On Fri, Jan 24, 2020 at 12:24 AM Peter Zijlstra <peterz@xxxxxxxxxxxxx> wrote: > > On Thu, Jan 23, 2020 at 11:07:59AM -0800, Nick Desaulniers wrote: > > > Good thing it's the variable being modified was not declared const; I > > get spooked when I see -Wdiscarded-qualifiers because of Section > > 6.7.3.6 of the ISO C11 draft spec: > > > > ``` > > If an attempt is made to modify an object defined with a > > const-qualified type through use > > of an lvalue with non-const-qualified type, the behavior is undefined. > > If an attempt is > > made to refer to an object defined with a volatile-qualified type > > through use of an lvalue > > with non-volatile-qualified type, the behavior is undefined.133) > > > > 133) This applies to those objects that behave as if they were defined > > with qualified types, even if they are > > never actually defined as objects in the program (such as an object at > > a memory-mapped input/output > > address). > > ``` > > > > Which is about the modification of a const-declared variable (explicit > > UB which Clang actively exploits), > > Just for curiosity's sake. What does clang actually do in that case? > https://bugs.llvm.org/show_bug.cgi?id=42763 I was playing around with this in godbolt but couldn't quickly come up with a simple reproducer. IIRC, I've fixed maybe 3 instances of this recently in code though. -- Thanks, ~Nick Desaulniers
