# `vmaxfp` — Vector Maximum Floating Point
> **Category:** [VMX (Altivec)](../categories/vmx.md) · **Form:** [VX](../forms/VX.md) · **Opcode:** `0x1000040a`
## Assembler Mnemonics
| Mnemonic | XML entry | Flags | Description |
| --- | --- | --- | --- |
| `vmaxfp` | `vmaxfp` | — | Vector Maximum Floating Point |
| `vmaxfp128` | `vmaxfp128` | — | Vector128 Maximum Floating Point |
## Syntax
```asm
vmaxfp [VD], [VA], [VB]
vmaxfp128 [VD], [VA], [VB]
```
## Encoding
### `vmaxfp` — form `VX`
- **Opcode word:** `0x1000040a`
- **Primary opcode (bits 0–5):** `4`
- **Extended opcode:** `1034`
- **Synchronising:** no
| Bits | Field | Meaning |
| --- | --- | --- |
| 0–5 | `OPCD` | primary opcode (4) |
| 6–10 | `VRT/VD` | destination vector register |
| 11–15 | `VRA/VA` | source A vector register |
| 16–20 | `VRB/VB` | source B vector register |
| 21–31 | `XO` | extended opcode (11 bits) |
### `vmaxfp128` — form `VX128`
- **Opcode word:** `0x18000280`
- **Primary opcode (bits 0–5):** `6`
- **Extended opcode:** `640`
- **Synchronising:** no
| Bits | Field | Meaning |
| --- | --- | --- |
| 0–5 | `OPCD` | primary opcode (4 or 5) |
| 6–10 | `VD128l` | destination low 5 bits |
| 11–15 | `VA128l` | source A low 5 bits |
| 16–20 | `VB128l` | source B low 5 bits |
| 21 | `VA128H` | source A high bit |
| 22 | `—` | reserved |
| 23–25 | `VC` | optional VC / XO sub-field |
| 26 | `VA128h` | source A middle bit |
| 27 | `—` | reserved |
| 28–29 | `VD128h` | destination high 2 bits |
| 30–31 | `VB128h` | source B high 2 bits |
## Operands
| Field | Role | Description |
| --- | --- | --- |
| `VA` | vmaxfp: read; vmaxfp128: read | Source A vector register. |
| `VB` | vmaxfp: read; vmaxfp128: read | Source B vector register. |
| `VD` | vmaxfp: write; vmaxfp128: write | Destination vector register. |
## Register Effects
### `vmaxfp`
- **Reads (always):** `VA`, `VB`
- **Reads (conditional):** _none_
- **Writes (always):** `VD`
- **Writes (conditional):** _none_
### `vmaxfp128`
- **Reads (always):** `VA`, `VB`
- **Reads (conditional):** _none_
- **Writes (always):** `VD`
- **Writes (conditional):** _none_
## Status-Register Effects
_No condition-register or status-register effects._
## Operation (pseudocode)
```
; Pseudocode derives directly from the xenia-rs interpreter
; arm (see Implementation References). Operation semantics:
; - Read source operands from the fields listed under Operands.
; - Apply the arithmetic / logical / memory action described
; in the Description field above.
; - Write results to the destination register(s); update any
; status bits enumerated under Status-Register Effects.
; Consult the IBM AIX reference link under IBM Reference for
; canonical PPC-style pseudocode where xenia's expression is
; terse.
