Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Binary file added AMDGPU/1-908.gpubin
Binary file not shown.
Binary file added AMDGPU/2-90a.gpubin
Binary file not shown.
Binary file added AMDGPU/3-940.gpubin
Binary file not shown.
9 changes: 9 additions & 0 deletions AMDGPU/README.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,9 @@
# Test Binaires for AMDGPU

| File Name | Architecture | Source from | Original Intend Test |
| :------------ | :----------: | :---------: | -----------------------------: |
| 1-gfx908.gpubin | gfx908 | John | Indirect Control Flow Analysis |
| 2-gfx90a.gpubin | gfx90a | John | Indirect Control Flow Analysis |
| 3-gfx940.gpubin | gfx940 | John | Indirect Control Flow Analysis |
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

  1. It's interesting how different the file sizes are. Is there something special about the 908 one to make it so much smaller?
  2. Are there tests to go into Dyninst to use these?
  3. Is 'John' Mellor-Crummey?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

  1. It's interesting how different the file sizes are. Is there something special about the 908 one to make it so much smaller?

These were binaries which previously trigger bugs in Dyninst, reported by John.

  1. Are there tests to go into Dyninst to use these?

I personally have an example that prints out the indirect control flow targets, and I used them to check if anything breaks.

  1. Is 'John' Mellor-Crummey?

Yes, I was being lazy, I can probably fix this.
TBH, I can probably find the GitHub issues / PR that were associated with the fix for these binaries, when I’m not lazy.