[PATCH] vt: Fix non-blinking cursor regression

[PATCH] vt: Fix non-blinking cursor regression

[Lukas Wunner]

Since commit a6dbe4427559 ("vt: perform safe console erase in the right
order"), when userspace clears both the scrollback buffer and the screen
by writing "\e[3J" to an fbdev virtual console, the cursor stops blinking
if that virtual console is not in the foreground.  I'm witnessing this
on every boot of Raspbian since updating to v4.19.37+ because agetty
writes the sequence to /dev/tty6 while the console is still switched to
/dev/tty1.  Switching consoles once makes the cursor blink again.

The commit added an invocation of ->con_switch() to flush_scrollback().
Normally this is only invoked from switch_screen() to switch consoles.
switch_screen() updates *vc->vc_display_fg to the new console and
fbcon_switch() updates ops->currcon.  Because the commit only invokes
fbcon_switch() but doesn't update *vc->vc_display_fg, it performs an
incomplete console switch.

When fb_flashcursor() subsequently blinks the cursor, it retrieves the
foreground console from ops->currcon.  Because *vc->vc_display_fg wasn't
updated, con_is_visible() incorrectly returns false and as a result,
fb_flashcursor() bails out without blinking the cursor.

The invocation of ->con_switch() appears to have been erroneous.  After
all, why should a console switch be performed when clearing the screen?
The commit message doesn't provide a rationale either.  So delete it.

Fixes: a6dbe4427559 ("vt: perform safe console erase in the right order")
Signed-off-by: Lukas Wunner <lukas@wunner.de>
Cc: stable@vger.kernel.org # v5.1+
Cc: Nicolas Pitre <nico@fluxnic.net>
Cc: Mikulas Patocka <mpatocka@redhat.com>
Cc: Matthew Whitehead <tedheadster@gmail.com>
---
 drivers/tty/vt/vt.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/tty/vt/vt.c b/drivers/tty/vt/vt.c
index 35d21cdb60d0..eb25fb4ca05f 100644
--- a/drivers/tty/vt/vt.c
+++ b/drivers/tty/vt/vt.c
@@ -938,8 +938,6 @@ static void flush_scrollback(struct vc_data *vc)
 	set_origin(vc);
 	if (vc->vc_sw->con_flush_scrollback)
 		vc->vc_sw->con_flush_scrollback(vc);
-	else
-		vc->vc_sw->con_switch(vc);
 }
 
 /*
-- 
2.24.0

Re: [PATCH] vt: Fix non-blinking cursor regression

[Nicolas Pitre]

On Wed, 22 Jan 2020, Lukas Wunner wrote:

> Since commit a6dbe4427559 ("vt: perform safe console erase in the right
> order"), when userspace clears both the scrollback buffer and the screen
> by writing "\e[3J" to an fbdev virtual console, the cursor stops blinking
> if that virtual console is not in the foreground.  I'm witnessing this
> on every boot of Raspbian since updating to v4.19.37+ because agetty
> writes the sequence to /dev/tty6 while the console is still switched to
> /dev/tty1.  Switching consoles once makes the cursor blink again.
> 
> The commit added an invocation of ->con_switch() to flush_scrollback().
> Normally this is only invoked from switch_screen() to switch consoles.
> switch_screen() updates *vc->vc_display_fg to the new console and
> fbcon_switch() updates ops->currcon.  Because the commit only invokes
> fbcon_switch() but doesn't update *vc->vc_display_fg, it performs an
> incomplete console switch.
> 
> When fb_flashcursor() subsequently blinks the cursor, it retrieves the
> foreground console from ops->currcon.  Because *vc->vc_display_fg wasn't
> updated, con_is_visible() incorrectly returns false and as a result,
> fb_flashcursor() bails out without blinking the cursor.
> 
> The invocation of ->con_switch() appears to have been erroneous.  After
> all, why should a console switch be performed when clearing the screen?
> The commit message doesn't provide a rationale either.  So delete it.

