[PATCH] vt: Do not clear UTF when resetting console

[PATCH] vt: Do not clear UTF when resetting console

[Antonino A. Daplas]

Resetting the console, either by ANSI escape sequences or by the reset utility,
will drop the console back to legacy (non-UTF-8) mode. Fix this by leaving the
field vc_data.vc_utf untouched in reset_terminal(). In addition, a global
variable (default_utf8) which defines system-wide UTF-8 setting is created.
This variable can be adjusted via sysfs.

This is based from patches by Jan Engelhardt and Paul LeoNerd Evans.

Signed-off-by: Antonino Daplas <adaplas@gmail.com>
---

 drivers/char/vt.c |    4 +++-
 1 files changed, 3 insertions(+), 1 deletions(-)

diff --git a/drivers/char/vt.c b/drivers/char/vt.c
index 1bbb45b..fdd548f 100644
--- a/drivers/char/vt.c
+++ b/drivers/char/vt.c
@@ -157,6 +157,8 @@ static void blank_screen_t(unsigned long
 static void set_palette(struct vc_data *vc);
 
 static int printable;		/* Is console ready for printing? */
+static int default_utf8;
+module_param(default_utf8, int, S_IRUGO | S_IWUSR);
 
 /*
  * ignore_poke: don't unblank the screen when things are typed.  This is
@@ -1497,7 +1499,6 @@ static void reset_terminal(struct vc_dat
 	vc->vc_charset		= 0;
 	vc->vc_need_wrap	= 0;
 	vc->vc_report_mouse	= 0;
-	vc->vc_utf		= 0;
 	vc->vc_utf_count	= 0;
 
 	vc->vc_disp_ctrl	= 0;
@@ -2590,6 +2591,7 @@ static void vc_init(struct vc_data *vc, 
 	vc->vc_rows = rows;
 	vc->vc_size_row = cols << 1;
 	vc->vc_screenbuf_size = vc->vc_rows * vc->vc_size_row;
+	vc->vc_utf = default_utf8;
 
 	set_origin(vc);
 	vc->vc_pos = vc->vc_origin;

Re: [PATCH] vt: Do not clear UTF when resetting console

[H. Peter Anvin]

Antonino A. Daplas wrote:
> Resetting the console, either by ANSI escape sequences or by the reset utility,
> will drop the console back to legacy (non-UTF-8) mode. Fix this by leaving the
> field vc_data.vc_utf untouched in reset_terminal(). In addition, a global
> variable (default_utf8) which defines system-wide UTF-8 setting is created.
> This variable can be adjusted via sysfs.

If you're going to introduce a system-wide default, instead of issuing 
the appropriate escape code, then I would argue it should still be 
forced (to the default) when issuing a console reset.

	-hpa

Re: [PATCH] vt: Do not clear UTF when resetting console

[Antonino A. Daplas]

On Mon, 2007-04-02 at 10:35 -0700, H. Peter Anvin wrote:
> Antonino A. Daplas wrote:
> > Resetting the console, either by ANSI escape sequences or by the reset utility,
> > will drop the console back to legacy (non-UTF-8) mode. Fix this by leaving the
> > field vc_data.vc_utf untouched in reset_terminal(). In addition, a global
> > variable (default_utf8) which defines system-wide UTF-8 setting is created.
> > This variable can be adjusted via sysfs.
> 
> If you're going to introduce a system-wide default, instead of issuing 
> the appropriate escape code, then I would argue it should still be 
> forced (to the default) when issuing a console reset.
> 

That would be the cleanest and purest behavior. But it's possible to set
one console to UTF-8 and another to legacy mode. So one can corrupt the
user's console just by issuing a reset or echo -e '\033c'. (Although one
can argue that users who know what UTF-8 is also knows how to set the
encoding back)

Until userspace is more capable of setting back the terminal to its
previous configuration, I would tend to agree with Jan, that we should
leave the current utf setting of that particular vc alone.

