e6d43a23ac
Bundles state that lives OUTSIDE the xenia-rs repo so a fresh clone on
another machine can be brought up to identical configuration via
migration/setup.sh:
- claude-memory/ ~/.claude/projects/-home-fabi-RE-Project-Sylpheed/memory/
(103 files, 1.1 MB - MEMORY.md + every
project_xenia_rs_*.md from audits
addis_signext through audit-058)
- project-root/dot-claude/ <project-root>/.claude/settings.json
(Stop hook + permissions)
- project-root/ppc-manual/ <project-root>/ppc-manual/
(PowerPC reference docs, 397 files, 3.7 MB)
- project-root/run-canary.sh <project-root>/run-canary.sh
- README.md Human-readable setup checklist
- setup.sh Idempotent installer (also reclones
xenia-canary at pinned HEAD 6de80dffe)
- MANIFEST.md Per-file mapping + per-file-not-bundled
restoration recipe
Excluded from bundle (not shippable via git):
- Sylpheed ISO (7.8 GB; copyright; manual copy required)
- sylpheed.db (395 MB; regenerable from XEX via analysis tooling)
- target/ build artifacts (rebuild on target)
- audit-runs probe firehoses (.log/.stdout/.stderr ~11 GB; rerun if needed)
- audit-runs memory dumps (.bin ~4.5 GB; rerun audit-026/027/029 if needed)
- xenia-canary checkout (setup.sh reclones from
git.mc02.dev/fabi/Xenia-Canary.git at HEAD 6de80dffe)
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
137 lines
6.1 KiB
Markdown
137 lines
6.1 KiB
Markdown
# `fmaddx` — Floating Multiply-Add
|
||
|
||
> **Category:** [Floating-Point](../categories/fpu.md) · **Form:** [A](../forms/A.md) · **Opcode:** `0xfc00003a`
|
||
|
||
<!-- GENERATED: BEGIN -->
|
||
|
||
## Assembler Mnemonics
|
||
|
||
| Mnemonic | XML entry | Flags | Description |
|
||
| --- | --- | --- | --- |
|
||
| `fmadd` | `fmaddx` | — | Floating Multiply-Add |
|
||
| `fmadd.` | `fmaddx` | Rc=1 | Floating Multiply-Add |
|
||
|
||
## Syntax
|
||
|
||
```asm
|
||
fmadd[Rc] [FD], [FA], [FC], [FB]
|
||
```
|
||
|
||
## Encoding
|
||
|
||
### `fmaddx` — form `A`
|
||
|
||
- **Opcode word:** `0xfc00003a`
|
||
- **Primary opcode (bits 0–5):** `63`
|
||
- **Extended opcode:** `29`
|
||
- **Synchronising:** no
|
||
|
||
| Bits | Field | Meaning |
|
||
| --- | --- | --- |
|
||
| 0–5 | `OPCD` | primary opcode (59 or 63) |
|
||
| 6–10 | `FRT` | destination FPR |
|
||
| 11–15 | `FRA` | source A FPR |
|
||
| 16–20 | `FRB` | source B FPR |
|
||
| 21–25 | `FRC` | source C FPR (multiplier for madd-style ops) |
|
||
| 26–30 | `XO` | extended opcode (5 bits) |
|
||
| 31 | `Rc` | record-form flag (updates CR1) |
|
||
|
||
## Operands
|
||
|
||
| Field | Role | Description |
|
||
| --- | --- | --- |
|
||
| `FA` | fmaddx: read | Source A floating-point register (`fr0`–`fr31`). |
|
||
| `FC` | fmaddx: read | Source C floating-point register (for madd-style ops). |
|
||
| `FB` | fmaddx: read | Source B floating-point register. |
|
||
| `FD` | fmaddx: write | Destination floating-point register. |
|
||
| `CR` | fmaddx: write (conditional) | Condition-register update. When `Rc=1`, CR field 0 (or CR6 for vector compares, CR1 for FPU) is updated from the result. |
|
||
| `FPSCR` | fmaddx: write | Floating-Point Status and Control Register. |
|
||
|
||
## Register Effects
|
||
|
||
### `fmaddx`
|
||
|
||
- **Reads (always):** `FA`, `FC`, `FB`
|
||
- **Reads (conditional):** _none_
|
||
- **Writes (always):** `FD`, `FPSCR`
|
||
- **Writes (conditional):** `CR`
|
||
|
||
## Status-Register Effects
|
||
|
||
- `fmaddx`: **CR1** ← FPSCR[FX, FEX, VX, OX] when `Rc=1`.; **FPSCR** updated per IEEE-754 flags (FX, FEX, FPRF, FR, FI, exceptions).
