ZTS: Apply small changes for speeding up the tests

[mcmilk]

Motivation and Context

The Github Action Runner got some new hardware metrics.
We should use the provided and empty disk which is pre-mounted at /mnt now.

Disk1: 89GiB -> rootfs + bootfs with ~80MB/s -> don't care
Disk2: 64GiB -> /mnt with 420MB/s -> new testing ssd

Description

This commit will mount the new disk to /var/tmp and provide hopefully some speedups within our testings.

How Has This Been Tested?

We will see, how much seepup we can get... when the ZTS is done with it ;-)

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Performance enhancement (non-breaking change which improves efficiency)
• Code cleanup (non-breaking change which makes code smaller or more readable)
• Breaking change (fix or feature that would cause existing functionality to change)
• Library ABI change (libzfs, libzfs_core, libnvpair, libuutil and libzfsbootenv)
• Documentation (a change to man pages or other documentation)

Checklist:

• My code follows the OpenZFS code style requirements.
• I have updated the documentation accordingly.
• I have read the contributing document.
• I have added tests to cover my changes.
• I have run the ZFS Test Suite with this change applied.
• All commit messages are properly formatted and contain Signed-off-by.

[andrewc12]

Do you have any performance comparisons for this change?
From what I could see it seemed a little bit slower.

[mcmilk]

Since last week, the runner hardware improved a lot.
Only disk io heavy testings will see a huge performance increase, like the zloop testings.

This is the comparison for a freshly created runner:

[mcmilk@lydia ~]$ ssh [email protected] -p 29936
Welcome to Ubuntu 22.04.3 LTS (GNU/Linux 6.2.0-1018-azure x86_64)

 * Documentation:  https://help.ubuntu.com
 * Management:     https://landscape.canonical.com
 * Support:        https://ubuntu.com/advantage

  System information as of Wed Jan 24 06:25:46 UTC 2024

  System load:  0.05126953125      Users logged in:          0
  Usage of /:   64.0% of 83.16GB   IPv4 address for docker0: 172.17.0.1
  Memory usage: 25%                IPv4 address for eth0:    10.1.0.99
  Swap usage:   0%                 IPv4 address for virbr0:  192.168.122.1
  Processes:    154

Expanded Security Maintenance for Applications is not enabled.

33 updates can be applied immediately.
21 of these updates are standard security updates.
To see these additional updates run: apt list --upgradable

19 additional security updates can be applied with ESM Apps.
Learn more about enabling ESM Apps service at https://ubuntu.com/esm

Last login: Wed Jan 24 06:25:47 2024 from 127.0.0.1
runner@fv-az1240-812:~$ sudo bash

root@fv-az1240-812:/home/runner# hdparm -t /dev/sda
/dev/sda:
 Timing buffered disk reads: 288 MB in  3.31 seconds =  87.06 MB/sec

root@fv-az1240-812:/home/runner# hdparm -t /dev/sdb
/dev/sdb:
 Timing buffered disk reads: 1278 MB in  3.01 seconds = 424.40 MB/sec

root@fv-az1240-812:/home/runner# mount | grep sd[ab]
/dev/sda1 on / type ext4 (rw,relatime,discard,errors=remount-ro)
/dev/sda15 on /boot/efi type vfat (rw,relatime,fmask=0077,dmask=0077,codepage=437,iocharset=iso8859-1,shortname=mixed,errors=remount-ro)
/dev/sdb1 on /mnt type ext4 (rw,relatime)

[gmelikov]

If we already format and remount ext4 - maybe:

• mount with noatime,barrier=0 ?
• disable journal?
• maybe compilation will be faster too from this ssd? If it's so - we may move code onto it
• other speedups I've missed?

[mcmilk]

mount with noatime,barrier=0 ?
disable journal?

These two are done .... thanks for pointing out.

maybe compilation will be faster too from this ssd? If it's so - we may move code onto it

Compiling is done once on startup in ~10..20 minutes. This should be okay.

[pcd1193182]

I've noticed sporadic test failures in the code added in this PR; I'm not sure if it's a bug that's always been there, or if something has changed with the github workers, but a nontrivial number of test runs are failing as a result.

[mcmilk]

I noticed these sporadic test failures also.

I am in doubt if the other disk space is the reason for that - but we could test this, by reverting this PR maybe...
@behlendorf @gmelikov - should we try a revert here?