labath added a comment.
In addition to that, it breaks tests with ubsan. :)
I raised a similar concern in one of the other patches. I'm of the opinion that
it's not worth pretending like the scripted processes (platforms, etc.) are
plugins. I don't think that it can be achieved without some sort of a leakage,
and I don't think it makes sense given that the part that's really interesting
to have pluggable is being able to plug a different scripting language into the
scripted process. I don't exactly have an opinion on this metadata argument,
but it definitely looks sub-optimal.
Given this uncertainty, and the ubsan problem, maybe we can revert this for the
time being?
================
Comment at: lldb/unittests/Platform/PlatformTest.cpp:90
+ []() { Debugger::Initialize(nullptr); });
+ list = new PlatformList(*m_debugger_sp.get());
+ }
----------------
m_debugger_sp is never initialized. I guess this kinda works because the test
does not actually dereference it, but UBSAN lights it up straight away.
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D139249/new/
https://reviews.llvm.org/D139249
_______________________________________________
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
