Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.debian.user > #182562 > unrolled thread

Stretch upgrade from Jessie -- several issues

Back to article view | Back to linux.debian.user

Contents

  Stretch upgrade from Jessie -- several issues Ralph Katz <ralph@ralphkatz.com> - 2017-06-23 23:50 +0200
    Re: Stretch upgrade from Jessie -- several issues Felix Miata <mrmazda@earthlink.net> - 2017-06-24 00:50 +0200
      Re: Stretch upgrade from Jessie -- several issues Ralph Katz <ralph@ralphkatz.com> - 2017-06-24 01:40 +0200
        Re: Stretch upgrade from Jessie -- several issues Felix Miata <mrmazda@earthlink.net> - 2017-06-24 03:20 +0200
          Re: Stretch upgrade from Jessie -- several issues Ralph Katz <ralph@ralphkatz.com> - 2017-06-24 03:30 +0200
            Re: Stretch upgrade from Jessie -- several issues Felix Miata <mrmazda@earthlink.net> - 2017-06-24 09:30 +0200
              Re: Stretch upgrade from Jessie -- several issues Ralph Katz <ralph@ralphkatz.com> - 2017-06-24 17:50 +0200
                Re: Stretch upgrade from Jessie -- several issues Ralph Katz <ralph@ralphkatz.com> - 2017-06-25 19:20 +0200
                  Re: Stretch upgrade from Jessie -- several issues Mike Kupfer <mkupfer@alum.berkeley.edu> - 2017-06-25 23:20 +0200
                    Re: Stretch upgrade from Jessie -- several issues Ralph Katz <ralph@ralphkatz.com> - 2017-06-25 23:50 +0200
                      light-locker (was Stretch upgrade from Jessie -- several issues) Mike Kupfer <mkupfer@alum.berkeley.edu> - 2017-06-26 01:20 +0200
                        Re: light-locker (was Stretch upgrade from Jessie -- several issues) Ralph Katz <ralph@ralphkatz.com> - 2017-06-26 03:30 +0200
                          Re: light-locker (was Stretch upgrade from Jessie -- several issues) Mike Kupfer <mkupfer@alum.berkeley.edu> - 2017-06-26 04:30 +0200
                          Re: light-locker (was Stretch upgrade from Jessie -- several issues) Greg Wooledge <wooledg@eeg.ccf.org> - 2017-06-26 14:10 +0200
                            Re: light-locker (was Stretch upgrade from Jessie -- several issues) Ralph Katz <ralph@ralphkatz.com> - 2017-06-26 20:00 +0200
                  Re: Stretch upgrade from Jessie -- several issues Floris <jkfloris@dds.nl> - 2017-06-26 00:30 +0200
                    Re: Stretch upgrade from Jessie -- several issues Ralph Katz <ralph@ralphkatz.com> - 2017-06-26 03:30 +0200
                      Re: Stretch upgrade from Jessie -- several issues Floris <jkfloris@dds.nl> - 2017-06-26 09:00 +0200
                        Re: Stretch upgrade from Jessie -- several issues Ralph Katz <ralph@ralphkatz.com> - 2017-06-26 20:00 +0200
                          Re: Stretch upgrade from Jessie -- several issues Dejan Jocic <jodejka@gmail.com> - 2017-06-26 21:00 +0200
                            SOLVED xfce volume keys [was: Re: Stretch upgrade from Jessie --  several issues] Ralph Katz <ralph@ralphkatz.com> - 2017-06-26 21:50 +0200
    Re: Stretch upgrade from Jessie -- several issues Jimmy Johnson <field.engineer@gmail.com> - 2017-06-24 05:30 +0200
      Re: Stretch upgrade from Jessie -- several issues Ralph Katz <ralph@ralphkatz.com> - 2017-06-24 06:10 +0200
        Re: Stretch upgrade from Jessie -- several issues Jimmy Johnson <field.engineer@gmail.com> - 2017-06-24 07:40 +0200

Page 1 of 2  [1] 2  Next page →

#182562 — Stretch upgrade from Jessie -- several issues

[Multipart message — attachments visible in raw view] — view raw

Since upgrading my old Acer laptop from jessie to stretch this week, I
have several unresolved issues.  Any help would be appreciated.

