Skip to contentSkip to conversation

Serial driver (serial_core.c) status messages should be set to KERN_INFO

DISCUSSION Alon Bar-Lev 4 participants 7 messages Sep 1, 2005 – Sep 2, 2005

lore
Share & tools
lore
#0 lore raw

Hello,

When upgrading to 2.6.13 I've noticed that serial driver reports it status with unknown severity, causing the boot-splash to be overridden.

Please consider this modification.

Best Regards,

Alon Bar-Lev.

At drivers/serial/serial_core.c

static inline void

uart_report_port(struct uart_driver *drv, struct uart_port *port) {
- printk("%s%d", drv->dev_name, port->line);
+ printk(KERN_INFO + "%s%d", drv->dev_name, port->line);

        printk(" at ");
        switch (port->iotype) {
        case UPIO_PORT:
                printk("I/O 0x%x", port->iobase);
                break;
        case UPIO_HUB6:
                printk("I/O 0x%x offset 0x%x", port->iobase, port->hub6);
                break;
        case UPIO_MEM:
        case UPIO_MEM32:
                printk("MMIO 0x%lx", port->mapbase);
                break;
        }
        printk(" (irq = %d) is a %s\n", port->irq, uart_type(port));
}
#1 lore raw

Alon Bar-Lev napsal(a):

Quoted 21 lines

Hello,

When upgrading to 2.6.13 I've noticed that serial driver reports it status with unknown severity, causing the boot-splash to be overridden.

Please consider this modification.

Best Regards,

Alon Bar-Lev.

At drivers/serial/serial_core.c

static inline void

uart_report_port(struct uart_driver *drv, struct uart_port *port) {

Review

Quoted 2 lines
(quoted diff) +1 -1
-        printk("%s%d", drv->dev_name, port->line);
+      printk(KERN_INFO + "%s%d", drv->dev_name, port->line);

plus sign between that?

Quoted 2 lines
        printk(" at ");

why the fellows didn't put this to the line above?

Quoted 3 lines
        switch (port->iotype) {
        case UPIO_PORT:
                printk("I/O 0x%x", port->iobase);

And what about these?

Quoted 11 lines
                break;
        case UPIO_HUB6:
                printk("I/O 0x%x offset 0x%x", port->iobase, port->hub6);
                break;
        case UPIO_MEM:
        case UPIO_MEM32:
                printk("MMIO 0x%lx", port->mapbase);
                break;
        }
        printk(" (irq = %d) is a %s\n", port->irq, uart_type(port));
}

regards,

Signature · 3 lines
Jiri Slaby         www.fi.muni.cz/~xslaby
~\-/~      [email protected]      ~\-/~
241B347EC88228DE51EE A49C4A73A25004CB2A10
#2 lore raw
Quoted 30 lines · Jiri Slaby

Alon Bar-Lev napsal(a):

Quoted 22 lines

Hello,

When upgrading to 2.6.13 I've noticed that serial driver reports it status with unknown severity, causing the boot-splash to be overridden.

Please consider this modification.

Best Regards,

Alon Bar-Lev.

At drivers/serial/serial_core.c

static inline void

uart_report_port(struct uart_driver *drv, struct uart_port *port) {

Review

Quoted 2 lines
(quoted diff) +1 -1
-        printk("%s%d", drv->dev_name, port->line);
+      printk(KERN_INFO + "%s%d", drv->dev_name, port->line);

plus sign between that?

Quoted 1 line
Quoted 8 lines
Quoted 1 line
       printk(" at ");

why the fellows didn't put this to the line above?

Quoted 3 lines
       switch (port->iotype) {
       case UPIO_PORT:
               printk("I/O 0x%x", port->iobase);

And what about these?

looks like they're not on a newline, so need no severity.

Quoted 11 lines
Quoted 11 lines
               break;
       case UPIO_HUB6:
               printk("I/O 0x%x offset 0x%x", port->iobase, port->hub6);
               break;
       case UPIO_MEM:
       case UPIO_MEM32:
               printk("MMIO 0x%lx", port->mapbase);
               break;
       }
       printk(" (irq = %d) is a %s\n", port->irq, uart_type(port));
}
Signature · 3 lines
Ben ([email protected], http://www.fluff.org/)

  'a smiley only costs 4 bytes'
#3 lore raw

Ben Dooks napsal(a):

Quoted 29 lines · Jiri Slaby
Quoted 28 lines

Alon Bar-Lev napsal(a):

Review

Quoted 8 lines

static inline void

uart_report_port(struct uart_driver *drv, struct uart_port *port) {

(quoted diff) +1 -1
-        printk("%s%d", drv->dev_name, port->line);
+      printk(KERN_INFO + "%s%d", drv->dev_name, port->line);

plus sign between that?

Quoted 3 lines
      printk(" at ");

why the fellows didn't put this to the line above?

Quoted 5 lines
      switch (port->iotype) {
      case UPIO_PORT:
              printk("I/O 0x%x", port->iobase);
Quoted 7 lines
Quoted 3 lines

And what about these?

looks like they're not on a newline, so need no severity.

ok, ok, but isn't this a little bit racy (so you can see dev_name and line, then another driver's info and then " at ", then something else...)

regards,

Signature · 3 lines
Jiri Slaby         www.fi.muni.cz/~xslaby
~\-/~      [email protected]      ~\-/~
241B347EC88228DE51EE A49C4A73A25004CB2A10
#4 lore raw
Quoted 2 lines · Jiri Slaby

ok, ok, but isn't this a little bit racy (so you can see dev_name and line, then another driver's info and then " at ", then something else...)

I think you'll find things are serialised.

Signature · 3 lines
Russell King
 Linux kernel    2.6 ARM Linux   - http://www.arm.linux.org.uk/
 maintainer of:  2.6 Serial core
#5 lore raw
In reply to Jiri Slaby

Hi!

Quoted 21 lines · Jiri Slaby

Alon Bar-Lev napsal(a):

Quoted 19 lines

Hello,

When upgrading to 2.6.13 I've noticed that serial driver reports it status with unknown severity, causing the boot-splash to be overridden.

Please consider this modification.

Best Regards,

Alon Bar-Lev.

At drivers/serial/serial_core.c

static inline void

Quoted 7 lines

Review

Quoted 4 lines

uart_report_port(struct uart_driver *drv, struct uart_port *port) {

(quoted diff) +1 -1
-        printk("%s%d", drv->dev_name, port->line);
+      printk(KERN_INFO + "%s%d", drv->dev_name, port->line);

plus sign between that?

You are right!!! the + is mistake.
The KERN_INFO is the main fix.

Quoted 6 lines
Quoted 3 lines

        printk(" at ");

why the fellows didn't put this to the line above?

Regarding the other comments... I really don't know... this is how the driver is written.... I would have constructed a string and only then printk it...

Regards,
Alon Bar-Lev.

#6 lore raw
In reply to Alon Bar-Lev
Quoted 2 lines · Alon Bar-Lev

When upgrading to 2.6.13 I've noticed that serial driver reports it status with unknown severity, causing the boot-splash to be overridden.

Please don't submit patches to bugzilla as a way to get them into the kernel. Instead, please send them direct to the person/mailing list responsible for the area, as per the SubmittingPatches document in the kernel tree.

Thanks.

Signature · 3 lines
Russell King
 Linux kernel    2.6 ARM Linux   - http://www.arm.linux.org.uk/
 maintainer of:  2.6 Serial core