-
Notifications
You must be signed in to change notification settings - Fork 4.7k
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
Assertion failed 'change >= 0' #100350
Comments
Tagging subscribers to this area: @JulieLeeMSFT, @jakobbotsch |
The old regex IL could produce some pretty messy flow. I think this is exposing lack of a cyclic cap for the irreducible (improper) loop cases:
Seems like any realistic flow we'll see should have counts that top out at 10^12 or so, so perhaps I'll just stop if we see something that large. |
Another option would be (assuming we were in repair mode) switch to blend, perhaps a hard blend, and then retry ... worth considering if we value consistency over accuracy. |
FYI this was triggered by stress putting in random profile data, which was wildly inconsistent, and triggered a repair. Normally we would not see profile data like this in a Tier0 method.
|
In large methods with lots of irreducible loops we may find reconstructed counts reaching very large values. Since profile counts in practice won't ever be much larger than say 10^12, detect when reconstructed counts exceed this value, and stop the algorithm. We may eventually decide to rerun in "hard blend" mode where we intentionally limit the edge likelihood ranges. But this should do for now. Closes dotnet#100350.
In large methods with lots of irreducible loops we may find reconstructed counts reaching very large values. Since profile counts in practice won't ever be much larger than say 10^12, detect when reconstructed counts exceed this value, and stop the algorithm. We may eventually decide to rerun in "hard blend" mode where we intentionally limit the edge likelihood ranges. But this should do for now. Closes #100350.
In large methods with lots of irreducible loops we may find reconstructed counts reaching very large values. Since profile counts in practice won't ever be much larger than say 10^12, detect when reconstructed counts exceed this value, and stop the algorithm. We may eventually decide to rerun in "hard blend" mode where we intentionally limit the edge likelihood ranges. But this should do for now. Closes dotnet#100350.
In large methods with lots of irreducible loops we may find reconstructed counts reaching very large values. Since profile counts in practice won't ever be much larger than say 10^12, detect when reconstructed counts exceed this value, and stop the algorithm. We may eventually decide to rerun in "hard blend" mode where we intentionally limit the edge likelihood ranges. But this should do for now. Closes dotnet#100350.
https://dev.azure.com/dnceng-public/public/_build/results?buildId=620358&view=ms.vss-test-web.build-test-results-tab
libraries-jitstress test failure
System.Text.RegularExpressions.Tests Work Item
seemingly every platform
The text was updated successfully, but these errors were encountered: