[PATCH 2.6.29.2] PXA I2C: Define log level for i2c PXA error report

[PATCH 2.6.29.2] PXA I2C: Define log level for i2c PXA error report

[Michael Abbott]

All the i2c pxa error messages should probably be KERN_DEBUG

At least, all the printk messages should have a log level specfied,
and KERN_DEBUG seems appropriate.

Signed-off-by: Michael Abbott <michael.abbott@diamond.ac.uk>
---
 drivers/i2c/busses/i2c-pxa.c |    8 ++++----
 1 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/i2c/busses/i2c-pxa.c b/drivers/i2c/busses/i2c-pxa.c
index bdb1f75..8cb9160 100644
--- a/drivers/i2c/busses/i2c-pxa.c
+++ b/drivers/i2c/busses/i2c-pxa.c
@@ -210,11 +210,11 @@ static irqreturn_t i2c_pxa_handler(int this_irq, void *dev_id);
 static void i2c_pxa_scream_blue_murder(struct pxa_i2c *i2c, const char *why)
 {
 	unsigned int i;
-	printk("i2c: error: %s\n", why);
-	printk("i2c: msg_num: %d msg_idx: %d msg_ptr: %d\n",
+	printk(KERN_DEBUG "i2c: error: %s\n", why);
+	printk(KERN_DEBUG "i2c: msg_num: %d msg_idx: %d msg_ptr: %d\n",
 		i2c->msg_num, i2c->msg_idx, i2c->msg_ptr);
-	printk("i2c: ICR: %08x ISR: %08x\n"
-	       "i2c: log: ", readl(_ICR(i2c)), readl(_ISR(i2c)));
+	printk(KERN_DEBUG "i2c: ICR: %08x ISR: %08x\n"
+	       KERN_DEBUG "i2c: log: ", readl(_ICR(i2c)), readl(_ISR(i2c)));
 	for (i = 0; i < i2c->irqlogidx; i++)
 		printk("[%08x:%08x] ", i2c->isrlog[i], i2c->icrlog[i]);
 	printk("\n");
-- 
1.6.1.3

Re: [PATCH 2.6.29.2] PXA I2C: Define log level for i2c PXA error report

[Jean Delvare]

On Wed, 6 May 2009 08:32:09 +0100 (BST), Michael Abbott wrote:
> All the i2c pxa error messages should probably be KERN_DEBUG
> 
> At least, all the printk messages should have a log level specfied,
> and KERN_DEBUG seems appropriate.
> 
> Signed-off-by: Michael Abbott <michael.abbott@diamond.ac.uk>
> ---
>  drivers/i2c/busses/i2c-pxa.c |    8 ++++----
>  1 files changed, 4 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/i2c/busses/i2c-pxa.c b/drivers/i2c/busses/i2c-pxa.c
> index bdb1f75..8cb9160 100644
> --- a/drivers/i2c/busses/i2c-pxa.c
> +++ b/drivers/i2c/busses/i2c-pxa.c
> @@ -210,11 +210,11 @@ static irqreturn_t i2c_pxa_handler(int this_irq, void *dev_id);
>  static void i2c_pxa_scream_blue_murder(struct pxa_i2c *i2c, const char *why)
>  {
>  	unsigned int i;
> -	printk("i2c: error: %s\n", why);
> -	printk("i2c: msg_num: %d msg_idx: %d msg_ptr: %d\n",
> +	printk(KERN_DEBUG "i2c: error: %s\n", why);
> +	printk(KERN_DEBUG "i2c: msg_num: %d msg_idx: %d msg_ptr: %d\n",
>  		i2c->msg_num, i2c->msg_idx, i2c->msg_ptr);
> -	printk("i2c: ICR: %08x ISR: %08x\n"
> -	       "i2c: log: ", readl(_ICR(i2c)), readl(_ISR(i2c)));
> +	printk(KERN_DEBUG "i2c: ICR: %08x ISR: %08x\n"
> +	       KERN_DEBUG "i2c: log: ", readl(_ICR(i2c)), readl(_ISR(i2c)));
>  	for (i = 0; i < i2c->irqlogidx; i++)
>  		printk("[%08x:%08x] ", i2c->isrlog[i], i2c->icrlog[i]);
>  	printk("\n");

Already fixed in Linus' tree:
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=154d22b04ae1741c5fcfd5d747b813a9a279abff#patch4

-- 
Jean Delvare

Re: [PATCH 2.6.29.2] PXA I2C: Define log level for i2c PXA error report

[Michael Abbott]

On Wed, 6 May 2009, Jean Delvare wrote:
> On Wed, 6 May 2009 08:32:09 +0100 (BST), Michael Abbott wrote:
> > All the i2c pxa error messages should probably be KERN_DEBUG
> > 
> > At least, all the printk messages should have a log level specfied,
> > and KERN_DEBUG seems appropriate.
> > 
> > Signed-off-by: Michael Abbott <michael.abbott@diamond.ac.uk>
> > ---
> >  drivers/i2c/busses/i2c-pxa.c |    8 ++++----
> >  1 files changed, 4 insertions(+), 4 deletions(-)
> > 
> > diff --git a/drivers/i2c/busses/i2c-pxa.c b/drivers/i2c/busses/i2c-pxa.c
> > index bdb1f75..8cb9160 100644
> > --- a/drivers/i2c/busses/i2c-pxa.c
> > +++ b/drivers/i2c/busses/i2c-pxa.c
> > @@ -210,11 +210,11 @@ static irqreturn_t i2c_pxa_handler(int this_irq, void *dev_id);
> >  static void i2c_pxa_scream_blue_murder(struct pxa_i2c *i2c, const char *why)
> >  {
> >  	unsigned int i;
> > -	printk("i2c: error: %s\n", why);
> > -	printk("i2c: msg_num: %d msg_idx: %d msg_ptr: %d\n",
> > +	printk(KERN_DEBUG "i2c: error: %s\n", why);
> > +	printk(KERN_DEBUG "i2c: msg_num: %d msg_idx: %d msg_ptr: %d\n",
> >  		i2c->msg_num, i2c->msg_idx, i2c->msg_ptr);
> > -	printk("i2c: ICR: %08x ISR: %08x\n"
> > -	       "i2c: log: ", readl(_ICR(i2c)), readl(_ISR(i2c)));
> > +	printk(KERN_DEBUG "i2c: ICR: %08x ISR: %08x\n"
> > +	       KERN_DEBUG "i2c: log: ", readl(_ICR(i2c)), readl(_ISR(i2c)));
> >  	for (i = 0; i < i2c->irqlogidx; i++)
> >  		printk("[%08x:%08x] ", i2c->isrlog[i], i2c->icrlog[i]);
> >  	printk("\n");
> 
> Already fixed in Linus' tree:
> http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=154d22b04ae1741c5fcfd5d747b813a9a279abff#patch4

Now this is very odd.  I'd assumed the error messages I was getting were 
just noise, but the commit you point me to uses KERN_ERR instead -- and so 
I get a brief storm of these messages on boot on my target:

[   20.667923] i2c: error: exhausted retries
[   20.671962] i2c: msg_num: 0 msg_idx: -2000 msg_ptr: 0
[   20.677113] i2c: ICR: 000007e0 ISR: 00000002
[   20.677128] i2c: log: [00000446:000007e0]

This is repeated nine times in total with exactly the same numbers.  Any 
thoughts as to what this means?  Other than these messages on startup my 
I2C sensors seem to work perfectly well.

Re: [PATCH 2.6.29.2] PXA I2C: Define log level for i2c PXA error report

[Jean Delvare]

On Mon, 11 May 2009 07:38:59 +0100 (BST), Michael Abbott wrote:
> On Wed, 6 May 2009, Jean Delvare wrote:
> > On Wed, 6 May 2009 08:32:09 +0100 (BST), Michael Abbott wrote:
> > > All the i2c pxa error messages should probably be KERN_DEBUG
> > > 
> > > At least, all the printk messages should have a log level specfied,
> > > and KERN_DEBUG seems appropriate.
> > > 
> > > Signed-off-by: Michael Abbott <michael.abbott@diamond.ac.uk>
> > > ---
> > >  drivers/i2c/busses/i2c-pxa.c |    8 ++++----
> > >  1 files changed, 4 insertions(+), 4 deletions(-)
> > > 
> > > diff --git a/drivers/i2c/busses/i2c-pxa.c b/drivers/i2c/busses/i2c-pxa.c
> > > index bdb1f75..8cb9160 100644
> > > --- a/drivers/i2c/busses/i2c-pxa.c
> > > +++ b/drivers/i2c/busses/i2c-pxa.c
> > > @@ -210,11 +210,11 @@ static irqreturn_t i2c_pxa_handler(int this_irq, void *dev_id);
> > >  static void i2c_pxa_scream_blue_murder(struct pxa_i2c *i2c, const char *why)
> > >  {
> > >  	unsigned int i;
> > > -	printk("i2c: error: %s\n", why);
> > > -	printk("i2c: msg_num: %d msg_idx: %d msg_ptr: %d\n",
> > > +	printk(KERN_DEBUG "i2c: error: %s\n", why);
> > > +	printk(KERN_DEBUG "i2c: msg_num: %d msg_idx: %d msg_ptr: %d\n",
> > >  		i2c->msg_num, i2c->msg_idx, i2c->msg_ptr);
> > > -	printk("i2c: ICR: %08x ISR: %08x\n"
> > > -	       "i2c: log: ", readl(_ICR(i2c)), readl(_ISR(i2c)));
> > > +	printk(KERN_DEBUG "i2c: ICR: %08x ISR: %08x\n"
> > > +	       KERN_DEBUG "i2c: log: ", readl(_ICR(i2c)), readl(_ISR(i2c)));
> > >  	for (i = 0; i < i2c->irqlogidx; i++)
> > >  		printk("[%08x:%08x] ", i2c->isrlog[i], i2c->icrlog[i]);
> > >  	printk("\n");
> > 
> > Already fixed in Linus' tree:
> > http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=154d22b04ae1741c5fcfd5d747b813a9a279abff#patch4
> 
> Now this is very odd.  I'd assumed the error messages I was getting were 
> just noise, but the commit you point me to uses KERN_ERR instead -- and so 
> I get a brief storm of these messages on boot on my target:
> 
> [   20.667923] i2c: error: exhausted retries
> [   20.671962] i2c: msg_num: 0 msg_idx: -2000 msg_ptr: 0
> [   20.677113] i2c: ICR: 000007e0 ISR: 00000002
> [   20.677128] i2c: log: [00000446:000007e0]
> 
> This is repeated nine times in total with exactly the same numbers.  Any 
> thoughts as to what this means?  Other than these messages on startup my 
> I2C sensors seem to work perfectly well.

I don't know anything about the i2c-pxa driver, sorry.

-- 
Jean Delvare

Re: [PATCH 2.6.29.2] PXA I2C: Define log level for i2c PXA error report

[伊泽]

2009/5/11 Michael Abbott <michael@araneidae.co.uk>
>
> On Wed, 6 May 2009, Jean Delvare wrote:
> > On Wed, 6 May 2009 08:32:09 +0100 (BST), Michael Abbott wrote:
> > > All the i2c pxa error messages should probably be KERN_DEBUG
> > >
> > > At least, all the printk messages should have a log level specfied,
> > > and KERN_DEBUG seems appropriate.
> > >
> > > Signed-off-by: Michael Abbott <michael.abbott@diamond.ac.uk>
> > > ---
> > >  drivers/i2c/busses/i2c-pxa.c |    8 ++++----
> > >  1 files changed, 4 insertions(+), 4 deletions(-)
> > >
> > > diff --git a/drivers/i2c/busses/i2c-pxa.c b/drivers/i2c/busses/i2c-pxa.c
> > > index bdb1f75..8cb9160 100644
> > > --- a/drivers/i2c/busses/i2c-pxa.c
> > > +++ b/drivers/i2c/busses/i2c-pxa.c
> > > @@ -210,11 +210,11 @@ static irqreturn_t i2c_pxa_handler(int this_irq, void *dev_id);
> > >  static void i2c_pxa_scream_blue_murder(struct pxa_i2c *i2c, const char *why)
> > >  {
> > >     unsigned int i;
> > > -   printk("i2c: error: %s\n", why);
> > > -   printk("i2c: msg_num: %d msg_idx: %d msg_ptr: %d\n",
> > > +   printk(KERN_DEBUG "i2c: error: %s\n", why);
> > > +   printk(KERN_DEBUG "i2c: msg_num: %d msg_idx: %d msg_ptr: %d\n",
> > >             i2c->msg_num, i2c->msg_idx, i2c->msg_ptr);
> > > -   printk("i2c: ICR: %08x ISR: %08x\n"
> > > -          "i2c: log: ", readl(_ICR(i2c)), readl(_ISR(i2c)));
> > > +   printk(KERN_DEBUG "i2c: ICR: %08x ISR: %08x\n"
> > > +          KERN_DEBUG "i2c: log: ", readl(_ICR(i2c)), readl(_ISR(i2c)));
> > >     for (i = 0; i < i2c->irqlogidx; i++)
> > >             printk("[%08x:%08x] ", i2c->isrlog[i], i2c->icrlog[i]);
> > >     printk("\n");
> >
> > Already fixed in Linus' tree:
> > http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=154d22b04ae1741c5fcfd5d747b813a9a279abff#patch4
>
> Now this is very odd.  I'd assumed the error messages I was getting were
> just noise, but the commit you point me to uses KERN_ERR instead -- and so
> I get a brief storm of these messages on boot on my target:
>
> [   20.667923] i2c: error: exhausted retries
> [   20.671962] i2c: msg_num: 0 msg_idx: -2000 msg_ptr: 0
> [   20.677113] i2c: ICR: 000007e0 ISR: 00000002
> [   20.677128] i2c: log: [00000446:000007e0]
>
> This is repeated nine times in total with exactly the same numbers.  Any
> thoughts as to what this means?  Other than these messages on startup my
> I2C sensors seem to work perfectly well.
If all error messages seem like "exhausted retries",not "timeout or
master error",there maybe some operation conflicts,and i2c controller
cannot succeed operateing after max times retry.  And which mode do
your i2c work?  Poll or Interrupt?

