Dmitry Vyukov

That's the child process that will exit. I don't immediately see why it should lead to any failures in the parent processes.

> @dvyukov do you have some more ideas? Nothing concrete.

We could also switch to own DWARF parsing: https://github.com/google/syzkaller/issues/4585#issuecomment-2007874590 After I switched objdump to manual binary parsing, addr2line is the last slow part. Manual parsing of all DWARF looks super...

We won't need CGUs nor compile units, we just map all callback PCs to file:line info. Basically scan whole DWARF and extract all PC -> file:line info.

> Do we want to call CleanPath() from readTextRanges? Since it's a hack already, I would say call whatever is necessary to make it work. > These same files are...

As far as I remember, such duping was not supposed to happen. Just opening visibility for the duped bug creates several problems: - upstream developers never seen the duped report,...

If you feel it's the right thing to do, let's remove restrictions on duping across stages and use canonical bug for accessibility.

Hard to comment, only dogfooding it for lots of cases can lead to an option that's not painful to use. I am also thinking about higher-level story. Nobody will access...

Are there any alternatives? Where do these commits come from? Coverage reports look like the lesser problem if commits disappear. It also means we report bugs on commits that nobody...

I remember we also discussed pushing all tested commits to a single git repo, which would preserve them for other developers, bisection, etc. How bad is the problem? Another possible...