1 |
On Fri, Jan 15, 2010 at 4:24 AM, Robin H. Johnson <robbat2@g.o> wrote: |
2 |
> On Fri, Jan 15, 2010 at 03:59:00AM +0530, Nirbheek Chauhan wrote: |
3 |
>> ChangeLog (-13,000 which would be redundant, and should be |
4 |
>> auto-generated alongwith metadata prior to distribution via rsync. |
5 |
>> Hey, this is something that needs to be done too!) |
6 |
> ChangeLog != commit logs. |
7 |
> |
8 |
> There is frequently additional information in the CVS commit messages |
9 |
> that isn't in the ChangeLogs. ChangeLogs aren't always updated reliably |
10 |
> either, esp for ebuild cleanups (hi vapier). |
11 |
> |
12 |
|
13 |
All the more reason to just chuck manual ChangeLogs in favour of |
14 |
auto-generated ones. Atleast, that's what gnome projects do since they |
15 |
moved to git [ http://live.gnome.org/Git/ChangeLog ]. |
16 |
|
17 |
However, this will entail a change in how commit messages are |
18 |
formatted; git commit messages need to be very different from CVS/svn |
19 |
ones. |
20 |
|
21 |
> The actual performance of git itself isn't the largest problem. |
22 |
> The migration issues, esp. the speed of the conversion are. |
23 |
> |
24 |
> My status of the migration side itself hasn't changed since the end of |
25 |
> October: |
26 |
> http://archives.gentoo.org/gentoo-scm/msg_e0a0a41200c1fc6a0fda68b4ff9d2c61.xml |
27 |
> |
28 |
> That top item is the largest blocker. The actual conversion time is down |
29 |
> to 9 hours, but with more than that again in setting it up. I'd like to |
30 |
> get the conversion time down to UNDER 4 hours. It's mostly |
31 |
> single-threaded, and we've got lots of cores available, it just needs |
32 |
> parallelization. We're basically dead in the water during the |
33 |
> conversion, there is NO incremental support at all. |
34 |
> |
35 |
|
36 |
Actually, this has confused me for a while. Sorry if this is a dumb |
37 |
question, but why do we care about the conversion speed if we can just |
38 |
convert it once, make the old cvs repo read-only and be done with it? |
39 |
Are we concerned about the window during which cvs access will have to |
40 |
be blocked and devs will sit around twiddling thumbs? |
41 |
|
42 |
> Side-project to the above: Is there anything link Psyco for Python |
43 |
> acceleration that works on 64-bit machines? Psyco itself has a warning |
44 |
> on the frontpage of no 64-bit support. |
45 |
> |
46 |
|
47 |
I had written Pysco off as dead and started looking forward to Unladen |
48 |
Swallow :) |
49 |
|
50 |
http://code.google.com/p/unladen-swallow/wiki/ProjectPlan |
51 |
|
52 |
> We DO still need somebody that cares about CVS access to test with |
53 |
> git-cvssserver against the existing conversion. |
54 |
> |
55 |
|
56 |
This will fit in quite badly with the proposed changes to make |
57 |
Manifest have only distfile manifests when using git, and to not have |
58 |
ChangeLogs for the simple reason that they (Manifests and ChangeLogs) |
59 |
invariably cause merge conflicts. And then there was also the plan to |
60 |
not edit headers for files to prevent extra commits (which are even |
61 |
more useless in git). |
62 |
|
63 |
|
64 |
-- |
65 |
~Nirbheek Chauhan |
66 |
|
67 |
Gentoo GNOME+Mozilla Team |