thanks
--wxc200

> --
> To unsubscribe from this list: send the line "unsubscribe linux-i2c" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 2.6.29.2] PXA I2C: Define log level for i2c PXA error report

[Michael Abbott]

[-- Attachment #1: Type: TEXT/PLAIN, Size: 1526 bytes --]

On Mon, 11 May 2009, ÒÁÔó wrote:
> 2009/5/11 Michael Abbott <michael@araneidae.co.uk>
> > Now this is very odd.  I'd assumed the error messages I was getting were
> > just noise, but the commit you point me to uses KERN_ERR instead -- and so
> > I get a brief storm of these messages on boot on my target:
> >
> > [   20.667923] i2c: error: exhausted retries
> > [   20.671962] i2c: msg_num: 0 msg_idx: -2000 msg_ptr: 0
> > [   20.677113] i2c: ICR: 000007e0 ISR: 00000002
> > [   20.677128] i2c: log: [00000446:000007e0]
> >
> > This is repeated nine times in total with exactly the same numbers.  Any
> > thoughts as to what this means?  Other than these messages on startup my
> > I2C sensors seem to work perfectly well.
> If all error messages seem like "exhausted retries",not "timeout or
> master error",there maybe some operation conflicts,and i2c controller
> cannot succeed operateing after max times retry.  And which mode do
> your i2c work?  Poll or Interrupt?

Well, to be honest I don't actually know (I've inherited the driver and I 
haven't figured out how the I2C drivers work), but here's how the i2c is 
initialised:

static struct i2c_pxa_platform_data xcep_i2c_platform_data  = {
        .class = I2C_CLASS_HWMON
};

...
        pxa_set_i2c_info(&xcep_i2c_platform_data);

This is called in the .init_machine method, and the target system is a 
PXA255.

However, after the error messages on startup, all the sensors seem to be 
detected and seem to be operating normally.

Re: [PATCH 2.6.29.2] PXA I2C: Define log level for i2c PXA error report

[Mark Brown]

On Mon, May 11, 2009 at 11:31:51AM +0100, Michael Abbott wrote:

> static struct i2c_pxa_platform_data xcep_i2c_platform_data  = {
>         .class = I2C_CLASS_HWMON
> };

> ...
>         pxa_set_i2c_info(&xcep_i2c_platform_data);

> This is called in the .init_machine method, and the target system is a 
> PXA255.

> However, after the error messages on startup, all the sensors seem to be 
> detected and seem to be operating normally.

These errors occur when attempting to talk to non-existant devices which
is almost certainly going to happen with the class based device probing
since the kernel will probe for each device at each possible address on
each I2C bus.

Re: [PATCH 2.6.29.2] PXA I2C: Define log level for i2c PXA error report

[Michael Abbott]

On Mon, 11 May 2009, Mark Brown wrote:
> On Mon, May 11, 2009 at 11:31:51AM +0100, Michael Abbott wrote:
> 
> > static struct i2c_pxa_platform_data xcep_i2c_platform_data  = {
> >         .class = I2C_CLASS_HWMON
> > };
> 
> > ...
> >         pxa_set_i2c_info(&xcep_i2c_platform_data);
> 
> > This is called in the .init_machine method, and the target system is a 
> > PXA255.
> 
> > However, after the error messages on startup, all the sensors seem to be 
> > detected and seem to be operating normally.
> 
> These errors occur when attempting to talk to non-existant devices which
> is almost certainly going to happen with the class based device probing
> since the kernel will probe for each device at each possible address on
> each I2C bus.

