-
Notifications
You must be signed in to change notification settings - Fork 2
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
Sylvan : Crash in Lib-BDD reconstruction #138
Comments
Could increase the stacksize through |
The only operation that is used during reconstruction of the input (serialized BDDs) is the ITE . Simultaneously, the BDD is built bottom-up, meaning the ITE should take O(1) . Hence, the issue is something much more subtle and weird. |
The issue lies in the fact, that Sylvan's garbage collection has the following undocumented assumption:
Hence, Sylvan breaks when the |
Partially resolves #138 by decreasing the number of concurrent BDD roots. Yet, this does not as such solve the problem but only push it to further. Sylvan will still crash, if a single level is wider than the Lace stack (or prior, if BDD nodes from multiple levels are still alive).
Partially resolves #138 by decreasing the number of concurrent BDD roots. Yet, this does not as such solve the problem but only push it to further. Sylvan will still crash, if a single level is wider than the Lace stack (or prior, if BDD nodes from multiple levels are still alive).
Merging that PR does not fully resolve the issue - just push the boundary of what Sylvan can deserialize. Closing this was not the intention. |
Sylvan seems to crash due to a LACE Stack overflow during the reconstruction of the input for the RelProd benchmark. This would presumably also happen in the Apply benchmark.
The text was updated successfully, but these errors were encountered: