On Wed, Nov 02, 2022 at 06:54:22PM +0100, Paolo Bonzini wrote: > On 10/31/22 21:54, Peter Zijlstra wrote: > > > PeterZ is contending that this isn't actually undefined behavior given how the > > > kernel is compiled[*]. That said, I would be in favor of replacing the open-coded > > > shift with BIT() to make the code a bit more self-documenting, and that would > > > naturally fix this maybe-undefined-behavior issue. > > > > > > [*]https://lore.kernel.org/all/Y1%2FAaJOcgIc%2FINtv@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx > > I'm definitely in favour of updating this code; both your suggestion and > > hpa's suggestion look like sane changes. But I do feel that whatever > > UBSAN thing generated this warning needs to be fixed too. > > > > I'm fine with the compiler warning about this code -- but it must not > > claim undefined behaviour given the compiler flags we use. > > Yes, the compiler is buggy here (see old bug report for GCC, now fixed, at > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=68418). > > I cannot even reproduce the problem with the simple userspace testcase > > #include <stdlib.h> > int main(int argc) { > int i = argc << 31; > exit(i < 0); > } > > on either GCC 12 or clang 15. Perhaps we should have the UBSAN splat include the compiler-version used... because clearly someone is using ancient crap here.
