Skip to contentSkip to conversation

vt: keyboard, use GENMAASK()/BIT() macros instead of open coded variants

PATCH Superseded Andy Shevchenko 4 participants 14 messages Nov 6, 2020 – Nov 9, 2020 TTY layer and serial drivers

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore
#0 lore raw

PATCH 1/3 vt: keyboard, use GENMAASK()/BIT() macros instead of open coded variants

There are few places when GENMASK() or BIT() macro is suitable and makes code easier to understand.

---
 drivers/tty/vt/keyboard.c | 48 +++++++++++++++++++--------------------
 1 file changed, 24 insertions(+), 24 deletions(-)
drivers/tty/vt/keyboard.c +24 -24
index 56b5e8f8fe88..bfe54b9822af 100644
@@ -53,7 +53,7 @@
  * Exported functions/variables
  */
 
-#define KBD_DEFMODE ((1 << VC_REPEAT) | (1 << VC_META))
+#define KBD_DEFMODE (BIT(VC_REPEAT) | BIT(VC_META))
 
 #if defined(CONFIG_X86) || defined(CONFIG_PARISC)
 #include <asm/kbdleds.h>
@@ -423,8 +423,8 @@ static unsigned int handle_diacr(struct vc_data *vc, unsigned int ch)
 
 	diacr = 0;
 
-	if ((d & ~0xff) == BRL_UC_ROW) {
-		if ((ch & ~0xff) == BRL_UC_ROW)
+	if ((d & ~GENMASK(7, 0)) == BRL_UC_ROW) {
+		if ((ch & ~GENMASK(7, 0)) == BRL_UC_ROW)
 			return d | ch;
 	} else {
 		for (i = 0; i < accent_table_size; i++)
@@ -857,16 +857,16 @@ static void k_shift(struct vc_data *vc, unsigned char value, char up_flag)
 		shift_down[value]++;
 
 	if (shift_down[value])
-		shift_state |= (1 << value);
+		shift_state |= BIT(value);
 	else
-		shift_state &= ~(1 << value);
+		shift_state &= ~BIT(value);
 
 	/* kludge */
 	if (up_flag && shift_state != old_state && npadch_active) {
 		if (kbd->kbdmode == VC_UNICODE)
 			to_utf8(vc, npadch_value);
 		else
-			put_queue(vc, npadch_value & 0xff);
+			put_queue(vc, npadch_value & GENMASK(7, 0));
 		npadch_active = false;
 	}
 }
@@ -880,7 +880,7 @@ static void k_meta(struct vc_data *vc, unsigned char value, char up_flag)
 		put_queue(vc, '\033');
 		put_queue(vc, value);
 	} else
-		put_queue(vc, value | 0x80);
+		put_queue(vc, value | BIT(7));
 }
 
 static void k_ascii(struct vc_data *vc, unsigned char value, char up_flag)
@@ -976,7 +976,7 @@ static void k_brl(struct vc_data *vc, unsigned char value, char up_flag)
 		return;
 
 	if (!up_flag) {
-		pressed |= 1 << (value - 1);
+		pressed |= BIT(value - 1);
 		if (!brl_timeout)
 			committing = pressed;
 	} else if (brl_timeout) {
@@ -986,7 +986,7 @@ static void k_brl(struct vc_data *vc, unsigned char value, char up_flag)
 			committing = pressed;
 			releasestart = jiffies;
 		}
-		pressed &= ~(1 << (value - 1));
+		pressed &= ~BIT(value - 1);
 		if (!pressed && committing) {
 			k_brlcommit(vc, committing, 0);
 			committing = 0;
@@ -996,7 +996,7 @@ static void k_brl(struct vc_data *vc, unsigned char value, char up_flag)
 			k_brlcommit(vc, committing, 0);
 			committing = 0;
 		}
-		pressed &= ~(1 << (value - 1));
+		pressed &= ~BIT(value - 1);
 	}
 }
 