Tony

Re: [PATCH] vt: Do not clear UTF when resetting console

[Jan Engelhardt]

On Apr 3 2007 08:16, Antonino A. Daplas wrote:
>
>That would be the cleanest and purest behavior. But it's possible to set
>one console to UTF-8 and another to legacy mode.

The question would be: why would you want to have mixed consoles?
Switching to UTF8 IMO does not take away any characters, and I mean
no-framebuffer 80x25 that is limited to 256 glyphs.


Jan
-- 

Re: [PATCH] vt: Do not clear UTF when resetting console

[Antonino A. Daplas]

On Tue, 2007-04-03 at 02:23 +0200, Jan Engelhardt wrote:
> On Apr 3 2007 08:16, Antonino A. Daplas wrote:
> >
> >That would be the cleanest and purest behavior. But it's possible to set
> >one console to UTF-8 and another to legacy mode.
> 
> The question would be: why would you want to have mixed consoles?
> Switching to UTF8 IMO does not take away any characters, and I mean
> no-framebuffer 80x25 that is limited to 256 glyphs.

As long as we provide the users the capability to support mixed
encodings, the why is not important, it will happen. If we want to be
more restrictive, I guess, we can remove support for

echo -e '\033%G' and '\033%@'

Tony

Re: [PATCH] vt: Do not clear UTF when resetting console

[H. Peter Anvin]

Jan Engelhardt wrote:
> On Apr 3 2007 08:16, Antonino A. Daplas wrote:
>> That would be the cleanest and purest behavior. But it's possible to set
>> one console to UTF-8 and another to legacy mode.
> 
> The question would be: why would you want to have mixed consoles?
> Switching to UTF8 IMO does not take away any characters, and I mean
> no-framebuffer 80x25 that is limited to 256 glyphs.
> 

512, not 256.  However, the reason would be because you have an 
application (which might actually be running on another system 
entirely!) which expects the other behaviour.

Antonio wrote:
> That would be the cleanest and purest behavior. But it's possible to set
> one console to UTF-8 and another to legacy mode. So one can corrupt the
> user's console just by issuing a reset or echo -e '\033c'. (Although one
> can argue that users who know what UTF-8 is also knows how to set the
> encoding back)
> 
> Until userspace is more capable of setting back the terminal to its
> previous configuration, I would tend to agree with Jan, that we should
> leave the current utf setting of that particular vc alone.

I think you're missing the whole point of console reset.  Its purpose is 
to force the console into a known-good state.  The fewer pieces of state 
it leaves unset, the better.  To some degree it's less important what 
that state actually is.

	-hpa

Re: [PATCH] vt: Do not clear UTF when resetting console

[Paul LeoNerd Evans]

[-- Attachment #1: Type: text/plain, Size: 1033 bytes --]

On Mon, 02 Apr 2007 17:55:43 -0700
"H. Peter Anvin" <hpa@zytor.com> wrote:

> I think you're missing the whole point of console reset.  Its purpose is 
> to force the console into a known-good state.  The fewer pieces of state 
> it leaves unset, the better.  To some degree it's less important what 
> that state actually is.

Ultimately, does the patch actually change that?

Without this change, utf mode gets reset to zero because that's what the
kernel source code says.

With this change, utf mode gets reset to zero because that's what the
module parameter says and the user hasn't changed it..

.. only with the change, users get the ability to change that policy.

For the last 2 years since I originally submitted that patch, I've been
running it on my box. For those last 2 years, finally, UTF-8 input
actually worked on the VTs. Previous to that, lots of things broke.

-- 
Paul "LeoNerd" Evans

leonerd@leonerd.org.uk
ICQ# 4135350       |  Registered Linux# 179460
http://www.leonerd.org.uk/

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [PATCH] vt: Do not clear UTF when resetting console

[Jan Engelhardt]

On Apr 2 2007 22:13, Antonino A. Daplas wrote:
>Resetting the console, either by ANSI escape sequences or by the reset utility,
>will drop the console back to legacy (non-UTF-8) mode. Fix this by leaving the
>field vc_data.vc_utf untouched in reset_terminal(). In addition, a global
>variable (default_utf8) which defines system-wide UTF-8 setting is created.
>This variable can be adjusted via sysfs.
>
>This is based from patches by Jan Engelhardt and Paul LeoNerd Evans.
>
>Signed-off-by: Antonino Daplas <adaplas@gmail.com>

Signed-off-by: Jan Engelhardt <jengelh@gmx.de>


>---
>
> drivers/char/vt.c |    4 +++-
> 1 files changed, 3 insertions(+), 1 deletions(-)


BTW. Is it feasible to make utf8 the default (static int default_utf8 = 1)
or is that likely to break some installs?


> static int printable;		/* Is console ready for printing? */
>+static int default_utf8;
>+module_param(default_utf8, int, S_IRUGO | S_IWUSR);

It's probably just me, but I would have used an unsiged int.



Jan
-- 

Re: [PATCH] vt: Do not clear UTF when resetting console

[Antonino A. Daplas]

On Mon, 2007-04-02 at 21:10 +0200, Jan Engelhardt wrote:
> On Apr 2 2007 22:13, Antonino A. Daplas wrote:
> >Resetting the console, either by ANSI escape sequences or by the reset utility,
> >will drop the console back to legacy (non-UTF-8) mode. Fix this by leaving the
> >field vc_data.vc_utf untouched in reset_terminal(). In addition, a global
> >variable (default_utf8) which defines system-wide UTF-8 setting is created.
> >This variable can be adjusted via sysfs.
> >
> >This is based from patches by Jan Engelhardt and Paul LeoNerd Evans.
> >
> >Signed-off-by: Antonino Daplas <adaplas@gmail.com>
> 
> Signed-off-by: Jan Engelhardt <jengelh@gmx.de>
> 
> 
> >---
> >
> > drivers/char/vt.c |    4 +++-
> > 1 files changed, 3 insertions(+), 1 deletions(-)
> 
> 
> BTW. Is it feasible to make utf8 the default (static int default_utf8 = 1)
> or is that likely to break some installs?

I guess it would, but most if not all major distribs are moving/have
moved to utf-8.

Tony

Re: [PATCH] vt: Do not clear UTF when resetting console

[Martin Mares]

Hello!

> Resetting the console, either by ANSI escape sequences or by the reset utility,
> will drop the console back to legacy (non-UTF-8) mode.

Yes, and as far as I understand the logic behind these escape sequences,
it's the intended behavior, not a bug.

The escape sequence for terminal reset should reset as much as possible,
I see no reason for making an exception for the UTF-8 mode.

				Have a nice fortnight
-- 
Martin `MJ' Mares                          <mj@ucw.cz>   http://mj.ucw.cz/
Faculty of Math and Physics, Charles University, Prague, Czech Rep., Earth
Lottery -- a tax on people who can't do math.

Re: [PATCH] vt: Do not clear UTF when resetting console

[Paul LeoNerd Evans]

[-- Attachment #1: Type: text/plain, Size: 949 bytes --]

On Tue, Apr 03, 2007 at 01:41:27PM +0200, Martin Mares wrote:
> > Resetting the console, either by ANSI escape sequences or by the reset utility,
> > will drop the console back to legacy (non-UTF-8) mode.
> 
> Yes, and as far as I understand the logic behind these escape sequences,
> it's the intended behavior, not a bug.
> 
> The escape sequence for terminal reset should reset as much as possible,
> I see no reason for making an exception for the UTF-8 mode.

Does whatever defines what these escapes mean, have any comment to make
about UTF-8? If not, why can't we declare that UTF-8 mode is the "reset"
mode, the default that would be dropped to on a full reset, and if
anyone wanted to switch that out to non-default not-UTF-8 mode, they
could, but that a reset will always reenable it again..?

