[Bug 2081334] Re: further performance improvement for dracut-install 060
Andreas Hasenack
2081334 at bugs.launchpad.net
Fri Sep 20 18:39:54 UTC 2024
** Description changed:
[ Impact ]
This is a follow-up for bug #2065180 implementing the suggestion from
comment #44 there.
When compared to Ubuntu 23.10, creating intramfs files with update-
initramfs takes 2 to 5 times more time on ARM devices.
IIUC, dracut-install usage was added to initramfs-tools to speed up the
process. But now its way slower. Even running update-initramfs on jammy,
which doesn't use dracut-install, is way faster then the time taken on
Noble.
first bad commit -
https://github.com/dracutdevs/dracut/commit/3de4c7313260fb600507c9b87f780390b874c870
Updating the initrd on a Raspberry Pi Zero 2W on Ubuntu 24.04 (noble)
with initramfs-tools 0.142ubuntu25.1 takes over six minutes:
```
bdrung at zero2w:~$ sudo hyperfine --warmup 1 -r 10 "update-initramfs -u"
Benchmark 1: update-initramfs -u
Time (mean ± σ): 402.751 s ± 5.592 s [User: 166.316 s, System: 228.909 s]
Range (min … max): 394.380 s … 411.445 s 10 runs
```
[ Test Plan ]
1. Measure `update-initramfs -u` before the update.
2. Log the content of the initrd before the update: `lsinitramfs /boot/initrd.img`
3. update dracut-install / initramfs-tools-core
4. Measure `update-initramfs -u`. It should be faster (the performance improvements on amd64 should be very small and might be within the measurement uncertainty).
5. Check with lsinitramfs that the content of the newly generated initrd hasn't changed.
[ Where problems could occur ]
The code that is responsible for including the kernel modules into the
initrd is touched. Negative consequences could be that some needed
kernel modules will not be included any more (should be covered by the
test case) or that building new initrds will fail.
The initramfs-tools fix changes how manual_add_modules behaves.
`manual_add_modules` does not copy kernel modules, but queues them for
being copied when the newly added function `apply_add_modules` is
called.
I checked all instances of calls to `manual_add_modules` for possible
- regressions (see comment #15). Only miniramfs needs to be adjusted to
- also call `apply_add_modules`. But this change could break consumers of
- the `manual_add_modules` function that are outside of the Ubuntu
- archive. I googled for `apply_add_modules` but found no public outside
- users.
+ regressions (see
+ https://bugs.launchpad.net/ubuntu/+source/dracut/+bug/2065180/comments/15).
+ Only miniramfs needs to be adjusted to also call `apply_add_modules`.
+ But this change could break consumers of the `manual_add_modules`
+ function that are outside of the Ubuntu archive. I googled for
+ `apply_add_modules` but found no public outside users.
[ Benchmarks ]
Stock noble on a Raspberry Pi Zero 2W:
```
bdrung at zero2w:~$ sudo hyperfine -r 5 "update-initramfs -u"
Benchmark 1: update-initramfs -u
Time (mean ± σ): 415.664 s ± 6.015 s [User: 166.728 s, System: 232.523 s]
Range (min … max): 409.139 s … 422.632 s 5 runs
```
noble with dracut-install 060+5-1ubuntu3.1 (with linux 6.8.0-1006.6 on
2024-07-01):
```
bdrung at zero2w:~$ sudo hyperfine --warmup 1 -r 10 "update-initramfs -u"
Benchmark 1: update-initramfs -u
Time (mean ± σ): 248.054 s ± 5.569 s [User: 67.410 s, System: 169.412 s]
Range (min … max): 238.909 s … 257.384 s 10 runs
```
noble with dracut-install 060+5-1ubuntu3.2 (with linux 6.8.0-1007 on
2024-07-11):
```
bdrung at zero2w:~$ sudo hyperfine --warmup 1 -r 10 "update-initramfs -u"
Benchmark 1: update-initramfs -u
Time (mean ± σ): 232.001 s ± 5.678 s [User: 55.456 s, System: 166.510 s]
Range (min … max): 222.120 s … 239.610 s 10 runs
```
noble with dracut-install 060+5-1ubuntu3.2 (with linux 6.8.0-1011.12 and
all updates on 2024-09-19):
```
bdrung at zero2w:~$ sudo hyperfine --warmup 1 -r 10 "update-initramfs -u"
Time (mean ± σ): 219.439 s ± 4.869 s [User: 45.626 s, System: 163.172 s]
Range (min … max): 213.340 s … 227.411 s 10 runs
```
[ Reduce manual_add_modules calls ]
Besides making the dracut-install calls faster, group the dracut-install
calls. Since the fix in oracular can cause regressions in custom hooks
that rely on the current behavior, the SRU takes a safe approach which
includes following packages (stating how many dracut-install calls are
used):
* cryptsetup: 2 -> 1
* lvm2: 8 -> 1
* thin-provisioning-tools: 3 -> 1
* open-iscsi: 9 -> 1
* cloud-initramfs-tools: 5 -> 1
dracut-install calls on a Raspberry Pi Zero 2W:
| area | before | noble SRU | oracular |
|--------------------------------------|--------|-----------|----------|
| auto_add_modules + apply_add_modules | 8 | 5 | 5 |
| calls by hooks + apply_add_modules | 42 | 20 | 2 |
| hidden_dep_add_modules | 1 | 1 | 1 |
| total | 51 | 26 | 8 |
--
You received this bug notification because you are a member of Ubuntu
Foundations Bugs, which is subscribed to initramfs-tools in Ubuntu.
https://bugs.launchpad.net/bugs/2081334
Title:
further performance improvement for dracut-install 060
Status in initramfs-tools package in Ubuntu:
Fix Released
Status in initramfs-tools source package in Noble:
New
Bug description:
[ Impact ]
This is a follow-up for bug #2065180 implementing the suggestion from
comment #44 there.