@@ -1096,9 +1096,9 @@ static int kbd_update_leds_helper(struct input_handle *handle, void *data)
 	unsigned int leds = *(unsigned int *)data;
 
 	if (test_bit(EV_LED, handle->dev->evbit)) {
-		input_inject_event(handle, EV_LED, LED_SCROLLL, !!(leds & 0x01));
-		input_inject_event(handle, EV_LED, LED_NUML,    !!(leds & 0x02));
-		input_inject_event(handle, EV_LED, LED_CAPSL,   !!(leds & 0x04));
+		input_inject_event(handle, EV_LED, LED_SCROLLL, !!(leds & BIT(0)));
+		input_inject_event(handle, EV_LED, LED_NUML,    !!(leds & BIT(1)));
+		input_inject_event(handle, EV_LED, LED_CAPSL,   !!(leds & BIT(2)));
 		input_inject_event(handle, EV_SYN, SYN_REPORT, 0);
 	}
 
@@ -1125,14 +1125,14 @@ static void kbd_init_leds(void)
  */
 static unsigned char getledstate(void)
 {
-	return ledstate & 0xff;
+	return ledstate & GENMASK(7, 0);
 }
 
 void setledstate(struct kbd_struct *kb, unsigned int led)
 {
         unsigned long flags;
         spin_lock_irqsave(&led_lock, flags);
-	if (!(led & ~7)) {
+	if (!(led & ~GENMASK(2, 0))) {
 		ledioctl = led;
 		kb->ledmode = LED_SHOW_IOCTL;
 	} else
@@ -1338,7 +1338,7 @@ static int emulate_raw(struct vc_data *vc, unsigned int keycode,
 
 		if (code & 0x100)
 			put_queue(vc, 0xe0);
-		put_queue(vc, (code & 0x7f) | up_flag);
+		put_queue(vc, (code & GENMASK(6, 0)) | up_flag);
 
 		break;
 	}
@@ -1355,7 +1355,7 @@ static inline bool kbd_is_hw_raw(const struct input_dev *dev)
 
 static int emulate_raw(struct vc_data *vc, unsigned int keycode, unsigned char up_flag)
 {
-	if (keycode > 127)
+	if (keycode >= BIT(7))
 		return -1;
 
 	put_queue(vc, keycode | up_flag);
@@ -1423,12 +1423,12 @@ static void kbd_keycode(unsigned int keycode, int down, bool hw_raw)
 		 * applications. This allows for 16384 different keycodes,
 		 * which should be enough.
 		 */
-		if (keycode < 128) {
+		if (keycode < BIT(7)) {
 			put_queue(vc, keycode | (!down << 7));
 		} else {
 			put_queue(vc, !down << 7);
-			put_queue(vc, (keycode >> 7) | 0x80);
-			put_queue(vc, keycode | 0x80);
+			put_queue(vc, (keycode >> 7) | BIT(7));
+			put_queue(vc, keycode | BIT(7));
 		}
 		raw_mode = true;
 	}
@@ -1487,7 +1487,7 @@ static void kbd_keycode(unsigned int keycode, int down, bool hw_raw)
 	if (type == KT_LETTER) {
 		type = KT_LATIN;
 		if (vc_kbd_led(kbd, VC_CAPSLOCK)) {
-			key_map = key_maps[shift_final ^ (1 << KG_SHIFT)];
+			key_map = key_maps[shift_final ^ BIT(KG_SHIFT)];
 			if (key_map)
 				keysym = key_map[keycode];
 		}
@@ -2108,11 +2108,11 @@ int vt_do_kdskled(int console, int cmd, unsigned long arg, int perm)
 	case KDSKBLED:
 		if (!perm)
 			return -EPERM;
-		if (arg & ~0x77)
+		if (arg & ~(GENMASK(6, 4) | GENMASK(2, 0)))
 			return -EINVAL;
                 spin_lock_irqsave(&led_lock, flags);
-		kb->ledflagstate = (arg & 7);
-		kb->default_ledflagstate = ((arg >> 4) & 7);
+		kb->ledflagstate = arg & GENMASK(2, 0);
+		kb->default_ledflagstate = (arg & GENMASK(6, 4)) >> 4;
 		set_leds();
                 spin_unlock_irqrestore(&led_lock, flags);
 		return 0;
Signature · 1 line
2.28.0
#1 lore raw

PATCH 2/3 vt: keyboard, replace numbers with \r, \n where appropriate

Instead of 10, 13 use \n, \r respectively.

