Difference between revisions of "Google Summer of Code"
From gem5
Line 1: | Line 1: | ||
+ | # Build a direct execution CPU model based on the Linux Kernel Virtual Machine | ||
+ | #* http://kvm.qumranet.com/kvmwiki | ||
+ | # Parallelize M5 | ||
+ | #* Use the Wisconsin Wind Tunnel as a guide | ||
+ | # Crossbar network | ||
+ | # Mesh network | ||
+ | # Directory Protocol | ||
# Real F/S In-order core | # Real F/S In-order core | ||
#* Kevin has one that works to some degree in SE, it doesn't have functional units yet, but it does have variable latency stages and such | #* Kevin has one that works to some degree in SE, it doesn't have functional units yet, but it does have variable latency stages and such | ||
Line 6: | Line 13: | ||
# Instruction page decode cache | # Instruction page decode cache | ||
#* Kinda goes along with the fast simple core, but i think it's a more manageable task. Clearing it is as simple as looking for flush instructions in sparc and the correct pal trap in alpha | #* Kinda goes along with the fast simple core, but i think it's a more manageable task. Clearing it is as simple as looking for flush instructions in sparc and the correct pal trap in alpha | ||
− | |||
− | |||
# SMT | # SMT | ||
#* Fix O3 bugs/ Fix ROB Units | #* Fix O3 bugs/ Fix ROB Units | ||
#* It's a huge pile of code to understand before you get anywhere if they get that far | #* It's a huge pile of code to understand before you get anywhere if they get that far | ||
− | |||
− | |||
# Are there any other benchmarks we want? | # Are there any other benchmarks we want? | ||
#* That they could possible make work? | #* That they could possible make work? |
Revision as of 11:48, 11 March 2008
- Build a direct execution CPU model based on the Linux Kernel Virtual Machine
- Parallelize M5
- Use the Wisconsin Wind Tunnel as a guide
- Crossbar network
- Mesh network
- Directory Protocol
- Real F/S In-order core
- Kevin has one that works to some degree in SE, it doesn't have functional units yet, but it does have variable latency stages and such
- Korey has one he did at MIPS, I don't know about it's features, but it's SE only as well
- Fast Simple Core
- This largely disregards all the compartmentalization we've done. It touches a lot of things, so the time to start being productive would be high
- Instruction page decode cache
- Kinda goes along with the fast simple core, but i think it's a more manageable task. Clearing it is as simple as looking for flush instructions in sparc and the correct pal trap in alpha
- SMT
- Fix O3 bugs/ Fix ROB Units
- It's a huge pile of code to understand before you get anywhere if they get that far
- Are there any other benchmarks we want?
- That they could possible make work?
- Devices?
- Validate the DRAM model?
- Sampling/checkpointing/restarting
- Testing, fixing, etc... Not exactly exciting work
- Write a PLI interface to connect Verilog CPUs to the memory system.
- Sampling/fast-forwarding techniques (making sure ours works, maybe adding in some new ones)
- Different cache models (different replacement policies, etc.; would allow them to do some research into it, maybe get some work done for Lisa)
- Different prefetcher models (expand on what Ron has, also can do some research into it)
- Flash memory device model? (seems popular nowadays)