| 1 |
On 10/03/2011 12:37 AM, Chí-Thanh Christopher Nguyễn wrote: |
| 2 |
> Samuli Suominen schrieb: |
| 3 |
> |
| 4 |
>>> And again, downgrade of dependencies it is not against any rule which |
| 5 |
>>> would justify mask and removal. |
| 6 |
>>> |
| 7 |
>>> Another example from the X.org packages, installing the proprietary |
| 8 |
>>> ATI/NVidia drivers will cause downgrades for xorg-server on ~arch |
| 9 |
>>> systems. Nobody in his right mind is proposing to treeclean them because |
| 10 |
>>> of this. |
| 11 |
>>> |
| 12 |
>> |
| 13 |
>> The new xorg-servers could get package.masked until these major drivers |
| 14 |
>> are available. |
| 15 |
>> Albeit, I'm not intrested in pursuing this since with separate |
| 16 |
>> xorg-server package, it's the drivers that need rebuilding against it, |
| 17 |
>> and the VIDEO_CARDS="" setting is keeping it in certain version until |
| 18 |
>> the VIDEO_CARDS="" setting is satisfied. |
| 19 |
>> |
| 20 |
>> Poor example to make a case. |
| 21 |
> |
| 22 |
> VIDEO_CARDS is just for user convenience. run "emerge nvidia-drivers" on |
| 23 |
> any system with xorg-server-1.11 installed and it will downgrade, no |
| 24 |
> matter what VIDEO_CARDS is set to. |
| 25 |
|
| 26 |
And your point is? The drivers will need to be rebuilt everytime the |
| 27 |
xorg-server version changes. This does not come as a suprise, the |
| 28 |
.ebuild should print a message about rebuilding them. If it doesn't, |
| 29 |
then the .ebuild should get fixed. |
| 30 |
Leaving this particular case for X.org maintainers to decide sounds fine |
| 31 |
to me, given the relaxing factors. |
| 32 |
|
| 33 |
> |
| 34 |
>> The intresting part of that document is "You should also not cause an |
| 35 |
>> unnecessary downgrade for any "~arch" when ..." which also applies to |
| 36 |
>> setting dependencies just as well. |
| 37 |
> |
| 38 |
> The downgrade is necessary to avoid user-visible breakage. |
| 39 |
|
| 40 |
Avoiding one in non-system critical package (like qutecom), but |
| 41 |
introducing multiple new scenarios in what-could-be system-critical |
| 42 |
packages. |
| 43 |
|
| 44 |
> And the wording clearly does only apply to package removals. |
| 45 |
|
| 46 |
The fact that the *common sense* snippet was inserted in this document, |
| 47 |
but isn't documented else where... doesn't make it any less true. |
| 48 |
|
| 49 |
- Samuli |
Archives