On Wednesday 30 November 2011 11:34:05 Zac Medico wrote:
> On 11/30/2011 08:09 AM, Mike Frysinger wrote:
> > On Wednesday 30 November 2011 01:23:59 Zac Medico wrote:
> >> If it wasn't for implicit system dependencies, the system set and its
> >> dependencies wouldn't need this kind of special treat
On Wed, Nov 30, 2011 at 11:34 AM, Zac Medico wrote:
> Ignoring circular dependencies doesn't make them go away. Ignoring
> dependencies can lead to build failures that could have been avoided if
> they were expressed in a way that the dependency resolver could properly
> account for them.
++
One
On 11/30/2011 08:09 AM, Mike Frysinger wrote:
> On Wednesday 30 November 2011 01:23:59 Zac Medico wrote:
>> If it wasn't for implicit system dependencies, the system set and its
>> dependencies wouldn't need this kind of special treatment.
>
> if it wasn't for implicit system dependencies, we'd ha
On Wednesday 30 November 2011 01:23:59 Zac Medico wrote:
> If it wasn't for implicit system dependencies, the system set and its
> dependencies wouldn't need this kind of special treatment.
if it wasn't for implicit system dependencies, we'd have significant bloat in
the tree and circular deps ou
Zac Medico posted on Tue, 29 Nov 2011 22:23:59 -0800 as excerpted:
> On 11/29/2011 08:51 PM, Duncan wrote:
>> Zac Medico posted on Tue, 29 Nov 2011 18:29:20 -0800 as excerpted:
>>
>>> One nice thing about removing them from the system profile is that it
>>> allows for greater parallelization with
On 11/29/2011 08:51 PM, Duncan wrote:
> Zac Medico posted on Tue, 29 Nov 2011 18:29:20 -0800 as excerpted:
>
>> One nice thing about removing them from the system profile is that it
>> allows for greater parallelization with emerge --jobs, since packages
>> that's aren't in the system set don't tr
Zac Medico posted on Tue, 29 Nov 2011 18:29:20 -0800 as excerpted:
> One nice thing about removing them from the system profile is that it
> allows for greater parallelization with emerge --jobs, since packages
> that's aren't in the system set don't trigger the code for bug 256616
> [1] which lim
