[PATCH v1 0/2] tty/vt: Cleanups for keyboard driver

[PATCH v1 0/2] tty/vt: Cleanups for keyboard driver

[Alexey Gladkov]

This is a few minor cleanups I found in keybord driver. There's nothing
important here and these changes just slightly improve readability.

Alexey Gladkov (2):
  tty/vt: Use KVAL instead of use bit operation
  tty/vt: Gather the code that outputs char with utf8 in mind

 drivers/tty/vt/keyboard.c | 39 ++++++++++++++++-----------------------
 1 file changed, 16 insertions(+), 23 deletions(-)

-- 
2.48.1

[PATCH v1 1/2] tty/vt: Use KVAL instead of use bit operation

[Alexey Gladkov]

The K_HANDLERS always gets KVAL as an argument. It is better to use the
KVAL macro itself instead of bit operation.

Signed-off-by: Alexey Gladkov <legion@kernel.org>
---
 drivers/tty/vt/keyboard.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/tty/vt/keyboard.c b/drivers/tty/vt/keyboard.c
index 804355da46f5..7df041ac4d5c 100644
--- a/drivers/tty/vt/keyboard.c
+++ b/drivers/tty/vt/keyboard.c
@@ -885,7 +885,7 @@ static void k_shift(struct vc_data *vc, unsigned char value, char up_flag)
 		if (kbd->kbdmode == VC_UNICODE)
 			to_utf8(vc, npadch_value);
 		else
-			put_queue(vc, npadch_value & 0xff);
+			put_queue(vc, KVAL(npadch_value));
 		npadch_active = false;
 	}
 }
@@ -1519,7 +1519,7 @@ static void kbd_keycode(unsigned int keycode, int down, bool hw_raw)
 	if ((raw_mode || kbd->kbdmode == VC_OFF) && type != KT_SPEC && type != KT_SHIFT)
 		return;
 
-	(*k_handler[type])(vc, keysym & 0xff, !down);
+	(*k_handler[type])(vc, KVAL(keysym), !down);
 
 	param.ledstate = kbd->ledflagstate;
 	atomic_notifier_call_chain(&keyboard_notifier_list, KBD_POST_KEYSYM, &param);
-- 
2.48.1

Re: [PATCH v1 1/2] tty/vt: Use KVAL instead of use bit operation

[Jiri Slaby]

On 18. 02. 25, 13:29, Alexey Gladkov wrote:
> The K_HANDLERS always gets KVAL as an argument. It is better to use the
> KVAL macro itself instead of bit operation.
> 
> Signed-off-by: Alexey Gladkov <legion@kernel.org>
> ---
>   drivers/tty/vt/keyboard.c | 4 ++--
>   1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/tty/vt/keyboard.c b/drivers/tty/vt/keyboard.c
> index 804355da46f5..7df041ac4d5c 100644
> --- a/drivers/tty/vt/keyboard.c
> +++ b/drivers/tty/vt/keyboard.c
> @@ -885,7 +885,7 @@ static void k_shift(struct vc_data *vc, unsigned char value, char up_flag)
>   		if (kbd->kbdmode == VC_UNICODE)
>   			to_utf8(vc, npadch_value);
>   		else
> -			put_queue(vc, npadch_value & 0xff);
> +			put_queue(vc, KVAL(npadch_value));

While the mask is the same, this is not a kval, right?

>   		npadch_active = false;
>   	}
>   }
> @@ -1519,7 +1519,7 @@ static void kbd_keycode(unsigned int keycode, int down, bool hw_raw)
>   	if ((raw_mode || kbd->kbdmode == VC_OFF) && type != KT_SPEC && type != KT_SHIFT)
>   		return;
>   
> -	(*k_handler[type])(vc, keysym & 0xff, !down);
> +	(*k_handler[type])(vc, KVAL(keysym), !down);

This makes sense.

