Old 2.5.x BK snapshots on kernel.org are now moved into the "old"
sub-directory, instead of being deleted.
The script change is untested... so yell at me if things break when
Linus releases 2.5.64.
Jeff Garzik wrote:
> Old 2.5.x BK snapshots on kernel.org are now moved into the "old"
> sub-directory, instead of being deleted.
>
> The script change is untested... so yell at me if things break when
> Linus releases 2.5.64.
It may not be related, but
http://www.kernel.org/pub/linux/kernel/v2.4/testing/cset/
has "BitKeeper patches since v2.4.21-pre4: 354 Changesets"
(not latest -pre).
Jacek
Jacek Kawa wrote:
> Jeff Garzik wrote:
>
>
>>Old 2.5.x BK snapshots on kernel.org are now moved into the "old"
>>sub-directory, instead of being deleted.
>>
>>The script change is untested... so yell at me if things break when
>>Linus releases 2.5.64.
>
>
> It may not be related, but
> http://www.kernel.org/pub/linux/kernel/v2.4/testing/cset/
> has "BitKeeper patches since v2.4.21-pre4: 354 Changesets"
> (not latest -pre).
Yep, I know :)
The point of nightly snapshots is more for users use and testing
purposes. The csets are very useful for developers, but a bit of a
moving target for users. If users are using the per-cset snapshots,
then it becomes a non-trivial chore to developers to track down exactly
what version of the kernel a problem is reported against.
Jeff
Jeff Garzik wrote:
> >>Old 2.5.x BK snapshots on kernel.org are now moved into the "old"
> >>sub-directory, instead of being deleted.
[...]
> >It may not be related, but
> >http://www.kernel.org/pub/linux/kernel/v2.4/testing/cset/
> >has "BitKeeper patches since v2.4.21-pre4: 354 Changesets"
> >(not latest -pre).
> Yep, I know :)
> The point of nightly snapshots is more for users use and testing
> purposes. The csets are very useful for developers, but a bit of a
> moving target for users.
I'm just used to look there from time to time to see 'what is new', and
AFAIR changes were reported against latest prerelease.
>From my point of view, as long as patches are in chronological order
it's not a problem. :)
> If users are using the per-cset snapshots,
> then it becomes a non-trivial chore to developers to track down exactly
> what version of the kernel a problem is reported against.
Csets are useful for users- e.g. if the problem you want to report does
not exist in latest cset -but generaly you are right.
Jacek