Re: [2.6.31 PATCH] tracing/filters: always free pred on filter_add_subsystem_pred() failure

[Tom Zanussi <tzanussi@gmail.com>]

On Sat, 2009-08-08 at 18:02 +0200, Ingo Molnar wrote:
> * Tom Zanussi <tzanussi@gmail.com> wrote:
> 
> > If filter_add_subsystem_pred() fails due to ENOSPC or ENOMEM, the 
> > pred doesn't get freed, while as a side effect it does for other 
> > errors. Make it so the caller always frees the pred for any error.
> 
> thanks, i've queued this up for .31.
> 
> Just to make sure: tracing/core has the filter code changed 
> substantially, and this issue seems to be moot there - correct?
> 

Right, the subsystem preds are gone in tip, so nothing to fix there.

> So the total impact from both patches on -tip is the patch below - 
> right?
> 

Right.

Tom

> 	Ingo
> 
> diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c
> index 27c2dbe..490337a 100644
> --- a/kernel/trace/trace_events_filter.c
> +++ b/kernel/trace/trace_events_filter.c
> @@ -1050,6 +1050,8 @@ static int replace_preds(struct event_subsystem *system,
>  
>  		pred = create_pred(elt->op, operand1, operand2);
>  add_pred:
> +		if (!pred)
> +			return -ENOMEM;
>  		if (call)
>  			err = filter_add_pred(ps, call, pred, false);
>  		else