RFR: 8342103: C2 compiler support for Float16 type and associated operations

[Bhavana Kilambi]

On Wed, 16 Oct 2024 14:19:40 GMT, Bhavana Kilambi <bkilambi at openjdk.org> wrote:

>> Hi All,
>> 
>> This patch adds C2 compiler support for various Float16 operations added by [PR#22128](https://github.com/openjdk/jdk/pull/22128)
>> 
>> Following is the summary of changes included with this patch:-
>> 
>> 1. Detection of various Float16 operations through inline expansion or pattern folding idealizations.
>> 2. Float16 operations like add, sub, mul, div, max, and min are inferred through pattern folding idealization.
>> 3. Float16 SQRT and FMA operation are inferred through inline expansion and their corresponding entry points are defined in the newly added Float16Math class.
>>       -    These intrinsics receive unwrapped short arguments encoding IEEE 754 binary16 values.
>> 5. New specialized IR nodes for Float16 operations, associated idealizations, and constant folding routines.
>> 6. New Ideal type for constant and non-constant Float16 IR nodes. Please refer to [FAQs ](https://github.com/openjdk/jdk/pull/21490#issuecomment-2482867818)for more details.
>> 7. Since Float16 uses short as its storage type, hence raw FP16 values are always loaded into general purpose register, but FP16 ISA instructions generally operate over floating point registers, therefore compiler injectes reinterpretation IR before and after Float16 operation nodes to move short value to floating point register and vice versa.
>> 8. New idealization routines to optimize redundant reinterpretation chains. HF2S + S2HF = HF
>> 6. Auto-vectorization of newly supported scalar operations.
>> 7. X86 and AARCH64 backend implementation for all supported intrinsics.
>> 9. Functional and Performance validation tests.
>> 
>> **Missing Pieces:-**
>> **-  AARCH64 Backend.**
>> 
>> Kindly review and share your feedback.
>> 
>> Best Regards,
>> Jatin
>
> test/hotspot/jtreg/compiler/lib/ir_framework/IRNode.java line 122:
> 
>> 120:     public static final String VECTOR_SIZE_64  = VECTOR_SIZE + "64";
>> 121: 
>> 122:     private static final String TYPE_BYTE   = "byte";
> 
> Hi Jatin, why have these changes been made? The PrintIdeal output still prints the vector size of the node in this format - `#vectord<S,4>`. This test - `test/hotspot/jtreg/compiler/vectorization/TestFloatConversionsVectorNaN.java` was failing due to this mismatch ..

Infact many tests under test/hotspot fail due to this change.

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

PR Review Comment: https://git.openjdk.org/jdk/pull/21490#discussion_r1804557172