Files
Sylpheed/tools/ppc-manual/vmx/vpkuwus.md
sim 9bfe96e44d fix(ppc-manual): 543 dead links, from two generator bugs and wrong relative paths
- Category pages linked each family as `<slug>.md`, relative to categories/,
  where no family page lives. They now link `../<category>/<slug>.md`.
- Form pages linked a member into its *own* category directory, so every
  VMX128 sibling (`vsldoi128`) pointed at vmx128/ although its family page is
  under vmx/. They now link into the family's directory.
- Hand-written "Related" and sibling mentions linked other categories' pages
  as if they were in the same directory. 109 are retargeted through the page
  index; 29 that pointed a family page at itself (`vrefp128` on vrefp.md) and
  6 naming instructions the manual has no page for are plain text now.

Regenerated at the existing Canary pin (f21ebd49e): upstream has moved on, and
re-pinning belongs in its own change. The generator reports 0 family pages
changed and is idempotent; the only dead links left are TEMPLATE.md's
placeholders.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-16 22:37:12 +02:00

7.9 KiB
Raw Blame History

vpkuwus — Vector Pack Unsigned Word Unsigned Saturate

Category: VMX (Altivec) · Form: VX · Opcode: 0x100000ce

Assembler Mnemonics

Mnemonic XML entry Flags Description
vpkuwus vpkuwus Vector Pack Unsigned Word Unsigned Saturate
vpkuwus128 vpkuwus128 Vector128 Pack Unsigned Word Unsigned Saturate

Syntax

vpkuwus [VD], [VA], [VB]
vpkuwus128 [VD], [VA], [VB]

Encoding

vpkuwus — form VX

  • Opcode word: 0x100000ce
  • Primary opcode (bits 05): 4
  • Extended opcode: 206
  • Synchronising: no
Bits Field Meaning
05 OPCD primary opcode (4)
610 VRT/VD destination vector register
1115 VRA/VA source A vector register
1620 VRB/VB source B vector register
2131 XO extended opcode (11 bits)

vpkuwus128 — form VX128

  • Opcode word: 0x140003c0
  • Primary opcode (bits 05): 5
  • Extended opcode: 960
  • Synchronising: no
Bits Field Meaning
05 OPCD primary opcode (4 or 5)
610 VD128l destination low 5 bits
1115 VA128l source A low 5 bits
1620 VB128l source B low 5 bits
21 VA128H source A high bit
22 reserved
2325 VC optional VC / XO sub-field
26 VA128h source A middle bit
27 reserved
2829 VD128h destination high 2 bits
3031 VB128h source B high 2 bits

Operands

Field Role Description
VA vpkuwus: read; vpkuwus128: read Source A vector register.
VB vpkuwus: read; vpkuwus128: read Source B vector register.
VD vpkuwus: write; vpkuwus128: write Destination vector register.
VSCR vpkuwus: write; vpkuwus128: write Vector Status and Control Register (NJ/SAT bits).

Register Effects

vpkuwus

  • Reads (always): VA, VB
  • Reads (conditional): none
  • Writes (always): VD, VSCR
  • Writes (conditional): none

vpkuwus128

  • Reads (always): VA, VB
  • Reads (conditional): none
  • Writes (always): VD, VSCR
  • Writes (conditional): none

Status-Register Effects

  • vpkuwus: VSCR[SAT] may be stickied on saturating vector operations.
  • vpkuwus128: VSCR[SAT] may be stickied on saturating vector operations.

Operation (pseudocode)

; No hand-written pseudocode for this instruction yet.
; The authoritative semantics are the Canary emitter snapshot under
; Implementation References; about half of Canary's emitters open
; with the PPC-style definition as a comment (`RD <- (RA) + (RB)`).
; Every side effect is also enumerated in the Register Effects and
; Status-Register Effects tables above.

