Is it possible anyhow to compute an integral image for an image w/ pixels of type Ipp32f (i.e. float) with use of Intel IPP?
it's not possible to do with the current version.
It's possible with new IPP Async library - take a look at ipp_async_op.h:
HPPFUNC(hppStatus, hppiIntegral, (hppAccel accel, hppiMatrix* src, hppScalar val, hppiMatrix* dst, hppDataType dst_type));
supported data types are 8u,16u,16s,32f