Skip to contentSkip to conversation

tty: Set correct tty name in 'active' sysfs attribute

PATCH In review Hannes Reinecke 3 participants 6 messages Feb 24, 2014 – Feb 27, 2014 TTY layer and serial drivers

lore
Share & tools
lore
#0 lore raw

PATCH v3 1/1 tty: Set correct tty name in 'active' sysfs attribute

The 'active' sysfs attribute should refer to the currently active tty devices the console is running on, not the currently active console.
The console structure doesn't refer to any device in sysfs, only the tty the console is running on has.
So we need to print out the tty names in 'active', not the console names.
But we need to take care for the virtual console to always display 'tty0' so as not to break existing programs.

Cc Lennart Poettering <[email protected]>
Cc Kay Sievers <[email protected]>
Cc Greg Kroah-Hartman <[email protected]>
Cc Jiri Slaby <[email protected]>
---
 drivers/tty/tty_io.c | 24 ++++++++++++++++++------
 1 file changed, 18 insertions(+), 6 deletions(-)
drivers/tty/tty_io.c +18 -6
index c74a00a..96eb462 100644
@@ -1271,12 +1271,13 @@ static void pty_line_name(struct tty_driver *driver, int index, char *p)
  *
  *	Locking: None
  */
-static void tty_line_name(struct tty_driver *driver, int index, char *p)
+static ssize_t tty_line_name(struct tty_driver *driver, int index, char *p)
 {
 	if (driver->flags & TTY_DRIVER_UNNUMBERED_NODE)
-		strcpy(p, driver->name);
+		return sprintf(p, "%s", driver->name);
 	else
-		sprintf(p, "%s%d", driver->name, index + driver->name_base);
+		return sprintf(p, "%s%d", driver->name,
+			       index + driver->name_base);
 }
 
 /**
@@ -3545,9 +3546,20 @@ static ssize_t show_cons_active(struct device *dev,
 		if (i >= ARRAY_SIZE(cs))
 			break;
 	}
-	while (i--)
-		count += sprintf(buf + count, "%s%d%c",
-				 cs[i]->name, cs[i]->index, i ? ' ':'\n');
+	while (i--) {
+		struct tty_driver *driver;
+		const char *name = cs[i]->name;
+		int index = cs[i]->index;
+
+		driver = cs[i]->device(cs[i], &index);
+		/* Some programs rely on 'tty0' for console */
+		if (driver && (index > 0 || driver->major != TTY_MAJOR)) {
+			count += tty_line_name(driver, index, buf + count);
+			count += sprintf(buf + count, "%c", i ? ' ':'\n');
+		} else
+			count += sprintf(buf + count, "%s%d%c",
+					 name, cs[i]->index, i ? ' ':'\n');
+	}
 	console_unlock();
 
 	return count;
Signature · 1 line
1.7.12.4
#1 lore raw

Hi

Quoted 46 lines · Hannes Reinecke

The 'active' sysfs attribute should refer to the currently active tty devices the console is running on, not the currently active console.
The console structure doesn't refer to any device in sysfs, only the tty the console is running on has.
So we need to print out the tty names in 'active', not the console names.
But we need to take care for the virtual console to always display 'tty0' so as not to break existing programs.

Cc Lennart Poettering <[email protected]>
Cc Kay Sievers <[email protected]>
Cc Greg Kroah-Hartman <[email protected]>
Cc Jiri Slaby <[email protected]>
---
 drivers/tty/tty_io.c | 24 ++++++++++++++++++------
 1 file changed, 18 insertions(+), 6 deletions(-)
drivers/tty/tty_io.c +4 -4
index c74a00a..96eb462 100644
@@ -1271,12 +1271,13 @@ static void pty_line_name(struct tty_driver *driver, int index, char *p)
  *
  *     Locking: None
  */
-static void tty_line_name(struct tty_driver *driver, int index, char *p)
+static ssize_t tty_line_name(struct tty_driver *driver, int index, char *p)
 {
        if (driver->flags & TTY_DRIVER_UNNUMBERED_NODE)
-               strcpy(p, driver->name);
+               return sprintf(p, "%s", driver->name);
        else
-               sprintf(p, "%s%d", driver->name, index + driver->name_base);
+               return sprintf(p, "%s%d", driver->name,
+                              index + driver->name_base);
 }
 
 /**
@@ -3545,9 +3546,20 @@ static ssize_t show_cons_active(struct device *dev,
                if (i >= ARRAY_SIZE(cs))
                        break;
        }
-       while (i--)

drivers/tty/tty_io.c static ssize_t show_cons_active(struct device *dev, · line 3556

Quoted 10 lines
drivers/tty/tty_io.c +8 -2
-               count += sprintf(buf + count, "%s%d%c",
-                                cs[i]->name, cs[i]->index, i ? ' ':'\n');
+       while (i--) {
+               struct tty_driver *driver;
+               const char *name = cs[i]->name;
+               int index = cs[i]->index;
+
+               driver = cs[i]->device(cs[i], &index);
+               /* Some programs rely on 'tty0' for console */
+               if (driver && (index > 0 || driver->major != TTY_MAJOR)) {

As Ray mentioned in the previous discussion, this has to be cs[i]->index instead of "index" as ->device() changes the parameter. See drivers/tty/vt/vt.c vt_console_device().