The problem here is that only vgacon provides a con_flush_scrollback 
method. When not provided, the only way to flush the scrollback buffer 
is to invoke the switch method. If you remove it the scrollback buffer 
of the foreground console won't be flushed in the fb case and possibly 
others.

Originally the con_switch method was invoked via update_screen(). The 
code rationalization in commit a6dbe4427559 failed to carry over a few 
details though.

So I think the actual fix should instead be like this:

diff --git a/drivers/tty/vt/vt.c b/drivers/tty/vt/vt.c
index 34aa39d1ae..93314a2f26 100644
--- a/drivers/tty/vt/vt.c
+++ b/drivers/tty/vt/vt.c
@@ -936,10 +936,13 @@ static void flush_scrollback(struct vc_data *vc)
 	WARN_CONSOLE_UNLOCKED();
 
 	set_origin(vc);
-	if (vc->vc_sw->con_flush_scrollback)
+	if (vc->vc_sw->con_flush_scrollback) {
 		vc->vc_sw->con_flush_scrollback(vc);
-	else
+	} else if (con_is_visible(vc)) {
+		hide_cursor(vc);
 		vc->vc_sw->con_switch(vc);
+		set_cursor(vc);
+	}
 }
 
 /*


Nicolas

Re: [PATCH] vt: Fix non-blinking cursor regression

[Lukas Wunner]

On Wed, Jan 22, 2020 at 11:40:38AM -0500, Nicolas Pitre wrote:
> On Wed, 22 Jan 2020, Lukas Wunner wrote:
> > Since commit a6dbe4427559 ("vt: perform safe console erase in the right
> > order"), when userspace clears both the scrollback buffer and the screen
> > by writing "\e[3J" to an fbdev virtual console, the cursor stops blinking
> > if that virtual console is not in the foreground.  I'm witnessing this
> > on every boot of Raspbian since updating to v4.19.37+ because agetty
> > writes the sequence to /dev/tty6 while the console is still switched to
> > /dev/tty1.  Switching consoles once makes the cursor blink again.
> > 
> > The commit added an invocation of ->con_switch() to flush_scrollback().
> > Normally this is only invoked from switch_screen() to switch consoles.
> > switch_screen() updates *vc->vc_display_fg to the new console and
> > fbcon_switch() updates ops->currcon.  Because the commit only invokes
> > fbcon_switch() but doesn't update *vc->vc_display_fg, it performs an
> > incomplete console switch.
> > 
> > When fb_flashcursor() subsequently blinks the cursor, it retrieves the
> > foreground console from ops->currcon.  Because *vc->vc_display_fg wasn't
> > updated, con_is_visible() incorrectly returns false and as a result,
> > fb_flashcursor() bails out without blinking the cursor.
> > 
> > The invocation of ->con_switch() appears to have been erroneous.  After
> > all, why should a console switch be performed when clearing the screen?
> > The commit message doesn't provide a rationale either.  So delete it.
> 
> The problem here is that only vgacon provides a con_flush_scrollback 
> method. When not provided, the only way to flush the scrollback buffer 
> is to invoke the switch method. If you remove it the scrollback buffer 
> of the foreground console won't be flushed in the fb case and possibly 
> others.

Okay.  I guess it's somewhat counter-intuitive that ->con_switch()
is called only because it has the side effect of flushing scrollback.
In particular, this approach doesn't work for nonvisible consoles.

So the proper solution might be to amend the fb_con struct with a
->con_flush_scrollback() hook.  Which portions of fbcon_switch()
would have to be duplicated in that hook?  The softback code at the
top of the function would seem like an obvious candidate.  What about
the invocation of fb_set_var() (which in turn calls fb_pan_display())?
Anything else?

