linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH][REPOST] Check maxlen on strnlen_user usage
@ 2011-09-07 23:54 Ryan Mallon
  2011-09-08  0:06 ` Andrew Morton
  0 siblings, 1 reply; 3+ messages in thread
From: Ryan Mallon @ 2011-09-07 23:54 UTC (permalink / raw)
  To: Mark Salter, Alexander Viro
  Cc: linux-fsdevel, linux-kernel@vger.kernel.org, Andrew Morton,
	trivial@kernel.org

strnlen_user returns the length of the string including the nul 
terminator. In the case where maxlen is reached strnlen_user returns 
maxlen + 1. Most callsites already check for this condition. Fix the 
call to strnlen_user in fs/exec.c to check for the maxlen case.

Signed-off-by: Ryan Mallon <rmallon@gmail.com>
---

diff --git a/fs/exec.c b/fs/exec.c
index 25dcbe5..e19588c 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -481,7 +481,7 @@ static int copy_strings(int argc, struct 
user_arg_ptr argv,
  			goto out;

  		len = strnlen_user(str, MAX_ARG_STRLEN);
-		if (!len)
+		if (!len || len>  MAX_ARG_STRLEN)
  			goto out;

  		ret = -E2BIG;

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

* Re: [PATCH][REPOST] Check maxlen on strnlen_user usage
  2011-09-07 23:54 [PATCH][REPOST] Check maxlen on strnlen_user usage Ryan Mallon
@ 2011-09-08  0:06 ` Andrew Morton
  2011-09-08  0:20   ` Ryan Mallon
  0 siblings, 1 reply; 3+ messages in thread
From: Andrew Morton @ 2011-09-08  0:06 UTC (permalink / raw)
  To: Ryan Mallon
  Cc: Mark Salter, Alexander Viro, linux-fsdevel,
	linux-kernel@vger.kernel.org, trivial@kernel.org

On Thu, 08 Sep 2011 09:54:18 +1000
Ryan Mallon <rmallon@gmail.com> wrote:

> strnlen_user returns the length of the string including the nul 
> terminator. In the case where maxlen is reached strnlen_user returns 
> maxlen + 1. Most callsites already check for this condition. Fix the 
> call to strnlen_user in fs/exec.c to check for the maxlen case.
> 
> diff --git a/fs/exec.c b/fs/exec.c
> index 25dcbe5..e19588c 100644
> --- a/fs/exec.c
> +++ b/fs/exec.c
> @@ -481,7 +481,7 @@ static int copy_strings(int argc, struct 
> user_arg_ptr argv,
>   			goto out;
> 
>   		len = strnlen_user(str, MAX_ARG_STRLEN);
> -		if (!len)
> +		if (!len || len>  MAX_ARG_STRLEN)
>   			goto out;
> 
>   		ret = -E2BIG;

The following call to vald_arg_len() already does this?

This change will cause copy_strings() to incorrectly return -EFAULT,
rather than -E2BIG.

Your email client is wordwrapping and space-stuffing the patches, btw.

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

* Re: [PATCH][REPOST] Check maxlen on strnlen_user usage
  2011-09-08  0:06 ` Andrew Morton
@ 2011-09-08  0:20   ` Ryan Mallon
  0 siblings, 0 replies; 3+ messages in thread
From: Ryan Mallon @ 2011-09-08  0:20 UTC (permalink / raw)
  To: Andrew Morton
  Cc: Mark Salter, Alexander Viro, linux-fsdevel,
	linux-kernel@vger.kernel.org, trivial@kernel.org

On 08/09/11 10:06, Andrew Morton wrote:
> On Thu, 08 Sep 2011 09:54:18 +1000
> Ryan Mallon<rmallon@gmail.com>  wrote:
>
>> strnlen_user returns the length of the string including the nul
>> terminator. In the case where maxlen is reached strnlen_user returns
>> maxlen + 1. Most callsites already check for this condition. Fix the
>> call to strnlen_user in fs/exec.c to check for the maxlen case.
>>
>> diff --git a/fs/exec.c b/fs/exec.c
>> index 25dcbe5..e19588c 100644
>> --- a/fs/exec.c
>> +++ b/fs/exec.c
>> @@ -481,7 +481,7 @@ static int copy_strings(int argc, struct
>> user_arg_ptr argv,
>>    			goto out;
>>
>>    		len = strnlen_user(str, MAX_ARG_STRLEN);
>> -		if (!len)
>> +		if (!len || len>   MAX_ARG_STRLEN)
>>    			goto out;
>>
>>    		ret = -E2BIG;
> The following call to vald_arg_len() already does this?
>
> This change will cause copy_strings() to incorrectly return -EFAULT,
> rather than -E2BIG.
Hmm, missed that. How about the patch below instead which passes the 
correct length to strnlen_user?
> Your email client is wordwrapping and space-stuffing the patches, btw.
I'll try again. Hopefully it works this time.

---
Replace valid_arg_len function in fs/exec.c with max_arg_len function 
and pass the correct length to strnlen_user.

Signed-off-by: Ryan Mallon <rmallon@gmail.com>
---

diff --git a/fs/exec.c b/fs/exec.c
index e19588c..12d2f7f 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -296,9 +296,9 @@ err:
  	return err;
  }

-static bool valid_arg_len(struct linux_binprm *bprm, long len)
+static long max_arg_len(struct linux_binprm *bprm)
  {
-	return len<= MAX_ARG_STRLEN;
+	return MAX_ARG_STRLEN;
  }

  #else
@@ -354,9 +354,9 @@ static int __bprm_mm_init(struct linux_binprm *bprm)
  	return 0;
  }

-static bool valid_arg_len(struct linux_binprm *bprm, long len)
+static long max_arg_len(struct linux_binprm *bprm)
  {
-	return len<= bprm->p;
+	return bprm->p;
  }

  #endif /* CONFIG_MMU */
@@ -474,18 +474,19 @@ static int copy_strings(int argc, struct user_arg_ptr argv,
  		const char __user *str;
  		int len;
  		unsigned long pos;
+		long max_len = max_arg_len(bprm);

  		ret = -EFAULT;
  		str = get_user_arg_ptr(argv, argc);
  		if (IS_ERR(str))
  			goto out;

-		len = strnlen_user(str, MAX_ARG_STRLEN);
-		if (!len || len>  MAX_ARG_STRLEN)
+		len = strnlen_user(str, max_len);
+		if (!len)
  			goto out;

  		ret = -E2BIG;
-		if (!valid_arg_len(bprm, len))
+		if (len>  max_len)
  			goto out;

  		/* We're going to work our way backwords. */



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

end of thread, other threads:[~2011-09-08  0:20 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2011-09-07 23:54 [PATCH][REPOST] Check maxlen on strnlen_user usage Ryan Mallon
2011-09-08  0:06 ` Andrew Morton
2011-09-08  0:20   ` Ryan Mallon

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).