1)  Sometimes unresponsive.  In up-to-date jessie, I could have two
instances of firefox-esr open with multiple tabs and get reasonable
response.  In stretch, sometimes performance is completely normal, but
at random intervals, the desktop becomes unresponsive as disk activity
takes over.  Top shows firefox very active and ps shows a number of
kworker processes recently started.  But closing one browser and
reducing open tabs does not seem to matter.  How can I identify the
offending processes?  No hints appear in log files using the Linux 3.16
kernel.

2) Returning from xscreensaver screen blanking sometimes shows the
lightdm log in screen.  Ctrl-alt-F7 once or twice can get my session
back after an annoying pause.  This never happened on jessie.

3) With the new Linux 4.9.0-3-amd64 kernel standard with stretch, the
problems above were worse; response was periodically lagging, and every
return from screensaver caused this in the syslog:

> Jun 19 14:17:08 spike2 kernel: [ 2268.384149] [drm:drm_atomic_helper_commit_clea
> nup_done [drm_kms_helper]] *ERROR* [CRTC:26:pipe A] flip_done timed out
> Jun 19 14:17:08 spike2 kernel: [ 2268.484810] ------------[ cut here ]----------
> --
> Jun 19 14:17:08 spike2 kernel: [ 2268.484859] WARNING: CPU: 1 PID: 2491 at /build/linux-hZ3RQX/linux-4.9.30/drivers/gpu/drm/drm_irq.c:1254 drm_wait_one_vblank+0x197/0x1a0 [drm]
> Jun 19 14:17:08 spike2 kernel: [ 2268.484861] vblank wait timed out on crtc 0
[Call Trace snipped]

ralph@spike2 ~$ zgrep "Call Trace" /var/log/syslog* |wc
    222    1899   17536

Going back to Linux 3.16.0-4-amd64 avoids all that.  Should I file a
bug?  Against which package?

3) The sound volume up/down function keys [Fn up/down arrow] no longer
respond (xfce4).  However the screen brightness keys function normally
[Fn left/right arrow].  The mute sound function key is also ignored [Fn
F8].  I can control volume with the mouse using pavucontrol or
gkrellm-volume.

I find nothing in the docs to enable these keys.  On the net, I found a
suggestion that only caused the keyboard setting app to run 100% cpu.

https://www.quora.com/How-do-I-control-volume-using-keyboard-shortcut-keys-in-Debian-Xfce

Again, any suggestions are welcome.  Thanks!

[toc] | [next] | [standalone]

#182564

Ralph Katz composed on 2017-06-23 15:44 (UTC-0600):

> Since upgrading my old Acer laptop from jessie to stretch this week, I
> have several unresolved issues.  Any help would be appreciated....
Help us help you by providing information other than symptoms, e.g.:

upgrade steps you took
which WM/DE you use
installed version of Firefox-ESR (45 vs 52)
whether it helps if you use a virgin Firefox profile to run Firefox
whether it helps if you create a new user to run Firefox
URL of Acer laptop specifications
output from 'inxi -c0 -v4' ('apt install inxi' if necessary)
output from 'lspci -nnk | grep -A4 VGA'
output from 'dmesg | grep error' or any suspicious part of dmesg
any suspicious output from the tail of journalctl
content of /etc/X11/xorg.conf if it exists

especially:
URL of pastebinit Xorg.0.log ('apt install pastebinit' if necessary)
-- 
"The wise are known for their understanding, and pleasant
words are persuasive." Proverbs 16:21 (New Living Translation)

 Team OS/2 ** Reg. Linux User #211409 ** a11y rocks!

Felix Miata  ***  http://fm.no-ip.com/

[toc] | [prev] | [next] | [standalone]

#182565

[Multipart message — attachments visible in raw view] — view raw

On 06/23/2017 04:47 PM, Felix Miata wrote:
> Ralph Katz composed on 2017-06-23 15:44 (UTC-0600):
> 
>> Since upgrading my old Acer laptop from jessie to stretch this week, I
>> have several unresolved issues.  Any help would be appreciated....
> Help us help you by providing information other than symptoms, e.g.:
> 
> upgrade steps you took

From jessie:
sudo apt-get update && sudo apt-get dist-upgrade
sudo dpkg --audit
aptitude search "~ahold"
dpkg -l "linux-image*" | grep ^ii | grep -i meta
sudo apt-get purge $(dpkg -l | awk '/^rc/ { print $2 }')

changed apt.sources to stretch, then:
sudo apt-get update && sudo apt-get upgrade
sudo apt-get dist-upgrade
Fixed packages left half-configured with apt-get -f, dpkg -i [offending
package], dpkg --configure --pending, etc
sudo dpkg --audit to confirm all's well.

> which WM/DE you use

As originally posted, lightdm, xfce4

> installed version of Firefox-ESR (45 vs 52)

52.2.0esr-1~deb9u1

> whether it helps if you use a virgin Firefox profile to run Firefox
> whether it helps if you create a new user to run Firefox

good suggestions, will try

> URL of Acer laptop specifications
 Too old for url.  But does it matter since it's fine on jessie?

Acer Extensa 4620-6456, Intel(R) Core(TM)2 Duo CPU T5550  @ 1.83GHz
2 Gb, 120 Gb disk, would you like the output of lshw?

> output from 'inxi -c0 -v4' ('apt install inxi' if necessary)

will lshw do?

> output from 'lspci -nnk | grep -A4 VGA'

lspci -nnk | grep -A4 VGA

00:02.0 VGA compatible controller [0300]: Intel Corporation Mobile
GM965/GL960 Integrated Graphics Controller (primary) [8086:2a02] (rev 03)
	Subsystem: Acer Incorporated [ALI] Mobile GM965/GL960 Integrated
Graphics Controller (primary) [1025:011c]
	Kernel driver in use: i915
	Kernel modules: i915
00:02.1 Display controller [0380]: Intel Corporation Mobile GM965/GL960
Integrated Graphics Controller (secondary) [8086:2a03] (rev 03)


> output from 'dmesg | grep error' or any suspicious part of dmesg

Nothing of note since current boot, uptime 1 day, 9 hrs

> any suspicious output from the tail of journalctl

Nothing suspicious.

> content of /etc/X11/xorg.conf if it exists

No file.

> especially:
> URL of pastebinit Xorg.0.log ('apt install pastebinit' if necessary)

http://paste.debian.net/973017/

Thanks!
Ralph

[toc] | [prev] | [next] | [standalone]

#182567

Ralph Katz composed on 2017-06-23 17:37 (UTC-0600):

> Felix Miata wrote:
...
>> output from 'inxi -c0 -v4' ('apt install inxi' if necessary)
.
> will lshw do?
.
Generally its overly verbose for these things. Other info you provided probably
covers well enough.
.
> lspci -nnk | grep -A4 VGA
...
> 00:02.1 Display controller [0380]: Intel Corporation Mobile GM965/GL960
> Integrated Graphics Controller (secondary) [8086:2a03] (rev 03)
...
>> especially:
>> URL of pastebinit Xorg.0.log ('apt install pastebinit' if necessary)
.
> http://paste.debian.net/973017/
.
Quite a surprise. The modeset Xorg driver was integrated into the server since
Jessie's release. Supposedly it only works with approximately two generation
newer chipsets than your i965. AIUI it was intended primarily for the latest
generations of gfx chipsets regardless of vendor. I have newer Intel (desktop)
gfxchips (G31 & G33) that refuse to use the modeset driver, though I also have a
Q965 (desktop) that can use it. I suspect its development got little testing
with so old a gfxchip. Give the Intel driver a try. It should be as simple as
'apt install xserver-xorg-video-intel' and restarting X.
-- 
"The wise are known for their understanding, and pleasant
words are persuasive." Proverbs 16:21 (New Living Translation)

 Team OS/2 ** Reg. Linux User #211409 ** a11y rocks!

Felix Miata  ***  http://fm.no-ip.com/

[toc] | [prev] | [next] | [standalone]

#182568

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA512

On 06/23/2017 07:13 PM, Felix Miata wrote:
> apt install xserver-xorg-video-intel' and restarting X

It was already installed.


-----BEGIN PGP SIGNATURE-----
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
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=O2nv
-----END PGP SIGNATURE-----

[toc] | [prev] | [next] | [standalone]

#182581

Ralph Katz composed on 2017-06-23 19:29 (UTC-0600):
.
> Felix Miata wrote:
.
>> apt install xserver-xorg-video-intel' and restarting X
.
> It was already installed.
Right, upgrade, not new installation, so kept. :-p

Is firmware-misc-nonfree installed?

https://wiki.debian.org/NewInStretch (at the bottom):

Try 'apt install xserver-xorg-legacy'. Whether that is simply about setuid or
more I haven't figured out, and likewise, how Stretch decides whether to use the
intel or the modeset driver.

