linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH] firmware: ti_sci: Fix an off-by-one in ti_sci_debugfs_create()
@ 2023-10-29  7:52 Christophe JAILLET
  2023-10-30  8:00 ` Dan Carpenter
  0 siblings, 1 reply; 3+ messages in thread
From: Christophe JAILLET @ 2023-10-29  7:52 UTC (permalink / raw)
  To: Nishanth Menon, Tero Kristo, Santosh Shilimkar
  Cc: linux-kernel, kernel-janitors, Christophe JAILLET, Tero Kristo,
	linux-arm-kernel

The ending NULL is not taken into account by strncat(), so switch to
strlcat() to correctly compute the size of the available memory when
building 'debug_name'.

Because of the difference in the return type between strncat() and
strlcat(), some code shuffling is needed.

Fixes: aa276781a64a ("firmware: Add basic support for TI System Control Interface (TI-SCI) protocol")
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
---
 drivers/firmware/ti_sci.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/drivers/firmware/ti_sci.c b/drivers/firmware/ti_sci.c
index 7041befc756a..1036c6b0f3eb 100644
--- a/drivers/firmware/ti_sci.c
+++ b/drivers/firmware/ti_sci.c
@@ -181,10 +181,9 @@ static int ti_sci_debugfs_create(struct platform_device *pdev,
 	/* Setup NULL termination */
 	info->debug_buffer[info->debug_region_size] = 0;
 
-	info->d = debugfs_create_file(strncat(debug_name, dev_name(dev),
-					      sizeof(debug_name) -
-					      sizeof("ti_sci_debug@")),
-				      0444, NULL, info, &ti_sci_debug_fops);
+	strlcat(debug_name, dev_name(dev), sizeof(debug_name));
+	info->d = debugfs_create_file(debug_name, 0444, NULL, info,
+				      &ti_sci_debug_fops);
 	if (IS_ERR(info->d))
 		return PTR_ERR(info->d);
 
-- 
2.34.1


^ permalink raw reply related	[flat|nested] 3+ messages in thread

* Re: [PATCH] firmware: ti_sci: Fix an off-by-one in ti_sci_debugfs_create()
  2023-10-29  7:52 [PATCH] firmware: ti_sci: Fix an off-by-one in ti_sci_debugfs_create() Christophe JAILLET
@ 2023-10-30  8:00 ` Dan Carpenter
  2023-10-30  8:02   ` Dan Carpenter
  0 siblings, 1 reply; 3+ messages in thread
From: Dan Carpenter @ 2023-10-30  8:00 UTC (permalink / raw)
  To: Christophe JAILLET
  Cc: Nishanth Menon, Tero Kristo, Santosh Shilimkar, linux-kernel,
	kernel-janitors, Tero Kristo, linux-arm-kernel

On Sun, Oct 29, 2023 at 08:52:36AM +0100, Christophe JAILLET wrote:
> The ending NULL is not taken into account by strncat(), so switch to
> strlcat() to correctly compute the size of the available memory when
> building 'debug_name'.
> 
> Because of the difference in the return type between strncat() and
> strlcat(), some code shuffling is needed.
> 
> Fixes: aa276781a64a ("firmware: Add basic support for TI System Control Interface (TI-SCI) protocol")

The fixes tag isn't really required.

> Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
> ---
>  drivers/firmware/ti_sci.c | 7 +++----
>  1 file changed, 3 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/firmware/ti_sci.c b/drivers/firmware/ti_sci.c
> index 7041befc756a..1036c6b0f3eb 100644
> --- a/drivers/firmware/ti_sci.c
> +++ b/drivers/firmware/ti_sci.c
> @@ -181,10 +181,9 @@ static int ti_sci_debugfs_create(struct platform_device *pdev,
>  	/* Setup NULL termination */
>  	info->debug_buffer[info->debug_region_size] = 0;
>  
> -	info->d = debugfs_create_file(strncat(debug_name, dev_name(dev),
> -					      sizeof(debug_name) -
> -					      sizeof("ti_sci_debug@")),
> -				      0444, NULL, info, &ti_sci_debug_fops);
> +	strlcat(debug_name, dev_name(dev), sizeof(debug_name));

Eventually we are going to want to get rid for the strlcat() as well,
right?  Everyone hates strl* functions because they do a strlen() on
the src string.  It's fine if the src string is a string constant, I
suppose.  Here we basically trust dev_name() to be reasonable so either
strncat and strlcpy() are fine I guess...

But it's is more readable if the we use snprintf().

        snprintf(debug_name, sizeof(debug_name),
                 "ti_sci_debug@%s", dev_name(dev));

regards,
dan carpenter

> +	info->d = debugfs_create_file(debug_name, 0444, NULL, info,
> +				      &ti_sci_debug_fops);
>  	if (IS_ERR(info->d))
>  		return PTR_ERR(info->d);
>  
> -- 
> 2.34.1
> 

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [PATCH] firmware: ti_sci: Fix an off-by-one in ti_sci_debugfs_create()
  2023-10-30  8:00 ` Dan Carpenter
@ 2023-10-30  8:02   ` Dan Carpenter
  0 siblings, 0 replies; 3+ messages in thread
From: Dan Carpenter @ 2023-10-30  8:02 UTC (permalink / raw)
  To: Christophe JAILLET
  Cc: Nishanth Menon, Tero Kristo, Santosh Shilimkar, linux-kernel,
	kernel-janitors, Tero Kristo, linux-arm-kernel

On Mon, Oct 30, 2023 at 11:00:59AM +0300, Dan Carpenter wrote:
> On Sun, Oct 29, 2023 at 08:52:36AM +0100, Christophe JAILLET wrote:
> > The ending NULL is not taken into account by strncat(), so switch to
> > strlcat() to correctly compute the size of the available memory when
> > building 'debug_name'.
> > 
> > Because of the difference in the return type between strncat() and
> > strlcat(), some code shuffling is needed.
> > 
> > Fixes: aa276781a64a ("firmware: Add basic support for TI System Control Interface (TI-SCI) protocol")
> 
> The fixes tag isn't really required.

Ah sorry.  It's an off by one.  Yeah, that's a bug even if it doesn't
affect runtime.  I don't know why I didn't read this commit message
better...

regards,
dan carpenter


^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2023-10-30  8:02 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2023-10-29  7:52 [PATCH] firmware: ti_sci: Fix an off-by-one in ti_sci_debugfs_create() Christophe JAILLET
2023-10-30  8:00 ` Dan Carpenter
2023-10-30  8:02   ` Dan Carpenter

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).