[PATCH] D148355: [analyzer] Fix comparison logic in ArrayBoundCheckerV2

Balázs Benics via Phabricator via cfe-commits cfe-commits at lists.llvm.org
Tue Apr 25 05:03:29 PDT 2023


steakhal accepted this revision.
steakhal added a comment.
This revision is now accepted and ready to land.

In D148355#4294798 <https://reviews.llvm.org/D148355#4294798>, @steakhal wrote:

> In D148355#4294738 <https://reviews.llvm.org/D148355#4294738>, @donat.nagy wrote:
>
>> @steakhal I marked a few comments as Done (I accidentally missed some when I was creating the most recent patch) and now the only not-Done thing is the followup commit for refactoring the optionalness of RegionRawOffsetV2. Do you see anything else to do?
>
> I think it's good. I've just scheduled the measurement now for 180+ OSS projects. Stay tuned!

The results are all good. The effect is pretty much equal to the effect of our implementation except for a handful of absent reports - and we are talking about at most 20 issues in total.
I must say, the report diff looks much better than I anticipated.
Thank you for working on this. Feel free to merge this change whenever you want. Thanks.

BTW the issues themselves look terrible. One cannot understand and track the value of the index and the assumed buffer size etc.
So the checker will still remain in alpha.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D148355/new/

https://reviews.llvm.org/D148355



More information about the cfe-commits mailing list