MachinePrecision in LogPlot

I am encountering an error every time I call LogPlot for a compiled function:

 sw = Compile[{{s, _Real}}, s^2];
    LogPlot[sw[x], {x, 0, 1}]


CompiledFunction::cfsa: Argument x at position 1 should be a machine-size real number. >>

I don’t understand what’s the problem with LogPlot, everything works fine with Plot.

Interesting. I can replicate that. This one works:

sw = Compile[{s}, s^2.];
ListLogPlot[Table[sw[x], {x, 0., 1., 0.01}], Joined -> True]

Cheers,
Marco