1 |
On 17-05-23 at 22:16, Ian Zimmerman wrote: |
2 |
> So what are gentoo users' opinions on this matter of faith? |
3 |
I use an ext4 partition backed by zram. Gives me ~3x compression on the |
4 |
things I normally have lying around there (plain text files) and ensures |
5 |
that anything I throw there (or programs throw there) gets cleaned up on |
6 |
reboot. |
7 |
|
8 |
> I have long been in the camp that thinks tmpfs for /tmp has no |
9 |
> advantages (and may have disadvantages) over a normal filesystem like |
10 |
> ext3, because the files there are normally so small that they will stay |
11 |
> in the page cache 100% of the time. |
12 |
I've never actually benchmarked this. Most of the things I notice that |
13 |
tend to end up there are temporary build files generated during |
14 |
configure stages or temporary log files used by various programs (clang |
15 |
static analyzer). Even if the entire file stays in the page cache, it'll |
16 |
still generate IO overhead and extra seeks that might slow down the rest |
17 |
of your system (unless your /tmp is on a different hard drive) which on |
18 |
spinning rust will cause slowdowns while on an ssd it'll eat away at |
19 |
your writes (which you may or may not have to worry about). |
20 |
|
21 |
> But I see that tmpfs is the default with systemd. Surely they have a |
22 |
> good reason for this? :) |
23 |
Or someone decided they liked the idea and made it the default and |
24 |
nobody ever complained (or if they did were told to just change it on |
25 |
their system). |
26 |
|
27 |
Either way, it'd be nice if someone actually benchmarked this. |
28 |
|
29 |
-- |
30 |
Simon Thelen |