[RESEND][PATCH/RFC] OMAP2: serial.c: Fix number of uarts in early_init

[RESEND][PATCH/RFC] OMAP2: serial.c: Fix number of uarts in early_init

[Thomas Weber]

The omap_serial_early_init prints the following errors:

Could not get uart4_ick
Could not get uart4_fck

because all the uarts available in omap_uart[] will be initialized.
Only omap4430 and omap3630 have 4 uarts at the moment.
This patch reduces the number of uarts when cpu is not omap4430 or
omap3630.

Signed-off-by: Thomas Weber <weber@corscience.de>
---
 arch/arm/mach-omap2/serial.c |   15 ++++++++++-----
 1 files changed, 10 insertions(+), 5 deletions(-)

diff --git a/arch/arm/mach-omap2/serial.c b/arch/arm/mach-omap2/serial.c
index b79bc89..da77930 100644
--- a/arch/arm/mach-omap2/serial.c
+++ b/arch/arm/mach-omap2/serial.c
@@ -644,16 +644,21 @@ static void serial_out_override(struct uart_port *up, int offset, int value)
 }
 void __init omap_serial_early_init(void)
 {
-	int i;
+	int i, nr_ports;
 	char name[16];
 
+	if (!(cpu_is_omap3630() || cpu_is_omap4430()))
+		nr_ports = 3;
+	else
+		nr_ports = ARRAY_SIZE(omap_uart);
+
 	/*
 	 * Make sure the serial ports are muxed on at this point.
 	 * You have to mux them off in device drivers later on
 	 * if not needed.
 	 */
 
-	for (i = 0; i < ARRAY_SIZE(omap_uart); i++) {
+	for (i = 0; i < nr_ports; i++) {
 		struct omap_uart_state *uart = &omap_uart[i];
 		struct platform_device *pdev = &uart->pdev;
 		struct device *dev = &pdev->dev;
@@ -669,17 +674,17 @@ void __init omap_serial_early_init(void)
 			continue;
 		}
 
-		sprintf(name, "uart%d_ick", i+1);
+		sprintf(name, "uart%d_ick", i + 1);
 		uart->ick = clk_get(NULL, name);
 		if (IS_ERR(uart->ick)) {
-			printk(KERN_ERR "Could not get uart%d_ick\n", i+1);
+			printk(KERN_ERR "Could not get uart%d_ick\n", i + 1);
 			uart->ick = NULL;
 		}
 
 		sprintf(name, "uart%d_fck", i+1);
 		uart->fck = clk_get(NULL, name);
 		if (IS_ERR(uart->fck)) {
-			printk(KERN_ERR "Could not get uart%d_fck\n", i+1);
+			printk(KERN_ERR "Could not get uart%d_fck\n", i + 1);
 			uart->fck = NULL;
 		}
 
-- 
1.6.4.4

[RESEND][PATCH/RFC] OMAP2: serial.c: Fix number of uarts in early_init

[Tony Lindgren]

* Thomas Weber <swirl@gmx.li> [100225 01:36]:
> 
> The omap_serial_early_init prints the following errors:
> 
> Could not get uart4_ick
> Could not get uart4_fck
> 
> because all the uarts available in omap_uart[] will be initialized.
> Only omap4430 and omap3630 have 4 uarts at the moment.
> This patch reduces the number of uarts when cpu is not omap4430 or
> omap3630.

..and I only patched omap_serial_init and forgot about
the omap_serial_early_init..

Looks good to me, Sergio, care to test and Ack this? Also, sounds
like we still need to patch omap4 for uart4 for the clocks.

Regards,

Tony
 
> Signed-off-by: Thomas Weber <weber@corscience.de>
> ---
>  arch/arm/mach-omap2/serial.c |   15 ++++++++++-----
>  1 files changed, 10 insertions(+), 5 deletions(-)
> 
> diff --git a/arch/arm/mach-omap2/serial.c b/arch/arm/mach-omap2/serial.c
> index b79bc89..da77930 100644
> --- a/arch/arm/mach-omap2/serial.c
> +++ b/arch/arm/mach-omap2/serial.c
> @@ -644,16 +644,21 @@ static void serial_out_override(struct uart_port *up, int offset, int value)
>  }
>  void __init omap_serial_early_init(void)
>  {
> -	int i;
> +	int i, nr_ports;
>  	char name[16];
>  
> +	if (!(cpu_is_omap3630() || cpu_is_omap4430()))
> +		nr_ports = 3;
> +	else
> +		nr_ports = ARRAY_SIZE(omap_uart);
> +
>  	/*
>  	 * Make sure the serial ports are muxed on at this point.
>  	 * You have to mux them off in device drivers later on
>  	 * if not needed.
>  	 */
>  
> -	for (i = 0; i < ARRAY_SIZE(omap_uart); i++) {
> +	for (i = 0; i < nr_ports; i++) {
>  		struct omap_uart_state *uart = &omap_uart[i];
>  		struct platform_device *pdev = &uart->pdev;
>  		struct device *dev = &pdev->dev;
> @@ -669,17 +674,17 @@ void __init omap_serial_early_init(void)
>  			continue;
>  		}
>  
> -		sprintf(name, "uart%d_ick", i+1);
> +		sprintf(name, "uart%d_ick", i + 1);
>  		uart->ick = clk_get(NULL, name);
>  		if (IS_ERR(uart->ick)) {
> -			printk(KERN_ERR "Could not get uart%d_ick\n", i+1);
> +			printk(KERN_ERR "Could not get uart%d_ick\n", i + 1);
>  			uart->ick = NULL;
>  		}
>  
>  		sprintf(name, "uart%d_fck", i+1);
>  		uart->fck = clk_get(NULL, name);
>  		if (IS_ERR(uart->fck)) {
> -			printk(KERN_ERR "Could not get uart%d_fck\n", i+1);
> +			printk(KERN_ERR "Could not get uart%d_fck\n", i + 1);
>  			uart->fck = NULL;
>  		}
>  
> -- 
> 1.6.4.4
> 

