[tremor] A possible improvement for some of the processor where tables hit hard

[Segher Boessenkool]

David Willmore wrote:
> 
> > > What's wrong here?  0xffffffffUL >> (32 - 0) should be 0 like the lookup
> > > table.
> >
> > Not on Intel, which limit a shift to five bits.  Funny, I was pretty
> > certain ARM did too.  Shift 32 == shift 0.
> 
> And this from the company that produced the i432?
> 
> Is shift 32 == shift 0 in the standard because it's not logically
> equivalent.  ROT 32 on a 32 bit variable is equivalent to ROT 0,
> but not a shift.  shift 32 == 0 on a 32 bit variable, right?

Shifts by a number of bits greater or equal to the number of bits in the
integer data type to shift are not defined by the C standard.

FWIW, x86 has shifts from 0 to 31, and ARM has shifts from 1 to 32, and
PowerPC has shifts from 0 to 63.  Not sure about the others.

<p>Cheers,

Segher
--- >8 ----
List archives:  http://www.xiph.org/archives/
Ogg project homepage: http://www.xiph.org/ogg/
To unsubscribe from this list, send a message to 'tremor-request at xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is needed.
Unsubscribe messages sent to the list will be ignored/filtered.