1 / 11

Digital Integrated Circuits - week twelve -

Digital Integrated Circuits - week twelve -. Gheorghe M. Ş tefan http://arh.pub.ro/gstefan/ - 2014 -. Synthesis problems. 326 LUTs, f clock = 310 MHz. 97 LUTs. 33 LUTs, 506 MHz. Timing in toyRISC. The longest path is trough RALU , because in NextPC there is a 16-bit adder only

cleta
Télécharger la présentation

Digital Integrated Circuits - week twelve -

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Digital Integrated Circuits- week twelve - Gheorghe M. Ştefan http://arh.pub.ro/gstefan/ - 2014 -

  2. Synthesis problems 326 LUTs, fclock = 310 MHz Digital Integrated Circuits - week twelve

  3. 97 LUTs Digital Integrated Circuits - week twelve

  4. 33 LUTs, 506 MHz Digital Integrated Circuits - week twelve

  5. Timing in toyRISC The longest path is trough RALU, because in NextPC there is a 16-bit adder only Tclock = tclock_to_instruction + tleftAddr_to_leftOp + troughALU + ttroughMUX + tfileRegSU Digital Integrated Circuits - week twelve

  6. The solution to improve speed • Pipelining: • Between toyRISC and memory • On the RALU’s loop • The price for pipelining: various latencies • The problem will be solved in the course of Computer Architecture (Hennessy & Patterson: Computer Architecture. A Quantitative Approach) Digital Integrated Circuits - week twelve

  7. Four-Loop Systems • ROM on loop: • a computer with a dedicated function (embedded computing) • RAM on loop: • the generic computer • LIFO on loop: • processor with a language-oriented architecture (ex. LISP) • PROCESSOR on loop: • processor with a functional accelerator (ex. floating-point ops.) Digital Integrated Circuits - week twelve

  8. Computer: the strongest segregation between simple and complex COMPUTER = the simple PROCESSOR + the complex programs stored in RAM The function is given by: the Instruction Set Architecture The symbolic complex structure, stored in RAM, is interpreted by the simple structure of a machine Digital Integrated Circuits - week twelve

  9. Computing machine: a recursive approach • Computing machine: • The machine language (the architecture) • The storage for a program written the machine language • The machine to interpret the content of the storage: • The machine language • The storage • The machine to interpret: • … • … … • The machine to execute Digital Integrated Circuits - week twelve

  10. What about more loops? A computing machine requires at least 4 loops to be competent More loops are requested for a computing machine to improve the performance. Digital Integrated Circuits - week twelve

  11. toyMachine revisited • The behavioral description (chapter 5) is implemented with 2429 LUTs and works at 164 MHz • The structural description is implemented with 321 LUTs and works at 205 MHz • The size is reduced 7.56 times • The speed increases with 25% Digital Integrated Circuits - week twelve

More Related