---
 drivers/tty/vt/keyboard.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/tty/vt/keyboard.c +3 -3
index bfe54b9822af..647c343f61fb 100644
@@ -462,9 +462,9 @@ static void fn_enter(struct vc_data *vc)
 		diacr = 0;
 	}
 
-	put_queue(vc, 13);
+	put_queue(vc, '\r');
 	if (vc_kbd_mode(kbd, VC_CRLF))
-		put_queue(vc, 10);
+		put_queue(vc, '\n');
 }
 
 static void fn_caps_toggle(struct vc_data *vc)
@@ -827,7 +827,7 @@ static void k_pad(struct vc_data *vc, unsigned char value, char up_flag)
 
 	put_queue(vc, pad_chars[value]);
 	if (value == KVAL(K_PENTER) && vc_kbd_mode(kbd, VC_CRLF))
-		put_queue(vc, 10);
+		put_queue(vc, '\n');
 }
 
 static void k_shift(struct vc_data *vc, unsigned char value, char up_flag)
Signature · 1 line
2.28.0
maintainer Acked-by on v1 2/3
#2 lore raw

Re: [PATCH v1 2/3] vt: keyboard, replace numbers with \r, \n where appropriate

Quoted 3 lines · Andy Shevchenko

Instead of 10, 13 use \n, \r respectively.

Quoted 30 lines
---
  drivers/tty/vt/keyboard.c | 6 +++---
  1 file changed, 3 insertions(+), 3 deletions(-)
drivers/tty/vt/keyboard.c +3 -3
index bfe54b9822af..647c343f61fb 100644
@@ -462,9 +462,9 @@ static void fn_enter(struct vc_data *vc)
  		diacr = 0;
  	}
 
-	put_queue(vc, 13);
+	put_queue(vc, '\r');
  	if (vc_kbd_mode(kbd, VC_CRLF))
-		put_queue(vc, 10);
+		put_queue(vc, '\n');
  }
 
  static void fn_caps_toggle(struct vc_data *vc)
@@ -827,7 +827,7 @@ static void k_pad(struct vc_data *vc, unsigned char value, char up_flag)
 
  	put_queue(vc, pad_chars[value]);
  	if (value == KVAL(K_PENTER) && vc_kbd_mode(kbd, VC_CRLF))
-		put_queue(vc, 10);
+		put_queue(vc, '\n');
  }
 
  static void k_shift(struct vc_data *vc, unsigned char value, char up_flag)
Signature · 2 lines
js
suse labs
#3 lore raw

PATCH 3/3 vt: keyboard, make use of assign_bit() API

In reply to Andy Shevchenko

We have for some time the assign_bit() API to replace open coded

	if (foo)
		set_bit(n, bar);
	else
		clear_bit(n, bar);

Use this API in VT keyboard library code.

---
 drivers/tty/vt/keyboard.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)
drivers/tty/vt/keyboard.c +1 -4
index 647c343f61fb..b5132191b0ad 100644
@@ -1433,10 +1433,7 @@ static void kbd_keycode(unsigned int keycode, int down, bool hw_raw)
 		raw_mode = true;
 	}
 
-	if (down)
-		set_bit(keycode, key_down);
-	else
-		clear_bit(keycode, key_down);
+	assign_bit(keycode, key_down, down);
 
 	if (rep &&
 	    (!vc_kbd_mode(kbd, VC_REPEAT) ||
Signature · 1 line
2.28.0
maintainer Acked-by on v1 3/3
#4 lore raw

Re: [PATCH v1 3/3] vt: keyboard, make use of assign_bit() API

Quoted 10 lines · Andy Shevchenko

We have for some time the assign_bit() API to replace open coded

	if (foo)
		set_bit(n, bar);
	else
		clear_bit(n, bar);

Use this API in VT keyboard library code.

