[Lldb-commits] [PATCH] D90757: [lldb] Enable FreeBSDRemote plugin by default and update test status

[Ed Maste via Phabricator via lldb-commits]

emaste added inline comments.


================
Comment at: lldb/packages/Python/lldbsuite/test/dotest.py:955
 
-    # Don't do lldb-server (llgs) tests on anything except Linux and Windows.
+    # Don't do lldb-server (llgs) tests on platforms not supporting it.
     configuration.dont_do_llgs_test = not (
----------------
mgorny wrote:
> emaste wrote:
> > It's a shame that these variables and the comments are inverted sense / double negatives, but not an issue in this patch.
> Yes, I had to look at it for a while to make sure I'm doing it right. I'll change them in a followup commit.
Thanks.


================
Comment at: lldb/test/API/commands/register/register/register_command/TestRegisters.py:31
     @skipIf(archs=no_match(['amd64', 'arm', 'i386', 'x86_64']))
-    @expectedFailureNetBSD
+    @expectedFailureAll(oslist=["freebsd", "netbsd"])
     def test_register_commands(self):
----------------
mgorny wrote:
> emaste wrote:
> > there's no (existing) PR for the failure?
> No. I'm planning to file bugs during the next stage, after trying to fix the trivial ones, if that's ok with you.
Yes absolutely. I was just curious because this was already failing for NetBSD


================
Comment at: lldb/test/API/functionalities/inferior-crashing/TestInferiorCrashingStep.py:54-55
     # intended IMHO.
     @skipIfLinux
-    @skipIfFreeBSD
-    @expectedFailureNetBSD
+    @expectedFailureAll(oslist=["freebsd", "netbsd"])
     def test_inferior_crashing_expr_step_and_expr(self):
----------------
labath wrote:
> I'm pretty sure the root cause here is the same for net/free bsd as it is for linux (it comes down to macos catching the "crashes" specially, before they even get turned to a SEGV -- something that's not possible elsewhere). I marked it skip because that's not something we should support, ever. I don't care that much which decorator (skip vs. xfail) is used here, but I think they should be consistent.
Sounds reasonable, we may want to change the comment above to make it clear this is an explicit decision.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D90757/new/

https://reviews.llvm.org/D90757