[PATCH] D83152: llvm-nm: add flag to suppress no symbols warning

Fangrui Song via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Thu Feb 11 12:01:08 PST 2021


MaskRay added a comment.

In D83152#2556189 <https://reviews.llvm.org/D83152#2556189>, @jhenderson wrote:

> In D83152#2555229 <https://reviews.llvm.org/D83152#2555229>, @kastiglione wrote:
>
>> What, if any, is the community opinion on hidden flag use? Can this be a hidden flag and be done with it? I assume means hidden means no forward guarantees, rely on it at your own discretion.
>
> My opinion: in this context, I don't think it buys us enough, but I'm not opposed to hidden flags in general. I don't know what the LLVM policy on hidden flags generally is though - more often than not, I've found hidden flags to be hidden accidentally!
>
>>> I don't see how `--no-warning-for-no-symbols` is better than `--quiet`, but `--quiet` is an option name they could use for other purposes.
>>
>> seems like you answer your own question :)
>
> Exactly: if `--quiet` has the risk of being used for something else, we should avoid using it (just the same as for short single-letter aliases). If on the other hand, someone wants to coordinate with GNU and/or they happen to implement `--quiet` to cover this case, we can remove the `--no-warning-for-no-symbols` option at that point, and direct users towards `--quiet` instead. I don't think it should be a requirement to do this coordination, especially given how easy it would be to migrate in this case.

I don't agree. If the coordination is done first, we have a potential to use `--quiet` to right away. Then we don't need to have a temporary `--no-warning-for-no-symbols` which will eventually be removed (and potential to break users).
The `no symbols` has caused us lots of trouble (see the history). I imagined that binutils folks would have some opinions on it. Given the contentious previous discussions, I think it is entirely justified to ask them for an opinion.
I have said in some other reviews, their objection does not necessarily block us to add an option, as we could get a promise that they will not add an option to deliberately make us incompatible.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D83152/new/

https://reviews.llvm.org/D83152



More information about the llvm-commits mailing list