[PATCH] selftests/mm: compaction_test: Fix off by one in check_compaction()

[PATCH] selftests/mm: compaction_test: Fix off by one in check_compaction()

[Dan Carpenter]

The "initial_nr_hugepages" variable is unsigned long so it takes up to
20 characters to print, plus 1 more character for the NUL terminator.
Unfortunately, this buffer is not quite large enough for the terminator
to fit.  Also use snprintf() for a belt and suspenders approach.

Fixes: fb9293b6b015 ("selftests/mm: compaction_test: fix bogus test success and reduce probability of OOM-killer invocation")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
---
 tools/testing/selftests/mm/compaction_test.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/mm/compaction_test.c b/tools/testing/selftests/mm/compaction_test.c
index e140558e6f53..2c3a0eb6b22d 100644
--- a/tools/testing/selftests/mm/compaction_test.c
+++ b/tools/testing/selftests/mm/compaction_test.c
@@ -89,9 +89,10 @@ int check_compaction(unsigned long mem_free, unsigned long hugepage_size,
 	int fd, ret = -1;
 	int compaction_index = 0;
 	char nr_hugepages[20] = {0};
-	char init_nr_hugepages[20] = {0};
+	char init_nr_hugepages[24] = {0};
 
-	sprintf(init_nr_hugepages, "%lu", initial_nr_hugepages);
+	snprintf(init_nr_hugepages, sizeof(init_nr_hugepages),
+		 "%lu", initial_nr_hugepages);
 
 	/* We want to test with 80% of available memory. Else, OOM killer comes
 	   in to play */
-- 
2.43.0

Re: [PATCH] selftests/mm: compaction_test: Fix off by one in check_compaction()

[Shuah Khan]

On 8/9/24 06:32, Dan Carpenter wrote:
> The "initial_nr_hugepages" variable is unsigned long so it takes up to
> 20 characters to print, plus 1 more character for the NUL terminator.
> Unfortunately, this buffer is not quite large enough for the terminator
> to fit.  Also use snprintf() for a belt and suspenders approach.
> 
> Fixes: fb9293b6b015 ("selftests/mm: compaction_test: fix bogus test success and reduce probability of OOM-killer invocation")
> Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
> ---
>   tools/testing/selftests/mm/compaction_test.c | 5 +++--
>   1 file changed, 3 insertions(+), 2 deletions(-)
> 
> diff --git a/tools/testing/selftests/mm/compaction_test.c b/tools/testing/selftests/mm/compaction_test.c
> index e140558e6f53..2c3a0eb6b22d 100644
> --- a/tools/testing/selftests/mm/compaction_test.c
> +++ b/tools/testing/selftests/mm/compaction_test.c
> @@ -89,9 +89,10 @@ int check_compaction(unsigned long mem_free, unsigned long hugepage_size,
>   	int fd, ret = -1;
>   	int compaction_index = 0;
>   	char nr_hugepages[20] = {0};
> -	char init_nr_hugepages[20] = {0};
> +	char init_nr_hugepages[24] = {0};

Can we exceed this limit too? Can you make this a define?

>   
> -	sprintf(init_nr_hugepages, "%lu", initial_nr_hugepages);
> +	snprintf(init_nr_hugepages, sizeof(init_nr_hugepages),
> +		 "%lu", initial_nr_hugepages);
>   
>   	/* We want to test with 80% of available memory. Else, OOM killer comes
>   	   in to play */

With that change:

Reviewed-by: Shuah Khan <skhan@linuxfoundation.org>

thanks,
-- Shuah

Re: [PATCH] selftests/mm: compaction_test: Fix off by one in check_compaction()

[Dan Carpenter]

On Fri, Aug 09, 2024 at 11:20:48AM -0600, Shuah Khan wrote:
> On 8/9/24 06:32, Dan Carpenter wrote:
> > The "initial_nr_hugepages" variable is unsigned long so it takes up to
> > 20 characters to print, plus 1 more character for the NUL terminator.
> > Unfortunately, this buffer is not quite large enough for the terminator
> > to fit.  Also use snprintf() for a belt and suspenders approach.
> > 
> > Fixes: fb9293b6b015 ("selftests/mm: compaction_test: fix bogus test success and reduce probability of OOM-killer invocation")
> > Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
> > ---
> >   tools/testing/selftests/mm/compaction_test.c | 5 +++--
> >   1 file changed, 3 insertions(+), 2 deletions(-)
> > 
> > diff --git a/tools/testing/selftests/mm/compaction_test.c b/tools/testing/selftests/mm/compaction_test.c
> > index e140558e6f53..2c3a0eb6b22d 100644
> > --- a/tools/testing/selftests/mm/compaction_test.c
> > +++ b/tools/testing/selftests/mm/compaction_test.c
> > @@ -89,9 +89,10 @@ int check_compaction(unsigned long mem_free, unsigned long hugepage_size,
> >   	int fd, ret = -1;
> >   	int compaction_index = 0;
> >   	char nr_hugepages[20] = {0};
> > -	char init_nr_hugepages[20] = {0};
> > +	char init_nr_hugepages[24] = {0};
> 
> Can we exceed this limit too? Can you make this a define?
> 

It's based on counting the digits in U64_MAX.

18446744073709551615X
123456789012345678901

We don't have any defines for that kind of thing.  It's not a bad idea.
#define STRLEN_U64_MAX 20
char init_nr_hugepages[STRLEN_U64_MAX + 1];

But it should be done as part of a kernel wide clean up and not part of this
buffer overflow fix.  The line above it could be changed as well, for example.
Let me create a KTODO and hope the internet will take care of it.

KTODO: create defines for STRLEN_[SU]8/16/32/64_MIN/MAX.

Btw, I rounded up to 24 just because I like buffer sizes that are divisible
by 4 but the compiler is probably going to do that automatically either way.

regards,
dan carpenter