On 29-07-08 22:52, Pete Clements wrote:

> [Previously sent with wrong subject line (2.6.26-rc1...) applicable
> to post 2.6.26 series]
>
> Missed most of the 26-git series, git6 first one tried and oops
> showed in that one as well as those tried after. oops below.

http://marc.info/?l=linux-kernel&m=121736480005656&w=2

For now just adding a CC as I'm off to bed. Note -- snd-cs4236 has been
one of the most tested drivers with the new PNP code.

Rene.


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

On Tuesday 29 July 2008 6:02:46 pm Rene Herman wrote:
> On 29-07-08 22:52, Pete Clements wrote:
> > [Previously sent with wrong subject line (2.6.26-rc1...) applicable
> > to post 2.6.26 series]
> >
> > Missed most of the 26-git series, git6 first one tried and oops
> > showed in that one as well as those tried after. oops below.
>
> http://marc.info/?l=linux-kernel&m=121736480005656&w=2
>
> For now just adding a CC as I'm off to bed. Note -- snd-cs4236 has been
> one of the most tested drivers with the new PNP code.

Thanks very much for the report. Can I trouble you to test the
patch below and see whether it fixes the problem?



PNP: fix formatting of dbg_pnp_show_resources() output

From: Bjorn Helgaas

Each resource should be printed on its own line, so start snprintf'ing
at the beginning of the buffer every time through the loop.

Signed-off-by: Bjorn Helgaas
---

drivers/pnp/support.c | 3 ++-
1 files changed, 2 insertions(+), 1 deletions(-)


diff --git a/drivers/pnp/support.c b/drivers/pnp/support.c
index bbf78ef..57b775b 100644
--- a/drivers/pnp/support.c
+++ b/drivers/pnp/support.c
@@ -77,7 +77,7 @@ void dbg_pnp_show_resources(struct pnp_dev *dev, char *desc)
{
#ifdef DEBUG
char buf[128];
- int len = 0;
+ int len;
struct pnp_resource *pnp_res;
struct resource *res;

@@ -89,6 +89,7 @@ void dbg_pnp_show_resources(struct pnp_dev *dev, char *desc)
dev_dbg(&dev->dev, "%s: current resources:\n", desc);
list_for_each_entry(pnp_res, &dev->resources, list) {
res = &pnp_res->res;
+ len = 0;

len += snprintf(buf + len, sizeof(buf) - len, " %-3s ",
pnp_resource_type_name(res));
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

Quoting Bjorn Helgaas
> On Tuesday 29 July 2008 6:02:46 pm Rene Herman wrote:
> > On 29-07-08 22:52, Pete Clements wrote:
> > > [Previously sent with wrong subject line (2.6.26-rc1...) applicable
> > > to post 2.6.26 series]
> > >
> > > Missed most of the 26-git series, git6 first one tried and oops
> > > showed in that one as well as those tried after. oops below.
> >
>
> Thanks very much for the report. Can I trouble you to test the
> patch below and see whether it fixes the problem?
>

Patch clears the problem. Thanks.
--
Pete Clements

>
> PNP: fix formatting of dbg_pnp_show_resources() output
>
> From: Bjorn Helgaas
>
> Each resource should be printed on its own line, so start snprintf'ing
> at the beginning of the buffer every time through the loop.
>
> Signed-off-by: Bjorn Helgaas
> ---
>
> drivers/pnp/support.c | 3 ++-
> 1 files changed, 2 insertions(+), 1 deletions(-)
>
>
> diff --git a/drivers/pnp/support.c b/drivers/pnp/support.c
> index bbf78ef..57b775b 100644
> --- a/drivers/pnp/support.c
> +++ b/drivers/pnp/support.c
> @@ -77,7 +77,7 @@ void dbg_pnp_show_resources(struct pnp_dev *dev, char *desc)
> {
> #ifdef DEBUG
> char buf[128];
> - int len = 0;
> + int len;
> struct pnp_resource *pnp_res;
> struct resource *res;
>
> @@ -89,6 +89,7 @@ void dbg_pnp_show_resources(struct pnp_dev *dev, char *desc)
> dev_dbg(&dev->dev, "%s: current resources:\n", desc);
> list_for_each_entry(pnp_res, &dev->resources, list) {
> res = &pnp_res->res;
> + len = 0;
>
> len += snprintf(buf + len, sizeof(buf) - len, " %-3s ",
> pnp_resource_type_name(res));
>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

On 30-07-08 05:59, Bjorn Helgaas wrote:

> PNP: fix formatting of dbg_pnp_show_resources() output
>
> From: Bjorn Helgaas
>
> Each resource should be printed on its own line, so start snprintf'ing
> at the beginning of the buffer every time through the loop.
>
> Signed-off-by: Bjorn Helgaas

Good WARN_ON() that...

I was looking at why I hadn't seen this here even with the same hardware
when I _did_ see it here. Most definitely did not back when I was paying
attention -- although there seems to be little opportunity for that.

Just for my own education; did something recent happen to bring this on
or must this have been present always? Serious buglet it seems.

Rene
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

On Wednesday 30 July 2008 6:15:07 am Rene Herman wrote:
> On 30-07-08 05:59, Bjorn Helgaas wrote:
> > PNP: fix formatting of dbg_pnp_show_resources() output
> >
> > From: Bjorn Helgaas
> >
> > Each resource should be printed on its own line, so start snprintf'ing
> > at the beginning of the buffer every time through the loop.
> >
> > Signed-off-by: Bjorn Helgaas
>
> Good WARN_ON() that...
>
> I was looking at why I hadn't seen this here even with the same hardware
> when I _did_ see it here. Most definitely did not back when I was paying
> attention -- although there seems to be little opportunity for that.
>
> Just for my own education; did something recent happen to bring this on
> or must this have been present always? Serious buglet it seems.

That function used to printk the resource piece-meal -- first the
resource type, then the value, then the flags, etc. I made a late
change to snprintf everything into the buffer, then do a single
printk to keep everything atomic, and I didn't test it well enough.

Bjorn
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/