Extensions
Forms
6 (6 EVEX)
Data
Intel XED v2026.08.23

VCVTTPS2UQQS converts singles to unsigned quadwords, truncating toward zero, and saturates: 2^64 or more becomes all ones, -1 or less becomes 0, and a NaN becomes 0, where VCVTTPS2UQQ gives all ones, the unsigned indefinite, for each of them. The result is twice as wide as the source: two, four or eight singles from the low half of a register or from 8, 16 or 32 bytes of memory. It is the unsigned partner of VCVTTPS2QQS: opcode 6Ch in EVEX map 5 with a 66 prefix and W0, part of AVX10.2, with a write mask, a broadcast single and {sae} on the 512-bit form. NaNs, infinities and out-of-range values still set IE, and a lost fraction sets PE.

In our builds

GCC 14 has no AVX10.2 option. Clang 23 with -mavx10.2 compiles _mm_cvtts_ps_epu64 to vcvttps2uqqs xmm0, xmm0, two singles in and two quadwords out; a loop converting floats to uint64_t keeps VCVTTPS2UQQ. No AVX10.2 processor appears in this site’s recorded cases.

Forms

Every encoding of VCVTTPS2UQQS that Intel XED knows, one row per XED instruction form (iform). How to read the notation.
InstructionEncodingRequires
VCVTTPS2UQQS xmm {k}{z}, xmm VCVTTPS2UQQS_XMMu64_MASKmskw_XMMf32_AVX512EVEX.128.66.MAP5.W0 6C /rAVX10.2
VCVTTPS2UQQS xmm {k}{z}, m64/m32bcst VCVTTPS2UQQS_XMMu64_MASKmskw_MEMf32_AVX512EVEX.128.66.MAP5.W0 6C /rAVX10.2
VCVTTPS2UQQS ymm {k}{z}, xmm VCVTTPS2UQQS_YMMu64_MASKmskw_XMMf32_AVX512EVEX.256.66.MAP5.W0 6C /rAVX10.2
VCVTTPS2UQQS ymm {k}{z}, m128/m32bcst VCVTTPS2UQQS_YMMu64_MASKmskw_MEMf32_AVX512EVEX.256.66.MAP5.W0 6C /rAVX10.2
VCVTTPS2UQQS zmm {k}{z}, ymm, {sae} VCVTTPS2UQQS_ZMMu64_MASKmskw_YMMf32_AVX512EVEX.512.66.MAP5.W0 6C /rAVX10.2
VCVTTPS2UQQS zmm {k}{z}, m256/m32bcst VCVTTPS2UQQS_ZMMu64_MASKmskw_MEMf32_AVX512EVEX.512.66.MAP5.W0 6C /rAVX10.2

Data flow

Where the elements of the sources go, element 0 on the right; results that depend on a control value are drawn for an example. How to read.

Widen: each single becomes a quadword, converted, in the same order.
VCVTTPS2UQQS: each element widenedResult quadwords 7 to 0: a7, unchanged (bit 6 of k1 is 0), a5, a4, a3, a2, a1, a0.01234567ymm2or m256a11111110k1a0a1a2a3a4a5a7zmm1

Shown for the zmm form; the xmm and ymm forms work the same on 2 and 4 singles. The forms write under a mask: where its bit is 0, the element keeps its old value, or zeroes it with {z}.

Result quadwords 7 to 0: a7, unchanged (bit 6 of k1 is 0), a5, a4, a3, a2, a1, a0.

Operation

What the forms do, in pseudocode that our tests run against results recorded on real processors. How to read.

for i in range(VL // 64):
    if k1[i]:
        a = src2.f32[i]
        zmm1.q[i] = cvtts_uint(a, 64)
    elif zeroing:
        zmm1.q[i] = 0
    # else: zmm1.q[i] keeps its value
# m32bcst: src2.f32[i] is the m32 for every i
# {sae}: no exception flags
# zmm1 is zeroed from bit VL up

Not run on a processor: none of these processors has AVX10.2.

Operands

What each form reads and writes, as Intel XED records it: the operands of its syntax, in order, and the implicit ones that the syntax does not name. How to read.
OperandKindAction
VCVTTPS2UQQS xmm {k}{z}, xmm/m64/m32bcst VCVTTPS2UQQS ymm {k}{z}, xmm/m128/m32bcst VCVTTPS2UQQS zmm {k}{z}, ymm/m256/m32bcst, {sae}
xmm, ymm, zmmexplicitwrite (masked)
{k}explicitread (write mask)
xmm/m64/m32bcst, xmm/m128/m32bcst, ymm/m256/m32bcstexplicitread

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
AVX512-E2

Attributes

What else XED records about the forms. How to read.

