LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
* [PATCH v15] tty: Fix the keyboard led light display problem
@ 2021-11-26  2:44 lianzhi chang
  2021-11-26  9:55 ` Andy Shevchenko
  0 siblings, 1 reply; 2+ messages in thread
From: lianzhi chang @ 2021-11-26  2:44 UTC (permalink / raw)
  To: linux-kernel
  Cc: dmitry.torokhov, gregkh, jirislaby, andriy.shevchenko, 282827961,
	lianzhi chang

When the desktop and tty switch mutually, the led state of the
keyboard may be inconsistent with the state of the keyboard lock.
This is because the desktop environment (Xorg, etc.) is bound to
a tty, and the kb->kbdmode attribute of this tty is set to VC_OFF.
This leads to the fact that in the desktop environment, the bound
tty will not set the keyboard light state, so in the current tty
scene, the values of ledstate and kb->ledflagstate are always 0.
This leads to two situations: (1) When switching from the desktop
to another tty, the code inside VT still compares ledstate with
the kb->ledflagstate of the next tty. If they are equal, then
after the switch is completed, The keyboard light will maintain
the state of the previous desktop settings, and the state of the
keyboard lock may be inconsistent; (2) When switching from another
tty to the desktop, according to the code logic, it may still
trigger the desktop bound tty to set the keyboard light state.
After the switch is completed, the keyboard light is forcibly
turned off. I think in this case, the tty should not set the
keyboard light, and give control to Xorg etc. to handle it.
The current modification judges the value of kb->kbdmode.
In some modes, when switching VT, the current tty keyboard
light status is forcibly issued. And when switching to the
desktop, the tty no longer sets the keyboard light.

Signed-off-by: lianzhi chang <changlianzhi@uniontech.com>
Suggested-by: dmitry.torokhov <dmitry.torokhov@gmail.com>
---
 v15:
 (1) Modify the description information;
 (2) When switching VT, add the judgment on the two modes of
 VC_RAW and VC_MEDIUMRAW in the place where it is determined
 whether it is mandatory to set the led state of the
 keyboard, which is consistent with other places.
 
 drivers/tty/vt/keyboard.c | 19 ++++++++++++++++++-
 1 file changed, 18 insertions(+), 1 deletion(-)

diff --git a/drivers/tty/vt/keyboard.c b/drivers/tty/vt/keyboard.c
index c7fbbcdcc346..98ec12001368 100644
--- a/drivers/tty/vt/keyboard.c
+++ b/drivers/tty/vt/keyboard.c
@@ -153,6 +153,7 @@ static int shift_state = 0;
 
 static unsigned int ledstate = -1U;			/* undefined */
 static unsigned char ledioctl;
+static bool vt_switch;
 
 /*
  * Notifier list for console keyboard events
@@ -412,9 +413,20 @@ static void do_compute_shiftstate(void)
 /* We still have to export this method to vt.c */
 void vt_set_leds_compute_shiftstate(void)
 {
+	struct kbd_struct *kb;
 	unsigned long flags;
 
-	set_leds();
+	/* When switching VT, according to the value of kb->kbdmode,
+	 * judge whether it is necessary to force the keyboard light
+	 * state to be issued.
+	 */
+	kb = kbd_table + fg_console;
+	if (kb->kbdmode == VC_RAW ||
+	     kb->kbdmode == VC_MEDIUMRAW ||
+	     kb->kbdmode == VC_OFF) {
+		vt_switch = true;
+		set_leds();
+	}
 
 	spin_lock_irqsave(&kbd_event_lock, flags);
 	do_compute_shiftstate();
@@ -1255,6 +1267,11 @@ static void kbd_bh(struct tasklet_struct *unused)
 	leds |= (unsigned int)kbd->lockstate << 8;
 	spin_unlock_irqrestore(&led_lock, flags);
 
+	if (vt_switch) {
+		ledstate = ~leds;
+		vt_switch = false;
+	}
+
 	if (leds != ledstate) {
 		kbd_propagate_led_state(ledstate, leds);
 		ledstate = leds;
-- 
2.20.1




^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: [PATCH v15] tty: Fix the keyboard led light display problem
  2021-11-26  2:44 [PATCH v15] tty: Fix the keyboard led light display problem lianzhi chang
@ 2021-11-26  9:55 ` Andy Shevchenko
  0 siblings, 0 replies; 2+ messages in thread
From: Andy Shevchenko @ 2021-11-26  9:55 UTC (permalink / raw)
  To: lianzhi chang; +Cc: linux-kernel, dmitry.torokhov, gregkh, jirislaby, 282827961

On Fri, Nov 26, 2021 at 10:44:23AM +0800, lianzhi chang wrote:
> When the desktop and tty switch mutually, the led state of the
> keyboard may be inconsistent with the state of the keyboard lock.
> This is because the desktop environment (Xorg, etc.) is bound to
> a tty, and the kb->kbdmode attribute of this tty is set to VC_OFF.
> This leads to the fact that in the desktop environment, the bound
> tty will not set the keyboard light state, so in the current tty
> scene, the values of ledstate and kb->ledflagstate are always 0.
> This leads to two situations: (1) When switching from the desktop
> to another tty, the code inside VT still compares ledstate with
> the kb->ledflagstate of the next tty. If they are equal, then
> after the switch is completed, The keyboard light will maintain
> the state of the previous desktop settings, and the state of the
> keyboard lock may be inconsistent; (2) When switching from another
> tty to the desktop, according to the code logic, it may still
> trigger the desktop bound tty to set the keyboard light state.
> After the switch is completed, the keyboard light is forcibly
> turned off. I think in this case, the tty should not set the
> keyboard light, and give control to Xorg etc. to handle it.
> The current modification judges the value of kb->kbdmode.
> In some modes, when switching VT, the current tty keyboard
> light status is forcibly issued. And when switching to the
> desktop, the tty no longer sets the keyboard light.
> 
> Signed-off-by: lianzhi chang <changlianzhi@uniontech.com>

> Suggested-by: dmitry.torokhov <dmitry.torokhov@gmail.com>

I guess I have told you at least a couple of times that the format of these
lines should follow the pattern, where it's "Real Name <real@email.com>".
I truly believe there is no guy with _real_ name "dmitry.torochov".

...

>  /*
>   * Notifier list for console keyboard events

(1)

> @@ -412,9 +413,20 @@ static void do_compute_shiftstate(void)
>  /* We still have to export this method to vt.c */
>  void vt_set_leds_compute_shiftstate(void)
>  {

> +	/* When switching VT, according to the value of kb->kbdmode,
> +	 * judge whether it is necessary to force the keyboard light
> +	 * state to be issued.
> +	 */

Is it too hard to be consistent? See (1).

-- 
With Best Regards,
Andy Shevchenko



^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2021-11-26  9:57 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-11-26  2:44 [PATCH v15] tty: Fix the keyboard led light display problem lianzhi chang
2021-11-26  9:55 ` Andy Shevchenko

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).