Use of the intel driver can be forced (tested here on i965Q) by creating:

File:
/etc/X11/xorg.conf.d/20-use-intel-gfx-driver.conf

Containing:
    Section "Device"
	Identifier "somestring"
	Driver	"intel"
    EndSection

It may or may not need to include any of the following if you experience visible
disruption:
	Option 	"AccelMethod" 	"sna"
	Option 	"AccelMethod" 	"uxa"
	Option 	"TearFree" 	"true"
-- 
"The wise are known for their understanding, and pleasant
words are persuasive." Proverbs 16:21 (New Living Translation)

 Team OS/2 ** Reg. Linux User #211409 ** a11y rocks!

Felix Miata  ***  http://fm.no-ip.com/

[toc] | [prev] | [next] | [standalone]

#182602

[Multipart message — attachments visible in raw view] — view raw

On 06/24/2017 01:19 AM, Felix Miata wrote:
...
> 
> Is firmware-misc-nonfree installed?

Yes.

> 
> https://wiki.debian.org/NewInStretch (at the bottom):
> 
> Try 'apt install xserver-xorg-legacy'. Whether that is simply about setuid or
> more I haven't figured out, and likewise, how Stretch decides whether to use the
> intel or the modeset driver.
> 
> Use of the intel driver can be forced (tested here on i965Q) by creating:
> 
> File:
> /etc/X11/xorg.conf.d/20-use-intel-gfx-driver.conf
> 
> Containing:
>     Section "Device"
> 	Identifier "somestring"
> 	Driver	"intel"
>     EndSection
> 
> It may or may not need to include any of the following if you experience visible
> disruption:
> 	Option 	"AccelMethod" 	"sna"
> 	Option 	"AccelMethod" 	"uxa"
> 	Option 	"TearFree" 	"true"
> 

Felix, thanks for the tips!  I had missed the wiki page.  Since my
graphics performance is fine as before, I'll just hit ctl-alt-F7 when
coming back from blanked screen to resume my session.

At your suggestion, I'm re-doing my firefox setup, re-install, fresh
profile, eliminate extensions, etc.  Report to follow.

Anyone have ideas on the non-functioning sound volume keys (in original
post)?

Thanks!

[toc] | [prev] | [next] | [standalone]

#182643

[Multipart message — attachments visible in raw view] — view raw

On 06/24/2017 09:48 AM, Ralph Katz wrote:
...
> At your suggestion, I'm re-doing my firefox setup, re-install, fresh
> profile, eliminate extensions, etc.  Report to follow.

Update report (quoting original thread post):

> Since upgrading my old Acer laptop from jessie to stretch this week, I
> have several unresolved issues.  Any help would be appreciated.
> 
> 1)  Sometimes unresponsive.  In up-to-date jessie, I could have two
> instances of firefox-esr open with multiple tabs and get reasonable
> response.  In stretch, sometimes performance is completely normal, but
> at random intervals, the desktop becomes unresponsive as disk activity
> takes over.  Top shows firefox very active and ps shows a number of
> kworker processes recently started.  But closing one browser and
> reducing open tabs does not seem to matter.  How can I identify the
> offending processes?  No hints appear in log files using the Linux 3.16
> kernel.

Firefox fix:  Went back to Linux 4.09 kernel. Moved ~/.mozilla/ out of
the way, removed all firefox non-debian extensions (ghostery, noscript,
adblock), reinstalled firefox-esr and flashplayer-mozilla, installed
xul-ext-noscript and xul-ext-ublock-origin, manually reset preferences,
imported old bookmarks, history, passwords, noscript whitelist, etc.
Load average in top has gone from an unusable 2,3,5 even 7 to under .50.

> 2) Returning from xscreensaver screen blanking sometimes shows the
> lightdm log in screen.  Ctrl-alt-F7 once or twice can get my session
> back after an annoying pause.  This never happened on jessie.

This problem got worse, so rebooted into Linux 4.09 kernel to work it
out and maintain a supportable security os.  I've uploaded logs as follows:

kernel errors on boot, *ERROR* [CRTC:29:pipe B] flip_done timed out:
syslog:  http://paste.debian.net/973246/

Xorg.0.log:  http://paste.debian.net/973247/

lightdm.log: http://paste.debian.net/973248/

> 3) The sound volume up/down function keys [Fn up/down arrow] no longer
> respond (xfce4).  However the screen brightness keys function normally
> [Fn left/right arrow].  The mute sound function key is also ignored [Fn
> F8].  I can control volume with the mouse using pavucontrol or
> gkrellm-volume.