[RESEND][PATCH/RFC] OMAP2: serial.c: Fix number of uarts in early_init

[Shilimkar, Santosh]

> -----Original Message-----
> From: Tony Lindgren [mailto:tony at atomide.com]
> Sent: Friday, February 26, 2010 12:14 AM
> To: Thomas Weber; Aguirre, Sergio
> Cc: linux-omap at vger.kernel.org; Russell King; Kevin Hilman; Shilimkar, Santosh; Pandita, Vikram;
> Syed, Rafiuddin; linux-arm-kernel at lists.infradead.org; linux-kernel at vger.kernel.org; Thomas Weber
> Subject: Re: [RESEND][PATCH/RFC] OMAP2: serial.c: Fix number of uarts in early_init
> 
> * Thomas Weber <swirl@gmx.li> [100225 01:36]:
> >
> > The omap_serial_early_init prints the following errors:
> >
> > Could not get uart4_ick
> > Could not get uart4_fck
> >
> > because all the uarts available in omap_uart[] will be initialized.
> > Only omap4430 and omap3630 have 4 uarts at the moment.
> > This patch reduces the number of uarts when cpu is not omap4430 or
> > omap3630.
> 
> ..and I only patched omap_serial_init and forgot about
> the omap_serial_early_init..
> 
> Looks good to me, Sergio, care to test and Ack this? Also, sounds
> like we still need to patch omap4 for uart4 for the clocks.
> 
Clocks will be take care automatically by "nr_ports" for omap4 too
with this patch. Below clock changes are just cosmetic ones, isn't it ?

Am I missing something?

> Regards,
> 
> Tony
> 
> > Signed-off-by: Thomas Weber <weber@corscience.de>
> > ---
> >  arch/arm/mach-omap2/serial.c |   15 ++++++++++-----
> >  1 files changed, 10 insertions(+), 5 deletions(-)
> >
> > diff --git a/arch/arm/mach-omap2/serial.c b/arch/arm/mach-omap2/serial.c
> > index b79bc89..da77930 100644
> > --- a/arch/arm/mach-omap2/serial.c
> > +++ b/arch/arm/mach-omap2/serial.c
> > @@ -644,16 +644,21 @@ static void serial_out_override(struct uart_port *up, int offset, int value)
> >  }
> >  void __init omap_serial_early_init(void)
> >  {
> > -	int i;
> > +	int i, nr_ports;
> >  	char name[16];
> >
> > +	if (!(cpu_is_omap3630() || cpu_is_omap4430()))
> > +		nr_ports = 3;
> > +	else
> > +		nr_ports = ARRAY_SIZE(omap_uart);
> > +
> >  	/*
> >  	 * Make sure the serial ports are muxed on at this point.
> >  	 * You have to mux them off in device drivers later on
> >  	 * if not needed.
> >  	 */
> >
> > -	for (i = 0; i < ARRAY_SIZE(omap_uart); i++) {
> > +	for (i = 0; i < nr_ports; i++) {
> >  		struct omap_uart_state *uart = &omap_uart[i];
> >  		struct platform_device *pdev = &uart->pdev;
> >  		struct device *dev = &pdev->dev;
> > @@ -669,17 +674,17 @@ void __init omap_serial_early_init(void)
> >  			continue;
> >  		}
> >
> > -		sprintf(name, "uart%d_ick", i+1);
> > +		sprintf(name, "uart%d_ick", i + 1);
> >  		uart->ick = clk_get(NULL, name);
> >  		if (IS_ERR(uart->ick)) {
> > -			printk(KERN_ERR "Could not get uart%d_ick\n", i+1);
> > +			printk(KERN_ERR "Could not get uart%d_ick\n", i + 1);
> >  			uart->ick = NULL;
> >  		}
> >
> >  		sprintf(name, "uart%d_fck", i+1);
> >  		uart->fck = clk_get(NULL, name);
> >  		if (IS_ERR(uart->fck)) {
> > -			printk(KERN_ERR "Could not get uart%d_fck\n", i+1);
> > +			printk(KERN_ERR "Could not get uart%d_fck\n", i + 1);
> >  			uart->fck = NULL;
> >  		}
> >
> > --
> > 1.6.4.4
> >

