Re: Pentium 4 and misaligned doubles

[Rici Lake <lua@...>]

> Note that the constant 6 is probably also misaligned. Also, there is a 
> difference between the layout of Lua 5.0.2 and Lua 5.1w6, so although 
> the periodicity should be the same, it will probably have a different 
> offset.

And actually, that was the sole cause of the difference: (Now I'm 
puzzled again.)

local a = {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}; for i = 1, 1e7 do a[1] = 
a[16] + 1 end
        1.35 real         1.33 user         0.00 sys
local a = {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}; for i = 1, 1e7 do a[2] = 
a[1] + 1 end
        1.35 real         1.33 user         0.01 sys
local a = {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}; for i = 1, 1e7 do a[3] = 
a[2] + 1 end
        1.35 real         1.34 user         0.00 sys
local a = {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}; for i = 1, 1e7 do a[4] = 
a[3] + 1 end
        1.35 real         1.34 user         0.00 sys
local a = {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}; for i = 1, 1e7 do a[5] = 
a[4] + 1 end
        1.35 real         1.34 user         0.00 sys
local a = {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}; for i = 1, 1e7 do a[6] = 
a[5] + 1 end
        1.34 real         1.31 user         0.01 sys
local a = {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}; for i = 1, 1e7 do a[7] = 
a[6] + 1 end
        1.36 real         1.34 user         0.00 sys
local a = {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}; for i = 1, 1e7 do a[8] = 
a[7] + 1 end
        1.36 real         1.35 user         0.00 sys
local a = {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}; for i = 1, 1e7 do a[9] = 
a[8] + 1 end
        1.35 real         1.33 user         0.01 sys