No progress.

Thanks again for any help.

Ralph

[toc] | [prev] | [next] | [standalone]

#182650

Ralph Katz wrote:

> > 2) Returning from xscreensaver screen blanking sometimes shows the
> > lightdm log in screen.  Ctrl-alt-F7 once or twice can get my session
> > back after an annoying pause.  This never happened on jessie.

This is a bit of a shot in the dark, but is light-locker running?

mike

[toc] | [prev] | [next] | [standalone]

#182651

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA512

On 06/25/2017 02:39 PM, Mike Kupfer wrote:
> Ralph Katz wrote:
> 
>>> 2) Returning from xscreensaver screen blanking sometimes shows the
>>> lightdm log in screen.  Ctrl-alt-F7 once or twice can get my session
>>> back after an annoying pause.  This never happened on jessie.
> 
> This is a bit of a shot in the dark, but is light-locker running?
> 
> mike

Yes!  apt history shows light-locker was installed by the dist-upgrade.
 I've just purged xscreensaver and can confirm that the time-out errors
in syslog occur on resume from screen blanking.

Other than pushing out the screen lock time which I've done in xfce
settings, or disabling the lock, what do you suggest?  light-locker is a
"depends" of task-xfce-desktop so not removable.

Thanks Mike!

-----BEGIN PGP SIGNATURE-----
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
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=uf7s
-----END PGP SIGNATURE-----

[toc] | [prev] | [next] | [standalone]

#182656 — light-locker (was Stretch upgrade from Jessie -- several issues)

Ralph Katz wrote:

> Yes!  apt history shows light-locker was installed by the dist-upgrade.
>  I've just purged xscreensaver and can confirm that the time-out errors
> in syslog occur on resume from screen blanking.

Did you also kill the xscreensaver daemon?  You might have been getting
odd behavior due to xscreensaver and light-locker/lightdm fighting over
the display.

I don't know enough about X to know whether those timeout errors are
important.  When I've used light-locker, the display would typically
blank briefly before showing the lightdm screen.  (That's how
light-locker works: it starts a lightdm instance, which you "log in" to
in order to unlock the session.)

> Other than pushing out the screen lock time which I've done in xfce
> settings, or disabling the lock, what do you suggest?  light-locker is a
> "depends" of task-xfce-desktop so not removable.

If light-locker running by itself still isn't behaving properly, you
could disable the daemon via the Xfce session control panel.  Untick the
box in the list of startup applications, and make sure it's not listed
in the list of currently running applications.

That would leave you without a screen lock, but I suppose you could
reinstall xscreensaver.

mike

[toc] | [prev] | [next] | [standalone]

#182662 — Re: light-locker (was Stretch upgrade from Jessie -- several issues)

[Multipart message — attachments visible in raw view] — view raw

On 06/25/2017 05:18 PM, Mike Kupfer wrote:
> Ralph Katz wrote:
> 
>> Yes!  apt history shows light-locker was installed by the dist-upgrade.
>>  I've just purged xscreensaver and can confirm that the time-out errors
>> in syslog occur on resume from screen blanking.
> 
> Did you also kill the xscreensaver daemon?  You might have been getting
> odd behavior due to xscreensaver and light-locker/lightdm fighting over
> the display.

Yes, apt-get purge killed the daemon as well.

> I don't know enough about X to know whether those timeout errors are
> important.  When I've used light-locker, the display would typically
> blank briefly before showing the lightdm screen.  (That's how
> light-locker works: it starts a lightdm instance, which you "log in" to
> in order to unlock the session.)

Yes, but resuming from blank causes the kernel errors in syslog.

> 
>> Other than pushing out the screen lock time which I've done in xfce
>> settings, or disabling the lock, what do you suggest?  light-locker is a
>> "depends" of task-xfce-desktop so not removable.
> 
> If light-locker running by itself still isn't behaving properly, you
> could disable the daemon via the Xfce session control panel.  Untick the
> box in the list of startup applications, and make sure it's not listed
> in the list of currently running applications.
> 
> That would leave you without a screen lock, but I suppose you could
> reinstall xscreensaver.

While purging xscreensaver removed 2 or 3 packages, installing now would
bring in a host of packages, for whatever reason.  I don't need a
screensaver as much as I need the lock.

