Class
AVX512-E6 in Intel XED
Applies to
EVEX encodings, 128 to 512 bits.
Instructions
37 mnemonics, 125 forms
Data
Intel XED v2026.08.23

Exception conditions

When an instruction of the class raises each exception, and in which processor modes. Protected mode includes compatibility mode. How to read the table.
ExceptionRealVirtual-8086Protected64-bitCondition
#UDyesyesnonoAlways: real and virtual-8086 mode have no EVEX prefix.
#UDnonoyesyesThe operating system has not enabled the state the instruction uses: CR4.OSXSAVE is 0, or XCR0 lacks one of the bits Table 2-39 requires (for EVEX, bits 1 and 2 and the opmask and ZMM bits 5 to 7).
#UDnonoyesyesAn EVEX field has a value that the manual’s general rules forbid: a bit with a fixed value or the opcode map (Table 2-40), a register field (Table 2-41), the opmask and zeroing fields (Table 2-42), EVEX.b where the instruction gives it no meaning (Table 2-43), or a vector length (EVEX.L’L) that the instruction lacks.
#UDnonoyesyesThe instruction has a LOCK prefix (F0).
#UDnonoyesyesA REX, F2, F3 or 66 prefix comes before the EVEX prefix.
#UDnonoyesyesThe processor lacks the instruction’s extension: its CPUID feature flag is 0.
#NMnonoyesyesCR0.TS (task switched, bit 3) is 1.
#SS(0)nonoyesnoAn element that the mask enables lies outside the stack segment (a memory operand addressed through SS).
#SS(0)nononoyesAn element that the mask enables has a non-canonical address (a memory operand addressed through SS).
#GP(0)nonoyesnoAn element that the mask enables has an effective address that its segment (CS, DS, ES, FS or GS) does not allow, such as one beyond the limit.
#GP(0)nononoyesAn element that the mask enables has a non-canonical address.
#PF(fault-code)nonoyesyesAccessing an element that the mask enables causes a page fault.
#AC(0)nonoyesyesCode at privilege level 3 makes a misaligned 2-, 4- or 8-byte memory access while alignment checking is on.

Notes

The manual’s table lists these conditions for protected and 64-bit mode only. Memory faults come only from the elements that the mask enables.

XED puts here the forms of the down-converting moves VPMOVDB to VPMOVUSWB that store to memory, and their forms with a register destination in AVX512-E6NF; the manual’s pages refer both to Type E6 and add #UD for an EVEX.vvvv other than 1111b. The broadcasts from a general-purpose register are AVX512-E7NM. The manual also lists VFPCLASSSD and VFPCLASSSS here, which XED puts in AVX512-E10.

Instructions

Intel XED puts forms of 37 mnemonics in this class, listed here by extension. Where a mnemonic also has forms of other classes, its link leads to its first form of this one.

AVX10.1

AVX10_V2_AUX

SM4

Sources

  1. Intel 64 and IA-32 Architectures Software Developer's Manual, Volume 2 (325383-093, September 2026): Section 2.8.6, Table 2-55; Section 2.8, Tables 2-44 and 2-45; Section 2.7.11, Tables 2-39 to 2-43; Chapter 5, VPMOVDB and the other down-converting moves.
  2. Intel 64 and IA-32 Architectures Software Developer's Manual, Volume 1 (253665-093, September 2026): Section 15.6.1.
  3. Intel XED v2026.08.23 (commit 0bcb6237345c): the exception class of each instruction form.

The list of instructions is derived from Intel XED, Copyright Intel Corporation, licensed under the Apache License 2.0; x86-64.net converted and reformatted it. The text is our own.