2014-06-18 09:35:43

by Tomasz Torcz

[permalink] [raw]
Subject: [PATCH] vt: disable console blanking by default


>From ea7cf47e3230eda63aa6c46092719437f9bbae8e Mon Sep 17 00:00:00 2001
From: Tomasz Torcz <[email protected]>
Date: Wed, 18 Jun 2014 11:18:06 +0200
Subject: [PATCH] vt: disable console blanking by default

Remove default 10 minute blank interval. Instead: never blank by
default.
"Screensaving" is no longer useful. Today it only provides
obstacle when interacting with text console, especially through
remote lights-out management solutions.

Signed-off-by: Tomasz Torcz <[email protected]>
---
Documentation/kernel-parameters.txt | 4 ++--
drivers/tty/vt/vt.c | 2 +-
2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 6eaa9cd..0618648 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -710,8 +710,8 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
For now, only VisioBraille is supported.

consoleblank= [KNL] The console blank (screen saver) timeout in
- seconds. Defaults to 10*60 = 10mins. A value of 0
- disables the blank timer.
+ seconds. A value of 0 disables the blank timer.
+ Defaults to never blank.

coredump_filter=
[KNL] Change the default value for
diff --git a/drivers/tty/vt/vt.c b/drivers/tty/vt/vt.c
index 5e0f6ff..dbd4f23 100644
--- a/drivers/tty/vt/vt.c
+++ b/drivers/tty/vt/vt.c
@@ -176,7 +176,7 @@ int console_blanked;

static int vesa_blank_mode; /* 0:none 1:suspendV 2:suspendH 3:powerdown */
static int vesa_off_interval;
-static int blankinterval = 10*60;
+static int blankinterval;
core_param(consoleblank, blankinterval, int, 0444);

static DECLARE_WORK(console_work, console_callback);
--
1.9.3

--
Tomasz Torcz Only gods can safely risk perfection,
xmpp: [email protected] it's a dangerous thing for a man. -- Alia


2014-06-22 16:37:32

by Pavel Machek

[permalink] [raw]
Subject: Re: [PATCH] vt: disable console blanking by default

On Wed 2014-06-18 11:27:31, Tomasz Torcz wrote:
>
> >From ea7cf47e3230eda63aa6c46092719437f9bbae8e Mon Sep 17 00:00:00 2001
> From: Tomasz Torcz <[email protected]>
> Date: Wed, 18 Jun 2014 11:18:06 +0200
> Subject: [PATCH] vt: disable console blanking by default
>
> Remove default 10 minute blank interval. Instead: never blank by
> default.
> "Screensaving" is no longer useful. Today it only provides
> obstacle when interacting with text console, especially through
> remote lights-out management solutions.
>
> Signed-off-by: Tomasz Torcz <[email protected]>

Dunno. You'll kill an LCD if you let same image there for too
long. Make it 10hours?

Pavel

> Documentation/kernel-parameters.txt | 4 ++--
> drivers/tty/vt/vt.c | 2 +-
> 2 files changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
> index 6eaa9cd..0618648 100644
> --- a/Documentation/kernel-parameters.txt
> +++ b/Documentation/kernel-parameters.txt
> @@ -710,8 +710,8 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
> For now, only VisioBraille is supported.
>
> consoleblank= [KNL] The console blank (screen saver) timeout in
> - seconds. Defaults to 10*60 = 10mins. A value of 0
> - disables the blank timer.
> + seconds. A value of 0 disables the blank timer.
> + Defaults to never blank.
>
> coredump_filter=
> [KNL] Change the default value for
> diff --git a/drivers/tty/vt/vt.c b/drivers/tty/vt/vt.c
> index 5e0f6ff..dbd4f23 100644
> --- a/drivers/tty/vt/vt.c
> +++ b/drivers/tty/vt/vt.c
> @@ -176,7 +176,7 @@ int console_blanked;
>
> static int vesa_blank_mode; /* 0:none 1:suspendV 2:suspendH 3:powerdown */
> static int vesa_off_interval;
> -static int blankinterval = 10*60;
> +static int blankinterval;
> core_param(consoleblank, blankinterval, int, 0444);
>
> static DECLARE_WORK(console_work, console_callback);

--
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

2014-06-25 11:41:14

by Tomasz Torcz

[permalink] [raw]
Subject: Re: [PATCH] vt: disable console blanking by default

On Sun, Jun 22, 2014 at 06:37:29PM +0200, Pavel Machek wrote:
> On Wed 2014-06-18 11:27:31, Tomasz Torcz wrote:
> >
> > >From ea7cf47e3230eda63aa6c46092719437f9bbae8e Mon Sep 17 00:00:00 2001
> > From: Tomasz Torcz <[email protected]>
> > Date: Wed, 18 Jun 2014 11:18:06 +0200
> > Subject: [PATCH] vt: disable console blanking by default
> >
> > Remove default 10 minute blank interval. Instead: never blank by
> > default.
> > "Screensaving" is no longer useful. Today it only provides
> > obstacle when interacting with text console, especially through
> > remote lights-out management solutions.
> >
> > Signed-off-by: Tomasz Torcz <[email protected]>
>
> Dunno. You'll kill an LCD if you let same image there for too
> long. Make it 10hours?

There is no practical difference between 10 minutes and 10 hours. Both
are more than none. This means that If I need to access console, it
will probably be blanked. For some servers, console access is needed
less than once per year, if ever. But IF the console need to be accessed,
there's a strong reason to it (some kind of fault), and having to press
anything to unblank may be invasive.