Thanks
David

Quoted 12 lines
drivers/tty/tty_io.c +6 -1
+                       count += tty_line_name(driver, index, buf + count);
+                       count += sprintf(buf + count, "%c", i ? ' ':'\n');
+               } else
+                       count += sprintf(buf + count, "%s%d%c",
+                                        name, cs[i]->index, i ? ' ':'\n');
+       }
        console_unlock();
 
        return count;
--

1.7.12.4

#2 lore raw
Quoted 55 lines · David Herrmann

Hi

Quoted 52 lines · Hannes Reinecke

The 'active' sysfs attribute should refer to the currently active tty devices the console is running on, not the currently active console.
The console structure doesn't refer to any device in sysfs, only the tty the console is running on has.
So we need to print out the tty names in 'active', not the console names.
But we need to take care for the virtual console to always display 'tty0' so as not to break existing programs.

Cc Lennart Poettering <[email protected]>
Cc Kay Sievers <[email protected]>
Cc Greg Kroah-Hartman <[email protected]>
Cc Jiri Slaby <[email protected]>
---
 drivers/tty/tty_io.c | 24 ++++++++++++++++++------
 1 file changed, 18 insertions(+), 6 deletions(-)
drivers/tty/tty_io.c +8 -6
index c74a00a..96eb462 100644
@@ -1271,12 +1271,13 @@ static void pty_line_name(struct tty_driver *driver, int index, char *p)
  *
  *     Locking: None
  */
-static void tty_line_name(struct tty_driver *driver, int index, char *p)
+static ssize_t tty_line_name(struct tty_driver *driver, int index, char *p)
 {
        if (driver->flags & TTY_DRIVER_UNNUMBERED_NODE)
-               strcpy(p, driver->name);
+               return sprintf(p, "%s", driver->name);
        else
-               sprintf(p, "%s%d", driver->name, index + driver->name_base);
+               return sprintf(p, "%s%d", driver->name,
+                              index + driver->name_base);
 }
 
 /**
@@ -3545,9 +3546,20 @@ static ssize_t show_cons_active(struct device *dev,
                if (i >= ARRAY_SIZE(cs))
                        break;
        }
-       while (i--)
-               count += sprintf(buf + count, "%s%d%c",
-                                cs[i]->name, cs[i]->index, i ? ' ':'\n');
+       while (i--) {
+               struct tty_driver *driver;
+               const char *name = cs[i]->name;
+               int index = cs[i]->index;
Quoted 9 lines

drivers/tty/tty_io.c static ssize_t show_cons_active(struct device *dev, · line 3556

Quoted 4 lines
drivers/tty/tty_io.c +4 -0
+
+               driver = cs[i]->device(cs[i], &index);
+               /* Some programs rely on 'tty0' for console */
+               if (driver && (index > 0 || driver->major != TTY_MAJOR)) {

As Ray mentioned in the previous discussion, this has to be cs[i]->index instead of "index" as ->device() changes the parameter. See drivers/tty/vt/vt.c vt_console_device().

Positive?
I thought this was precisely the problem, ->device() changing the index '0' into something non-zero.
The reports we had were that the line 'tty0' changed into 'tty1'. Hence ->device() converted cs[i]->index (which is '0') into index (which is '1').
Hence the check would be correct, wouldn't it?

Cheers,

Hannes

Signature · 4 lines
Dr. Hannes Reinecke		      zSeries & Storage
[email protected]			      +49 911 74053 688
SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: J. Hawn, J. Guild, F. Imendörffer, HRB 16746 (AG Nürnberg)
#3 lore raw

Hi

Quoted 65 lines · Hannes Reinecke
Quoted 55 lines · David Herrmann

Hi

Quoted 52 lines · Hannes Reinecke

The 'active' sysfs attribute should refer to the currently active tty devices the console is running on, not the currently active console.
The console structure doesn't refer to any device in sysfs, only the tty the console is running on has.
So we need to print out the tty names in 'active', not the console names.
But we need to take care for the virtual console to always display 'tty0' so as not to break existing programs.

