Instruction
TOP4BUSD
ACE outer product of a ZMM register of unsigned bytes with one of signed bytes, four per dword, accumulated into a tile of 32-bit integers.
- Extensions
- Forms
- 1 (1 EVEX)
- Data
- Intel XED v2026.08.23
TOP4BUSD is TOP4BSSD with unsigned bytes in its first source and signed bytes in its second: element (i, j) of the tile gains the four products of the unsigned bytes of dword i of zmm2 with the signed bytes of dword j of zmm3, added modulo 2^32. The order is that of AMX’s TDPBUSD and of VPDPBUSD, the original VNNI instruction: unsigned first, signed second.
The transpose of TOP4BSUD
Both sources of an ACE outer product have the same layout, 16 dwords of
four bytes, one dword for each row or each column. Swapping them
therefore transposes the update: top4busd tmm0, zmm1, zmm2 adds to
element (i, j) exactly what top4bsud tmm0, zmm2, zmm1 adds to element
(j, i). The choice between TOP4BSUD and TOP4BUSD
decides only whether the unsigned matrix runs along the rows or along
the columns of the result, and the registers stay as they were loaded.
With AMX, swapping TDPBUSD for TDPBSUD means repacking both matrices, as
TDPBSUD’s page explains.
The products span the same range as TOP4BSUD’s, -32,640 to 32,385, so an accumulator that starts at zero again stays inside the signed 32-bit range for 16,448 instructions. Nothing saturates: a longer sum wraps around silently.
In our builds
GNU as 2.44 and Clang 23.1.2 reject the mnemonic, and neither GCC 14.2.0
nor Clang 23.1.2 has ACE’s _tile_top4busd intrinsic. The encoding is
EVEX opcode 5Eh in map 0F38 with W0 and the 66 prefix, top4busd tmm0, zmm1, zmm2 being 62 F2 6D 48 5E C1 by the specification’s encoding,
which objdump 2.44 decodes as (bad); the exception class is
ACE-E4. No processor at hand has ACE, so
this site’s recorded cases include none of TOP4BUSD.
Forms
| Instruction | Encoding | Requires |
|---|---|---|
TOP4BUSD tmm, zmm, zmm 64-bit mode only
TOP4BUSD_TMMu32_ZMMu8_ZMMi8 | EVEX.512.66.0F38.W0 5E /r | ACE v1 |
Operation
What the forms do, in pseudocode that our tests run against results recorded on real processors. How to read.
on = xcr0 if cr4 >> 18 & 1 else 0 # XCR0, where CR4.OSXSAVE is set
if on >> 17 & 0xB != 0xB: # TILECFG, TILEDATA, SCALEDATA
fault(UD)
if on & 0xE6 != 0xE6: # AVX-512's state, AVX's and
fault(UD) # SSE's
if palette != 2: # ACE's palette alone
fault(UD)
if cr0 >> 3 & 1 or xfd >> 18 & 1: # CR0.TS, or the OS keeps the
fault(NM) # tiles from this thread
for i in range(16): # row i: dword i of zmm2
for j in range(16): # column j: dword j of zmm3
s = tmm1.row[i].d[j]
for k in range(4):
a = zmm2.b[4 * i + k]
s += a * zmm3.sb[4 * j + k]
tmm1.row[i].d[j] = s # modulo 2^32
Not run on a processor: none of these processors has ACE.
Operands
| Operand | Kind | Action |
|---|---|---|
TOP4BUSD tmm, zmm, zmm 64-bit mode only | ||
tmm | explicit | read, write |
zmm | explicit | read |
zmm | explicit | read |
Flags
None: XED records no flag that the forms test or change.
Exceptions
The exception class is XED's name for the set of exception conditions that a form follows. How to read.
- All forms
ACE-E4
Attributes
What else XED records about the forms. How to read.
- All forms
- may abort a TSX transaction
CPUID and processors
Each form belongs to an XED ISA set. A form can be used when the processor reports every CPUID bit of one of its ISA set's alternatives.
| ISA set | CPUID | Processors |
|---|---|---|
ACE_1 | ACE v1 | None in XED’s chip model, which lists these instructions only under its placeholder for future processors. |
More ACE instructions
Sources
- Intel XED v2026.08.23 (commit
0bcb6237345c): forms, encodings, operands, flags, exception classes, attributes, ISA sets, CPUID bits and chip model. - x86 Ecosystem Advisory Group: AI Compute Extensions (ACE) Specification, version 1.16.2 (July 2026)
The tables are derived from Intel XED, Copyright Intel Corporation, licensed under the Apache License 2.0; x86-64.net converted and reformatted them. The text is our own.