C Translation Example

/* No hand-written C yet. Translate the Canary emitter snapshot   */
/* under Implementation References; its HIR maps directly:        */
/*   f.LoadGPR(n) / f.StoreGPR(n, v)  -> r[n] / r[n] = v          */
/*   f.LoadFPR / StoreFPR, f.LoadVR / StoreVR -> f[n], v[n]        */
/*   f.Load(ea, T), f.Store(ea, v) -> raw read / write; emitters   */
/*     wrap them in f.ByteSwap for the big-endian guest value      */
/*   f.UpdateCR(n, v)  -> CR field n from v's LOW 32 BITS vs 0     */
/*   f.LoadCA / f.StoreCA -> xer.CA;  f.StoreSAT -> vscr.SAT       */
/*   i.XO.RA, i.D.DS, ... -> the bit-fields listed under Operands  */
/* The Register Effects and Status-Register Effects tables above  */
/* enumerate every side effect a faithful translation must emit.  */

Implementation References

vpkuwus

Canary emitter (frozen snapshot @ f21ebd49e9)
int InstrEmit_vpkuwus(PPCHIRBuilder& f, const InstrData& i) {
  return InstrEmit_vpkuwus_(f, i.VX.VD, i.VX.VA, i.VX.VB);
}

// ── delegates to (src/xenia/cpu/ppc/ppc_emit_altivec.cc:1966) ──
int InstrEmit_vpkuwus_(PPCHIRBuilder& f, uint32_t vd, uint32_t va,
                       uint32_t vb) {
  // Vector Pack Unsigned Word Unsigned Saturate
  // Convert VA and VB from unsigned int words to unsigned saturated shorts then
  // concat:
  // for each i in VA + VB:
  //   i = uint16_t(Clamp(EXTZ(uint32_t(t)), 0, 2^16-1))
  // dest = VA | VB (lower 16bit values)
  Value* v = f.Pack(f.LoadVR(va), f.LoadVR(vb),
                    PACK_TYPE_16_IN_32 | PACK_TYPE_IN_UNSIGNED |
                        PACK_TYPE_OUT_UNSIGNED | PACK_TYPE_OUT_SATURATE);
  f.StoreSAT(f.DidSaturate(v));
  f.StoreVR(vd, v);
  return 0;
}

vpkuwus128

Canary emitter (frozen snapshot @ f21ebd49e9)
int InstrEmit_vpkuwus128(PPCHIRBuilder& f, const InstrData& i) {
  return InstrEmit_vpkuwus_(f, VX128_VD128, VX128_VA128, VX128_VB128);
}

// ── delegates to (src/xenia/cpu/ppc/ppc_emit_altivec.cc:1966) ──
int InstrEmit_vpkuwus_(PPCHIRBuilder& f, uint32_t vd, uint32_t va,
                       uint32_t vb) {
  // Vector Pack Unsigned Word Unsigned Saturate
  // Convert VA and VB from unsigned int words to unsigned saturated shorts then
  // concat:
  // for each i in VA + VB:
  //   i = uint16_t(Clamp(EXTZ(uint32_t(t)), 0, 2^16-1))
  // dest = VA | VB (lower 16bit values)
  Value* v = f.Pack(f.LoadVR(va), f.LoadVR(vb),
                    PACK_TYPE_16_IN_32 | PACK_TYPE_IN_UNSIGNED |
                        PACK_TYPE_OUT_UNSIGNED | PACK_TYPE_OUT_SATURATE);
  f.StoreSAT(f.DidSaturate(v));
  f.StoreVR(vd, v);
  return 0;
}

Special Cases & Edge Conditions

  • Unsigned word → unsigned half-word saturating pack. Each of the 8 input word lanes (interpreted as uint32) is clamped to [0, 65535]. Overflow sticky-sets VSCR[SAT].
  • Lane-count doubling. 8 word lanes → 8 half-word lanes; VA then VB.
  • Pair with vpkuwum when a modulo wrap is required.
  • VSCR[SAT] is sticky.
  • No Rc, no XER.
  • VMX128 sibling vpkuwus128.

IBM Reference