[lldb-dev] Enabling tests on the Windows LLVM buildbot

Ed Maste via lldb-dev lldb-dev at lists.llvm.org
Wed May 18 10:49:21 PDT 2016

On 18 May 2016 at 05:08, Pavel Labath via lldb-dev
<lldb-dev at lists.llvm.org> wrote:
> Sounds reasonable. I'd like to add a clarifying point (2.5): If you
> have added a new test, and this test fails on some other platform AND
> there is no reason to believe that this is due to a problem in the
> test (like the python3 bytes thingy, etc.), then you can just xfail
> the test for the relevant architecture is fine.

That sounds reasonable to me.

I hope to re-enable tests on the FreeBSD buildbot shortly as well. I
have a "temporary" build-only buildbot I put into service when the
previous ones needed to be decommissioned.

Since FreeBSD's currently the only platform still using the old-style
POSIX in-process debug support it's quite likely we could run into a
failure when a test is added. I'd prefer to have the test marked XFAIL
on FreeBSD with a bug report (or at least a post to the mailing list)
than for it to be backed out pending investigation.

A bit of a tangent but for reference, on FreeBSD 10 I currently see
the following set of undesired test results:

ERROR: test_with_run_command_dwarf
ERROR: test_with_run_command_dwarf
ERROR: test_with_run_command_dwarf
ERROR: [EXCEPTIONAL EXIT 10 (SIGBUS)] test_python_os_plugin_dwarf
UNEXPECTED SUCCESS: test_and_run_command_dwarf
UNEXPECTED SUCCESS: test_and_run_command_dwarf
TIMEOUT: test_asm_int_3
TIMEOUT: test_with_dsym_and_python_api_dwarf

More information about the lldb-dev mailing list