EECS 470: Computer Architecture

Homework, Project, and Verilog handouts can be found under homework/projects.

Class Wiki

• EECS 470 Wiki on ctools

Lecture notes

Slides are all powerpoint.

• Lecture 1: Class intro, basics, and pipelining review
• Lecture 2 and 3: Performance and goals, in order pipelining, start of out-of-order.
• Lecture 4: (done on board) Finish T1 in detail, start prediction/BTB
• Lecture 5 (and 6?): Prediction, some on T2.
• Lecture 6/7: T2, ILP
• Lecture 8: T2, ILP, Project, start of T3
• Lecture 9: T3, Tclock
• Lecture 10: Exam stuff, Tclk, Mem, T3 with back pointers, Pentium 4
• Lecture 11: Superscalar, early branch, LSQ,
• Lecture 12: VM, Caches
• Lecture 13: More on caches and memory
• Power lecture.
• Multi-processor stuff done mostly on the board, but also: Parallel overview/review
• Compliers:

Papers

Note that many of these links go to the ACM or IEEE digital libraries, which require subscriptions to access. You will need to access them from a umich.edu IP address to take advantage of the University's subscription.

• Trace Cache: A Low Latency Approach to High Bandwidth Instruction Fetching by E. Rotenberg, S. Bennett, and J.E. Smith, Proceedings of the 29th Annual International Symposium on Microarchitecture, November 1996

  First paper on trace caches.

• Combining Branch Predictors, S. McFarling, WRL Technical Note TN-36, June 1993.

  Proposes the gshare branch predictor, covers a few others. See also the paper by Yeh and Patt (below).

• Checkpoint processing and recovery: Towards scalable large instruction window processors. By H. Akkary, R. Rajwar, and S. T. Srinivasan. In MICRO 36, December 2003.

  Reordering without the reorder buffer.

• Implementation of precise interrupts in pipelined processors by J. E. Smith and A. R. Pleszkun. Proceedings of the 12th Annual International Symposium on Computer Architecture, June 1985, pp. 36-44.

  The original paper on reorder buffers and their alternatives.

• The Mips R10000 superscalar microprocessor by K. C. Yeager, IEEE Micro, April 1996.

  One of the first out-of-order microprocessors. Uses a merged physical register file (unlike the P6).

• The Alpha 21264 microprocessor by R. E. Kessler, IEEE Micro, Mar/Apr 1999.

  Another out-of-order microprocessor that also uses a merged physical register file. The 21264 was easily the fastest processor available when it came out. The "dual cluster" design that uses two copies of the register file to reduce the complexity and latency of the bypass network is particularly interesting. This paper also has a substantial discussion of the 21264 tournament branch predictor that's also described in the textbook.

• Alternative Implementations of Two-Level Adaptive Branch Prediction by T.-Y. Yeh and Y. N. Patt. Proceedings of the 19th Annual International Symposium on Computer Architecture, June 1992, pp. 124-134.

  The classic reference on two-level branch prediction.

• Understanding the detailed Architecture of AMD's 64 bit Core by Hans de Vries, www.chip-architect.com.

  Some guy's attempt to reverse engineer the Opteron based on published documents, patents, and some speculation. The most detailed Opteron description I know of, but not 100% accurate.

• The Microarchitecture of the PentiumŪ 4 Processor by Glenn Hinton et al. Intel Technology Journal, Vol. 5 Issue 1 (February 2001).

  Description of the Pentium 4 microarchitecture by the chief designers. ncludes some comparisons with P6 and some justification of the deep pipeline/high frequency deisgn goal.

On-line stuff

• Ars Technica is probably the best for a student of computer architecture. Their CPU/chipselt overview is pretty good. (link dead at the moment).
• Tom's hardware has a few interesting articles but tends to be unreadable with the ads and chopped-up articles.
• Lots of papers can be found though the so called Computer Architecture home page. This is a LOT more academic than the above.
• Also, the usenet group comp.arch can be a source for latest opinions and thoughts, but take it with a grain of salt. You can also read it via google. It varies widely from pure academic stuff to hands-on, to random spam...

Files from discussion examples

None at this time.