On 20/05/2026 13:18, Andrew Pinski via Gcc wrote:
On Wed, May 20, 2026 at 7:30 AM Jonathan Wakely via Gcc <[email protected]> wrote:
The false positive rate has been unacceptably high for years, with
very little progress on reducing that rate.
Is the false positive rate so bad that it reduces the the utility of the
true positives that potentially get caught through -Wall? It's not like
all code is perfect now and that the true positives are no longer
valuable enough to keep in -Wall.
With the introduction of predictively devirtualization that can
include more than one candidate, the warnings are way too late. and
have become useless.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122197 is one example (it
blocks a few others too).
So either we start marking whole sections not to warn (while inlining)
or we disable the warnings.
Since I don't see an easy solution to mark while inining, I vote for
removing them from -Wall (and NOT adding them to -Wextra).
Alternatively, could this also be an indication that we easily let
transformations lose original program context and that when a pass does
this (e.g. address aliasing), we should have it try harder retain the
lost information somehow?
The middle end bounds warnings aren't the only ones that struggle with
this problem; objsz has similar issues and we have hacks to accommodate
for some of it, mainly because it directly affects codegen. We still
have the ugly early_objsz hack that doesn't quite work and needs a real
solution to recognize subobjects.
Thanks,
Sid
