[Lldb-commits] [PATCH] D66858: POSIX DYLD: add workaround for android L loader

[Pavel Labath via Phabricator via lldb-commits]

labath added a comment.

My other concerns about this patch are:

- the base_addr field is the *difference* between the preferred load address encoded in the object file, and the actual load address. This means that this workaround will fire every time the object is loaded at it's preferred load address, which is definitely not what this patch advertises to do...
- it looks like it should be possible to test this using the "gdb-client" test suite. I am hoping that it all it would take is to mock the appropriate responses to the `qXfer:libraries` and `qFileLoadAddress` packets. We should at least try to do that...
- just because this patch was fast-tracked into the swift fork of lldb without proper scrutiny (it seems), it does not mean it should be fast-tracked here (in fact, I would say the opposite is true...)


Repository:
  rLLDB LLDB

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D66858/new/

https://reviews.llvm.org/D66858