Is there a CPU half data type?

The GPU can use 16-bit floats by using the "half" data type. Do the CPU have a 16-bit float?


My actual issue is that I want to use "half" in my GPU uniform structs - how can I initialize such values on the CPU?

Up vote post of Jakob Schou Down vote post of Jakob Schou
3.1k views
Posted by
Jakob Schou
Reply
Add a Comment

Accepted Reply

For storage, you can use the 

__fp16
type, or any type that has the same size and alignment as 
half
, such as 
uint16_t
.


Assignments from 

float
to 
__fp16
will emit the appropriate 
fcvt
instruction or intrinsic, depending on the target platform.


If you elect to use 

uint16_t
as your storage type, you can still type-pun through the 
__fp16
type to load and store:


static float loadFromF16(const uint16_t *pointer) { return *(const __fp16 *)pointer; } 
static void storeAsF16(float value, uint16_t *pointer) { *(__fp16 *)pointer = value; }


If you have a long list of floats you need to convert to half, rather than a few struct members, using the 

vImageConvert_PlanarFtoPlanar16F
function may be more efficient.
Posted by
wcm
Up vote reply of wcm Down vote reply of wcm
Post marked as solved
Add a Comment

Replies

For storage, you can use the 

__fp16
type, or any type that has the same size and alignment as 
half
, such as 
uint16_t
.


Assignments from 

float
to 
__fp16
will emit the appropriate 
fcvt
instruction or intrinsic, depending on the target platform.


If you elect to use 

uint16_t
as your storage type, you can still type-pun through the 
__fp16
type to load and store:


static float loadFromF16(const uint16_t *pointer) { return *(const __fp16 *)pointer; } 
static void storeAsF16(float value, uint16_t *pointer) { *(__fp16 *)pointer = value; }


If you have a long list of floats you need to convert to half, rather than a few struct members, using the 

vImageConvert_PlanarFtoPlanar16F
function may be more efficient.
Posted by
wcm
Up vote reply of wcm Down vote reply of wcm
Post marked as solved
Add a Comment

hi Jakob,


You might also be interested in the OpenEXR project's Half type which a C++ class representing a half precision float.. http://www.openexr.com


OpenEXR is focused on the image file format of that name but the Half type and the associated utilities (in the Ilmbase library) can be used entirely separately from the OpenEXR library which is built on top of that.


Cheers,

- James

Posted by
essexs
Up vote reply of essexs Down vote reply of essexs
Add a Comment

How to do it in Swift? Should Swift have a 16 bit float type?

Posted by
yvsong
Up vote reply of yvsong Down vote reply of yvsong
Add a Comment
Note that _fp16 will only emit the hardware conversion on Intel if you pass -mf16c to the compiler. Fortunately, all BigSur supported machines support the conversion, but I don't believe it is on by default yet for x8664 targeting that OS. It is an IvyBridge or later instruction. If you don't pass -mf16c, then you'll get a software conversion, which is not very cheap.

On Apple Silicon, the appropriate instructions have been there since Cortex A9 and should be there on any arm64 machine.

If you need to convert a whole bunch of half float values in an array, please see vImageConvertPlanarFtoPlanar16F and vImageConvertPlanar16FtoPlanarF.

If you need to test to see if the machine supports the instruction, you can use sysctlbyname( "hw.optional.f16c", ...)
Posted by
IanOllmann
Up vote reply of IanOllmann Down vote reply of IanOllmann
Add a Comment