When compared to Ubuntu 23.10, creating intramfs files with update-
initramfs takes 2 to 5 times more time on ARM devices.
IIUC, dracut-install usage was added to initramfs-tools to speed up
the process. But now its way slower. Even running update-initramfs on
jammy, which doesn't use dracut-install, is way faster then the time
taken on Noble.
first bad commit -
https://github.com/dracutdevs/dracut/commit/3de4c7313260fb600507c9b87f780390b874c870
Updating the initrd on a Raspberry Pi Zero 2W on Ubuntu 24.04 (noble)
with initramfs-tools 0.142ubuntu25.1 takes over six minutes:
```
bdrung at zero2w:~$ sudo hyperfine --warmup 1 -r 10 "update-initramfs -u"
Benchmark 1: update-initramfs -u
Time (mean ± σ): 402.751 s ± 5.592 s [User: 166.316 s, System: 228.909 s]
Range (min … max): 394.380 s … 411.445 s 10 runs
```
[ Test Plan ]
1. Measure `update-initramfs -u` before the update.
2. Log the content of the initrd before the update: `lsinitramfs /boot/initrd.img`
3. update dracut-install / initramfs-tools-core
4. Measure `update-initramfs -u`. It should be faster (the performance improvements on amd64 should be very small and might be within the measurement uncertainty).
5. Check with lsinitramfs that the content of the newly generated initrd hasn't changed.
[ Where problems could occur ]
The code that is responsible for including the kernel modules into the
initrd is touched. Negative consequences could be that some needed
kernel modules will not be included any more (should be covered by the
test case) or that building new initrds will fail.
The initramfs-tools fix changes how manual_add_modules behaves.
`manual_add_modules` does not copy kernel modules, but queues them for
being copied when the newly added function `apply_add_modules` is
called.
I checked all instances of calls to `manual_add_modules` for possible
regressions (see
https://bugs.launchpad.net/ubuntu/+source/dracut/+bug/2065180/comments/15).
Only miniramfs needs to be adjusted to also call `apply_add_modules`.
But this change could break consumers of the `manual_add_modules`
function that are outside of the Ubuntu archive. I googled for
`apply_add_modules` but found no public outside users.
[ Benchmarks ]
Stock noble on a Raspberry Pi Zero 2W:
```
bdrung at zero2w:~$ sudo hyperfine -r 5 "update-initramfs -u"
Benchmark 1: update-initramfs -u
Time (mean ± σ): 415.664 s ± 6.015 s [User: 166.728 s, System: 232.523 s]
Range (min … max): 409.139 s … 422.632 s 5 runs
```
noble with dracut-install 060+5-1ubuntu3.1 (with linux 6.8.0-1006.6 on
2024-07-01):
```
bdrung at zero2w:~$ sudo hyperfine --warmup 1 -r 10 "update-initramfs -u"
Benchmark 1: update-initramfs -u
Time (mean ± σ): 248.054 s ± 5.569 s [User: 67.410 s, System: 169.412 s]
Range (min … max): 238.909 s … 257.384 s 10 runs
```
noble with dracut-install 060+5-1ubuntu3.2 (with linux 6.8.0-1007 on
2024-07-11):
```
bdrung at zero2w:~$ sudo hyperfine --warmup 1 -r 10 "update-initramfs -u"
Benchmark 1: update-initramfs -u
Time (mean ± σ): 232.001 s ± 5.678 s [User: 55.456 s, System: 166.510 s]
Range (min … max): 222.120 s … 239.610 s 10 runs
```
noble with dracut-install 060+5-1ubuntu3.2 (with linux 6.8.0-1011.12
and all updates on 2024-09-19):
```
bdrung at zero2w:~$ sudo hyperfine --warmup 1 -r 10 "update-initramfs -u"
Time (mean ± σ): 219.439 s ± 4.869 s [User: 45.626 s, System: 163.172 s]
Range (min … max): 213.340 s … 227.411 s 10 runs
```
[ Reduce manual_add_modules calls ]
Besides making the dracut-install calls faster, group the dracut-
install calls. Since the fix in oracular can cause regressions in
custom hooks that rely on the current behavior, the SRU takes a safe
approach which includes following packages (stating how many dracut-
install calls are used):
* cryptsetup: 2 -> 1
* lvm2: 8 -> 1
* thin-provisioning-tools: 3 -> 1
* open-iscsi: 9 -> 1
* cloud-initramfs-tools: 5 -> 1
dracut-install calls on a Raspberry Pi Zero 2W:
| area | before | noble SRU | oracular |
|--------------------------------------|--------|-----------|----------|
| auto_add_modules + apply_add_modules | 8 | 5 | 5 |
| calls by hooks + apply_add_modules | 42 | 20 | 2 |
| hidden_dep_add_modules | 1 | 1 | 1 |
| total | 51 | 26 | 8 |
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/initramfs-tools/+bug/2081334/+subscriptions
More information about the foundations-bugs
mailing list