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

[Yuri Tikhonov <yur@emcraft.com>]

=0D=0AHello Scott,

On Wednesday, December 10, 2008 you wrote:

> On Wed, Dec 10, 2008 at 01:29:12PM +0300, Yuri Tikhonov wrote:
>> On Wednesday, December 10, 2008 you wrote:
>> > x * y / z is parsed as (x * y) / z, not x * (y / z).
>>=20
>>  Here we believe in preprocessor: since all PAGE_SIZE, 8, and=20
>> THREAD_SIZE are the constants we expect it will calculate this.
>>=20
>>  E.g. here is the result from this line as produced by cross-gcc=20
>> 4.2.2:
>>=20
>>         lis     r9,0
>>         rlwinm  r29,r29,2,16,29
>>         stw     r29,0(r9)
>>=20
>>  As you see - only rotate-left, i.e. multiplication to the constant.

> Yes, and also note that it is masking the result by 0xfffc, to preserve
> the effect of any overflow in (x * y).

 Right, such a mask was the result of missed brackets. Now (see=20
"[PATCH][v2] fork_init: fix division by zero" I've just posted) the=20
situation is better:

        lis     r9,0
        rlwinm  r29,r29,2,0,29
        stw     r29,0(r9)

 I.e. the mask is 0xFFFFFFFC.

 Regards, Yuri

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