-
Notifications
You must be signed in to change notification settings - Fork 88
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
SIGSEGV: running basic/clone looping with 2 ethreads and OE debug allocator #740
Comments
The cancelbuf field is gone from the lthread structure in the cleaned up layering. This may be a UAF of something allocated with the oe malloc allocator (so something aliases the lthread), or it may be a bug in code that is about to be deleted. |
I am hitting this with every run right now of the clone-loop test |
Are you running with lthread UAF checks enabled? |
I am. UAF, OE Debug Heap, DEBUG and LKL_DEBUG. Running on 4-core machine with 2 ETHREADS and running in the debugger |
@paulcallen are you actively working on this? Should I assign you to it? |
@SeanTAllen I am not sure I have the deep understanding of this code yet to be able to fix this. |
@paulcallen can you leave your complete build command for sgx-lkl so I can try to reproduce on my machine? |
@paulcallen is this issue now outdated and can be closed? The code in question above has been removed due to the relayering work. |
Since we haven't had further reports of this (and the clone test is part of the CI and passes), I am closing this issue. |
I enabled OE debug allocator (-DUSE_DEBUG_MALLOC=ON) for building OE
Running on 4-core ACC VM with 2 ETHREADS
Running test tests/basic/clone with command
sudo make DEBUG=1 run-hw-gdb-clone-loop
Crash is here:
All threads:
cannot include full log as it is too big!
The text was updated successfully, but these errors were encountered: