User:Agebhard

2022

Week 1 (5/31 - 6/3)

Tuesday (5/31)

• Attended REU orientation [2hrs]
• Out for the rest of the day

Wednesday (6/1)

• Began to merge branches of past semesters of XINU work [2hrs]
• Attended REU meeting on "How to conduct research" [1.5hrs]
• Lab meeting [3hrs]

Thursday (6/2)

• Setup the RISC-V 64-bit compiler and QEMU [1hr]
• Read and took notes on the RISC Unprivileged Spec [6hrs]

Friday (6/3)

• Work on writing start.S for RISC-V [2 hrs]
• Team Lunch [2 hrs]
• Team meeting [2 hrs]

Week 2 (6/6 - 6/10)

Monday (6/6)

• Finished rough draft of start.S [1 hr]
• Wrote serial driver [5 hrs]

Tuesday (6/7)

• Began implementing context switch [5 hrs]

Wednesday (6/8)

• Attended REU group meeting/lunch [2 hrs]
• Discussed project roadmap with Dr. Brylow and Jack [3 hrs]

Thursday (6/9)

• Finished context switch implementation [5 hrs]
• Tested context switch [1 hr]

Friday (6/10)

• Minor changes to ensure context switch works with 8+ arguments [1 hr]
• Attended team lunch [2 hrs]
• Attended team meeting to discuss weekly progress with the group [2 hrs]

2020

Week 1 (6/1 - 6/5)

• Attended Data Science Bootcamp
• Met daily with Dr. Brylow, Alex R., John A., and Rade L. to discuss project ideas for this summer
• Read the following papers
  • Using Embedded Xinu and the Raspberry Pi 3 to Teach Operating Systems (not yet published)
  • Using Embedded Xinu and the Raspberry Pi 3 to Teach Parallel Computing in Assembly Programming
  • XinuPi3: Teaching Multicore Concepts Using Embedded Xinu
• Attended Dr. Brylow’s talk on good research practices
• Worked on fixing the mailbox system in Xinu for homework assignment #9

Week 2 (6/8 - 6/12)

• Completed 3 CITI Sessions for RCR
• Attended Dr. Brylow's RCR Training
• Attended Dr. Brylow's Technical Writing discussion
• Read most of the xv6 O/S book
  • Looked into how xv6 uses locks and synchronization primitives
• Read the following papers
  • Implementing Lock-Free Queues
  • A Lock-Free Multiprocessor OS Kernel
• Continued to work on fixing the mailbox system in Xinu for homework assignment #9
  • I would like to discuss this more with the team next week

Week 3 (6/15 - 6/19)

• Looked into core atomic operation we need in Xinu
• Submitted a PR for a race condition semalloc
• Audited the use of synchronization primitives in Xinu
• Read the following papers
  • Wait-Free Synchronization
  • Teaching Operating Systems Using Code Review
  • Xest: an automated framework for regression testing of embedded software
  • Hands-on networking labs with embedded routers
• Weekly Xinu Team Meetings
• Met with Dr. Brylow, John A, and Rade L to discuss the Wait-Free Synchronization paper

Week 4 (6/22 - 6/26)

• Attempted to implement a lock-free array-based queue in XINU
• Submitted a PR for general code cleanup in various areas of XINU
• Worked on draft of presentation to give next week to fellow REU students
• Read the following papers
  • Simple, fast, and practical non-blocking and blocking concurrent queue algorithms
  • (Dis)Advantages of Lock-free Synchronization Mechanisms for Multicore Embedded Systems
  • Efficient agglomerative hierarchical clustering
• Weekly Xinu Team Meetings
• Weekly REU Meeting

Week 5 (6/29 - 7/3)

• Created presentation to give to fellow REU students and mentors regarding my current progress
  • Delivered this presentation on Wednesday 5/1
• Researched implementing a dynamic lock-free linked list based queue by Maged Michael and Michael Scott
  • Ran into issues with memory reclamation
• Read the following papers
  • Simple, fast, and practical non-blocking and blocking concurrent queue algorithms
  • Hazard Pointers: Safe Memory Reclamation for Lock-Free Objects
  • Chapters 5 and 6 of xv6: a simple, Unix-like teaching operating system
• Weekly Xinu Team Meetings

Week 6 (7/6 - 7/10)

• Studied XINU's USB subsystem
• Started to convert shell (xhs) to multicore
• Read the following paper:
  • mTCP: a Highly Scalable User-level TCP Stackfor Multicore Systems
• Weekly Xinu Team Meetings

Week 7 (7/13 - 7/17)

• Converted XINU's shell and device infrastructure to multicore
  • Merged into main branch
• Started to work on poster
• Weekly Xinu Team Meetings

Week 8 (7/20 - 7/24)

• Replaced current atomic primitives to use compare and swap
  • Merged into main branch
• Finished first draft of poster
• Started to convert Xinu's USB subsystem to support multicore
• Created a code map of the USB subsystem
• Read the following paper:
  • Fixing deadlocks via lock pre-acquisitions
• Weekly Xinu Team Meetings

Week 9 (7/20 - 7/24)

• Finished poster
• Began revising/editing the paper
• Weekly Xinu Team Meetings

Week 10 (8/3 - 8/7)

• Presented poster
• Finished paper
• Submitted materials