1 |
Il 04/09/2012 19:15, Zac Medico ha scritto: |
2 |
> On 09/04/2012 04:00 AM, Walter Dnes wrote: |
3 |
>> On Sat, Sep 01, 2012 at 05:20:02PM -0700, Brian Harring wrote |
4 |
>> |
5 |
>>> This approach is fine imo, although I'd *potentially* look at adding a |
6 |
>>> magic $PROC_COUNT var that is the # of cpu threads on the system; |
7 |
>>> either that or defaulting jobs to it. |
8 |
>>> |
9 |
>>> I rather dislike requiring users to go jam a 2/4/8 in there when it's |
10 |
>>> easy to compute. That said, it's minor. |
11 |
>>> |
12 |
>>> Either way, yes, I think EJOBS should be in EAPI5. |
13 |
>> One question about the suggested EJOBS variable; will it over-ride |
14 |
>> MAKEOPTS? Every so often on the Gentoo-user list, someone comes along |
15 |
>> with a mysterious build failure. The first suggestion is to reset |
16 |
>> MAKEOPTS to -j1. And on some occasions, that is indeed the solution to |
17 |
>> the mysterious build failure. |
18 |
> That would be due to a missing dependency in the Makefiles, and using |
19 |
> -j1 is just a workaround. The ebuild can be hardcoded to use emake -j1 |
20 |
> until the Makefile gets fixed. |
21 |
> |
22 |
>> I set -j1 and leave it that way. Yes, the builds take longer, but the |
23 |
>> resulting binary is just as fast. And the amount of time I "save" will |
24 |
>> be blown away the first time I end up screwing around a couple of hours |
25 |
>> trying to fix a mysterious build failure. That's why I want the user to |
26 |
>> have the option of over-riding EJOBS, should it ever be implemented. |
27 |
> You could use EXTRA_EMAKE for that. You can do per-package settings via |
28 |
> /etc/portage/package.env. |
29 |
|
30 |
Dunno where to place this request, but if we go for something like EJOBS |
31 |
could we also make it phase specific? |
32 |
So compile, install and test could have a different number of jobs running. |
33 |
Possibly three different variables that override a predefined EJOBS. |
34 |
|
35 |
TIA |