I would prefer to fix the issue at the source, instead of sticking "consoleblank=0"
in grub2.cfg on houndreds of boxes. BTW, kernel already contains "consoleblank=0" in
arch/arc/boot/dts/nsimosci.dts and arch/arm/boot/dts/omap3-lilly-a83x.dtsi

Is LCD damage for currently produced LCDs real issue, still?

--
Tomasz Torcz To co nierealne -- tutaj jest normalne.
xmpp: [email protected] Ziomale na życie mają tu patenty specjalne.

2014-06-25 12:40:08

by Alan Cox

[permalink] [raw]
Subject: Re: [PATCH] vt: disable console blanking by default

On Wed, 18 Jun 2014 11:27:31 +0200
Tomasz Torcz <[email protected]> wrote:

>
> >From ea7cf47e3230eda63aa6c46092719437f9bbae8e Mon Sep 17 00:00:00 2001
> From: Tomasz Torcz <[email protected]>
> Date: Wed, 18 Jun 2014 11:18:06 +0200
> Subject: [PATCH] vt: disable console blanking by default
>
> Remove default 10 minute blank interval. Instead: never blank by
> default.
> "Screensaving" is no longer useful. Today it only provides
> obstacle when interacting with text console, especially through
> remote lights-out management solutions.

Screensaving is extremely useful as is power management, and power prices
are going up not down. We don't customise the upstream kernel just
because J Random User happens to want his defaults.

There are a couple of things you can sensibly do

1. Just turn it off in software. It's run time configurable via setterm

2. If you have a reliable way of detecting the presence of a remote
console device and that device supports reporting when a connection is
made/broken you could submit a patch to make it unblank on connect.

Alan

2014-06-25 19:21:44

by Arkadiusz Miskiewicz

[permalink] [raw]
Subject: Re: [PATCH] vt: disable console blanking by default

On Wednesday 25 of June 2014, One Thousand Gnomes wrote:
> On Wed, 18 Jun 2014 11:27:31 +0200
>
> Tomasz Torcz <[email protected]> wrote:
> > >From ea7cf47e3230eda63aa6c46092719437f9bbae8e Mon Sep 17 00:00:00 2001
> >
> > From: Tomasz Torcz <[email protected]>
> > Date: Wed, 18 Jun 2014 11:18:06 +0200
> > Subject: [PATCH] vt: disable console blanking by default
> >
> > Remove default 10 minute blank interval. Instead: never blank by
> > default.
> > "Screensaving" is no longer useful. Today it only provides
> > obstacle when interacting with text console, especially through
> > remote lights-out management solutions.
>
> Screensaving is extremely useful as is power management, and power prices
> are going up not down. We don't customise the upstream kernel just
> because J Random User happens to want his defaults.
>
> There are a couple of things you can sensibly do
>
> 1. Just turn it off in software. It's run time configurable via setterm
>
> 2. If you have a reliable way of detecting the presence of a remote
> console device and that device supports reporting when a connection is
> made/broken you could submit a patch to make it unblank on connect.

3. unblank & disable blanking on any oops? (probably problematic)

Anyway I dislike blanking here, too.

> Alan


--
Arkadiusz Miśkiewicz, arekm / maven.pl

2014-06-26 21:35:49

by Alan Cox

[permalink] [raw]
Subject: Re: [PATCH] vt: disable console blanking by default

> 3. unblank & disable blanking on any oops? (probably problematic)

We already try to do that.

Alan

2014-06-29 20:35:08

by Pavel Machek

[permalink] [raw]
Subject: Re: [PATCH] vt: disable console blanking by default

On Wed 2014-06-25 13:41:09, Tomasz Torcz wrote:
> On Sun, Jun 22, 2014 at 06:37:29PM +0200, Pavel Machek wrote:
> > On Wed 2014-06-18 11:27:31, Tomasz Torcz wrote:
> > >
> > > >From ea7cf47e3230eda63aa6c46092719437f9bbae8e Mon Sep 17 00:00:00 2001
> > > From: Tomasz Torcz <[email protected]>
> > > Date: Wed, 18 Jun 2014 11:18:06 +0200
> > > Subject: [PATCH] vt: disable console blanking by default
> > >
> > > Remove default 10 minute blank interval. Instead: never blank by
> > > default.
> > > "Screensaving" is no longer useful. Today it only provides
> > > obstacle when interacting with text console, especially through
> > > remote lights-out management solutions.
> > >
> > > Signed-off-by: Tomasz Torcz <[email protected]>
> >
> > Dunno. You'll kill an LCD if you let same image there for too
> > long. Make it 10hours?
>
> There is no practical difference between 10 minutes and 10 hours. Both
> are more than none. This means that If I need to access console, it
> will probably be blanked. For some servers, console access is needed
> less than once per year, if ever. But IF the console need to be accessed,
> there's a strong reason to it (some kind of fault), and having to press
> anything to unblank may be invasive.
>
> I would prefer to fix the issue at the source, instead of sticking "consoleblank=0"
> in grub2.cfg on houndreds of boxes. BTW, kernel already contains "consoleblank=0" in
> arch/arc/boot/dts/nsimosci.dts and arch/arm/boot/dts/omap3-lilly-a83x.dtsi
>
> Is LCD damage for currently produced LCDs real issue, still?

Probably yes. Plus, people are still running CRTs and blanked screen
still eats significantly less power. So "consoleblank=0" seems right
solution for you, and current default still sounds right.
Pavel
--
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html