steakhal planned changes to this revision. steakhal added a comment. In D121214#3372091 <https://reviews.llvm.org/D121214#3372091>, @whisperity wrote:
> In D121214#3369871 <https://reviews.llvm.org/D121214#3369871>, @steakhal > wrote: > >> Drop the alias-related changes and preserve the note in the >> `bugprone-shared-ptr-array-mismatch.rst` stating this relationship with the >> cert rule? >> If we do it like that, then this will be again NFC. > > I would suggest definitely doing that. Perhaps with a bit more emphasis on > this one catching the `shared_ptr` case only. Okay, I'll pick this direction. > If you could create the check for the `unique_ptr` case, that would also be > great. We will put it on the roadmap, but no promises :D > I think some of MEM51-CPP is covered by the Static Analyser, right? (Things > like `free()`ing a `new`-created pointer.) Exactly. CSA checkers `unix.MismatchedDeallocator`,`cplusplus.NewDelete`,`cplusplus.NewDeleteLeaks` together can catch all but the last two cases. > In D121214#3369871 <https://reviews.llvm.org/D121214#3369871>, @steakhal > wrote: > >> How shall I proceed? > > Let's ask @aaron.ballman or @njames93 with regards to that. The long-term > solution would be implementing the `1-to-N` check aliasing in the > infrastructure, but I understand it might be an out-of-scope work right now. This is definitely out of scope for me ATM. > However, I am also in favour of creating a mapping of "partially implemented > guidelines" in the documentation somewhere... maybe in the `list.rst`, maybe > on a separate page. There, we could start documenting cases like this one... > It's less direct than an alias, but more accessible to users who wish to > cover as many rules of thumb as possible than having to iterate through > hundreds of distinct documentation pages for the checks and trying to deduce > //"Hey, so this one makes me catch some of that guideline!"//. We should also consider the documentation hell, and carefully design a solution that mitigates the already existing burden. It's already hard to keep everything in sync. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D121214/new/ https://reviews.llvm.org/D121214 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
