| Class |
Date |
Description |
Reading |
Assignments |
| 1 |
1/15 |
Introduction, Local Opts |
D9.1 |
|
| 2 |
1/17 |
Dataflow, SSA |
*D9.2,9.4, 6.2.4; M8.1-4, 8.10, 8.11; C9.1-3; A4.4 |
1 out |
| 3 |
1/22 |
LLVM |
LLVM Documentation
|
|
| 4 |
1/24 |
SSA Optimizations cont. |
M8.11, 7.3; D6.2.4, 9.6.1; *C9.3; A4.4.4
|
|
| 5 |
1/29 |
Loop Optimizations |
*D9.6.2-6; M7.4, 14.1, 9.3 |
|
| 6 |
1/31 |
Dependency Analysis |
*A2.1-2; 3.1-2; M9.3-4; D11.3,4,6 |
1 due, 2 out |
| 7 |
2/5 |
Dependence Cont. |
*A3.3,4,6; M9.3,4; D11.6 |
|
| 8 |
2/7 |
CANCELLED |
|
|
| 9 |
2/12 |
Locality Optimization |
A9.1-3; *D11.1.4,11.2,11.5,11.10; M20.4; **Paper |
| |
| 10 |
2/14 |
Locality cont. |
|
Pick readings, 2 due, 3 out |
| 11 |
2/19 |
Paper Presentations by Seth and Dave |
|
|
| 12 |
2/21 |
Paper Presentations [srosenth, tjmurray] |
Adaptive Compilation GPUs |
|
| 13 |
2/26 |
Paper Presentations [pramanuj, msanphy] |
Pointer Analysis |
|
| 14 |
2/28 |
Paper Presentations [jcipar, ppoosank] |
Predicting Unroll Factors
Necula and Lee |
3 checkpoint due |
| 15 |
3/4 |
Register Allocation |
Linear Scan D8.8; A8.2; *C13.5; M16.3-5 |
|
| 16 |
3/6 |
Paper Presentation [saagarp] |
Transactional Memory
Multicore |
3 due; Initial Project Proposal due 3/7 |
| SPRING BREAK |
| 17 |
3/18 |
Local Scheduling |
A10.1,10.2.1-4; D10.1-10.3; C12.1-3 |
|
| 18 |
3/20 |
Software Pipelining |
A10.2.5; M17.4; D10.5* |
Final Project Proposal |
| 19 |
3/25 |
Dataflow and Interval Analysis |
M8.2-4,8.7.1,8.8*, D9.3,9.7.1-6 |
|
| 20 |
3/27 |
Partial Redundancy Elimination |
M13.3*,D9.5 |
|
| 21 |
4/1 |
Pointer Analysis |
M10* |
|
| 22 |
4/5 |
Shape Analysis |
Shape Analysis |
|
| 23 |
4/8 |
Power and Energy |
The Design, ...
for CPU Energy
Reduction
Compiler Opts... for Low Power ... |
|
| 24 |
4/10 |
Exam |
|
|
| 25 |
4/15 |
Project checkpoint presentations |
|
Project checkpoint |
| 26 |
4/17 |
No Class |
|
|
| 27 |
4/22 |
Power and Energy (2) |
Profile-Driven Energy Reduction in Network-on-Chips |
|
| 28 |
4/24 |
Parallel Compilers: Paradigm |
paradigm |
|
| 29 |
4/29 |
Speculative Multi-threading |
Compiler Optimization of Scalar Value Communication Between Speculative Threads |
|
| 30 |
5/1 |
StreamIt |
Exploiting Coarse-Grained Task, Data,
and Pipeline Parallelism in Stream Programs
|
|
| |
5/12 |
Project Poster Presentations |
|
|
| |
5/13 |
Projects Due |
|
|