1 |
On Wed, 24 May 2017 15:45:45 +0300 |
2 |
Andrew Savchenko <bircoph@g.o> wrote: |
3 |
|
4 |
> - smaller CPU overhead: not every i/o is being compressed, e.g. if |
5 |
> there is sill enough RAM available it is used without compression |
6 |
> overhead as usual, but if memory is not enough, swapped out pages |
7 |
> are being compressed instead of swapping out to disk; |
8 |
|
9 |
I found the opposite problem somehow. CPU started becomming frequently pegged |
10 |
in zswap for no obvious reason, while the underlying IO that zswap was doing |
11 |
was only measurable in kb/s , far, far, far below the noise thresholds and |
12 |
by no means a strain on even my crappy spinning rust based swap. |
13 |
|
14 |
And to add to that, zswap introduced general protection faults and kernel panics. |
15 |
|
16 |
So nah, I'm glad I turned that off, it was a huge mistake. |