Ralph

[toc] | [prev] | [next] | [standalone]

#182665 — Re: light-locker (was Stretch upgrade from Jessie -- several issues)

Ralph Katz wrote:

> While purging xscreensaver removed 2 or 3 packages, installing now would
> bring in a host of packages, for whatever reason.  I don't need a
> screensaver as much as I need the lock.

Take a look at /usr/bin/xflock4.  It's just a script that runs through a
fixed list of lock programs until it finds one that runs successfully.
Maybe you can install one of the other commands that it tries.

mike

[toc] | [prev] | [next] | [standalone]

#182672 — Re: light-locker (was Stretch upgrade from Jessie -- several issues)

On Sun, Jun 25, 2017 at 07:23:10PM -0600, Ralph Katz wrote:
> While purging xscreensaver removed 2 or 3 packages, installing now would
> bring in a host of packages, for whatever reason.  I don't need a
> screensaver as much as I need the lock.

I use i3lock, from a terminal, exactly as I used to use "xlock" when
Debian still had that.  (I don't use any other part of i3.)

[toc] | [prev] | [next] | [standalone]

#182701 — Re: light-locker (was Stretch upgrade from Jessie -- several issues)

[Multipart message — attachments visible in raw view] — view raw

On 06/26/2017 06:09 AM, Greg Wooledge wrote:
> On Sun, Jun 25, 2017 at 07:23:10PM -0600, Ralph Katz wrote:
>> While purging xscreensaver removed 2 or 3 packages, installing now would
>> bring in a host of packages, for whatever reason.  I don't need a
>> screensaver as much as I need the lock.
> 
> I use i3lock, from a terminal, exactly as I used to use "xlock" when
> Debian still had that.  (I don't use any other part of i3.)
> 

light-locker works fine until the kernel errors make it hard to re-enter
my open session.  This morning I had to restart lightdm because of the
errors:

> ralph@spike2 ~$ grep ERROR /var/log/syslog 
> grep: warning: GREP_OPTIONS is deprecated; please use an alias or script
> Jun 26 06:00:19 spike2 kernel: [49759.456083] [drm:drm_atomic_helper_commit_cleanup_done [drm_kms_helper]] *ERROR* [CRTC:26:pipe A] flip_done timed out
> Jun 26 06:01:59 spike2 kernel: [49859.296184] [drm:drm_atomic_helper_commit_cleanup_done [drm_kms_helper]] *ERROR* [CRTC:26:pipe A] flip_done timed out
> Jun 26 06:07:08 spike2 kernel: [50168.544120] [drm:drm_atomic_helper_commit_cleanup_done [drm_kms_helper]] *ERROR* [CRTC:26:pipe A] flip_done timed out
> Jun 26 06:10:02 spike2 kernel: [50342.624077] [drm:drm_atomic_helper_commit_cleanup_done [drm_kms_helper]] *ERROR* [CRTC:26:pipe A] flip_done timed out
...
> ralph@spike2 ~$ grep ERROR /var/log/syslog |wc -l
> grep: warning: GREP_OPTIONS is deprecated; please use an alias or script
> 24
> ralph@spike2 ~$ 

The workaround is don't blank or sleep. ;-)

[toc] | [prev] | [next] | [standalone]

#182654

Op Sun, 25 Jun 2017 19:10:40 +0200 schreef Ralph Katz  
<ralph@ralphkatz.com>:
>
>> 3) The sound volume up/down function keys [Fn up/down arrow] no longer
>> respond (xfce4).  However the screen brightness keys function normally
>> [Fn left/right arrow].  The mute sound function key is also ignored [Fn
>> F8].  I can control volume with the mouse using pavucontrol or
>> gkrellm-volume.
>
> No progress.

if the keys are recognized by the kernel, then you can remap them with an  
udev hwdb rule.

Test with evtest if the keys get a scancode and a keycode. Something like:
	type 4 (EV_MSC), code 4 (MSC_SCAN), value 700e0
	type 1 (EV_KEY), code 29 (KEY_LEFTCTRL), value 1
In this case I pressed the Left Ctrl key, which has a value of 700e0

Create a new rule in /etc/udev/hwdb.d with something like:

evdev:atkbd:dmi:bvn*:bvr*:bd*:svnAcer*:pn*
  KEYBOARD_KEY_ed=volumeup
  KEYBOARD_KEY_ee=volumedown
  etc...

use evemu-describe to find your evdev:... line

also read:
https://wiki.archlinux.org/index.php/Map_scancodes_to_keycodes
https://wiki.archlinux.org/index.php/Extra_keyboard_keys
/lib/udev/hwdb.d/60-keyboard.hwdb
for more information and examples

