[PATCH 1/1] kernel/workqueue.c: pr_warning/pr_warn & printk/pr_info

[PATCH 1/1] kernel/workqueue.c: pr_warning/pr_warn & printk/pr_info

[Fabian Frederick]

Cc: Tejun Heo <tj@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Fabian Frederick <fabf@skynet.be>
---
 kernel/workqueue.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index 0ee63af..6824afc 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -4098,8 +4098,8 @@ static void wq_update_unbound_numa(struct workqueue_struct *wq, int cpu,
 	/* create a new pwq */
 	pwq = alloc_unbound_pwq(wq, target_attrs);
 	if (!pwq) {
-		pr_warning("workqueue: allocation failed while updating NUMA affinity of \"%s\"\n",
-			   wq->name);
+		pr_warn("workqueue: allocation failed while updating NUMA affinity of \"%s\"\n",
+			wq->name);
 		goto out_unlock;
 	}
 
@@ -4548,7 +4548,7 @@ void print_worker_info(const char *log_lvl, struct task_struct *task)
 		probe_kernel_read(desc, worker->desc, sizeof(desc) - 1);
 
 	if (fn || name[0] || desc[0]) {
-		printk("%sWorkqueue: %s %pf", log_lvl, name, fn);
+		pr_info("%sWorkqueue: %s %pf", log_lvl, name, fn);
 		if (desc[0])
 			pr_cont(" (%s)", desc);
 		pr_cont("\n");
-- 
1.9.1

Re: [PATCH 1/1] kernel/workqueue.c: pr_warning/pr_warn & printk/pr_info

[Tejun Heo]

On Mon, May 12, 2014 at 07:56:23PM +0200, Fabian Frederick wrote:
> 
> Cc: Tejun Heo <tj@kernel.org>
> Cc: Andrew Morton <akpm@linux-foundation.org>
> Signed-off-by: Fabian Frederick <fabf@skynet.be>

Applied to wq/for-3.16.  Patch didn't apply and refreshed.  Please
consider basing further patches on top of subsystem devel branches.

Thanks.

-- 
tejun

Re: [PATCH 1/1] kernel/workqueue.c: pr_warning/pr_warn & printk/pr_info

[Joe Perches]

On Mon, 2014-05-12 at 19:56 +0200, Fabian Frederick wrote:
> Cc: Tejun Heo <tj@kernel.org>
> Cc: Andrew Morton <akpm@linux-foundation.org>
> Signed-off-by: Fabian Frederick <fabf@skynet.be>
> ---
>  kernel/workqueue.c | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/kernel/workqueue.c b/kernel/workqueue.c
> index 0ee63af..6824afc 100644
> --- a/kernel/workqueue.c
> +++ b/kernel/workqueue.c
> @@ -4098,8 +4098,8 @@ static void wq_update_unbound_numa(struct workqueue_struct *wq, int cpu,
>  	/* create a new pwq */
>  	pwq = alloc_unbound_pwq(wq, target_attrs);
>  	if (!pwq) {
> -		pr_warning("workqueue: allocation failed while updating NUMA affinity of \"%s\"\n",
> -			   wq->name);
> +		pr_warn("workqueue: allocation failed while updating NUMA affinity of \"%s\"\n",
> +			wq->name);
>  		goto out_unlock;
>  	}
>  
> @@ -4548,7 +4548,7 @@ void print_worker_info(const char *log_lvl, struct task_struct *task)
>  		probe_kernel_read(desc, worker->desc, sizeof(desc) - 1);
>  
>  	if (fn || name[0] || desc[0]) {
> -		printk("%sWorkqueue: %s %pf", log_lvl, name, fn);
> +		pr_info("%sWorkqueue: %s %pf", log_lvl, name, fn);

Not correct.

log_lvl already emits the appropriate level
and this overrides it with KERN_INFO.

The output will now be incorrect emitting
at KERN_INFO followed by an SOH and the
desired log level.

