Re: [PATCH] kobject_set_name_vargs memory leak

From: Eric W. Biederman
Date: Fri Jun 26 2009 - 22:10:40 EST


Sergey Senozhatsky <sergey.senozhatsky@xxxxxxx> writes:

> On (06/26/09 07:49), Greg KH wrote:
>> We've been through this before (search lkml archives). If kvasprintf
>> fails, then we don't want to free old_name, as the caller might want to
>> do something with it.
>>
>> Or something along those lines, I can't remember the exact reasoning
>> this early in the morning.
>>
>> Kay, do you remember?
>>
> I found.
> http://lkml.org/lkml/2009/5/11/11
>
>>kobject with name set before should not come into this function,
>>kobject_rename should be used instead.
>
> It's just would be safer to kfree or restore I guess.

Yes. There does seem to be a good point in there that the code should be:
BUG_ON(kobj->name);

And otherwise simply not handle old_name at all.

Eric
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/