FWIW, even without any call to ->con_switch, writing "\e[3J" does
flush scrollback.  Works both on the foreground console as well as
nonvisible consoles.  I've tested this with bcm2708_fb which isn't
upstream yet, it's in the Raspberry Pi Foundation's downstream tree:

https://github.com/raspberrypi/linux/blob/rpi-4.19.y/drivers/video/fbdev/bcm2708_fb.c


> @@ -936,10 +936,13 @@ static void flush_scrollback(struct vc_data *vc)
>  	WARN_CONSOLE_UNLOCKED();
>  
>  	set_origin(vc);
> -	if (vc->vc_sw->con_flush_scrollback)
> +	if (vc->vc_sw->con_flush_scrollback) {
>  		vc->vc_sw->con_flush_scrollback(vc);
> -	else
> +	} else if (con_is_visible(vc)) {
> +		hide_cursor(vc);
>  		vc->vc_sw->con_switch(vc);
> +		set_cursor(vc);
> +	}

A dumb question perhaps, but why is it necessary to hide the cursor?

Thanks,

Lukas

Re: [PATCH] vt: Fix non-blinking cursor regression

[Nicolas Pitre]

On Sun, 26 Jan 2020, Lukas Wunner wrote:

> On Wed, Jan 22, 2020 at 11:40:38AM -0500, Nicolas Pitre wrote:
> > On Wed, 22 Jan 2020, Lukas Wunner wrote:
> > > Since commit a6dbe4427559 ("vt: perform safe console erase in the right
> > > order"), when userspace clears both the scrollback buffer and the screen
> > > by writing "\e[3J" to an fbdev virtual console, the cursor stops blinking
> > > if that virtual console is not in the foreground.  I'm witnessing this
> > > on every boot of Raspbian since updating to v4.19.37+ because agetty
> > > writes the sequence to /dev/tty6 while the console is still switched to
> > > /dev/tty1.  Switching consoles once makes the cursor blink again.
> > > 
> > > The commit added an invocation of ->con_switch() to flush_scrollback().
> > > Normally this is only invoked from switch_screen() to switch consoles.
> > > switch_screen() updates *vc->vc_display_fg to the new console and
> > > fbcon_switch() updates ops->currcon.  Because the commit only invokes
> > > fbcon_switch() but doesn't update *vc->vc_display_fg, it performs an
> > > incomplete console switch.
> > > 
> > > When fb_flashcursor() subsequently blinks the cursor, it retrieves the
> > > foreground console from ops->currcon.  Because *vc->vc_display_fg wasn't
> > > updated, con_is_visible() incorrectly returns false and as a result,
> > > fb_flashcursor() bails out without blinking the cursor.
> > > 
> > > The invocation of ->con_switch() appears to have been erroneous.  After
> > > all, why should a console switch be performed when clearing the screen?
> > > The commit message doesn't provide a rationale either.  So delete it.
> > 
> > The problem here is that only vgacon provides a con_flush_scrollback 
> > method. When not provided, the only way to flush the scrollback buffer 
> > is to invoke the switch method. If you remove it the scrollback buffer 
> > of the foreground console won't be flushed in the fb case and possibly 
> > others.
> 
> Okay.  I guess it's somewhat counter-intuitive that ->con_switch()
> is called only because it has the side effect of flushing scrollback.
> In particular, this approach doesn't work for nonvisible consoles.

Normally, only the foreground console has a scrollback, so by not being 
visible it therefore has no scrollback to flush. When the scrollback is 
persistent (only vgacon implements that) then there must be a 
con_flush_scrollback method.

> So the proper solution might be to amend the fb_con struct with a
> ->con_flush_scrollback() hook.

Yes, and every other console drivers as well.

> Which portions of fbcon_switch()
> would have to be duplicated in that hook?  The softback code at the
> top of the function would seem like an obvious candidate.

Most likely, yes, after making sure the scrollback does correspond to 
the vc argument.

> What about the invocation of fb_set_var() (which in turn calls 
> fb_pan_display())? Anything else?