-- 
js
suse labs

Re: [PATCH v1 1/2] tty/vt: Use KVAL instead of use bit operation

[Alexey Gladkov]

On Wed, Feb 19, 2025 at 07:24:52AM +0100, Jiri Slaby wrote:
> On 18. 02. 25, 13:29, Alexey Gladkov wrote:
> > The K_HANDLERS always gets KVAL as an argument. It is better to use the
> > KVAL macro itself instead of bit operation.
> > 
> > Signed-off-by: Alexey Gladkov <legion@kernel.org>
> > ---
> >   drivers/tty/vt/keyboard.c | 4 ++--
> >   1 file changed, 2 insertions(+), 2 deletions(-)
> > 
> > diff --git a/drivers/tty/vt/keyboard.c b/drivers/tty/vt/keyboard.c
> > index 804355da46f5..7df041ac4d5c 100644
> > --- a/drivers/tty/vt/keyboard.c
> > +++ b/drivers/tty/vt/keyboard.c
> > @@ -885,7 +885,7 @@ static void k_shift(struct vc_data *vc, unsigned char value, char up_flag)
> >   		if (kbd->kbdmode == VC_UNICODE)
> >   			to_utf8(vc, npadch_value);
> >   		else
> > -			put_queue(vc, npadch_value & 0xff);
> > +			put_queue(vc, KVAL(npadch_value));
> 
> While the mask is the same, this is not a kval, right?

I'm pretty sure it's KVAL, but to be honest I don't understand why it is
not done for to_utf8() as well. All values passed to to_utf8() must be
kval.

We call to_utf8() in k_unicode, fn_enter (through k_spec), handle_diacr
(through k_deadunicode or k_unicode). All K_HANDLERS take KVAL as value.

If I understand this code correctly, it is more correct to write it like
this:

--- a/drivers/tty/vt/keyboard.c
+++ b/drivers/tty/vt/keyboard.c
@@ -882,10 +882,11 @@ static void k_shift(struct vc_data *vc, unsigned char value, char up_flag)

        /* kludge */
        if (up_flag && shift_state != old_state && npadch_active) {
+               u32 kval = KVAL(npadch_value);
                if (kbd->kbdmode == VC_UNICODE)
-                       to_utf8(vc, npadch_value);
+                       to_utf8(vc, kval);
                else
-                       put_queue(vc, npadch_value & 0xff);
+                       put_queue(vc, kval);
                npadch_active = false;
        }
 }

But I may be wrong because the code about npadch_value is very old and I
may be missing something.

> >   		npadch_active = false;
> >   	}
> >   }
> > @@ -1519,7 +1519,7 @@ static void kbd_keycode(unsigned int keycode, int down, bool hw_raw)
> >   	if ((raw_mode || kbd->kbdmode == VC_OFF) && type != KT_SPEC && type != KT_SHIFT)
> >   		return;
> >   
> > -	(*k_handler[type])(vc, keysym & 0xff, !down);
> > +	(*k_handler[type])(vc, KVAL(keysym), !down);
> 
> This makes sense.
> 
> -- 
> js
> suse labs
> 

-- 
Rgrds, legion

Re: [PATCH v1 1/2] tty/vt: Use KVAL instead of use bit operation

[Jiri Slaby]

On 19. 02. 25, 10:23, Alexey Gladkov wrote:
> On Wed, Feb 19, 2025 at 07:24:52AM +0100, Jiri Slaby wrote:
>> On 18. 02. 25, 13:29, Alexey Gladkov wrote:
>>> The K_HANDLERS always gets KVAL as an argument. It is better to use the
>>> KVAL macro itself instead of bit operation.
>>>
>>> Signed-off-by: Alexey Gladkov <legion@kernel.org>
>>> ---
>>>    drivers/tty/vt/keyboard.c | 4 ++--
>>>    1 file changed, 2 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/drivers/tty/vt/keyboard.c b/drivers/tty/vt/keyboard.c
>>> index 804355da46f5..7df041ac4d5c 100644
>>> --- a/drivers/tty/vt/keyboard.c
>>> +++ b/drivers/tty/vt/keyboard.c
>>> @@ -885,7 +885,7 @@ static void k_shift(struct vc_data *vc, unsigned char value, char up_flag)
>>>    		if (kbd->kbdmode == VC_UNICODE)
>>>    			to_utf8(vc, npadch_value);
>>>    		else
>>> -			put_queue(vc, npadch_value & 0xff);
>>> +			put_queue(vc, KVAL(npadch_value));
>>
>> While the mask is the same, this is not a kval, right?
> 
> I'm pretty sure it's KVAL, but to be honest I don't understand why it is
> not done for to_utf8() as well. All values passed to to_utf8() must be
> kval.

