[Lldb-commits] [PATCH] D139833: [LLDB][LoongArch] Add branch instructions for EmulateInstructionLoongArch

[Hui Li via Phabricator via lldb-commits]

lh03061238 added a comment.

In D139833#3997386 <https://reviews.llvm.org/D139833#3997386>, @DavidSpickett wrote:

> Is the further work in the commit message work you intended to include in this patch? I appreciate the status update might be useful for some but commit messages are for describing what's in a commit (among other things).
>
> If you think people might wrongly assume that this commit contains support for "floating-point branch instructions", I would write something more explicit. For example "note that this does not include <things>, that will come in a later patch".

The code submissions for the EmulateInstruction section refer to some ways of riscv and sometimes record further work.  You are right, should only describe what this patch does.

In D139833#3997409 <https://reviews.llvm.org/D139833#3997409>, @DavidSpickett wrote:

> This is all very repetitive, but that's to be expected. Whatever you can do to reduce boilerplate in future patches would be great. For example, can 32 and 64 bit variants of some instructions be emulated in the same way? Perhaps with a template parameter for the return types.
>
> This LGTM with a few comments you can do/not do/keep in mind for future patches.

This patch simply separates 32 and 64 bit processing, make sure the 64-bit works properly. During the continuous 
improvement of EmulateInstruction code,  the details or some ways to reduce repetition you mentioned will be given 
priority consideration.

Thank you very much for your detailed and reasonable suggestions.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D139833/new/

https://reviews.llvm.org/D139833