[PATCH] D117717: [clang] Ignore -fconserve-stack

Aaron Ballman via Phabricator via cfe-commits cfe-commits at lists.llvm.org
Thu Jan 20 06:43:28 PST 2022

aaron.ballman added reviewers: steakhal, NoQ, martong, Szelethus.
aaron.ballman added a comment.
Herald added a subscriber: rnkovacs.

Adding some reviewers from the Clang Static Analyzer side of things.

In D117717#3256243 <https://reviews.llvm.org/D117717#3256243>, @nathanchance wrote:

> Should we just tell people to run `scan-build` only with `clang`? How `ccc-analyzer` works seems weird to me but I assume some of the flag machinery that is present here is for that reason?

FWIW, I don't have good impressions of using scan-build. It attempts to interpose the analysis compilation with the real execution of the compiler, but it has effectively no intelligence with that interposition and ends up passing along all the options to Clang. Other tools I've worked on in this realm do far more complicated modelling of compiler options to make that interposition work for arbitrary compilers. I'm not convinced scan-build works for anything but Clang or extremely simple GCC invocations (but doesn't work very well at all with other compilers like ICC, etc because the compiler options get farther away from what Clang understands).

I'm not certain whether that warrants telling people it only works with Clang or not, but I (personally) wouldn't be opposed to that. (I also wouldn't be opposed to ridding ourselves of this script -- it's the only Perl script we have in the project which is a significant pain for those of us on Windows.)

  rG LLVM Github Monorepo



More information about the cfe-commits mailing list