```
## C Translation Example
```c
/* C translation: the xenia-rs interpreter arm below in */
/* Implementation References is the authoritative semantic */
/* snapshot. Translate it line-by-line: */
/* - ctx.gpr[N] -> r[N] (or f[]/v[] for FPRs/VRs) */
/* - mem.read_u*/write_u* -> mem_read_u*_be / mem_write_u*_be */
/* - ctx.update_cr_signed(fld, v) -> update_cr_signed(fld, v) */
/* - ctx.xer_ca / xer_ov / xer_so -> xer.CA / xer.OV / xer.SO */
/* The Register Effects and Status-Register Effects tables above */
/* enumerate every side effect a faithful translation must emit. */
```
## Implementation References
**`vmaxfp`**
- xenia-canary XML: [`tools/ppc-instructions.xml` — search for `mnem="vmaxfp"`](../../xenia-canary/tools/ppc-instructions.xml)
- xenia-canary emit: [`src/xenia/cpu/ppc/ppc_emit_altivec.cc:831`](../../xenia-canary/src/xenia/cpu/ppc/ppc_emit_altivec.cc#L831)
- xenia-rs opcode: [`crates/xenia-cpu/src/opcode.rs:101`](../../xenia-rs/crates/xenia-cpu/src/opcode.rs#L101)
- xenia-rs decoder: [`crates/xenia-cpu/src/decoder.rs:522`](../../xenia-rs/crates/xenia-cpu/src/decoder.rs#L522)
- xenia-rs interpreter: [`crates/xenia-cpu/src/interpreter.rs:2121-2128`](../../xenia-rs/crates/xenia-cpu/src/interpreter.rs#L2121-L2128)
xenia-rs interpreter body (frozen snapshot)
```rust
PpcOpcode::vmaxfp => {
let a = ctx.vr[instr.ra()].as_f32x4();
let b = ctx.vr[instr.rb()].as_f32x4();
let mut r = [0f32; 4];
for i in 0..4 { r[i] = vmx::max_nan(a[i], b[i]); }
ctx.vr[instr.rd()] = xenia_types::Vec128::from_f32x4_array(r);
ctx.pc += 4;
}
```
**`vmaxfp128`**
- xenia-canary XML: [`tools/ppc-instructions.xml` — search for `mnem="vmaxfp128"`](../../xenia-canary/tools/ppc-instructions.xml)
- xenia-canary emit: [`src/xenia/cpu/ppc/ppc_emit_altivec.cc:834`](../../xenia-canary/src/xenia/cpu/ppc/ppc_emit_altivec.cc#L834)
- xenia-rs opcode: [`crates/xenia-cpu/src/opcode.rs:101`](../../xenia-rs/crates/xenia-cpu/src/opcode.rs#L101)
- xenia-rs decoder: [`crates/xenia-cpu/src/decoder.rs:696`](../../xenia-rs/crates/xenia-cpu/src/decoder.rs#L696)
- xenia-rs interpreter: [`crates/xenia-cpu/src/interpreter.rs:2129-2136`](../../xenia-rs/crates/xenia-cpu/src/interpreter.rs#L2129-L2136)
xenia-rs interpreter body (frozen snapshot)
```rust
PpcOpcode::vmaxfp128 => {
let a = ctx.vr[instr.va128()].as_f32x4();
let b = ctx.vr[instr.vb128()].as_f32x4();
let mut r = [0f32; 4];
for i in 0..4 { r[i] = vmx::max_nan(a[i], b[i]); }
ctx.vr[instr.vd128()] = xenia_types::Vec128::from_f32x4_array(r);
ctx.pc += 4;
}
```
## Special Cases & Edge Conditions
- **Per-lane IEEE max.** Four word lanes; `VD[i] = (VA[i] > VB[i]) ? VA[i] : VB[i]`.
- **NaN propagation surprise.** Xenia uses `if a > b { a } else { b }`, so any NaN comparison evaluates false and the result is `VB`. The IBM manual specifies "the larger of `VA[i]` and `VB[i]`, with NaN handling such that any NaN input yields a NaN result" — this is *not* what xenia does. Hardware's `vmaxfp(NaN, x) = NaN` while xenia returns `x`. **Worth checking against `vmx.rs` for any future correctness fixes.**
- **Sign of zero.** `vmaxfp(+0, -0)` returns `-0` in xenia (since `+0 > -0` is false → returns `b = -0`). The hardware likely returns the sign-positive zero — also worth verifying.
- **`VSCR[NJ]` denormals.** With `NJ = 1` (Xenon default), denormal inputs are flushed to `±0` before comparison.
- **No `VSCR[SAT]` change, no XER change, no exceptions.**
- **Big-endian word lanes.** Lane 0 is the most-significant word.
- **Aliasing legal.** `vmaxfp v3, v3, v4` is the standard "clamp from below by `v4`" idiom.
- **VMX128 sibling (`vmaxfp128`).** Identical comparator semantics with the extended encoding.
## Related Instructions
- [`vminfp`](vminfp.md) — the per-lane minimum.
- [`vcmpgtfp`](vcmpgtfp.md), [`vcmpgefp`](vcmpgefp.md) — separate compare-and-mask path.
- [`vsel`](vsel.md) — combine masks with arbitrary alternatives.
- [`vmaddfp`](vmaddfp.md) — fused multiply-add when the max is part of a polynomial.
- [`vmaxsw`](vmaxsw.md) — integer-word max if the lanes are signed integers.
## IBM Reference
- [AIX 7.3 — `vmaxfp` (Vector Maximum Floating Point)](https://www.ibm.com/docs/en/aix/7.3.0?topic=set-vmaxfp-vector-maximum-floating-point-instruction)
- [IBM AltiVec Technology Programmer's Interface Manual, Chapter 5 — Floating-Point Min/Max](https://www.nxp.com/docs/en/reference-manual/ALTIVECPIM.pdf)