On Wed, 2011-06-29 at 20:47 +0200, Nick Copeland wrote:
The ARM
softfloat overhead is not that great and the coding required to get
access to the GPUs is suitable that developers will implement them
for optimisations.
1) There are several hw floating point implemntations on the "ARM
platform" as such ... It is meaningless to generalize
2) What GPU vendor/implementation are you talking about here?