1 |
Yes. The manager can still parallelize prefetching, only consuming a build |
2 |
job slot post fetch. |
3 |
On Sep 6, 2012 11:49 AM, "Michał Górny" <mgorny@g.o> wrote: |
4 |
|
5 |
> On Thu, 6 Sep 2012 09:49:13 -0700 |
6 |
> Brian Harring <ferringb@×××××.com> wrote: |
7 |
> |
8 |
> > One additional thought- re: the scenarios where we don't fetch to an |
9 |
> > intermediate location, then transfer that contents into ${WORKDIR}, |
10 |
> > while a better name is needed, something along the lines of |
11 |
> > RESTRICT=fetches-into-workdir comes to mind. |
12 |
> > |
13 |
> > Basically that restriction would be interpretted as "$WORKDIR must be |
14 |
> > setup/preserved from invocation of src_fetch to actual building". |
15 |
> > |
16 |
> > Via that restrict, both scenarios should be addressed in full. |
17 |
> |
18 |
> Does separate src_fetch() provide any benefit in that scenario? |
19 |
> |
20 |
> -- |
21 |
> Best regards, |
22 |
> Michał Górny |
23 |
> |