>Number:         32951
>Category:       lib
>Synopsis:       printf(3) "%.*f" display problem for some numbers with 0 precision
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    lib-bug-people
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Tue Feb 28 10:25:00 +0000 2006
>Originator:     Nicolas Joly
>Release:        NetBSD 3.99.15
>Organization:
Insitut Pasteur, Paris.
>Environment:
System: NetBSD calamity.sis.pasteur.fr 3.99.15 NetBSD 3.99.15 (CALAMITY) #9: Mon Feb 27 16:40:49 CET 2006 njoly@calamity.sis.pasteur.fr:/local/src/NetBSD/obj/i386/sys/arch/i386/compile/CALAMITY i386
Architecture: i386
Machine: i386
>Description:
I just discovered, a small printf(3) family functions problem when trying to
display some float numbers with 0 precision. With numbers in ]0.0;0.5], the
resul has an extra period `.' where it should not.

This effect can be easily seen :
njoly@calamity [XFree86/xc]> sysstat -w 1 vm
[...]
Namei         Sys-cache     Proc-cache           FPU synch IPI        497 ndcpy
    Calls     hits    %     hits     %        98 TLB shootdown I      296 fltcp
    31229    30234   97        8    0.           cpu1 softnet        5913 zfod
                                             100 cpu1 timer            59 cow
[...]
>How-To-Repeat:
njoly@calamity [tmp/sysstat]> cat xx.c
#include <stdio.h>

int main() {
  float f;

  for (f = 0.0; f < 1.2; f += 0.1) {
    printf("%f ... %.*f\n", f, 0, f); }

  return 0; }
njoly@calamity [tmp/sysstat]> cc xx.c
njoly@calamity [tmp/sysstat]> ./a.out
0.000000 ... 0
0.100000 ... 0.
0.200000 ... 0.
0.300000 ... 0.
0.400000 ... 0.
0.500000 ... 0.
0.600000 ... 1
0.700000 ... 1
0.800000 ... 1
0.900000 ... 1
1.000000 ... 1
1.100000 ... 1
>Fix: