These are already skipped on Darwin because they cause build bot failures. Both on the build bots as well as in our testing we have seen a number of these tests fail and hang. This change skips the failing/hanging tests on Linux and also fixes one of the test - the test needs the thread library to build.
Details
Diff Detail
- Repository
- rL LLVM
Event Timeline
packages/Python/lldbsuite/test/tools/lldb-vscode/attach/TestVSCode_attach.py | ||
---|---|---|
50 ↗ | (On Diff #162430) | "is well as" -> "as well as" (same for the other comments). |
packages/Python/lldbsuite/test/tools/lldb-vscode/attach/TestVSCode_attach.py | ||
---|---|---|
50 ↗ | (On Diff #162430) | Apparently I can't spell today. Thanks for catching this! |
Looks like we *really* need to figure out what the underlying bug is here, because there's not many platforms left after this patch.
This patch is fine as a stopgap, but we can't have a feature like this be in the source code while being virtually untested, that's just asking for trouble.
Greg, were you able to reproduce this?
Stella, do you have any logging available that could help Greg in tracking down the root cause?
It's fairly easy for me to reproduce, so if Greg has time to investigate, I can collect the logs he needs.
I spent some more time on this to isolate the failing tests (and reduce the number of disabled tests). I was also able to figure out what one of the failures is - the test needs the thread library to build. The update change reflects that.