Re[2]: [PATCH][v2] fork_init: fix division by zero

[Yuri Tikhonov <yur@emcraft.com>]

=0D=0AHello Andrew,

On Thursday, December 11, 2008 you wrote:

[snip]

> The expression you've chosen here can be quite inacccurate, because
> ((PAGE_SIZE / (8 * THREAD_SIZE)) is a small number.=20

 But why is it bad? We do multiplication to 'mempages', not division.=20
All the numbers in the multiplier are the power of 2, so both=20
expressions:

mempages * (PAGE_SIZE / (8 * THREAD_SIZE))

and

max_threads =3D (mempages * PAGE_SIZE) / (8 * THREAD_SIZE)

are finally equal.=20

>  The way to preserve accuracy is

>         max_threads =3D (mempages * PAGE_SIZE) / (8 * THREAD_SIZE);

> so how about avoiding the nasty ifdefs and doing

 I'm OK with the approach below, but, leading resulting to the same,=20
this involves some overhead to the code where there was no this=20
overhead before this patch: e.g. your implementation is finally boils=20
down to ~5 times more processor instructions than there were before,
plus operations with stack for the 'm' variable.

 On the other hand, my approach with nasty (I agree) ifdefs doesn't=20
lead to overheads to the code which does not need this: i.e. the most=20
common situation of small PAGE_SIZEs. Big PAGE_SIZE is the exception,=20
so I believe that the more common cases should not suffer because of=20
this.

> --- a/kernel/fork.c~fork_init-fix-division-by-zero
> +++ a/kernel/fork.c
> @@ -69,6 +69,7 @@
>  #include <asm/mmu_context.h>
>  #include <asm/cacheflush.h>
>  #include <asm/tlbflush.h>
> +#include <asm/div64.h>
> =20
>  /*
>   * Protected counters by write_lock_irq(&tasklist_lock)
> @@ -185,10 +186,15 @@ void __init fork_init(unsigned long memp
> =20
>         /*
>          * The default maximum number of threads is set to a safe
> -        * value: the thread structures can take up at most half
> -        * of memory.
> +        * value: the thread structures can take up at most
> +        * (1/8) part of memory.
>          */
> -       max_threads =3D mempages / (8 * THREAD_SIZE / PAGE_SIZE);
> +       {
> +               /* max_threads =3D (mempages * PAGE_SIZE) / THREAD_SIZE /=
 8; */
> +               u64 m =3D mempages * PAGE_SIZE;
> +               do_div(m, THREAD_SIZE * 8);
> +               max_threads =3D m;
> +       }
> =20
>         /*
>          * we need to allow at least 20 threads to boot a system
> _

> ?


> The code is also inaccurate because it assumes that <whatever allocator
is used for threads>> will pack the thread_structs into pages with best
> possible density, which isn't necessarily the case.  Let's not worry
> about that.




> OT:

> max_threads is widly wrong anyway.

> - the caller passes in num_physpages, which includes highmem.  And we
>   can't allocate thread structs from highmem.

> - num_physpages includes kernel pages and other stuff which can never
>   be allocated via the page allocator.

> A suitable fix would be to switch the caller to the strangely-named
> nr_free_buffer_pages().

> If you grep the tree for `num_physpages', you will find a splendid
> number of similar bugs.  num_physpages should be unexported, burnt,
> deleted, etc.  It's just an invitation to write buggy code.


 Regards, Yuri

 --
 Yuri Tikhonov, Senior Software Engineer
 Emcraft Systems, www.emcraft.com