Quoted 21 lines
---
  drivers/tty/vt/keyboard.c | 5 +----
  1 file changed, 1 insertion(+), 4 deletions(-)
drivers/tty/vt/keyboard.c +1 -4
index 647c343f61fb..b5132191b0ad 100644
@@ -1433,10 +1433,7 @@ static void kbd_keycode(unsigned int keycode, int down, bool hw_raw)
  		raw_mode = true;
  	}
 
-	if (down)
-		set_bit(keycode, key_down);
-	else
-		clear_bit(keycode, key_down);
+	assign_bit(keycode, key_down, down);
 
  	if (rep &&
  	    (!vc_kbd_mode(kbd, VC_REPEAT) ||
Signature · 2 lines
js
suse labs
#5 lore raw
In reply to Andy Shevchenko

From: Andy Shevchenko

Quoted 5 lines

Sent: 06 November 2020 14:36

There are few places when GENMASK() or BIT() macro is suitable and makes code easier to understand.

...

drivers/tty/vt/keyboard.c static unsigned int handle_diacr(struct vc_data *vc, unsigned int ch) · line 428

Quoted 5 lines
drivers/tty/vt/keyboard.c +2 -2
-	if ((d & ~0xff) == BRL_UC_ROW) {
-		if ((ch & ~0xff) == BRL_UC_ROW)
+	if ((d & ~GENMASK(7, 0)) == BRL_UC_ROW) {
+		if ((ch & ~GENMASK(7, 0)) == BRL_UC_ROW)
 			return d | ch;

Do you really think that makes it more readable?

David

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK Registration No: 1397386 (Wales)

#6 lore raw
Quoted 7 lines · David Laight

From: Andy Shevchenko

Quoted 5 lines

Sent: 06 November 2020 14:36

There are few places when GENMASK() or BIT() macro is suitable and makes code easier to understand.

Quoted 8 lines

...

drivers/tty/vt/keyboard.c static unsigned int handle_diacr(struct vc_data *vc, unsigned int ch) · line 428

Quoted 5 lines
drivers/tty/vt/keyboard.c +2 -2
-     if ((d & ~0xff) == BRL_UC_ROW) {
-             if ((ch & ~0xff) == BRL_UC_ROW)
+     if ((d & ~GENMASK(7, 0)) == BRL_UC_ROW) {
+             if ((ch & ~GENMASK(7, 0)) == BRL_UC_ROW)
                      return d | ch;

Do you really think that makes it more readable?

Yes. Because this tells explicitly how many bits are used for metadata vs. data.

Signature · 2 lines
With Best Regards,
Andy Shevchenko
maintainer
#7 lore raw
Quoted 15 lines · Andy Shevchenko
Quoted 14 lines · David Laight

From: Andy Shevchenko

Quoted 5 lines

Sent: 06 November 2020 14:36

There are few places when GENMASK() or BIT() macro is suitable and makes code easier to understand.

...

Quoted 5 lines
drivers/tty/vt/keyboard.c +2 -2
-     if ((d & ~0xff) == BRL_UC_ROW) {
-             if ((ch & ~0xff) == BRL_UC_ROW)
+     if ((d & ~GENMASK(7, 0)) == BRL_UC_ROW) {
+             if ((ch & ~GENMASK(7, 0)) == BRL_UC_ROW)
                       return d | ch;
Quoted 4 lines
Quoted 1 line

Do you really think that makes it more readable?

Yes. Because this tells explicitly how many bits are used for metadata vs. data.

No, because ~0xff is clearly what it is. GENMASK(7, 0) is:
1) longer to read & parse by brain with result: "GENMASK undefined"
2) terrible in this particular use case

Another instance of an even worse switch:
- if (arg & ~0x77)
+ if (arg & ~(GENMASK(6, 4) | GENMASK(2, 0)))

OTOH, the switch to BIT is legit in all cases except the comparisons with keycode:
- if (keycode > 127)
+ if (keycode >= BIT(7))
- if (keycode < 128) {
+ if (keycode < BIT(7)) {

That's horrid and non-sense too.

sorry,

Signature · 2 lines
js
suse labs
#8 lore raw
Quoted 7 lines · Jiri Slaby
Quoted 6 lines · Andy Shevchenko
Quoted 5 lines · David Laight

From: Andy Shevchenko

Quoted 4 lines

Sent: 06 November 2020 14:36

There are few places when GENMASK() or BIT() macro is suitable and makes code easier to understand.