Not at all, it handles multibyte chars.

> We call to_utf8() in k_unicode, fn_enter (through k_spec), handle_diacr
> (through k_deadunicode or k_unicode). All K_HANDLERS take KVAL as value.

Yes, but pass unicode multibyte to to_utf8().

> If I understand this code correctly, it is more correct to write it like
> this:
> 
> --- a/drivers/tty/vt/keyboard.c
> +++ b/drivers/tty/vt/keyboard.c
> @@ -882,10 +882,11 @@ static void k_shift(struct vc_data *vc, unsigned char value, char up_flag)
> 
>          /* kludge */
>          if (up_flag && shift_state != old_state && npadch_active) {
> +               u32 kval = KVAL(npadch_value);
>                  if (kbd->kbdmode == VC_UNICODE)
> -                       to_utf8(vc, npadch_value);
> +                       to_utf8(vc, kval);

Definitely not, as you want to pass that multibyte char in.

-- 
js
suse labs

Re: [PATCH v1 1/2] tty/vt: Use KVAL instead of use bit operation

[Alexey Gladkov]

On Wed, Feb 19, 2025 at 10:33:38AM +0100, Jiri Slaby wrote:
> On 19. 02. 25, 10:23, Alexey Gladkov wrote:
> > On Wed, Feb 19, 2025 at 07:24:52AM +0100, Jiri Slaby wrote:
> >> On 18. 02. 25, 13:29, Alexey Gladkov wrote:
> >>> The K_HANDLERS always gets KVAL as an argument. It is better to use the
> >>> KVAL macro itself instead of bit operation.
> >>>
> >>> Signed-off-by: Alexey Gladkov <legion@kernel.org>
> >>> ---
> >>>    drivers/tty/vt/keyboard.c | 4 ++--
> >>>    1 file changed, 2 insertions(+), 2 deletions(-)
> >>>
> >>> diff --git a/drivers/tty/vt/keyboard.c b/drivers/tty/vt/keyboard.c
> >>> index 804355da46f5..7df041ac4d5c 100644
> >>> --- a/drivers/tty/vt/keyboard.c
> >>> +++ b/drivers/tty/vt/keyboard.c
> >>> @@ -885,7 +885,7 @@ static void k_shift(struct vc_data *vc, unsigned char value, char up_flag)
> >>>    		if (kbd->kbdmode == VC_UNICODE)
> >>>    			to_utf8(vc, npadch_value);
> >>>    		else
> >>> -			put_queue(vc, npadch_value & 0xff);
> >>> +			put_queue(vc, KVAL(npadch_value));
> >>
> >> While the mask is the same, this is not a kval, right?
> > 
> > I'm pretty sure it's KVAL, but to be honest I don't understand why it is
> > not done for to_utf8() as well. All values passed to to_utf8() must be
> > kval.
> 
> Not at all, it handles multibyte chars.
> 
> > We call to_utf8() in k_unicode, fn_enter (through k_spec), handle_diacr
> > (through k_deadunicode or k_unicode). All K_HANDLERS take KVAL as value.
> 
> Yes, but pass unicode multibyte to to_utf8().
> 
> > If I understand this code correctly, it is more correct to write it like
> > this:
> > 
> > --- a/drivers/tty/vt/keyboard.c
> > +++ b/drivers/tty/vt/keyboard.c
> > @@ -882,10 +882,11 @@ static void k_shift(struct vc_data *vc, unsigned char value, char up_flag)
> > 
> >          /* kludge */
> >          if (up_flag && shift_state != old_state && npadch_active) {
> > +               u32 kval = KVAL(npadch_value);
> >                  if (kbd->kbdmode == VC_UNICODE)
> > -                       to_utf8(vc, npadch_value);
> > +                       to_utf8(vc, kval);
> 
> Definitely not, as you want to pass that multibyte char in.

