[Lldb-commits] [lldb] [lldb] Flush delayed breakpoints before eagerly changing one (PR #195815)

[David Spickett via lldb-commits]

DavidSpickett wrote:

Latest build has a log you can look at - https://lab.llvm.org/buildbot/#/builders/18/builds/25884

I see more tests failing, @luporl do you know which ones are caused by this PR or is it all mixed up at this point?

This specific test looks like an Arm/Thumb mode difference:
```
======================================================================
FAIL: test_eager_breakpoints_dwarf (TestDelayedBreakpoint.TestDelayedBreakpoint)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/home/tcwg-buildbot/worker/lldb-arm-ubuntu/llvm-project/lldb/packages/Python/lldbsuite/test/lldbtest.py", line 2032, in test_method
    return attrvalue(self)
  File "/home/tcwg-buildbot/worker/lldb-arm-ubuntu/llvm-project/lldb/test/API/functionalities/breakpoint/delayed_breakpoints/TestDelayedBreakpoint.py", line 79, in test_eager_breakpoints
    self.assertIn(addr, breakpoint_lines)
AssertionError: '1234567' not found in 'python3.10       <  16> send packet: $Z0,1234566,2#7fpython3.10       <  62> send packet: $jMultiBreakpoint:{"breakpoint_requests":["Z0,34c0618,4"]}]#46'
Config=arm-/home/tcwg-buildbot/worker/lldb-arm-ubuntu/build/bin/clang
```
'123456**7**' not found in 'python3.10       <  16> send packet: $Z0,123456**6**,2#

I assume 1234567 is an arbitrary fake address, and something in the Arm code is stripping the bottom bit because that's the Arm/Thumb mode bit. As in: it's not a real PC in either mode, so the real breakpoint address is one less.

https://github.com/llvm/llvm-project/pull/195815