Skip to contentSkip to conversation

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

PATCH Superseded Hannes Reinecke 4 participants 7 messages Feb 5, 2014 – Feb 6, 2014 TTY layer and serial drivers

v1 of 3 Superseded Latest is v3 →

lore
Share & tools
lore
#0 lore raw

PATCH 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.

---
 drivers/tty/tty_io.c | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)
drivers/tty/tty_io.c +12 -2
index c74a00a..17db8ca 100644
@@ -3545,9 +3545,19 @@ static ssize_t show_cons_active(struct device *dev,
 		if (i >= ARRAY_SIZE(cs))
 			break;
 	}
-	while (i--)
+	while (i--) {
+		const struct tty_driver *driver;
+		const char *name = cs[i]->name;
+		int index = cs[i]->index;
+
+		driver = cs[i]->device(cs[i], &index);
+		if (driver) {
+			index += driver->name_base;
+			name = driver->name;
+		}
 		count += sprintf(buf + count, "%s%d%c",
-				 cs[i]->name, cs[i]->index, i ? ' ':'\n');
+				 name, index, i ? ' ':'\n');
+	}
 	console_unlock();
 
 	return count;
Signature · 1 line
1.7.12.4
#1 lore raw

Hi

Quoted 29 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.

---
 drivers/tty/tty_io.c | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)
drivers/tty/tty_io.c +4 -1
index c74a00a..17db8ca 100644
@@ -3545,9 +3545,19 @@ static ssize_t show_cons_active(struct device *dev,
                if (i >= ARRAY_SIZE(cs))
                        break;
        }
-       while (i--)
+       while (i--) {
+               const struct tty_driver *driver;
+               const char *name = cs[i]->name;
+               int index = cs[i]->index;

drivers/tty/tty_io.c static ssize_t show_cons_active(struct device *dev, · line 3560

Quoted 10 lines
drivers/tty/tty_io.c +8 -1
+
+               driver = cs[i]->device(cs[i], &index);
+               if (driver) {
+                       index += driver->name_base;
+                       name = driver->name;
+               }
                count += sprintf(buf + count, "%s%d%c",
-                                cs[i]->name, cs[i]->index, i ? ' ':'\n');
+                                name, index, i ? ' ':'\n');
+       }

Nice catch and indeed, systemd already relies on these names to be identical to their char-dev name. Fortunately, VTs and most serial devices register the console with the same name as the TTY, so we're fine.
Two minor nitpicks:
1) Could you use tty_line_name() instead of sprintf()? It's in the same file and avoids duplicating the name_base logic.
2) Does it make sense to print the console-name if ->device() returns NULL? Seems weird if we print console-names and tty-names in the same attribute. It's unlikely that it causes problems, but there might be conflicts.

Thanks
David

Quoted 10 lines
drivers/tty/tty_io.c +0 -1
        console_unlock();
 
        return count;
--

1.7.12.4

_______________________________________________
systemd-devel mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/systemd-devel

#2 lore raw
Quoted 39 lines · David Herrmann

Hi

Quoted 36 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.

---
  drivers/tty/tty_io.c | 14 ++++++++++++--
  1 file changed, 12 insertions(+), 2 deletions(-)
drivers/tty/tty_io.c +10 -1
index c74a00a..17db8ca 100644
@@ -3545,9 +3545,19 @@ static ssize_t show_cons_active(struct device *dev,
                 if (i >= ARRAY_SIZE(cs))
                         break;
         }
-       while (i--)
+       while (i--) {
+               const struct tty_driver *driver;
+               const char *name = cs[i]->name;
+               int index = cs[i]->index;
+
+               driver = cs[i]->device(cs[i], &index);
+               if (driver) {
+                       index += driver->name_base;
+                       name = driver->name;
+               }
                 count += sprintf(buf + count, "%s%d%c",
Quoted 8 lines

drivers/tty/tty_io.c static ssize_t show_cons_active(struct device *dev, · line 3560

Quoted 3 lines
drivers/tty/tty_io.c +2 -1
-                                cs[i]->name, cs[i]->index, i ? ' ':'\n');
+                                name, index, i ? ' ':'\n');
+       }

Nice catch and indeed, systemd already relies on these names to be identical to their char-dev name. Fortunately, VTs and most serial devices register the console with the same name as the TTY, so we're fine.