Ok. So I misunderstood this code. I will remove this change from the next
version.

-- 
Rgrds, legion

[PATCH v1 2/2] tty/vt: Gather the code that outputs char with utf8 in mind

[Alexey Gladkov]

When we putting character to the tty, we take into account the keyboard
mode to properly handle utf8. This code is duplicated few times.

Signed-off-by: Alexey Gladkov <legion@kernel.org>
---
 drivers/tty/vt/keyboard.c | 35 ++++++++++++++---------------------
 1 file changed, 14 insertions(+), 21 deletions(-)

diff --git a/drivers/tty/vt/keyboard.c b/drivers/tty/vt/keyboard.c
index 7df041ac4d5c..eb32bf00fe7b 100644
--- a/drivers/tty/vt/keyboard.c
+++ b/drivers/tty/vt/keyboard.c
@@ -376,6 +376,17 @@ static void to_utf8(struct vc_data *vc, uint c)
 	}
 }
 
+static void put_queue_utf8(struct vc_data *vc, u32 value)
+{
+	if (kbd->kbdmode == VC_UNICODE)
+		to_utf8(vc, value);
+	else {
+		int c = conv_uni_to_8bit(value);
+		if (c != -1)
+			put_queue(vc, c);
+	}
+}
+
 /* FIXME: review locking for vt.c callers */
 static void set_leds(void)
 {
@@ -454,13 +465,7 @@ static unsigned int handle_diacr(struct vc_data *vc, unsigned int ch)
 	if (ch == ' ' || ch == (BRL_UC_ROW|0) || ch == d)
 		return d;
 
-	if (kbd->kbdmode == VC_UNICODE)
-		to_utf8(vc, d);
-	else {
-		int c = conv_uni_to_8bit(d);
-		if (c != -1)
-			put_queue(vc, c);
-	}
+	put_queue_utf8(vc, d);
 
 	return ch;
 }
@@ -471,13 +476,7 @@ static unsigned int handle_diacr(struct vc_data *vc, unsigned int ch)
 static void fn_enter(struct vc_data *vc)
 {
 	if (diacr) {
-		if (kbd->kbdmode == VC_UNICODE)
-			to_utf8(vc, diacr);
-		else {
-			int c = conv_uni_to_8bit(diacr);
-			if (c != -1)
-				put_queue(vc, c);
-		}
+		put_queue_utf8(vc, diacr);
 		diacr = 0;
 	}
 
@@ -685,13 +684,7 @@ static void k_unicode(struct vc_data *vc, unsigned int value, char up_flag)
 		diacr = value;
 		return;
 	}
-	if (kbd->kbdmode == VC_UNICODE)
-		to_utf8(vc, value);
-	else {
-		int c = conv_uni_to_8bit(value);
-		if (c != -1)
-			put_queue(vc, c);
-	}
+	put_queue_utf8(vc, value);
 }
 
 /*
-- 
2.48.1

Re: [PATCH v1 2/2] tty/vt: Gather the code that outputs char with utf8 in mind

[Jiri Slaby]

On 18. 02. 25, 13:29, Alexey Gladkov wrote:
> When we putting character to the tty, we take into account the keyboard
> mode to properly handle utf8. This code is duplicated few times.
> 
> Signed-off-by: Alexey Gladkov <legion@kernel.org>

Reviewed-by: Jiri Slaby <jirislaby@kernel.org>

thanks,
-- 
js
suse labs

[PATCH v2 0/2] tty/vt: Cleanups for keyboard driver

[Alexey Gladkov]

This is a few minor cleanups I found in keybord driver. There's nothing
important here and these changes just slightly improve readability.

---
v2:
* Removed the faulty use of KVAL for npadch_value.

Alexey Gladkov (2):
  tty/vt: Use KVAL instead of use bit operation
  tty/vt: Gather the code that outputs char with utf8 in mind

 drivers/tty/vt/keyboard.c | 37 +++++++++++++++----------------------
 1 file changed, 15 insertions(+), 22 deletions(-)

-- 
2.48.1

[PATCH v2 1/2] tty/vt: Use KVAL instead of use bit operation

[Alexey Gladkov]

The K_HANDLERS always gets KVAL as an argument. It is better to use the
KVAL macro itself instead of bit operation.

