instruction.   
      
   CRAY 1 vector sequencer used such a "shift register" approach.   
      
   > > The scheduling is done at Rename and time slots assigned for each resource   
   > > needed, source operand read ports, FU's, result buses.   
   > > If a load later misses L1 it triggers a replay of all younger instructions.   
   >   
   > It's the same that was refrenced in the presentation; the drawings   
   > also match.   
   >   
   > > They claim it is simpler but I question that.   
   >   
   > Patents and marketing often claim advantages which are, let's say,   
   > dubious :-)   
   >   
   > [...]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)