Re: [PATCH 2/2] tracing: Fix kmemleak in set_trigger_filter

[Chunyu Hu]

----- Original Message -----
> From: "Steven Rostedt" <rostedt@xxxxxxxxxxx>
> To: "Chunyu Hu" <chuhu@xxxxxxxxxx>
> Cc: mingo@xxxxxxxxxx, linux-kernel@xxxxxxxxxxxxxxx
> Sent: Wednesday, August 23, 2017 12:52:49 PM
> Subject: Re: [PATCH 2/2] tracing: Fix kmemleak in set_trigger_filter
> 
> On Wed, 23 Aug 2017 10:41:55 -0400
> Steven Rostedt <rostedt@xxxxxxxxxxx> wrote:
> 
> >  * On success, returns 0 and *@filterp points to the new filter.  On
> >  * failure, returns -errno and *@filterp may point to %NULL or to a new
> >  * filter.  In the latter case, the returned filter contains error
> >  * information if @set_str is %true and the caller is responsible for
> >  * freeing it.
> > 
> > So filter contains an error string when it fails. It seems that we
> > should somehow propagate that up the chain to display. I'll look more
> > into this.
> 
> The bug is in create_filter(), because "set_str" is set to false, and
> the filter should not be passed back allocated on error.

Thanks for all the analysis. I think you are right. I'll try to have a test on it
in case we miss something. But please don't block on my test.

> 
> The correct fix is below.
> 
> Thanks!
> 
> -- Steve
> 
> 
> From 9975be0b2dccaea8ec3574d69a3504e11659a6ea Mon Sep 17 00:00:00 2001
> From: "Steven Rostedt (VMware)" <rostedt@xxxxxxxxxxx>
> Date: Wed, 23 Aug 2017 12:46:27 -0400
> Subject: [PATCH] tracing: Fix freeing of filter in create_filter() when
>  set_str is false
> 
> Performing the following task with kmemleak enabled:
> 
>  # cd /sys/kernel/tracing/events/irq/irq_handler_entry/
>  # echo 'enable_event:kmem:kmalloc:3 if irq >' > trigger
>  # echo 'enable_event:kmem:kmalloc:3 if irq > 31' > trigger
>  # echo scan > /sys/kernel/debug/kmemleak
>  # cat /sys/kernel/debug/kmemleak
> unreferenced object 0xffff8800b9290308 (size 32):
>   comm "bash", pid 1114, jiffies 4294848451 (age 141.139s)
>   hex dump (first 32 bytes):
>     00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
>     00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
>   backtrace:
>     [<ffffffff81cef5aa>] kmemleak_alloc+0x4a/0xa0
>     [<ffffffff81357938>] kmem_cache_alloc_trace+0x158/0x290
>     [<ffffffff81261c09>] create_filter_start.constprop.28+0x99/0x940
>     [<ffffffff812639c9>] create_filter+0xa9/0x160
>     [<ffffffff81263bdc>] create_event_filter+0xc/0x10
>     [<ffffffff812655e5>] set_trigger_filter+0xe5/0x210
>     [<ffffffff812660c4>] event_enable_trigger_func+0x324/0x490
>     [<ffffffff812652e2>] event_trigger_write+0x1a2/0x260
>     [<ffffffff8138cf87>] __vfs_write+0xd7/0x380
>     [<ffffffff8138f421>] vfs_write+0x101/0x260
>     [<ffffffff8139187b>] SyS_write+0xab/0x130
>     [<ffffffff81cfd501>] entry_SYSCALL_64_fastpath+0x1f/0xbe
>     [<ffffffffffffffff>] 0xffffffffffffffff
> 
> The function create_filter() is passed a 'filterp' pointer that gets
> allocated, and if "set_str" is true, it is up to the caller to free it, even
> on error. The problem is that the pointer is not freed by create_filter()
> when set_str is false. This is a bug, and it is not up to the caller to free
> the filter on error if it doesn't care about the string.
> 
> Link:
> http://lkml.kernel.org/r/1502705898-27571-2-git-send-email-chuhu@xxxxxxxxxx
> 
> Reported-by: Chunyu Hu <chuhu@xxxxxxxxxx>
> Signed-off-by: Steven Rostedt (VMware) <rostedt@xxxxxxxxxxx>
> ---
>  kernel/trace/trace_events_filter.c | 4 ++++
>  1 file changed, 4 insertions(+)
> 
> diff --git a/kernel/trace/trace_events_filter.c
> b/kernel/trace/trace_events_filter.c
> index 59a411f..181e139 100644
> --- a/kernel/trace/trace_events_filter.c
> +++ b/kernel/trace/trace_events_filter.c
> @@ -1959,6 +1959,10 @@ static int create_filter(struct trace_event_call
> *call,
>  		if (err && set_str)
>  			append_filter_err(ps, filter);
>  	}
> +	if (err && !set_str) {
> +		free_event_filter(filter);
> +		filter = NULL;
> +	}
>  	create_filter_finish(ps);
>  
>  	*filterp = filter;
> --
> 2.9.5
> 
> 

-- 
Regards,
Chunyu Hu