All forms
write masking; follows MXCSR
Memory forms
embedded broadcast (tuple Half); memory fault suppression

Intrinsics

The C intrinsics of Clang 23.1.2 that compile to VCVTTPS2UQQS, each with the form it compiles to and the Clang options it needs. The forms come from compiling every intrinsic and decoding the result with Intel XED; all intrinsics A–Z explains the method.

IntrinsicCompiles to
_mm_cvtts_ps_epu64 __m128i _mm_cvtts_ps_epu64(__m128 __A)

VCVTTPS2UQQS xmm, xmm EVEX.128 -mavx10.2

_mm_mask_cvtts_ps_epu64 __m128i _mm_mask_cvtts_ps_epu64(__m128i __W, __mmask8 __U, __m128 __A)

VCVTTPS2UQQS xmm {k}, xmm EVEX.128 -mavx10.2

_mm_maskz_cvtts_ps_epu64 __m128i _mm_maskz_cvtts_ps_epu64(__mmask8 __U, __m128 __A)

VCVTTPS2UQQS xmm {k}{z}, xmm EVEX.128 -mavx10.2

_mm256_cvtts_ps_epu64 __m256i _mm256_cvtts_ps_epu64(__m128 __A)

VCVTTPS2UQQS ymm, xmm EVEX.256 -mavx10.2

_mm256_mask_cvtts_ps_epu64 __m256i _mm256_mask_cvtts_ps_epu64(__m256i __W, __mmask8 __U, __m128 __A)

VCVTTPS2UQQS ymm {k}, xmm EVEX.256 -mavx10.2

_mm256_maskz_cvtts_ps_epu64 __m256i _mm256_maskz_cvtts_ps_epu64(__mmask8 __U, __m128 __A)

VCVTTPS2UQQS ymm {k}{z}, xmm EVEX.256 -mavx10.2

_mm512_cvtts_ps_epu64 __m512i _mm512_cvtts_ps_epu64(__m256 __A)

VCVTTPS2UQQS zmm, ymm EVEX.512 -mavx10.2

_mm512_cvtts_roundps_epu64 __m512i _mm512_cvtts_roundps_epu64(__m256 __A, const int __R)

VCVTTPS2UQQS zmm, ymm, {sae} EVEX.512 -mavx10.2

_mm512_mask_cvtts_ps_epu64 __m512i _mm512_mask_cvtts_ps_epu64(__m512i __W, __mmask8 __U, __m256 __A)

VCVTTPS2UQQS zmm {k}, ymm EVEX.512 -mavx10.2

_mm512_mask_cvtts_roundps_epu64 __m512i _mm512_mask_cvtts_roundps_epu64(__m512i __W, __mmask8 __U, __m256 __A, const int __R)

VCVTTPS2UQQS zmm {k}, ymm, {sae} EVEX.512 -mavx10.2

_mm512_maskz_cvtts_ps_epu64 __m512i _mm512_maskz_cvtts_ps_epu64(__mmask8 __U, __m256 __A)

VCVTTPS2UQQS zmm {k}{z}, ymm EVEX.512 -mavx10.2

_mm512_maskz_cvtts_roundps_epu64 __m512i _mm512_maskz_cvtts_roundps_epu64(__mmask8 __U, __m256 __A, const int __R)

VCVTTPS2UQQS zmm {k}{z}, ymm, {sae} EVEX.512 -mavx10.2

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.

CPUID requirements and processors of the XED ISA sets above. Alternatives are separated by “or”; processors follow XED’s chip model, whose AMD processors end with Zen 2, and the CPU data of this site for Zen 3 to Zen 6.
ISA setCPUIDProcessors
AVX512_SAT_CVT_DS_128

AVX10.2
CPUID.(EAX=07H,ECX=1):EDX[19] AVX10
CPUID.(EAX=24H,ECX=0):EBX[7:0] ≥ 2 AVX10 version

Nova Lake, Diamond Rapids
AVX512_SAT_CVT_DS_256
Same as AVX512_SAT_CVT_DS_128Same as AVX512_SAT_CVT_DS_128
AVX512_SAT_CVT_DS_512
Same as AVX512_SAT_CVT_DS_128Same as AVX512_SAT_CVT_DS_128

More AVX10.2 instructions

All 96 AVX10.2 instructions

Sources

  1. Intel XED v2026.08.23 (commit 0bcb6237345c): forms, encodings, operands, flags, exception classes, attributes, ISA sets, CPUID bits and chip model.
  2. Clang's x86 intrinsic headers, LLVM 23.1.2 (commit 85ac56026243): the intrinsics, their signatures and doc comments, and the Clang of that release, which compiled each of them.
  3. Intel Advanced Vector Extensions 10.2 Architecture Specification, revision 7.0 (361050-007, April 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 intrinsics and their signatures come from Clang, part of the LLVM Project, licensed under the Apache License 2.0 with LLVM Exceptions. The text is our own.