I don't know enough about the fbcon code to be sure of all the 
implications here.

Still, I'd prefer to get back to the same functional state from before 
commit a6dbe44275 with the switch method first. Can you confirm that the 
patch I propose does fix it for you?

> > @@ -936,10 +936,13 @@ static void flush_scrollback(struct vc_data *vc)
> >  	WARN_CONSOLE_UNLOCKED();
> >  
> >  	set_origin(vc);
> > -	if (vc->vc_sw->con_flush_scrollback)
> > +	if (vc->vc_sw->con_flush_scrollback) {
> >  		vc->vc_sw->con_flush_scrollback(vc);
> > -	else
> > +	} else if (con_is_visible(vc)) {
> > +		hide_cursor(vc);
> >  		vc->vc_sw->con_switch(vc);
> > +		set_cursor(vc);
> > +	}
> 
> A dumb question perhaps, but why is it necessary to hide the cursor?

Many console implements the cursor by changing the background color of 
the cursor position. If the switch occurs while the cursor is in its 
visible period, the rest of the code will assume that the cursor is the 
actual background color, effectively leaving the drawn cursor there 
after it moved.


Nicolas

Re: [PATCH] vt: Fix non-blinking cursor regression

[Lukas Wunner]

On Sun, Jan 26, 2020 at 12:32:10PM -0500, Nicolas Pitre wrote:
> On Sun, 26 Jan 2020, Lukas Wunner wrote:
> > On Wed, Jan 22, 2020 at 11:40:38AM -0500, Nicolas Pitre wrote:
> > > On Wed, 22 Jan 2020, Lukas Wunner wrote:
> > > > Since commit a6dbe4427559 ("vt: perform safe console erase in the right
> > > > order"), when userspace clears both the scrollback buffer and the screen
> > > > by writing "\e[3J" to an fbdev virtual console, the cursor stops blinking
> > > > if that virtual console is not in the foreground.  I'm witnessing this
> > > > on every boot of Raspbian since updating to v4.19.37+ because agetty
> > > > writes the sequence to /dev/tty6 while the console is still switched to
> > > > /dev/tty1.  Switching consoles once makes the cursor blink again.
> > > > 
> > > > The commit added an invocation of ->con_switch() to flush_scrollback().
> > > > Normally this is only invoked from switch_screen() to switch consoles.
> > > > switch_screen() updates *vc->vc_display_fg to the new console and
> > > > fbcon_switch() updates ops->currcon.  Because the commit only invokes
> > > > fbcon_switch() but doesn't update *vc->vc_display_fg, it performs an
> > > > incomplete console switch.
> > > > 
> > > > When fb_flashcursor() subsequently blinks the cursor, it retrieves the
> > > > foreground console from ops->currcon.  Because *vc->vc_display_fg wasn't
> > > > updated, con_is_visible() incorrectly returns false and as a result,
> > > > fb_flashcursor() bails out without blinking the cursor.
> > > > 
> > > > The invocation of ->con_switch() appears to have been erroneous.  After
> > > > all, why should a console switch be performed when clearing the screen?
> > > > The commit message doesn't provide a rationale either.  So delete it.
> > > 
> > > The problem here is that only vgacon provides a con_flush_scrollback 
> > > method. When not provided, the only way to flush the scrollback buffer 
> > > is to invoke the switch method. If you remove it the scrollback buffer 
> > > of the foreground console won't be flushed in the fb case and possibly 
> > > others.
[...]
> Still, I'd prefer to get back to the same functional state from before 
> commit a6dbe44275 with the switch method first. Can you confirm that the 
> patch I propose does fix it for you?

Yes, your patch is

Reported-and-tested-by: Lukas Wunner <lukas@wunner.de>

I'm withdrawing my own patch because further testing has shown that while it
fixes the non-blinking cursor issue, it doesn't flush scrollback if "\e[3J"
is written to the foreground console.

