Skipping previous view doesn't let performance degrade

[klfyt]

Something is wrong with previous view
killing performance and responsiveness, etc.
The only obvious fix seems to be to skip it.

:))

# for example if you skip to previous
# a > b (previous) fix it with
# b > c > b

Edit:
if you skip to previous view with Shift key performance
doesn't degrade
https://bugs.launchpad.net/xfwm4/+bug/1837783

It's ignored at the moment, however it looks to be
significant performance, at least in terms
of idle power.

[Sebastien Bacher (seb128)]

Thank you for your bug report, could you provide some details?

- what Ubuntu/nautilus version are you using?
- what sort of performance improvement are you speaking about (and how slow is it by default for you/with what content)
- could you maybe also report it upstream on network-manager:upstream

[klfyt (klfyt)]

> what sort of performance improvement are you speaking about

Like you started on a 1st page.
I believe window history is broken, I don't know how.

Just revisit history. :))

[fakenxid (fakenxid-redhat-bugs)]

Description of problem:
Anywhere where you need to travel to previous view
performance drops visibly.

Version-Release number of selected component (if applicable):

How reproducible:

Steps to Reproduce:
1. go to previous view
2.
3.

Actual results:
Performance starts to degrade.

Expected results:
Performance is like before. :))

Additional info:
Documents
Documents/a
Documents < performance degrades (workaround go anywhere else besides
                                       Documents or Documents/.. and only
                                       then go to Documents)

[fakenxid (fakenxid-redhat-bugs)]

*** Bug 1729681 has been marked as a duplicate of this bug. ***

[klfyt (klfyt)]

Just skip previous directly seems the rule, dunno what it is. :))

[fakenxid (fakenxid-redhat-bugs)]

Think the only obvious so-called workaround is avoiding
getting into previous through whatever means possible.

This way you can keep it on the level. :-)

Simple rule is:
previous view ? skip to something else :))

>:]

[fakenxid (fakenxid-redhat-bugs)]

*** Bug 1670504 has been marked as a duplicate of this bug. ***

[klfyt (klfyt)]

Maybe previous view contains recursive function that keeps fighting for resources?

:))

[fakenxid (fakenxid-redhat-bugs)]

(In reply to klfyt from comment #2)
> Think the only obvious so-called workaround is avoiding
> getting into previous through whatever means possible.
>
> This way you can keep it on the level. :-)
>
> Simple rule is:
> previous view ? skip to something else :))
>
> >:]

a > b

b == previous?

b > c > b

[klfyt (klfyt)]

https://bugs.launchpad.net/xfwm4/+bug/1837783

:-)

lol?

[Chris Guiver (guiverc)]

No answer yet to seb128's prior question.

[fakenxid (fakenxid-redhat-bugs)]

Fastest xfmw

alt tab > shift alt tab > alt tab

Fastest browser (Firefox, Chromium)

shift ctrl tab > ctrl tab

[fakenxid (fakenxid-redhat-bugs)]

Fastest text :-)

navigate left -1 letter keep typing.

For example:

Good m|f
Good morning|f

Good morning.

[fakenxid (fakenxid-redhat-bugs)]

*** Bug 1744930 has been marked as a duplicate of this bug. ***

[fakenxid (fakenxid-redhat-bugs)]

*** Bug 1724029 has been marked as a duplicate of this bug. ***

[fakenxid (fakenxid-redhat-bugs)]

*** Bug 1677161 has been marked as a duplicate of this bug. ***

[fakenxid (fakenxid-redhat-bugs)]

*** Bug 1674486 has been marked as a duplicate of this bug. ***

[fakenxid (fakenxid-redhat-bugs)]

*** Bug 1733819 has been marked as a duplicate of this bug. ***

[fakenxid (fakenxid-redhat-bugs)]

*** Bug 1729666 has been marked as a duplicate of this bug. ***

[fakenxid (fakenxid-redhat-bugs)]

*** Bug 1727998 has been marked as a duplicate of this bug. ***

[fakenxid (fakenxid-redhat-bugs)]

*** Bug 1740646 has been marked as a duplicate of this bug. ***

[fakenxid (fakenxid-redhat-bugs)]

*** Bug 1731759 has been marked as a duplicate of this bug. ***

[fakenxid (fakenxid-redhat-bugs)]

*** Bug 1730272 has been marked as a duplicate of this bug. ***

[fakenxid (fakenxid-redhat-bugs)]

*** Bug 1726250 has been marked as a duplicate of this bug. ***

[fakenxid (fakenxid-redhat-bugs)]

*** Bug 1725831 has been marked as a duplicate of this bug. ***

[fakenxid (fakenxid-redhat-bugs)]

*** Bug 1697062 has been marked as a duplicate of this bug. ***

[fakenxid (fakenxid-redhat-bugs)]

*** Bug 1738604 has been marked as a duplicate of this bug. ***

[fakenxid (fakenxid-redhat-bugs)]

*** Bug 1670504 has been marked as a duplicate of this bug. ***

[fakenxid (fakenxid-redhat-bugs)]

*** Bug 1677547 has been marked as a duplicate of this bug. ***

[fakenxid (fakenxid-redhat-bugs)]

*** Bug 1683074 has been marked as a duplicate of this bug. ***

[fakenxid (fakenxid-redhat-bugs)]

*** Bug 1684757 has been marked as a duplicate of this bug. ***

[fakenxid (fakenxid-redhat-bugs)]

*** Bug 1725310 has been marked as a duplicate of this bug. ***

[fakenxid (fakenxid-redhat-bugs)]

*** Bug 1693960 has been marked as a duplicate of this bug. ***

[fakenxid (fakenxid-redhat-bugs)]

*** Bug 1726553 has been marked as a duplicate of this bug. ***

[fakenxid (fakenxid-redhat-bugs)]

Magic combo is previous + next for a view. Affected: Firefox, Chromium, Thunar >:].

  :))

[fakenxid (fakenxid-redhat-bugs)]

Hi. Maybe it resets list of arrays, but I don't know
which array?

:))

[fakenxid (fakenxid-redhat-bugs)]

(In reply to klfyt from comment #28)
> Magic combo is previous + next for a view. Affected: Firefox, Chromium,
> Thunar >:].
>
> :))

previous + previous is even faster

as is typing text if you keep two empty [ ] spaces before text, example "text__"

:))

[fakenxid (fakenxid-redhat-bugs)]

Created attachment 1615120
little graphic (I can't get any closer to the answer - memory allocation? core affinity?)

[fakenxid (fakenxid-redhat-bugs)]

(In reply to klfyt from comment #31)
> Created attachment 1615120 [details]
> little graphic (I can't get any closer to the answer - memory allocation?
> core affinity?)

https://bugzilla.redhat.com/show_bug.cgi?id=1744930
This seems to improve latency.

[fakenxid (fakenxid-redhat-bugs)]

Scrolling is also affected by NumPad and CapsLock,
if you turn both of them on then scrolling is
faster than turning either of them on.

NUMPAD + CAPSLOCK -> FASTER THAN CAPSCLOCK

NUMPAD + CAPSLOCK -> FASTER THAN NUMPAD

>:]