Re: [PATCHSET v5 0/12] Add support for async buffered reads

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



[ ... ]

> > Hi Jens,
> >
> > I have pulled linux-block.git#async-buffered.5 on top of Linux v5.7-rc7.
> >
> > From first feelings:
> > The booting into the system (until sddm display-login-manager) took a
> > bit longer.
> > The same after login and booting into KDE/Plasma.
>
> There is no difference for "regular" use cases, only io_uring with
> buffered reads will behave differently. So I don't think you have longer
> boot times due to this.
>

Yupp, you are right.

The previous Linux v5.7-rc7 without your patchset shows the same symptoms.

I did some debugging and optimizing with systemd-analyze boot and time.

I optimized systemd-journald.service and systemd-journal-flush.service...

# cat /etc/systemd/journald.conf.d/00-journal-size.conf
[Journal]
SystemMaxUse=50M

...and reduced the time spent flushing systemd's journal from ~30s
down to 1,6s...

# journalctl -b --unit systemd-journald.service
-- Logs begin at Fri 2020-05-29 00:58:37 CEST, end at Fri 2020-05-29
11:42:18 CEST. --
Mai 29 11:34:52 iniza systemd-journald[281]: Journal started
Mai 29 11:34:52 iniza systemd-journald[281]: Runtime Journal
(/run/log/journal/566abbcb226b405db834b17a26fe4727) is 8.0M, max
78.5M, 70.5M free.
Mai 29 11:34:53 iniza systemd-journald[281]: Time spent on flushing to
/var/log/journal/566abbcb226b405db834b17a26fe4727 is 1.656233s for 765
entries.
Mai 29 11:34:53 iniza systemd-journald[281]: System Journal
(/var/log/journal/566abbcb226b405db834b17a26fe4727) is 56.2M, max
50.0M, 0B free.

Unfortunately, I upgraded some user-space stuff like udisks2 and
libblockdev packages.
Downgrading did not help and disabling the systemd-unit also.

As I saw stallings with e2scrub_reap.service and swap partition
(partly seen in the boot-process and noted the UUID 3f8e).
I disabled e2scrub_reap.service and deactivated swap partition in /etc/fstab.

Doing all above together did not help.

Finally, I checked the health of the HDD where my root-fs is.
smartmontools says everything is OK.

I have not checked the status of the Ext4-FS where my root-fs is.
Such things I do with a linux-live-system - as a Debianist I admit I
use an ArchLinux ISO on USB-stick :-).

Unsure, if I will contact the systemd (and mabye udisks) Debian folks
to hear their opinion.

Thanks Jens and your patchset.
I don't know when I last run systemd-analyze & stuff and investigated
so deeply :-).

A lot of Hygge (I love to write wrong Huegge - see English hugs) to you Jens.

- Sedat -

[1] http://ftp.halifax.rwth-aachen.de/archlinux/iso/



[Index of Archives]     [Linux Ext4 Filesystem]     [Union Filesystem]     [Filesystem Testing]     [Ceph Users]     [Ecryptfs]     [AutoFS]     [Kernel Newbies]     [Share Photos]     [Security]     [Netfilter]     [Bugtraq]     [Yosemite News]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux Cachefs]     [Reiser Filesystem]     [Linux RAID]     [Samba]     [Device Mapper]     [CEPH Development]

  Powered by Linux