1 |
I just logged a bug... |
2 |
|
3 |
https://bugs.gentoo.org/show_bug.cgi?id=487056 |
4 |
|
5 |
On 10/06/13 00:16, Alan Hourihane wrote: |
6 |
> No it's not cygwin, and as I said I already googled it. |
7 |
> |
8 |
> But I will say the current mechanism is not POSIX compliant. Gentoo is |
9 |
> using a Linux'ism here and it's not portable. Opening a fifo in |
10 |
> read/write mode has nondeterministic behavior in POSIX. |
11 |
> |
12 |
> Should I log this as a bug ? |
13 |
> |
14 |
> Alan. |
15 |
> |
16 |
> On 10/05/13 08:04, Greg Turner wrote: |
17 |
>> If you're running portage under cygwin, somehow, you'll want to |
18 |
>> replace the multijob eclass and multijob portage-internal bash script |
19 |
>> with equivalent noop stubs. Cygwin has broken named pipe handling, |
20 |
>> blah blah blah... Google it. |
21 |
>> |
22 |
>> But you'll need a LOT more than that to make it work! I tried pretty |
23 |
>> hard for a couple of years to get it working and at some points |
24 |
>> managed to get most of '@system' working -- but never without gross |
25 |
>> hacks. See my github if interested (but be advised it has not |
26 |
>> received any love for a long time and is surely of extremely limited |
27 |
>> utility by now). |
28 |
>> |
29 |
>> There probably are no cygwin people. The cygwin profile in portage is |
30 |
>> for a super old version if that's what you're using, btw -- what you |
31 |
>> get with that is more akin to "msys" than present-day cygwin. |
32 |
>> |
33 |
>> -gmt |
34 |
>> |
35 |
> |
36 |
> |