On Mon, Aug 21, 2017 at 6:55 AM, Joseph Myers wrote: > On Mon, 21 Aug 2017, H.J. Lu wrote: > >> On Mon, Aug 21, 2017 at 6:23 AM, Joseph Myers wrote: >> > On Sun, 20 Aug 2017, H.J. Lu wrote: >> > >> >> @@ -101,7 +101,7 @@ do_test (void) >> >> >> >> if (STRLEN (&adr[outer]) != (size_t) (inner - outer)) >> >> { >> >> - printf ("%s flunked for outer = %d, inner = %d\n", >> >> + printf ("%s flunked for outer = %Zd, inner = %Zd\n", >> > >> > I don't think we should be using legacy %Z in any new code. Use C99 %zu >> > for size_t (%zd only if the argument is the corresponding signed type >> > rather than size_t itself). >> > >> >> There are some %Zd in string/stratcliff.c. Should they be changed? >> Since All of them are size_t, I will replace %Zd with %zu. > > In my view it makes sense to clean up existing uses of %Z legacy formats, > provided there is still stdio test coverage that %Z behaves as expected. > Here is the updated patch. OK for master? Thanks. -- H.J.