| 1 |
On Wed, 2022-01-05 at 21:06 +0000, Roy Bamford wrote: |
| 2 |
> On 2022.01.05 20:22, Sam James wrote: |
| 3 |
> > |
| 4 |
> > |
| 5 |
> > > On 5 Jan 2022, at 19:02, Roy Bamford <neddyseagoon@g.o> |
| 6 |
> > wrote: |
| 7 |
> > > |
| 8 |
> > > Sam, |
| 9 |
> > > |
| 10 |
> > > Do users with FEATURES=distcc still have to opt out of this |
| 11 |
> > > MAKEOPTS clamping? |
| 12 |
> > > |
| 13 |
> > |
| 14 |
> > Great point! I think we could add an exemption for that and make it |
| 15 |
> > a |
| 16 |
> > noop or warning-only. |
| 17 |
> > |
| 18 |
> > Best, |
| 19 |
> > sam |
| 20 |
> > |
| 21 |
> > |
| 22 |
> |
| 23 |
> |
| 24 |
> Sam, |
| 25 |
> |
| 26 |
> You are building a better mousetrap here. That's not a reason to try. |
| 27 |
> |
| 28 |
> Do users of I_KNOW_WHAT_I_AM_DOING, who have already |
| 29 |
> opted to shoot themselves in both feet, get a free pass here? |
| 30 |
> |
| 31 |
> There are users who run emerge --jobs=X with MAKEOPTS='-jY" |
| 32 |
> and get firefox, thunderbird and libreoffice all building concurrently |
| 33 |
> as they allow X * Y MAKE threads, reduced by this proposed |
| 34 |
> throttling, still triggering the OOM. |
| 35 |
> |
| 36 |
> I don't think you can head that off beforehand. |
| 37 |
> |
| 38 |
|
| 39 |
What's your proposed alternative? |
Archives