[cairo] [PATCH] Rely less on fast double-precision FPUs

[Jonathan Morton]

On 1 June 2010 23:23, Bill Spitzak <spitzak at gmail.com> wrote:
> I'm sure this has already been discussed for cairo, but is there a reason
> double is used instead of 32-bit float? Many other graphic apis I have seen
> use float coordinates and matrices for precisely this reason.

I don't think single-precision would help much for ARM - at least not
for those particular functions.  The compare mechanism is basically
the same.  Other platforms have had competent DP support for a while,
so they would not benefit either.

For more demanding drawing, SP would run faster on Cortex-A8 because
the pipelined NFP unit can be used instead of the nonpipelined VFP.
This would tend to affect non-trivial transforms and path handling
rather than the simpler cases.

 - Jonathan Morton