Re: bit.lshift and performance - luabitop v.s. lua-5.2.0-work4

[David Manura <dm.lua@...>]

> On Wed, Oct 13, 2010 at 1:12 AM, KHMan <keinhong@gmail.com> wrote:
>> On 10/13/2010 10:34 AM, David Manura wrote:
>>>  local bits = bit.band(buffer, bit.lshift(1, nbits)-1)
>> ...Is there any actual technical problem with using this instead?
>>    bit.rshift(0xFFFFFFFF, 32 - nbits)

One downside is that the code

  local bits = bit.band(buffer, rshift(0xffffffff, 32 - nbits))
  buffer = bit.rshift(buffer, nbits)

now requires two special cases: nbits == 0 and nbits == 32.  Both
actually occur in my code.  I don't know if this is a rare case, but
it can affect code containing shifts by both n and 32 - n where n in
[0, 32].