[LLVMdev] Adding masked vector load and store intrinsics

dag at cray.com dag at cray.com
Mon Oct 27 10:19:45 PDT 2014

"Demikhovsky, Elena" <elena.demikhovsky at intel.com> writes:

>> The same problem exists with any potentially trapping instruction
>> (e.g. all floating point computations).  The need for intrinsics
>> goes way beyond loads and stores.
> We definitely looking at them, but decided to start from load and
> store. All FP + gather/scatter are in our long term plan. It will be
> about 20 intrinsics.
> But step-by-step.

Makes total sense.  Glad to hear the rest is coming!


More information about the llvm-dev mailing list