v_mfma_f32_16x16x4_4b_bf16 GPU Native ISA AMD Matrix

V MFMA F32 16X16X4 4B BF16 Vector Packed Arithmetic

v_mfma_f32_16x16x4_4b_bf16

Multiply the 16x4 matrix in the first input by the 4x16 matrix in the second input and add the 16x16 matrix in the third input using fused multiply…

Encoding

Verified bit-level encoding data is not yet available for this instruction. The instruction-format classification below (VOP3P) is well-documented and stable; exact per-target opcode/field bit positions have not yet been imported from a verified source.
Format VOP3P
Execution Unit

Operands

Operand details have not yet been curated for this instruction.

GFX Target Compatibility

TargetSupport
gfx942✅ Supported

This instruction performs 8 passes.

Related

More in Vector Packed Arithmetic

Reference

AMDGPU / GFX ISA

Description

Multiply the 16x4 matrix in the first input by the 4x16 matrix in the second input and add the 16x16 matrix in the third input using fused multiply add. Store the resulting matrix into vector registers.

Semantics

D = A (16x4) * B (4x16) + C (16x16) This instruction performs 4 matrix multiplies. Each operand contains 4 matrices back to back, and each matrix has elements distributed across all lanes of the wave. Each matrix multiple is computed and the row-column dot products are distributed across the vector ALU for higher performance. The result matrices are stored back-to-back in the destination vector registers. Matrices A and B are BF16 float format. Matrices C and D are single-precision float format.

Sources