What device did this trip over?

Also, this file is not private to systemd. Maybe these changes should be forked into a different sysfs attribute, "active_devices"?

Regards,
Peter Hurley

#3 lore raw

Hi

Quoted 55 lines · Peter Hurley
Quoted 44 lines · David Herrmann

Hi

Quoted 40 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.

---
  drivers/tty/tty_io.c | 14 ++++++++++++--
  1 file changed, 12 insertions(+), 2 deletions(-)
drivers/tty/tty_io.c +10 -2
index c74a00a..17db8ca 100644
@@ -3545,9 +3545,19 @@ static ssize_t show_cons_active(struct device
 *dev,
                 if (i >= ARRAY_SIZE(cs))
                         break;
         }
-       while (i--)
+       while (i--) {
+               const struct tty_driver *driver;
+               const char *name = cs[i]->name;
+               int index = cs[i]->index;
+
+               driver = cs[i]->device(cs[i], &index);
+               if (driver) {
+                       index += driver->name_base;
+                       name = driver->name;
+               }
                 count += sprintf(buf + count, "%s%d%c",
-                                cs[i]->name, cs[i]->index, i ? '
 ':'\n');
Quoted 8 lines

drivers/tty/tty_io.c static ssize_t show_cons_active(struct device *dev, · line 3560

Quoted 2 lines
drivers/tty/tty_io.c +2 -0
+                                name, index, i ? ' ':'\n');
+       }

Nice catch and indeed, systemd already relies on these names to be identical to their char-dev name. Fortunately, VTs and most serial devices register the console with the same name as the TTY, so we're fine.

Quoted 1 line

What device did this trip over?

I haven't seen one so far, but to me it's a coincident, not something we should rely on.

Quoted 2 lines

Also, this file is not private to systemd. Maybe these changes should be forked into a different sysfs attribute, "active_devices"?

What's the use-case to return the name of the console-driver? There is no way for user-space to read active console-drivers anywhere so I think returning the TTY makes more sense. We already have working user-space that can spawn gettys on active consoles via this file. I am open to change this to "active_devices" as the existing interface was clearly not designed to return the device-names.

However, given the fact that both matched so far, I think changing the existing interface to the only user I am aware of is better than adding a new interface just to keep this unused attribute. But obviously it's the maintainer's/your decision and you might know user-space which requires the console-names instead of the tty-names. So please let us know which way to go as we would like to see a reliable way to match active consoles to TTY devices for automated getty-startup.

Thanks
David

#4 lore raw

This patch won't get very far if not addressed to the actual maintainers

[ +cc Greg Kroah-Hartman, Jiri Slaby]

Quoted 73 lines · David Herrmann

Hi

Quoted 55 lines · Peter Hurley
Quoted 44 lines · David Herrmann

Hi

Quoted 40 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.

---
   drivers/tty/tty_io.c | 14 ++++++++++++--
   1 file changed, 12 insertions(+), 2 deletions(-)
drivers/tty/tty_io.c +10 -2
index c74a00a..17db8ca 100644
@@ -3545,9 +3545,19 @@ static ssize_t show_cons_active(struct device
 *dev,
                  if (i >= ARRAY_SIZE(cs))
                          break;
          }
-       while (i--)
+       while (i--) {
+               const struct tty_driver *driver;
+               const char *name = cs[i]->name;
+               int index = cs[i]->index;
+
+               driver = cs[i]->device(cs[i], &index);
+               if (driver) {
+                       index += driver->name_base;
+                       name = driver->name;
+               }
                  count += sprintf(buf + count, "%s%d%c",
-                                cs[i]->name, cs[i]->index, i ? '
 ':'\n');
Quoted 8 lines

drivers/tty/tty_io.c static ssize_t show_cons_active(struct device *dev, · line 3560

Quoted 2 lines
drivers/tty/tty_io.c +2 -0
+                                name, index, i ? ' ':'\n');
+       }

Nice catch and indeed, systemd already relies on these names to be identical to their char-dev name. Fortunately, VTs and most serial devices register the console with the same name as the TTY, so we're fine.

