1 |
On 2014.05.12 10:35, Tom Wijsman wrote: |
2 |
> On Sun, 11 May 2014 19:46:50 +0200 |
3 |
> Michał Górny <mgorny@g.o> wrote: |
4 |
> |
5 |
> > Rationale: xz-utils is quite widespread nowadays and it is a part |
6 |
> > of @system set. It can achieve better compression ratio than bzip2, |
7 |
> > and faster decompression at the same time. |
8 |
> |
9 |
> Some thoughts: |
10 |
> |
11 |
> What about putting multiple doc / man / info files in a single .xz |
12 |
> file |
13 |
> for each package? Would that further improve the situation? |
14 |
> |
15 |
> (As they can share dictionary, instead of having multiple |
16 |
> dictionaries) |
17 |
> |
18 |
> Some algorithms tend to work better for smaller files, whereas others |
19 |
> work better for larger files; might this be the case for bzip2 vs. |
20 |
> xz? |
21 |
> |
22 |
> -- |
23 |
> With kind regards, |
24 |
> |
25 |
> Tom Wijsman (TomWij) |
26 |
> Gentoo Developer |
27 |
> |
28 |
> E-mail address : TomWij@g.o |
29 |
> GPG Public Key : 6D34E57D |
30 |
> GPG Fingerprint : C165 AF18 AB4C 400B C3D2 ABF0 95B2 1FCD 6D34 E57D |
31 |
> |
32 |
|
33 |
Some more thoughts ... |
34 |
|
35 |
What about not compressing files smaller than the filesysem block size |
36 |
at all. In my case its 4k. Any file gets allocated 4k on disc anyway, |
37 |
so compression/decompression is just a waste of resource for files |
38 |
<=4k. |
39 |
|
40 |
I'm not suggesting dynamically determining the output filesystem block |
41 |
size (unless you really want to), choose a static limit below which |
42 |
compression will not be applied. |
43 |
|
44 |
That eliminates the discussion about small files. |
45 |
-- |
46 |
Regards, |
47 |
|
48 |
Roy Bamford |
49 |
(Neddyseagoon) a member of |
50 |
elections |
51 |
gentoo-ops |
52 |
forum-mods |
53 |
trustees |