[Devel] Re: [PATCH 1/1] cgroups: strcpy destination string overflow

[Paul Menage]

On Tue, Oct 5, 2010 at 12:48 PM, Andrew Morton
<akpm at linux-foundation.org> wrote:
> On Tue,  5 Oct 2010 12:38:05 +0400
> Evgeny Kuznetsov <EXT-Eugeny.Kuznetsov at nokia.com> wrote:
>
>> From: Evgeny Kuznetsov <ext-eugeny.kuznetsov at nokia.com>
>>
>> Function "strcpy" is used without check for maximum allowed source
>> string length and could cause destination string overflow.
>> Check for string length is added before using "strcpy".
>> Function now is return error if source string length is more than
>> a maximum.
>>
>> Signed-off-by: Evgeny Kuznetsov <EXT-Eugeny.Kuznetsov at nokia.com>
>> ---
>>  kernel/cgroup.c |    2 ++
>>  1 files changed, 2 insertions(+), 0 deletions(-)
>>
>> diff --git a/kernel/cgroup.c b/kernel/cgroup.c
>> index c9483d8..82bbede 100644
>> --- a/kernel/cgroup.c
>> +++ b/kernel/cgroup.c
>> @@ -1883,6 +1883,8 @@ static int cgroup_release_agent_write(struct cgroup *cgrp, struct cftype *cft,
>>                                     const char *buffer)
>>  {
>>       BUILD_BUG_ON(sizeof(cgrp->root->release_agent_path) < PATH_MAX);
>> +     if (strlen(buffer) >= PATH_MAX)
>> +             return -EINVAL;
>>       if (!cgroup_lock_live_group(cgrp))
>>               return -ENODEV;
>>       strcpy(cgrp->root->release_agent_path, buffer);
>
> I don't think this can happen, because cftype.max_write_len is
> PATH_MAX.

Yes, it shouldn't be possible.

>
> But it's pretty unobvious if this is actually true, and the code is
> fragile against future changes.

Fair enough - adding the check doesn't hurt anything.

Acked-by: Paul Menage <menage at google.com>
_______________________________________________
Containers mailing list
Containers at lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/containers