| 1 |
On Friday 11 May 2012 13:32:46 Zac Medico wrote: |
| 2 |
> On 05/11/2012 09:39 AM, Mike Frysinger wrote: |
| 3 |
> > +multijob_finish() { |
| 4 |
> > + local ret=0 |
| 5 |
> > + while [[ ${mj_num_jobs} -gt 0 ]] ; do |
| 6 |
> > + multijob_finish_one |
| 7 |
> > + : $(( ret += $? )) |
| 8 |
> > + done |
| 9 |
> > + # Let bash clean up its internal child tracking state. |
| 10 |
> > + wait |
| 11 |
> > + return ${ret} |
| 12 |
> > +} |
| 13 |
> |
| 14 |
> Wouldn't it be better to use $(( ret |= $? )) there, in order to avoid a |
| 15 |
> possible integer overflow? Other than that, the patch looks good to me. |
| 16 |
|
| 17 |
i meant to use |= like the other places ... |
| 18 |
|
| 19 |
not that i'm too worried about overflow here as the exit value is clamped to |
| 20 |
[0..255], so it'd take millions of failing processes to cause a problem :). |
| 21 |
-mike |
Archives