tgs-sc wrote: > Maybe something should be skipping forward to try and re-sync.
I think that, in general, a situation is always possible in which it is unclear how to decode the instruction length, so it is hard to imagine how this can be improved. Now, there are two phases: decode and disassemble. If the decode fails, disassembling makes no sense, and we will see a newly added warning. If the disassemble fails, we will see an `<unknown>` opcode. > Jonas is probably busy preparing for US LLVM Dev now, and at the event all > next week. > > You removed the "warning: " prefix from the message and that seemed to be the > crux of his comment. You're now following the local pattern at least, like > the "invalid target" message below it. > > So this LGTM and any other feedback can come post-commit. @DavidSpickett, ok, so can you probably merge this as I don't have commit access? https://github.com/llvm/llvm-project/pull/164413 _______________________________________________ lldb-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