-- 
Paul "LeoNerd" Evans

leonerd@leonerd.org.uk
ICQ# 4135350       |  Registered Linux# 179460
http://www.leonerd.org.uk/

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [PATCH] vt: Do not clear UTF when resetting console

[Martin Mares]

Hello!

> Does whatever defines what these escapes mean, have any comment to make
> about UTF-8? If not, why can't we declare that UTF-8 mode is the "reset"
> mode, the default that would be dropped to on a full reset, and if
> anyone wanted to switch that out to non-default not-UTF-8 mode, they
> could, but that a reset will always reenable it again..?

Maybe it would be a little bit more convenient, but I doubt that such
bit of convenience justifies breaking backward compatibility.

				Have a nice fortnight
-- 
Martin `MJ' Mares                          <mj@ucw.cz>   http://mj.ucw.cz/
Faculty of Math and Physics, Charles University, Prague, Czech Rep., Earth
A computer without Windows is like a chocolate cake without mustard.

Re: [PATCH] vt: Do not clear UTF when resetting console

[Antonino A. Daplas]

On Tue, 2007-04-03 at 17:37 +0200, Martin Mares wrote:
> Hello!
> 
> > Does whatever defines what these escapes mean, have any comment to make
> > about UTF-8? If not, why can't we declare that UTF-8 mode is the "reset"
> > mode, the default that would be dropped to on a full reset, and if
> > anyone wanted to switch that out to non-default not-UTF-8 mode, they
> > could, but that a reset will always reenable it again..?
> 
> Maybe it would be a little bit more convenient, but I doubt that such
> bit of convenience justifies breaking backward compatibility.

Please note that I have dropped this patch in favor of this one:

http://lkml.org/lkml/2007/4/2/422

This patch behaves exactly as the one Paul posted before.

Tony

Re: [PATCH] vt: Do not clear UTF when resetting console

[Jan Engelhardt]

On Apr 4 2007 13:42, Antonino A. Daplas wrote:
>> 
>> > Does whatever defines what these escapes mean, have any comment to make
>> > about UTF-8? If not, why can't we declare that UTF-8 mode is the "reset"
>> > mode, the default that would be dropped to on a full reset, and if
>> > anyone wanted to switch that out to non-default not-UTF-8 mode, they
>> > could, but that a reset will always reenable it again..?
>> 
>> Maybe it would be a little bit more convenient, but I doubt that such
>> bit of convenience justifies breaking backward compatibility.
>
>Please note that I have dropped this patch in favor of this one:
>
>http://lkml.org/lkml/2007/4/2/422

^^ Which is ok by me.



Jan
-- 

Re: [PATCH] vt: Do not clear UTF when resetting console

[Paul LeoNerd Evans]

[-- Attachment #1: Type: text/plain, Size: 450 bytes --]

On Wed, Apr 04, 2007 at 10:12:06AM +0200, Jan Engelhardt wrote:
> 
> On Apr 4 2007 13:42, Antonino A. Daplas wrote:
> >Please note that I have dropped this patch in favor of this one:
> >
> >http://lkml.org/lkml/2007/4/2/422
> 
> ^^ Which is ok by me.

I agree - Does The Right Thing as far as I'm concerned.

-- 
Paul "LeoNerd" Evans

leonerd@leonerd.org.uk
ICQ# 4135350       |  Registered Linux# 179460
http://www.leonerd.org.uk/

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [PATCH] vt: Do not clear UTF when resetting console

[Martin Mares]

Hello!

> Please note that I have dropped this patch in favor of this one:
> 
> http://lkml.org/lkml/2007/4/2/422

Yes, this sounds very reasonable.

				Have a nice fortnight
-- 
Martin `MJ' Mares                          <mj@ucw.cz>   http://mj.ucw.cz/
Faculty of Math and Physics, Charles University, Prague, Czech Rep., Earth
Ctrl and Alt keys stuck -- press Del to continue.