Is there any concern if I change the make "check-all" target to run the lldb test suite if lldb is checked out/built?
Currently "make check-all" seems to run the llvm tests and a those from a few other projects (clang, clang-extras) but running the lldb tests requires manually invoking the dotest.py script, which is somewhat obscure (and annoying).
The lldb test suite takes a really loooooong time, even with a pretty fast computer (it's single threaded, too).
I don't have a big problem with it, but other people may have.
Hi, thanks for the heads up. I can confirm that on Mac OS X the tests seem to take a really long time.
On Linux though, the tests are running much faster; a debug build of lldb can be tested (with two test configurations x86_64-clang and x86_64-gcc) finishes in around 5min. We do have some differences though; there is no debugserver on Linux, and a handful of test cases are being skipped).