Signed-off-by: Alexey Gladkov <legion@kernel.org>
---
 drivers/tty/vt/keyboard.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/tty/vt/keyboard.c b/drivers/tty/vt/keyboard.c
index 804355da46f5..38d4df932f73 100644
--- a/drivers/tty/vt/keyboard.c
+++ b/drivers/tty/vt/keyboard.c
@@ -1519,7 +1519,7 @@ static void kbd_keycode(unsigned int keycode, int down, bool hw_raw)
 	if ((raw_mode || kbd->kbdmode == VC_OFF) && type != KT_SPEC && type != KT_SHIFT)
 		return;
 
-	(*k_handler[type])(vc, keysym & 0xff, !down);
+	(*k_handler[type])(vc, KVAL(keysym), !down);
 
 	param.ledstate = kbd->ledflagstate;
 	atomic_notifier_call_chain(&keyboard_notifier_list, KBD_POST_KEYSYM, &param);
-- 
2.48.1

[PATCH v2 2/2] tty/vt: Gather the code that outputs char with utf8 in mind

[Alexey Gladkov]

When we putting character to the tty, we take into account the keyboard
mode to properly handle utf8. This code is duplicated few times.

Signed-off-by: Alexey Gladkov <legion@kernel.org>
Reviewed-by: Jiri Slaby <jirislaby@kernel.org>
---
 drivers/tty/vt/keyboard.c | 35 ++++++++++++++---------------------
 1 file changed, 14 insertions(+), 21 deletions(-)

diff --git a/drivers/tty/vt/keyboard.c b/drivers/tty/vt/keyboard.c
index 38d4df932f73..517db498beed 100644
--- a/drivers/tty/vt/keyboard.c
+++ b/drivers/tty/vt/keyboard.c
@@ -376,6 +376,17 @@ static void to_utf8(struct vc_data *vc, uint c)
 	}
 }
 
+static void put_queue_utf8(struct vc_data *vc, u32 value)
+{
+	if (kbd->kbdmode == VC_UNICODE)
+		to_utf8(vc, value);
+	else {
+		int c = conv_uni_to_8bit(value);
+		if (c != -1)
+			put_queue(vc, c);
+	}
+}
+
 /* FIXME: review locking for vt.c callers */
 static void set_leds(void)
 {
@@ -454,13 +465,7 @@ static unsigned int handle_diacr(struct vc_data *vc, unsigned int ch)
 	if (ch == ' ' || ch == (BRL_UC_ROW|0) || ch == d)
 		return d;
 
-	if (kbd->kbdmode == VC_UNICODE)
-		to_utf8(vc, d);
-	else {
-		int c = conv_uni_to_8bit(d);
-		if (c != -1)
-			put_queue(vc, c);
-	}
+	put_queue_utf8(vc, d);
 
 	return ch;
 }
@@ -471,13 +476,7 @@ static unsigned int handle_diacr(struct vc_data *vc, unsigned int ch)
 static void fn_enter(struct vc_data *vc)
 {
 	if (diacr) {
-		if (kbd->kbdmode == VC_UNICODE)
-			to_utf8(vc, diacr);
-		else {
-			int c = conv_uni_to_8bit(diacr);
-			if (c != -1)
-				put_queue(vc, c);
-		}
+		put_queue_utf8(vc, diacr);
 		diacr = 0;
 	}
 
@@ -685,13 +684,7 @@ static void k_unicode(struct vc_data *vc, unsigned int value, char up_flag)
 		diacr = value;
 		return;
 	}
-	if (kbd->kbdmode == VC_UNICODE)
-		to_utf8(vc, value);
-	else {
-		int c = conv_uni_to_8bit(value);
-		if (c != -1)
-			put_queue(vc, c);
-	}
+	put_queue_utf8(vc, value);
 }
 
 /*
-- 
2.48.1