Floris

[toc] | [prev] | [next] | [standalone]

#182663

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA512

On 06/25/2017 04:22 PM, Floris wrote:
> 
> Op Sun, 25 Jun 2017 19:10:40 +0200 schreef Ralph Katz 
> <ralph@ralphkatz.com>:
>> 
>>> 3) The sound volume up/down function keys [Fn up/down arrow] no
>>> longer respond (xfce4).  However the screen brightness keys
>>> function normally [Fn left/right arrow].  The mute sound
>>> function key is also ignored [Fn F8].  I can control volume
>>> with the mouse using pavucontrol or gkrellm-volume.
>> 
>> No progress.
> 
> if the keys are recognized by the kernel, then you can remap them
> with an udev hwdb rule.
> 
> Test with evtest if the keys get a scancode and a keycode.
> Something like: type 4 (EV_MSC), code 4 (MSC_SCAN), value 700e0 
> type 1 (EV_KEY), code 29 (KEY_LEFTCTRL), value 1 In this case I
> pressed the Left Ctrl key, which has a value of 700e0
> 
> Create a new rule in /etc/udev/hwdb.d with something like:
> 
> evdev:atkbd:dmi:bvn*:bvr*:bd*:svnAcer*:pn* 
> KEYBOARD_KEY_ed=volumeup KEYBOARD_KEY_ee=volumedown etc...
> 
> use evemu-describe to find your evdev:... line
> 
> also read: 
> https://wiki.archlinux.org/index.php/Map_scancodes_to_keycodes 
> https://wiki.archlinux.org/index.php/Extra_keyboard_keys 
> /lib/udev/hwdb.d/60-keyboard.hwdb for more information and
> examples
> 
> Floris

Thanks Floris!  I will look into this, but I am confounded as to why
on jessie and previous versions, recognizing these sound keys was
automatic.

Ralph


-----BEGIN PGP SIGNATURE-----
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
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=YzSP
-----END PGP SIGNATURE-----

[toc] | [prev] | [next] | [standalone]

#182669

Op Mon, 26 Jun 2017 03:23:32 +0200 schreef Ralph Katz  
<ralph@ralphkatz.com>:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA512
>
> On 06/25/2017 04:22 PM, Floris wrote:
>>
>> Op Sun, 25 Jun 2017 19:10:40 +0200 schreef Ralph Katz
>> <ralph@ralphkatz.com>:
>>>
>>>> 3) The sound volume up/down function keys [Fn up/down arrow] no
>>>> longer respond (xfce4).  However the screen brightness keys
>>>> function normally [Fn left/right arrow].  The mute sound
>>>> function key is also ignored [Fn F8].  I can control volume
>>>> with the mouse using pavucontrol or gkrellm-volume.
>>>
>>> No progress.
>>
>> if the keys are recognized by the kernel, then you can remap them
>> with an udev hwdb rule.
>>
>> Test with evtest if the keys get a scancode and a keycode.
>> Something like: type 4 (EV_MSC), code 4 (MSC_SCAN), value 700e0
>> type 1 (EV_KEY), code 29 (KEY_LEFTCTRL), value 1 In this case I
>> pressed the Left Ctrl key, which has a value of 700e0
>>
>> Create a new rule in /etc/udev/hwdb.d with something like:
>>
>> evdev:atkbd:dmi:bvn*:bvr*:bd*:svnAcer*:pn*
>> KEYBOARD_KEY_ed=volumeup KEYBOARD_KEY_ee=volumedown etc...
>>
>> use evemu-describe to find your evdev:... line
>>
>> also read:
>> https://wiki.archlinux.org/index.php/Map_scancodes_to_keycodes
>> https://wiki.archlinux.org/index.php/Extra_keyboard_keys
>> /lib/udev/hwdb.d/60-keyboard.hwdb for more information and
>> examples
>>
>> Floris
>
> Thanks Floris!  I will look into this, but I am confounded as to why
> on jessie and previous versions, recognizing these sound keys was
> automatic.
>
> Ralph
>

Maybe the switch from "the old" xserver-xorg-input-evdev to "the Wayland"  
xserver-xorg-input-libinput has something to do with it. But I don't know  
if you have the libinput version installed.

https://packages.debian.org/stretch/xserver-xorg-input-libinput
...
X.Org X server -- libinput input driver

This package provides the driver for input devices using libinput library.  
It can handle keyboards, mice and touchpads, and essentially replaces the  
separate -evdev and -synaptics drivers.
...