|
||
|
||
## Operation (pseudocode)
|
||
|
||
```
|
||
FRT <- (FRA × FRC) + FRB
|
||
```
|
||
|
||
## 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
|
||
|
||
**`fmaddx`**
|
||
- xenia-canary XML: [`tools/ppc-instructions.xml` — search for `mnem="fmaddx"`](../../xenia-canary/tools/ppc-instructions.xml)
|
||
- xenia-canary emit: [`src/xenia/cpu/ppc/ppc_emit_fpu.cc:186`](../../xenia-canary/src/xenia/cpu/ppc/ppc_emit_fpu.cc#L186)
|
||
- xenia-rs opcode: [`crates/xenia-cpu/src/opcode.rs:28`](../../xenia-rs/crates/xenia-cpu/src/opcode.rs#L28)
|
||
- xenia-rs decoder: [`crates/xenia-cpu/src/decoder.rs:928`](../../xenia-rs/crates/xenia-cpu/src/decoder.rs#L928)
|
||
- xenia-rs interpreter: [`crates/xenia-cpu/src/interpreter.rs:2640-2652`](../../xenia-rs/crates/xenia-cpu/src/interpreter.rs#L2640-L2652)
|
||
<details><summary>xenia-rs interpreter body (frozen snapshot)</summary>
|
||
|
||
```rust
|
||
PpcOpcode::fmaddx => {
|
||
// PPCBUG-202: VXISI from input properties (not from `a*c` which has wrong sign on overflow).
|
||
let a = ctx.fpr[instr.ra()];
|
||
let c = ctx.fpr[instr.rc()];
|
||
let b = ctx.fpr[instr.rb()];
|
||
fpscr::check_invalid_mul(ctx, a, c);
|
||
fpscr::check_invalid_fma_add(ctx, a, c, b, false);
|
||
let result = a.mul_add(c, b);
|
||
ctx.fpr[instr.rd()] = result;
|
||
fpscr::update_after_op(ctx, result, a.is_finite() && b.is_finite() && c.is_finite());
|
||
if instr.rc_bit() { update_cr1_from_fpscr(ctx); }
|
||
ctx.pc += 4;
|
||
}
|
||
```
|
||
</details>
|
||
|
||
<!-- GENERATED: END -->
|
||
|
||
## Special Cases & Edge Conditions
|
||
|
||
- **Single rounding step.** `fmadd` computes `(FRA × FRC) + FRB` with one IEEE-754 rounding at the end — strictly more accurate than separate multiply + add. xenia-rs uses Rust's `f64::mul_add`, which guarantees a true FMA on hosts with hardware FMA (x86_64 with FMA3, ARM with NEON-FMA); on hosts without it, Rust's stdlib falls back to a software FMA so the semantic match is preserved.
|
||
- **Operand layout.** A-form: `FRT, FRA, FRC, FRB`. Note the assembler order — `FRC` (multiplier) comes before `FRB` (addend). Encoding bit fields are `FRA` (11–15), `FRB` (16–20), `FRC` (21–25).
|
||
- **Invalid operations.** `0×∞ + finite` → `VXIMZ`; `∞×x + ∓∞` (after multiplication produces ±∞ that opposes addend sign) → `VXISI`. Quiet NaN result with `FPSCR[VX, FX]` set.
|
||
- **FPSCR side effects.** Hardware updates `FPRF`, `FR`, `FI`, `FX`, `OX`, `UX`, `XX`, `VXIMZ`, `VXISI`, `VXSNAN`. xenia-rs does not update FPSCR (xenia quirk).
|
||
- **`Rc=1` (`fmadd.`)** copies `FPSCR[FX, FEX, VX, OX]` into CR1.
|
||
- **NaN propagation.** Quiet-NaN result for any NaN operand; signalling NaNs are quietened.
|
||
- **Use case.** Dot products, polynomial evaluation (Horner's method), matrix multiplies, Newton-Raphson divide/sqrt refinement. Hot-path PPC code is dense with `fmadd`.
|
||
- **Denormal flush.** Xenon boots with `FPSCR[NI]=1`; xenia uses host IEEE behavior.
|
||
|
||
## Related Instructions
|
||
|
||
- [`fmaddsx`](fmaddsx.md) — single-precision sibling.
|
||
- [`fmsubx`](fmsubx.md), [`fnmaddx`](fnmaddx.md), [`fnmsubx`](fnmsubx.md) — the other three fused multiply-add variants:
|
||
- `fmsub` = `(A×C) − B`
|
||
- `fnmadd` = `−((A×C) + B)`
|
||
- `fnmsub` = `−((A×C) − B)`
|
||
- [`fmulx`](fmulx.md), [`faddx`](faddx.md) — non-fused decomposition (two rounding steps; less precise).
|
||
- [`fresx`](fresx.md), [`frsqrtex`](frsqrtex.md) — reciprocal helpers refined by `fmadd`/`fnmsub`.
|
||
|
||
## IBM Reference
|
||
|
||
- [AIX 7.3 — `fmadd` (Floating Multiply-Add)](https://www.ibm.com/docs/en/aix/7.3.0?topic=set-fma-fmadd-floating-multiply-add-instruction)
|
||
- [PowerISA v2.07B, Book I, Chapter 4 — Floating-Point Processor](https://openpowerfoundation.org/specifications/isa/) (single-rounding fused multiply-add definition).
|