Would you prefer me to submit your patch with your Signed-off-by or rather
submit it yourself with my Tested-by?  If the latter, please include a code
comment explaining that ->con_switch() has the side effect of flushing
scrollback.  That seems quite non-obvious to me.


> > A dumb question perhaps, but why is it necessary to hide the cursor?
> 
> Many console implements the cursor by changing the background color of 
> the cursor position. If the switch occurs while the cursor is in its 
> visible period, the rest of the code will assume that the cursor is the 
> actual background color, effectively leaving the drawn cursor there 
> after it moved.

I see, thanks for the explanation.

Lukas

Re: [PATCH] vt: Fix non-blinking cursor regression

[Nicolas Pitre]

On Tue, 28 Jan 2020, Lukas Wunner wrote:

> On Sun, Jan 26, 2020 at 12:32:10PM -0500, Nicolas Pitre wrote:
> > On Sun, 26 Jan 2020, Lukas Wunner wrote:
> > > On Wed, Jan 22, 2020 at 11:40:38AM -0500, Nicolas Pitre wrote:
> > > > On Wed, 22 Jan 2020, Lukas Wunner wrote:
> > > > > Since commit a6dbe4427559 ("vt: perform safe console erase in the right
> > > > > order"), when userspace clears both the scrollback buffer and the screen
> > > > > by writing "\e[3J" to an fbdev virtual console, the cursor stops blinking
> > > > > if that virtual console is not in the foreground.  I'm witnessing this
> > > > > on every boot of Raspbian since updating to v4.19.37+ because agetty
> > > > > writes the sequence to /dev/tty6 while the console is still switched to
> > > > > /dev/tty1.  Switching consoles once makes the cursor blink again.
> > > > > 
> > > > > The commit added an invocation of ->con_switch() to flush_scrollback().
> > > > > Normally this is only invoked from switch_screen() to switch consoles.
> > > > > switch_screen() updates *vc->vc_display_fg to the new console and
> > > > > fbcon_switch() updates ops->currcon.  Because the commit only invokes
> > > > > fbcon_switch() but doesn't update *vc->vc_display_fg, it performs an
> > > > > incomplete console switch.
> > > > > 
> > > > > When fb_flashcursor() subsequently blinks the cursor, it retrieves the
> > > > > foreground console from ops->currcon.  Because *vc->vc_display_fg wasn't
> > > > > updated, con_is_visible() incorrectly returns false and as a result,
> > > > > fb_flashcursor() bails out without blinking the cursor.
> > > > > 
> > > > > The invocation of ->con_switch() appears to have been erroneous.  After
> > > > > all, why should a console switch be performed when clearing the screen?
> > > > > The commit message doesn't provide a rationale either.  So delete it.
> > > > 
> > > > The problem here is that only vgacon provides a con_flush_scrollback 
> > > > method. When not provided, the only way to flush the scrollback buffer 
> > > > is to invoke the switch method. If you remove it the scrollback buffer 
> > > > of the foreground console won't be flushed in the fb case and possibly 
> > > > others.
> [...]
> > Still, I'd prefer to get back to the same functional state from before 
> > commit a6dbe44275 with the switch method first. Can you confirm that the 
> > patch I propose does fix it for you?
> 
> Yes, your patch is
> 
> Reported-and-tested-by: Lukas Wunner <lukas@wunner.de>

OK, thanks.

> I'm withdrawing my own patch because further testing has shown that while it
> fixes the non-blinking cursor issue, it doesn't flush scrollback if "\e[3J"
> is written to the foreground console.
> 
> Would you prefer me to submit your patch with your Signed-off-by or rather
> submit it yourself with my Tested-by?  If the latter, please include a code
> comment explaining that ->con_switch() has the side effect of flushing
> scrollback.  That seems quite non-obvious to me.

I'll submit it with a comment stating the non-obvious.


Nicolas