Thanks for the review, my answers below.

Quoted 7 lines
Quoted 7 lines
Quoted 7 lines

...

Quoted 5 lines
drivers/tty/vt/keyboard.c +2 -2
-     if ((d & ~0xff) == BRL_UC_ROW) {
-             if ((ch & ~0xff) == BRL_UC_ROW)
+     if ((d & ~GENMASK(7, 0)) == BRL_UC_ROW) {
+             if ((ch & ~GENMASK(7, 0)) == BRL_UC_ROW)
                       return d | ch;
Quoted 8 lines
Quoted 4 lines
Quoted 1 line

Do you really think that makes it more readable?

Yes. Because this tells explicitly how many bits are used for metadata vs. data.

No, because ~0xff is clearly what it is. GENMASK(7, 0) is:
1) longer to read & parse by brain with result: "GENMASK undefined"
2) terrible in this particular use case

Maybe #define with a proper name can bring some shed of light here?

drivers/tty/vt/keyboard.c int vt_do_kdskled(int console, int cmd, unsigned long arg, int perm) · line 2111

Quoted 3 lines

Another instance of an even worse switch:

drivers/tty/vt/keyboard.c +1 -1
-               if (arg & ~0x77)
+               if (arg & ~(GENMASK(6, 4) | GENMASK(2, 0)))

It exactly shows what bits we are accepting and what are not. 0x77 you need to translate to the bitmap and then figure out the bit numbers. This is error prone as shown in some cases.

Quoted 8 lines

OTOH, the switch to BIT is legit in all cases except the comparisons with keycode:

drivers/tty/vt/keyboard.c +2 -2
-       if (keycode > 127)
+       if (keycode >= BIT(7))
-               if (keycode < 128) {
+               if (keycode < BIT(7)) {

That's horrid and non-sense too.

Isn't it the exact threshold about keycodes that we only use 7-bit value?

Quoted 1 line

sorry,

Consider this then as RFC.
What about the rest of the series?

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#9 lore raw

From: Andy Shevchenko

Quoted 22 lines

Sent: 09 November 2020 10:10

Quoted 7 lines · Jiri Slaby
Quoted 6 lines · Andy Shevchenko
Quoted 5 lines · David Laight

From: Andy Shevchenko

Quoted 4 lines

Sent: 06 November 2020 14:36

There are few places when GENMASK() or BIT() macro is suitable and makes code easier to understand.

Thanks for the review, my answers below.

Quoted 9 lines
Quoted 9 lines
Quoted 8 lines

...

drivers/tty/vt/keyboard.c static unsigned int handle_diacr(struct vc_data *vc, unsigned int ch) · line 428

Quoted 5 lines
drivers/tty/vt/keyboard.c +2 -2
-     if ((d & ~0xff) == BRL_UC_ROW) {
-             if ((ch & ~0xff) == BRL_UC_ROW)
+     if ((d & ~GENMASK(7, 0)) == BRL_UC_ROW) {
+             if ((ch & ~GENMASK(7, 0)) == BRL_UC_ROW)
                       return d | ch;

Do you really think that makes it more readable?

Quoted 8 lines
Quoted 6 lines
Quoted 2 lines

Yes. Because this tells explicitly how many bits are used for metadata vs. data.

No, because ~0xff is clearly what it is. GENMASK(7, 0) is:
1) longer to read & parse by brain with result: "GENMASK undefined"
2) terrible in this particular use case

Maybe #define with a proper name can bring some shed of light here?

Possibly.