[RESEND][PATCH/RFC] OMAP2: serial.c: Fix number of uarts in early_init

[Tony Lindgren]

* Shilimkar, Santosh <santosh.shilimkar@ti.com> [100225 10:51]:
> 
> > -----Original Message-----
> > From: Tony Lindgren [mailto:tony at atomide.com]
> > Sent: Friday, February 26, 2010 12:14 AM
> > To: Thomas Weber; Aguirre, Sergio
> > Cc: linux-omap at vger.kernel.org; Russell King; Kevin Hilman; Shilimkar, Santosh; Pandita, Vikram;
> > Syed, Rafiuddin; linux-arm-kernel at lists.infradead.org; linux-kernel at vger.kernel.org; Thomas Weber
> > Subject: Re: [RESEND][PATCH/RFC] OMAP2: serial.c: Fix number of uarts in early_init
> > 
> > * Thomas Weber <swirl@gmx.li> [100225 01:36]:
> > >
> > > The omap_serial_early_init prints the following errors:
> > >
> > > Could not get uart4_ick
> > > Could not get uart4_fck
> > >
> > > because all the uarts available in omap_uart[] will be initialized.
> > > Only omap4430 and omap3630 have 4 uarts at the moment.
> > > This patch reduces the number of uarts when cpu is not omap4430 or
> > > omap3630.
> > 
> > ..and I only patched omap_serial_init and forgot about
> > the omap_serial_early_init..
> > 
> > Looks good to me, Sergio, care to test and Ack this? Also, sounds
> > like we still need to patch omap4 for uart4 for the clocks.
> > 
> Clocks will be take care automatically by "nr_ports" for omap4 too
> with this patch. Below clock changes are just cosmetic ones, isn't it ?
> 
> Am I missing something?

Yes sorry, I'm referring to "[RFC][PATCH 0/3] OMAP3630: UART4 startup"
series to fix the uart4 issues for 3630, not omap4.

Regards,

Tony

[RESEND][PATCH/RFC] OMAP2: serial.c: Fix number of uarts in early_init

[Shilimkar, Santosh]

> -----Original Message-----
> From: Tony Lindgren [mailto:tony at atomide.com]
> Sent: Friday, February 26, 2010 12:28 AM
> To: Shilimkar, Santosh
> Cc: Thomas Weber; Aguirre, Sergio; linux-omap at vger.kernel.org; Russell King; Kevin Hilman; Pandita,
> Vikram; Syed, Rafiuddin; linux-arm-kernel at lists.infradead.org; linux-kernel at vger.kernel.org; Thomas
> Weber
> Subject: Re: [RESEND][PATCH/RFC] OMAP2: serial.c: Fix number of uarts in early_init
> 
> * Shilimkar, Santosh <santosh.shilimkar@ti.com> [100225 10:51]:
> >
> > > -----Original Message-----
> > > From: Tony Lindgren [mailto:tony at atomide.com]
> > > Sent: Friday, February 26, 2010 12:14 AM
> > > To: Thomas Weber; Aguirre, Sergio
> > > Cc: linux-omap at vger.kernel.org; Russell King; Kevin Hilman; Shilimkar, Santosh; Pandita, Vikram;
> > > Syed, Rafiuddin; linux-arm-kernel at lists.infradead.org; linux-kernel at vger.kernel.org; Thomas Weber
> > > Subject: Re: [RESEND][PATCH/RFC] OMAP2: serial.c: Fix number of uarts in early_init
> > >
> > > * Thomas Weber <swirl@gmx.li> [100225 01:36]:
> > > >
> > > > The omap_serial_early_init prints the following errors:
> > > >
> > > > Could not get uart4_ick
> > > > Could not get uart4_fck
> > > >
> > > > because all the uarts available in omap_uart[] will be initialized.
> > > > Only omap4430 and omap3630 have 4 uarts at the moment.
> > > > This patch reduces the number of uarts when cpu is not omap4430 or
> > > > omap3630.
> > >
> > > ..and I only patched omap_serial_init and forgot about
> > > the omap_serial_early_init..
> > >
> > > Looks good to me, Sergio, care to test and Ack this? Also, sounds
> > > like we still need to patch omap4 for uart4 for the clocks.
> > >
> > Clocks will be take care automatically by "nr_ports" for omap4 too
> > with this patch. Below clock changes are just cosmetic ones, isn't it ?
> >
> > Am I missing something?
> 
> Yes sorry, I'm referring to "[RFC][PATCH 0/3] OMAP3630: UART4 startup"
> series to fix the uart4 issues for 3630, not omap4.
> 
OK

Regards,
Santosh