That makes sense.

However, then, does it make sense for the PXA i2c bus prober to be calling 
the "scream blue murder" routine in this frankly routine case?  I guess it 
could just say "no device found at address xx"

Re: [PATCH 2.6.29.2] PXA I2C: Define log level for i2c PXA error report

[Mark Brown]

On Mon, May 11, 2009 at 01:16:38PM +0100, Michael Abbott wrote:

> However, then, does it make sense for the PXA i2c bus prober to be calling 
> the "scream blue murder" routine in this frankly routine case?  I guess it 
> could just say "no device found at address xx"

You should really convert your board to specify explicitly which chips
are on the board rather than using the old style probe routines.

Re: [PATCH 2.6.29.2] PXA I2C: Define log level for i2c PXA error report

[伊泽]

2009/5/11 Mark Brown <broonie@sirena.org.uk>:
> On Mon, May 11, 2009 at 11:31:51AM +0100, Michael Abbott wrote:
>
>> static struct i2c_pxa_platform_data xcep_i2c_platform_data  = {
>>         .class = I2C_CLASS_HWMON
>> };
>
>> ...
>>         pxa_set_i2c_info(&xcep_i2c_platform_data);
>
>> This is called in the .init_machine method, and the target system is a
>> PXA255.
>
>> However, after the error messages on startup, all the sensors seem to be
>> detected and seem to be operating normally.
>
> These errors occur when attempting to talk to non-existant devices which
> is almost certainly going to happen with the class based device probing
> since the kernel will probe for each device at each possible address on
> each I2C bus.
>
that happen when the device driver registered but the hardware doesnot
exist.i once met this
when my camera driver could not find the hardware camera.

thanks
--wxc200

Re: [PATCH 2.6.29.2] PXA I2C: Define log level for i2c PXA error report

[Michael Abbott]

[-- Attachment #1: Type: TEXT/PLAIN, Size: 1444 bytes --]

On Mon, 11 May 2009, ÒÁÔó wrote:
> 2009/5/11 Mark Brown <broonie@sirena.org.uk>:
> > On Mon, May 11, 2009 at 11:31:51AM +0100, Michael Abbott wrote:
> >
> >> static struct i2c_pxa_platform_data xcep_i2c_platform_data  = {
> >>         .class = I2C_CLASS_HWMON
> >> };
> >
> >> ...
> >>         pxa_set_i2c_info(&xcep_i2c_platform_data);
> >
> >> This is called in the .init_machine method, and the target system is a
> >> PXA255.
> >
> >> However, after the error messages on startup, all the sensors seem to be
> >> detected and seem to be operating normally.
> >
> > These errors occur when attempting to talk to non-existant devices 
> > which is almost certainly going to happen with the class based device 
> > probing since the kernel will probe for each device at each possible 
> > address on each I2C bus.
> that happen when the device driver registered but the hardware doesnot 
> exist.i once met this when my camera driver could not find the hardware 
> camera.

Fair enough.  I have a bit of a challenge to specify the precise list of 
I2C devices at the board level, as the board support is for a CPU module 
with the I2C bus off-board ... and clearly I can live with the error 
messages, anyway.

Still, I think it would make more sense for these to be one-liners; I 
guess routine probing failures ought to be reported in a slightly 
different way.  I'll see if I can figure out a sensible patch.

Re: [PATCH 2.6.29.2] PXA I2C: Define log level for i2c PXA error report

[Wolfram Sang]

[-- Attachment #1: Type: text/plain, Size: 596 bytes --]


> Still, I think it would make more sense for these to be one-liners; I 
> guess routine probing failures ought to be reported in a slightly 
> different way.  I'll see if I can figure out a sensible patch.

You might want to have a look here as I also noted once that the
pxa-driver is a bit too talkative when writing to eeproms:

http://www.mail-archive.com/linux-i2c@vger.kernel.org/msg00324.html

Regards,

   Wolfram

-- 
Pengutronix e.K.                           | Wolfram Sang                |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 197 bytes --]