Cc Lennart Poettering <[email protected]>
Cc Kay Sievers <[email protected]>
Cc Greg Kroah-Hartman <[email protected]>
Cc Jiri Slaby <[email protected]>
---
 drivers/tty/tty_io.c | 24 ++++++++++++++++++------
 1 file changed, 18 insertions(+), 6 deletions(-)
drivers/tty/tty_io.c +8 -6
index c74a00a..96eb462 100644
@@ -1271,12 +1271,13 @@ static void pty_line_name(struct tty_driver *driver, int index, char *p)
  *
  *     Locking: None
  */
-static void tty_line_name(struct tty_driver *driver, int index, char *p)
+static ssize_t tty_line_name(struct tty_driver *driver, int index, char *p)
 {
        if (driver->flags & TTY_DRIVER_UNNUMBERED_NODE)
-               strcpy(p, driver->name);
+               return sprintf(p, "%s", driver->name);
        else
-               sprintf(p, "%s%d", driver->name, index + driver->name_base);
+               return sprintf(p, "%s%d", driver->name,
+                              index + driver->name_base);
 }
 
 /**
@@ -3545,9 +3546,20 @@ static ssize_t show_cons_active(struct device *dev,
                if (i >= ARRAY_SIZE(cs))
                        break;
        }
-       while (i--)
-               count += sprintf(buf + count, "%s%d%c",
-                                cs[i]->name, cs[i]->index, i ? ' ':'\n');
+       while (i--) {
+               struct tty_driver *driver;
+               const char *name = cs[i]->name;
+               int index = cs[i]->index;
Quoted 9 lines

drivers/tty/tty_io.c static ssize_t show_cons_active(struct device *dev, · line 3556

Quoted 4 lines
drivers/tty/tty_io.c +4 -0
+
+               driver = cs[i]->device(cs[i], &index);
+               /* Some programs rely on 'tty0' for console */
+               if (driver && (index > 0 || driver->major != TTY_MAJOR)) {

As Ray mentioned in the previous discussion, this has to be cs[i]->index instead of "index" as ->device() changes the parameter. See drivers/tty/vt/vt.c vt_console_device().

Quoted 7 lines

Positive?
I thought this was precisely the problem, ->device() changing the index '0' into something non-zero.
The reports we had were that the line 'tty0' changed into 'tty1'. Hence ->device() converted cs[i]->index (which is '0') into index (which is '1').
Hence the check would be correct, wouldn't it?

If "cs[i]" points to tty0, then cs[i]->index is 0. If you call ->device(), it will store 1 (or !=0) in "index". Thus, "(driver && (index > 0))" will be true and you will write tty1 into the file instead of tty0. So you don't want to check whether the new value is non-zero, but whether the *previous* value was 0, turning this into:

if (driver && (cs[i]->index > 0 || driver->major != TTY_MAJOR))

So loosely speaking, we use the new code only for devices which either are not a VT or have an idx > 0. Otherwise, we use our fallback.

Thanks
David

#4 lore raw
Quoted 16 lines · David Herrmann
Quoted 7 lines · Hannes Reinecke

Positive?
I thought this was precisely the problem, ->device() changing the index '0' into something non-zero.
The reports we had were that the line 'tty0' changed into 'tty1'. Hence ->device() converted cs[i]->index (which is '0') into index (which is '1').
Hence the check would be correct, wouldn't it?

If "cs[i]" points to tty0, then cs[i]->index is 0. If you call ->device(), it will store 1 (or !=0) in "index". Thus, "(driver && (index > 0))" will be true and you will write tty1 into the file instead of tty0. So you don't want to check whether the new value is non-zero, but whether the *previous* value was 0, turning this into:

Quoted 4 lines

if (driver && (cs[i]->index > 0 || driver->major != TTY_MAJOR))

So loosely speaking, we use the new code only for devices which either are not a VT or have an idx > 0. Otherwise, we use our fallback.

Hannes, David, care to update the patch to do that? It all sounds fine to me. And we should get this merged again.

Kay

#5 lore raw

Hi

Quoted 17 lines · Kay Sievers
Quoted 16 lines · David Herrmann
Quoted 7 lines · Hannes Reinecke

Positive?
I thought this was precisely the problem, ->device() changing the index '0' into something non-zero.
The reports we had were that the line 'tty0' changed into 'tty1'. Hence ->device() converted cs[i]->index (which is '0') into index (which is '1').
Hence the check would be correct, wouldn't it?

If "cs[i]" points to tty0, then cs[i]->index is 0. If you call ->device(), it will store 1 (or !=0) in "index". Thus, "(driver && (index > 0))" will be true and you will write tty1 into the file instead of tty0. So you don't want to check whether the new value is non-zero, but whether the *previous* value was 0, turning this into:

Quoted 7 lines
Quoted 4 lines

if (driver && (cs[i]->index > 0 || driver->major != TTY_MAJOR))

So loosely speaking, we use the new code only for devices which either are not a VT or have an idx > 0. Otherwise, we use our fallback.

Hannes, David, care to update the patch to do that? It all sounds fine to me. And we should get this merged again.

I have picked it up and resent a fixed patch. I screwed up the version field, so it's called v3 again.. sorry.

Thanks
David