Re: Bug in kernel code?

["Stephen C. Biggs" <s.biggs@softier.com>]

On 27 Aug 2002 at 17:42, David S. Miller wrote:

>    From: "Stephen C. Biggs" <s.biggs@softier.com>
>    Date: Tue, 27 Aug 2002 17:40:41 -0700
>    
>    You're misunderstanding me, I meant that the  first test is done
>    AFTER the first iteration is   executed, so my fix is correct
>    since, even if order is 0 because at least one iteration of the
>    loop  is done, and the post decrement makes sure that the test
>    succeeds if order was 0 going into the loop.
> 
> Order is always >= 0 when we enter the loop.
> 
> If we actually get the table allocated then the decrement of 'order'
> is not executed if we allocate the table successfully.
> 
> I don't understand what the problem is with my fix.
> 
[Trying again - my mail client is messing up/sending empty posts]

As I look at it, it can only be 0 <= order < 32, so your fix works, as mine does.  Yours is simpler 
and doesn't involve changing code, just declarations.

There need to be some sanity checks in this code: what if mempages is passed as some insanely huge 
number, e.g.