[code-reflection] RFR: [hat] Proposal for bfloat16 [v4]

[Gary Frost]

On Wed, 3 Dec 2025 13:57:21 GMT, Juan Fumero <jfumero at openjdk.org> wrote:

>> This PR introduces the type [`bfloat16`](https://cloud.google.com/blog/products/ai-machine-learning/bfloat16-the-secret-to-high-performance-on-cloud-tpus) for HAT.
>> 
>> Testing for OpenCL:
>> 
>> 
>> HAT=SHOW_CODE java -cp hat/job.jar hat.java test ffi-opencl hat.test.TestBFloat16Type
>> 
>> 
>> Testing for CUDA:
>> 
>> 
>> HAT=SHOW_CODE java -cp hat/job.jar hat.java test ffi-cuda hat.test.TestBFloat16Type
>> 
>> 
>> Some tests are expecting to fail due to precision error. We will need to improve the type conversion with round-to-nearest-even for example.
>
> Juan Fumero has updated the pull request incrementally with one additional commit since the last revision:
> 
>   [hat] bfloat16 codegen refactored

hat/core/src/main/java/hat/codebuilders/C99HATKernelBuilder.java line 126:

> 124:         if (OpTk.isAssignable(buildContext.lookup, javaType, MappableIface.class) && javaType instanceof ClassType classType) {
> 125:             globalPtrPrefix().suffix_t(classType).asterisk();
> 126:         } else if (javaType instanceof ClassType classType && classType.toClassName().equals(F16.class.getCanonicalName())) {

Can we leverage OpTk.isAssignable ? or at least avoid string comparisons on Class.getCanonicalName

Something like 
OpTK.isAssignable(lookup, classType, F16.class) ? 

I think we have access to the lookup in the context...

hat/core/src/main/java/hat/codebuilders/HATCodeBuilder.java line 84:

> 82:     }
> 83: 
> 84:     public T suffix_t(Class<?> klass) {

Can we compose using existing methods?

return suffix_t(class.getSimpleName());

hat/core/src/main/java/hat/codebuilders/HATCodeBuilder.java line 89:

> 87: 
> 88:     public T suffix_u(Class<?> klass) {
> 89:         return identifier(klass.getSimpleName()).identifier("_u");

Same as above

-------------

PR Review Comment: https://git.openjdk.org/babylon/pull/716#discussion_r2585272419
PR Review Comment: https://git.openjdk.org/babylon/pull/716#discussion_r2585257810
PR Review Comment: https://git.openjdk.org/babylon/pull/716#discussion_r2585259258