[Lldb-commits] [PATCH] D27088: [LLDB][MIPS] Fix TestLldbGdbServer failure for MIPS

[Greg Clayton via Phabricator via lldb-commits]

clayborg added a comment.

We should be able to handle this being thread specific. Each stop reply packet and qThreadStopInfo (asks for a complete stop reply packet for each thread), or the the "jThreadsInfo" packet (see "$trunk/docs/lldb-gdb-remote.txt" for detail) has the ability to return key/value pairs that are specific to each thread. So this is where this key would be specified, once for each thread. Let me know if anything isn't clear.

Overall we really shouldn't ever see anything MIPS specific in generic code. It is ok to have MIPS specific code in say "RegisterContextMIPSXXX" classes, but not in general classes. We really want to abstract out anything arch specific into arch agnostic code.


https://reviews.llvm.org/D27088