Quoted 7 lines

drivers/tty/vt/keyboard.c int vt_do_kdskled(int console, int cmd, unsigned long arg, int perm) · line 2111

Quoted 3 lines

Another instance of an even worse switch:

drivers/tty/vt/keyboard.c +1 -1
-               if (arg & ~0x77)
+               if (arg & ~(GENMASK(6, 4) | GENMASK(2, 0)))

It exactly shows what bits we are accepting and what are not. 0x77 you need to translate to the bitmap and then figure out the bit numbers. This is error prone as shown in some cases.

We all know what 0xff and 0x77 mean.
It is ingrained from years of writing software.

Now it may be that the 0x77 is related to masking off some other bit values.
In that case you could have a named constant based on the names of the other bit values.
But if you are putting in simple constants there is nothing wrong with hex.

David

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK Registration No: 1397386 (Wales)

#10 lore raw
Quoted 2 lines · David Laight
Quoted 2 lines · Jiri Slaby
Quoted 1 line · Andy Shevchenko

...

Quoted 9 lines
Quoted 7 lines

drivers/tty/vt/keyboard.c int vt_do_kdskled(int console, int cmd, unsigned long arg, int perm) · line 2111

Quoted 3 lines

Another instance of an even worse switch:

drivers/tty/vt/keyboard.c +1 -1
-               if (arg & ~0x77)
+               if (arg & ~(GENMASK(6, 4) | GENMASK(2, 0)))

It exactly shows what bits we are accepting and what are not. 0x77 you need to translate to the bitmap and then figure out the bit numbers. This is error prone as shown in some cases.

We all know what 0xff and 0x77 mean.

Oh, do you expect one with curiosity and absence of C/Linux kernel experience may not try to understand the code easily? We have real examples of such curious people, and honestly I admire them much more than people who knows what 0x77 or 0xff means in every case by heart (yes, at some point of time I used to program Z80 directly from my mind in assembly, but does it really a must for a curious reader / contributor?).

Quoted 8 lines

It is ingrained from years of writing software.

Now it may be that the 0x77 is related to masking off some other bit values.
In that case you could have a named constant based on the names of the other bit values.
But if you are putting in simple constants there is nothing wrong with hex.

It makes code harder to understand. Besides that GENMASK() and BIT() avoids UB which is often happen when programmers doesn't think about (yes, we may discuss about poorness of C standard etc, but this what we have now).

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#11 lore raw
In reply to Andy Shevchenko
Quoted 2 lines · Andy Shevchenko
Quoted 1 line · Jiri Slaby

On 06. 11. 20, 17:06, Andy Shevchenko wrote:

...

Quoted 4 lines
Quoted 1 line

sorry,

Consider this then as RFC.
What about the rest of the series?

I got the answer, thanks!
So, I will drop the first patch and resend the rest with your Ack.

Signature · 2 lines
With Best Regards,
Andy Shevchenko
maintainer
#12 lore raw
Quoted 6 lines · Andy Shevchenko
Quoted 2 lines · Andy Shevchenko
Quoted 1 line · Jiri Slaby

On 06. 11. 20, 17:06, Andy Shevchenko wrote:

...

Quoted 7 lines
Quoted 4 lines
Quoted 1 line

sorry,

Consider this then as RFC.
What about the rest of the series?

I got the answer, thanks!
So, I will drop the first patch and resend the rest with your Ack.

As I wrote the BIT pieces are mostly fine too…

Signature · 2 lines
js
suse labs
#13 lore raw
Quoted 9 lines · Jiri Slaby
Quoted 8 lines · Andy Shevchenko
Quoted 2 lines · Andy Shevchenko
Quoted 1 line · Jiri Slaby

On 06. 11. 20, 17:06, Andy Shevchenko wrote:

...

Quoted 2 lines
Quoted 1 line

sorry,

Quoted 7 lines
Quoted 5 lines
Quoted 2 lines

Consider this then as RFC.
What about the rest of the series?

I got the answer, thanks!
So, I will drop the first patch and resend the rest with your Ack.

As I wrote the BIT pieces are mostly fine too…

Okay, I will try again with BIT() conversion only.

Signature · 2 lines
With Best Regards,
Andy Shevchenko