[PATCH] Docs: Document that taint analyzer checker stops other checkers

David Malcolm dmalcolm@redhat.com
Tue Mar 22 23:15:02 GMT 2022


On Tue, 2022-03-22 at 23:43 +0530, Avinash Sonawane via Gcc-patches
wrote:
> Hello!
> 
> $ man gcc says that "Some checkers are disabled by default (even with
> -fanalyzer), such as the "taint" checker that implements
> -Wanalyzer-tainted-array-index, and this option is required to enable
> them." but still lists the -Wanalyzer-tainted-* checkers under
> -fanalyzer saying "Enabling this option effectively enables the
> following warnings:"
> 
> This patch comments out the tainted-* checkers so that when
> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103533 gets fixed we can
> remove the comment and list them under -fanalyzer
> 
> Also, the above wording suggests that `-fanalyzer
> -fanalyzer-checker=taint` enables the taint checkers in addition to
> the default checkers but currently, enabling -fanalyzer-checker=taint
> stops other checkers. This patch documents that too.
> 
> Please find the attached patch.
> 
Thanks for the patch.  I think it's a definite improvement in
readability for the end-user.  

The option and the docs refer to "checkers" e.g.
  "currently, -fanalyzer-checker=taint stops all other checkers"
but these checkers are an implementation detail of the analyzer; not
all of the analyzer warnings relate to a checker.

So it would be even better if the text listed the warnings that are
affected by this: basically all of the ones that are implemented in
checkers i.e. analyzer/sm-*.cc (apart from sm-taint, of course).

I can put that together if need be (but I'm about to sign off for today
and only just saw this patch).

Ideally, of course, we'd fix bug 103533, but that's GCC 13 material at
this point :)

Thanks again for the patch
Dave



More information about the Gcc-patches mailing list