Floris

[toc] | [prev] | [next] | [standalone]

#182702

[Multipart message — attachments visible in raw view] — view raw

On 06/26/2017 12:50 AM, Floris wrote:
> Op Mon, 26 Jun 2017 03:23:32 +0200 schreef Ralph Katz
> <ralph@ralphkatz.com>:
> 
>> -----BEGIN PGP SIGNED MESSAGE-----
>> Hash: SHA512
>>
>> On 06/25/2017 04:22 PM, Floris wrote:
>>>
>>> Op Sun, 25 Jun 2017 19:10:40 +0200 schreef Ralph Katz
>>> <ralph@ralphkatz.com>:
>>>>
>>>>> 3) The sound volume up/down function keys [Fn up/down arrow] no
>>>>> longer respond (xfce4).  However the screen brightness keys
>>>>> function normally [Fn left/right arrow].  The mute sound
>>>>> function key is also ignored [Fn F8].  I can control volume
>>>>> with the mouse using pavucontrol or gkrellm-volume.
>>>>
>>>> No progress.
>>>
>>> if the keys are recognized by the kernel, then you can remap them
>>> with an udev hwdb rule.
>>>
>>> Test with evtest if the keys get a scancode and a keycode.
>>> Something like: type 4 (EV_MSC), code 4 (MSC_SCAN), value 700e0
>>> type 1 (EV_KEY), code 29 (KEY_LEFTCTRL), value 1 In this case I
>>> pressed the Left Ctrl key, which has a value of 700e0
>>>
>>> Create a new rule in /etc/udev/hwdb.d with something like:
>>>
>>> evdev:atkbd:dmi:bvn*:bvr*:bd*:svnAcer*:pn*
>>> KEYBOARD_KEY_ed=volumeup KEYBOARD_KEY_ee=volumedown etc...
>>>
>>> use evemu-describe to find your evdev:... line
>>>
>>> also read:
>>> https://wiki.archlinux.org/index.php/Map_scancodes_to_keycodes
>>> https://wiki.archlinux.org/index.php/Extra_keyboard_keys
>>> /lib/udev/hwdb.d/60-keyboard.hwdb for more information and
>>> examples
>>>
>>> Floris
>>
>> Thanks Floris!  I will look into this, but I am confounded as to why
>> on jessie and previous versions, recognizing these sound keys was
>> automatic.
>>
>> Ralph
>>
> 
> Maybe the switch from "the old" xserver-xorg-input-evdev to "the
> Wayland" xserver-xorg-input-libinput has something to do with it. But I
> don't know if you have the libinput version installed.
> 
> https://packages.debian.org/stretch/xserver-xorg-input-libinput
> ...
> X.Org X server -- libinput input driver
> 
> This package provides the driver for input devices using libinput
> library. It can handle keyboards, mice and touchpads, and essentially
> replaces the separate -evdev and -synaptics drivers.
> ...
> 
> Floris

Yes, the libinput packages were pulled in by the dist-upgrade.  Nothing
in /usr/share/doc/xserver-xorg-input-libinput/ gives any clue on how to
use it.  There is no man page I can find.  There is
/usr/share/X11/xorg.conf.d/40-libinput.conf but again no clue on usage.

I think the workaround of using the mouse on pavucontrol or
gkrellm-volume will have to do for sound control.  Thanks!

[toc] | [prev] | [next] | [standalone]

#182709

On 26-06-17, Ralph Katz wrote:
> 
> I think the workaround of using the mouse on pavucontrol or
> gkrellm-volume will have to do for sound control.  Thanks!
> 

Well, not exactly solution, but as a work around, you can assign some
shortcuts to do your sound work in a same way volume keys would do. For
example, in case of the pulse audio you can use these:

pactl set-sink-volume 0 -- +5% to increase volume by 5%

pactl set-sink-volume 0 -- -5% to decrease volume by 5%

pactl set-sink-mute 0 toggle to mute/unmute

In case of ALSA, should be something like:

amixer set Master 5%+ to increase

amixer set Master 5%- to decrease

amixer set Master toggle to mute/unmute

[toc] | [prev] | [next] | [standalone]

Page 1 of 2  [1] 2  Next page →

Back to top | Article view | linux.debian.user

csiph-web