BFMMLA
BFloat16 floating-point matrix multiply-accumulate into 2x2 matrix.
If FEAT_EBF16 is not implemented or FPCR.EBF is 0, irrespective of the other control bits in the FPCR, this instruction:
- Performs two unfused sums-of-products within each two pairs of adjacent BFloat16 elements while multiplying the 2x4 matrix of BFloat16 values in the first source vector with the 4x2 matrix of BFloat16 values in the second source vector. The intermediate single-precision products are rounded before they are summed and the intermediate sum is rounded before accumulation into the 2x2 single-precision matrix in the destination vector. This is equivalent to accumulating two 2-way unfused dot products per destination element.
- Uses the non-IEEE 754 Round-to-Odd rounding mode, which forces bit 0 of an inexact result to 1, and rounds an overflow to an appropriately signed Infinity.
- Does not modify the cumulative FPSR exception bits (IDC, IXC, UFC, OFC, DZC, and IOC).
- Disables trapped floating-point exceptions, as if the FPCR trap enable bits (IDE, IXE, UFE, OFE, DZE, and IOE) are all zero.
- Flushes denormalized inputs and results to zero, as if FPCR.{FZ, FIZ} is {1, 1}.
- Generates only the default NaN, as if FPCR.DN is 1.
If FEAT_EBF16 is implemented and FPCR.EBF is 1, then this instruction:
- Performs two fused sums-of-products within each two pairs of adjacent BFloat16 elements while multiplying the 2x4 matrix of BFloat16 values in the first source vector with the 4x2 matrix of BFloat16 values in the second source vector. The intermediate single-precision products are not rounded before they are summed, but the intermediate sum is rounded before accumulation into the 2x2 single-precision matrix in the destination vector. This is equivalent to accumulating two 2-way fused dot products per destination element.
- Generates only the default NaN, as if FPCR.DN is 1.
- Follows all other floating-point behaviors that apply to single-precision arithmetic, as controlled by the effective value of the FPCR in the current execution mode, and captured in the FPSR.
Note
Arm expects that the BFMMLA instruction will deliver a peak BFloat16 multiply throughput that is at least as high as can be achieved using two BFDOT instructions, with a goal that it should have significantly higher throughput.
Vector
(FEAT_BF16)
31 | 30 | 29 | 28 | 27 | 26 | 25 | 24 | 23 | 22 | 21 | 20 | 19 | 18 | 17 | 16 | 15 | 14 | 13 | 12 | 11 | 10 | 9 | 8 | 7 | 6 | 5 | 4 | 3 | 2 | 1 | 0 |
0 | 1 | 1 | 0 | 1 | 1 | 1 | 0 | 0 | 1 | 0 | Rm | 1 | 1 | 1 | 0 | 1 | 1 | Rn | Rd |
if !HaveBF16Ext() then UNDEFINED;
integer n = UInt(Rn);
integer m = UInt(Rm);
integer d = UInt(Rd);
Assembler Symbols
<Vd> |
Is the name of the SIMD&FP third source and destination register, encoded in the "Rd" field.
|
<Vn> |
Is the name of the first SIMD&FP source register, encoded in the "Rn" field.
|
<Vm> |
Is the name of the second SIMD&FP source register, encoded in the "Rm" field.
|
Internal version only: isa v33.53, AdvSIMD v29.11, pseudocode v2022-09_rel, sve v2022-09_rel
; Build timestamp: 2022-09-30T16:37
Copyright © 2010-2022 Arm Limited or its affiliates. All rights reserved.
This document is Non-Confidential.