Quoted 1 line

What device did this trip over?

I haven't seen one so far, but to me it's a coincident, not something we should rely on.

Quoted 2 lines

Also, this file is not private to systemd. Maybe these changes should be forked into a different sysfs attribute, "active_devices"?

What's the use-case to return the name of the console-driver? There is no way for user-space to read active console-drivers anywhere so I think returning the TTY makes more sense. We already have working user-space that can spawn gettys on active consoles via this file. I am open to change this to "active_devices" as the existing interface was clearly not designed to return the device-names.

Quoted 8 lines

However, given the fact that both matched so far, I think changing the existing interface to the only user I am aware of is better than adding a new interface just to keep this unused attribute. But obviously it's the maintainer's/your decision and you might know user-space which requires the console-names instead of the tty-names. So please let us know which way to go as we would like to see a reliable way to match active consoles to TTY devices for automated getty-startup.

Sure, I get it. Just wanted to point out the obvious right up front so that if it turns out there is another userspace dependency and this gets rewound, possibly across future -stable kernels, the fallout could get ugly.

Regards,
Peter Hurley

PS - The "active" attribute won't be unused; old systemd's will still depend on it, yes?

#5 lore raw
In reply to David Herrmann
Quoted 43 lines · David Herrmann

Hi

Quoted 29 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.

---
 drivers/tty/tty_io.c | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)
drivers/tty/tty_io.c +4 -1
index c74a00a..17db8ca 100644
@@ -3545,9 +3545,19 @@ static ssize_t show_cons_active(struct device *dev,
                if (i >= ARRAY_SIZE(cs))
                        break;
        }
-       while (i--)
+       while (i--) {
+               const struct tty_driver *driver;
+               const char *name = cs[i]->name;
+               int index = cs[i]->index;

drivers/tty/tty_io.c static ssize_t show_cons_active(struct device *dev, · line 3560

Quoted 10 lines
drivers/tty/tty_io.c +8 -1
+
+               driver = cs[i]->device(cs[i], &index);
+               if (driver) {
+                       index += driver->name_base;
+                       name = driver->name;
+               }
                count += sprintf(buf + count, "%s%d%c",
-                                cs[i]->name, cs[i]->index, i ? ' ':'\n');
+                                name, index, i ? ' ':'\n');
+       }
Quoted 7 lines

Nice catch and indeed, systemd already relies on these names to be identical to their char-dev name. Fortunately, VTs and most serial devices register the console with the same name as the TTY, so we're fine.
Two minor nitpicks:
1) Could you use tty_line_name() instead of sprintf()? It's in the same file and avoids duplicating the name_base logic.

Ok. Not that it makes the patch nicer, but hey.

Quoted 5 lines

2) Does it make sense to print the console-name if ->device() returns NULL? Seems weird if we print console-names and tty-names in the same attribute. It's unlikely that it causes problems, but there might be conflicts.

This is basically a fallback; this is the old behaviour, which still might be called for when coming across a tty which just has a stub for the ->device callback.
It's not that the '->device' callback is used that frequently, so I wouldn't be surprised here.

Meanwhile I've sent a new patch, reviews are welcome there.

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)
maintainer
#6 lore raw
In reply to Hannes Reinecke
Quoted 8 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.

Quoted 7 lines
---
 drivers/tty/tty_io.c | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

Would have been nice to actually cc: the tty maintainer(s) :(

scripts/get_maintainer.pl is your friend...

greg k-h