[linux-lvm] [PATCH] dmsetup: fix sscanf return check

[linux-lvm] [PATCH] dmsetup: fix sscanf return check

[Matthew Booth]

---
 tools/dmsetup.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/tools/dmsetup.c b/tools/dmsetup.c
index 0ac970f..65d17f8 100644
--- a/tools/dmsetup.c
+++ b/tools/dmsetup.c
@@ -228,8 +228,7 @@ static int _parse_line(struct dm_task *dmt, char *buffer, const char *file,
 	if (!*ptr || *ptr == '#')
 		return 1;
 
-	if (sscanf(ptr, "%llu %llu %s %n",
-		   &start, &size, ttype, &n) < 3) {
+	if (sscanf(ptr, "%llu %llu %s %n", &start, &size, ttype, &n) != 4) {
 		err("Invalid format on line %d of table %s", line, file);
 		return 0;
 	}
-- 
1.7.11.4

Re: [linux-lvm] [PATCH] dmsetup: fix sscanf return check

[Mark van Dijk]

On Thu, 23 Aug 2012 14:31:02 +0100
Matthew Booth <mbooth@redhat.com> wrote:

> ---
>  tools/dmsetup.c | 3 +--
>  1 file changed, 1 insertion(+), 2 deletions(-)
> 
> diff --git a/tools/dmsetup.c b/tools/dmsetup.c
> index 0ac970f..65d17f8 100644
> --- a/tools/dmsetup.c
> +++ b/tools/dmsetup.c
> @@ -228,8 +228,7 @@ static int _parse_line(struct dm_task *dmt, char
> *buffer, const char *file, if (!*ptr || *ptr == '#')
>  		return 1;
>  
> -	if (sscanf(ptr, "%llu %llu %s %n",
> -		   &start, &size, ttype, &n) < 3) {
> +	if (sscanf(ptr, "%llu %llu %s %n", &start, &size, ttype,
> &n) != 4) { err("Invalid format on line %d of table %s", line, file);
>  		return 0;
>  	}

Ah. For non-coders like myself the purpose of this patch seems to be
clouded in mystery...

-- 
Stay in touch,
Mark van Dijk.                  ,------------------------------------
-------------------------------'            Thu Aug 23 20:00 UTC 2012
Today is Setting Orange, the 16th day of Bureaucracy in the YOLD 3178

Re: [linux-lvm] [PATCH] dmsetup: fix sscanf return check

[Alasdair G Kergon]

On Thu, Aug 23, 2012 at 02:31:02PM +0100, Matthew Booth wrote:
> -	if (sscanf(ptr, "%llu %llu %s %n",
> -		   &start, &size, ttype, &n) < 3) {
> +	if (sscanf(ptr, "%llu %llu %s %n", &start, &size, ttype, &n) != 4) {

Did you test this?

According to the sscanf man page:

       n      Nothing is expected; instead, the number of characters  consumed
              thus  far  from  the  input  is stored through the next pointer,
              which must be a pointer to  int.   This  is  not  a  conversion,
              although  it can be suppressed with the * assignment-suppression
              character.  The C standard says: "Execution of  a  %n  directive
              does  not increment the assignment count returned at the comple-
              tion of execution" but the Corrigendum seems to contradict this.
              Probably it is wise not to make any assumptions on the effect of
              %n conversions on the return value.

Alasdair

Re: [linux-lvm] [PATCH] dmsetup: fix sscanf return check

[Matthew Booth]

On 23/08/12 21:27, Alasdair G Kergon wrote:
> On Thu, Aug 23, 2012 at 02:31:02PM +0100, Matthew Booth wrote:
>> -	if (sscanf(ptr, "%llu %llu %s %n",
>> -		   &start, &size, ttype, &n) < 3) {
>> +	if (sscanf(ptr, "%llu %llu %s %n", &start, &size, ttype, &n) != 4) {
>
> Did you test this?

No. My git-fu isn't good enough to point that out in the email without 
cluttering the commit message with it, but I pointed it out on IRC.

> According to the sscanf man page:
>
>         n      Nothing is expected; instead, the number of characters  consumed
>                thus  far  from  the  input  is stored through the next pointer,
>                which must be a pointer to  int.   This  is  not  a  conversion,
>                although  it can be suppressed with the * assignment-suppression
>                character.  The C standard says: "Execution of  a  %n  directive
>                does  not increment the assignment count returned at the comple-
>                tion of execution" but the Corrigendum seems to contradict this.
>                Probably it is wise not to make any assumptions on the effect of
>                %n conversions on the return value.

Thanks for being thorough. I've now made another entry in my mental list 
of surprising interfaces :)

Matt
-- 
Matthew Booth, RHCA, RHCSS
Red Hat Engineering, Virtualisation Team

GPG ID:  D33C3490
GPG FPR: 3733 612D 2D05 5458 8A8A 1600 3441 EA19 D33C 3490