>  		if (desc[0])
>  			pr_cont(" (%s)", desc);
>  		pr_cont("\n");

Trivially, it likely would be a bit better to avoid
any possible output message interleaving by using:

	if (fn || name[0] || desc[0]) {
		if (desc[0])
			printk("%sWorkqueue: %s %pf (%s)\n",
			       log_lvl, name, fn, desc);
		else
			printk("%sWorkqueue: %s %pf\n",
			       log_lvl, name, fn);
	}

Re: [PATCH 1/1] kernel/workqueue.c: pr_warning/pr_warn & printk/pr_info

[Fabian Frederick]

On Mon, 12 May 2014 11:21:28 -0700
Joe Perches <joe@perches.com> wrote:

> On Mon, 2014-05-12 at 19:56 +0200, Fabian Frederick wrote:
> > Cc: Tejun Heo <tj@kernel.org>
> > Cc: Andrew Morton <akpm@linux-foundation.org>
> > Signed-off-by: Fabian Frederick <fabf@skynet.be>
> > ---
> >  kernel/workqueue.c | 6 +++---
> >  1 file changed, 3 insertions(+), 3 deletions(-)
> > 
> > diff --git a/kernel/workqueue.c b/kernel/workqueue.c
> > index 0ee63af..6824afc 100644
> > --- a/kernel/workqueue.c
> > +++ b/kernel/workqueue.c
> > @@ -4098,8 +4098,8 @@ static void wq_update_unbound_numa(struct workqueue_struct *wq, int cpu,
> >  	/* create a new pwq */
> >  	pwq = alloc_unbound_pwq(wq, target_attrs);
> >  	if (!pwq) {
> > -		pr_warning("workqueue: allocation failed while updating NUMA affinity of \"%s\"\n",
> > -			   wq->name);
> > +		pr_warn("workqueue: allocation failed while updating NUMA affinity of \"%s\"\n",
> > +			wq->name);
> >  		goto out_unlock;
> >  	}
> >  
> > @@ -4548,7 +4548,7 @@ void print_worker_info(const char *log_lvl, struct task_struct *task)
> >  		probe_kernel_read(desc, worker->desc, sizeof(desc) - 1);
> >  
> >  	if (fn || name[0] || desc[0]) {
> > -		printk("%sWorkqueue: %s %pf", log_lvl, name, fn);
> > +		pr_info("%sWorkqueue: %s %pf", log_lvl, name, fn);
> 
> Not correct.
> 
> log_lvl already emits the appropriate level
> and this overrides it with KERN_INFO.
> 
> The output will now be incorrect emitting
> at KERN_INFO followed by an SOH and the
> desired log level.
Such printk can't be converted to pr_ ?

> 
> >  		if (desc[0])
> >  			pr_cont(" (%s)", desc);
> >  		pr_cont("\n");
> 
> Trivially, it likely would be a bit better to avoid
> any possible output message interleaving by using:
> 
> 	if (fn || name[0] || desc[0]) {
> 		if (desc[0])
> 			printk("%sWorkqueue: %s %pf (%s)\n",
> 			       log_lvl, name, fn, desc);
> 		else
> 			printk("%sWorkqueue: %s %pf\n",
> 			       log_lvl, name, fn);
> 	}
> 
> 
> 


-- 
Fabian Frederick <fabf@skynet.be>

Re: [PATCH 1/1] kernel/workqueue.c: pr_warning/pr_warn & printk/pr_info

[Joe Perches]

On Mon, 2014-05-12 at 20:26 +0200, Fabian Frederick wrote:
> On Mon, 12 May 2014 11:21:28 -0700 Joe Perches <joe@perches.com> wrote:
> > On Mon, 2014-05-12 at 19:56 +0200, Fabian Frederick wrote:
[]
> > > diff --git a/kernel/workqueue.c b/kernel/workqueue.c
[]
> > > @@ -4548,7 +4548,7 @@ void print_worker_info(const char *log_lvl, struct task_struct *task)
> > >  		probe_kernel_read(desc, worker->desc, sizeof(desc) - 1);
> > >  
> > >  	if (fn || name[0] || desc[0]) {
> > > -		printk("%sWorkqueue: %s %pf", log_lvl, name, fn);
> > > +		pr_info("%sWorkqueue: %s %pf", log_lvl, name, fn);
> > 
> > Not correct.
> > 
> > log_lvl already emits the appropriate level
> > and this overrides it with KERN_INFO.
> > 
> > The output will now be incorrect emitting
> > at KERN_INFO followed by an SOH and the
> > desired log level.
>
> Such printk can't be converted to pr_ ?

Correct.