https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92486
--- Comment #12 from rguenther at suse dot de <rguenther at suse dot de> --- On Thu, 14 Nov 2019, ch3root at openwall dot com wrote: > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92486 > > --- Comment #9 from Alexander Cherepanov <ch3root at openwall dot com> --- > > Now as an exercise build a complete testcase for the DSE issue above. > > Source code: > > ---------------------------------------------------------------------- > #include <string.h> > #include <stdio.h> > > struct s { > char c; > int i; > }; > > __attribute__((noinline,noclone)) > void f(struct s *p, struct s *q) > { > struct s w; > > memset(&w, 0, sizeof(struct s)); > w = *q; > > memset(p, 0, sizeof(struct s)); > *p = w; > } > > int main() > { > struct s x; > memset(&x, 1, sizeof(struct s)); > > struct s y; > memset(&y, 2, sizeof(struct s)); > > f(&y, &x); > > for (unsigned char *p = (unsigned char *)&y; p < (unsigned char *)&y + > sizeof(struct s); p++) > printf("%d", *p); > printf("\n"); > } > ---------------------------------------------------------------------- > > Results: > > ---------------------------------------------------------------------- > $ gcc -std=c11 -pedantic -Wall -Wextra test.c && ./a.out > 11111111 > $ gcc -std=c11 -pedantic -Wall -Wextra -O3 test.c && ./a.out > 12221111 > ---------------------------------------------------------------------- > > gcc x86-64 version: gcc (GCC) 10.0.0 20191114 (experimental) > > > But from the C standard POV this case is much more clear: there is no problem > as stores into a struct make its padding unspecified (C11, 6.2.6.1p6). OTOH > this sample demonstrates the problem with trunc, so it could be more > convenient > for testing. Yeah, I think the C standard is clear here but the question is about GIMPLE IL semantics where making things "unspecified" is a bit awkward.
