Skip to content

powerpc: Bypass FP16 as BF16 on Power#5670

Merged
martin-frbg merged 1 commit intoOpenMathLib:developfrom
amritahs-ibm:handle_fp16_power
Mar 11, 2026
Merged

powerpc: Bypass FP16 as BF16 on Power#5670
martin-frbg merged 1 commit intoOpenMathLib:developfrom
amritahs-ibm:handle_fp16_power

Conversation

@amritahs-ibm
Copy link
Contributor

typedef FP16 as BF16 on Power as FP16 is
not yet supported

typedef FP16 as BF16 on Power as FP16 is
not yet supported
@martin-frbg
Copy link
Collaborator

Thanks - is uint16_t unsupported there too ?

@martin-frbg martin-frbg added this to the 0.3.32 milestone Mar 9, 2026
@RajalakshmiSR
Copy link

Thanks - is uint16_t unsupported there too ?

If the bfloat16 type is unavailable, the implementation defaults to uint16_t.

typedef uint16_t bfloat16;

@martin-frbg martin-frbg merged commit 1ff3a1a into OpenMathLib:develop Mar 11, 2026
98 of 102 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants