| 1 |
On 9/24/2012 9:08 AM, Zac Medico wrote: |
| 2 |
> On 09/24/2012 03:25 AM, Gregory M. Turner wrote: |
| 3 |
>> multijob_finish_one() { |
| 4 |
>> local pid ret |
| 5 |
>> - read -r -u ${mj_control_fd} pid ret |
| 6 |
>> + multijob_assert |
| 7 |
>> + if this_host_is_cygwin ; then |
| 8 |
>> + read -r pid ret < ${mj_control_pipe} |
| 9 |
>> + else |
| 10 |
>> + read -r -u ${mj_control_fd} pid ret |
| 11 |
>> + fi |
| 12 |
>> : $(( --mj_num_jobs )) |
| 13 |
>> return ${ret} |
| 14 |
>> } |
| 15 |
> Since you re-open the pipe for each read, there's a race condition which |
| 16 |
> I've explained here: |
| 17 |
> |
| 18 |
> http://permalink.gmane.org/gmane.linux.gentoo.devel/77528 |
| 19 |
|
| 20 |
Gross, thanks for this pointer, Zac. Plenty of solutions one could |
| 21 |
concoct, but no point dumping effort into this black-hole when I can |
| 22 |
just wait for somebody else to get fed up and fix the underlying problem |
| 23 |
for me :) |
| 24 |
|
| 25 |
Guess it's back to stubs, for now. |
| 26 |
|
| 27 |
-gmt |
Archives