Instruction
PABSD
Computes the absolute values of packed signed doublewords of an MMX or XMM operand; SSSE3.
- Extensions
- Forms
- 4 (4 Legacy)
- Data
- Intel XED v2026.08.23
PABSD replaces each doubleword with its absolute value: 4 results in
an XMM register. It is the SSSE3 form of VPABSD,
whose page covers the most negative doubleword, which stays negative
because it has no positive 32-bit counterpart, and the sign-mask
sequence that stands in for it without SSSE3. Unlike VPABSD, PABSD
also has a 64-bit MMX form: on Zen 5, the recorded pabsd mm3, qword ptr [mem] turned 7FFFFFFFh into itself and left 80000000h as
80000000h in MM3, the same edge VPABSD’s page describes.
What the legacy form adds
The 128-bit form is destructive, with the second operand an XMM
register or a 128-bit memory location that must be aligned to 16
bytes, and it leaves bits 255:128 of the register as they were,
unlike the VEX form, which zeros them (see mixing SSE and
AVX). SSSE3 is part of
x86-64-v2. In our builds at -O2, the sign-mask sequence
VPABSD’s page shows for the baseline compiles the same way whether
the code is written for PABSD or its VEX form; only
-march=x86-64-v2 or -mssse3 turns it into the single instruction.
AMD’s latency spreadsheet for Zen 5 gives PABSD, like VPABSD, one macro-op with a latency of two cycles on two of the four vector pipelines; the MMX form has no latency recorded, only a throughput of two.
Forms
| Instruction | Encoding | Requires |
|---|---|---|
PABSD mm, m64
PABSD_MMXq_MEMq | NP 0F 38 1E /r | SSSE3 + MMX |
PABSD mm, mm
PABSD_MMXq_MMXq | NP 0F 38 1E /r | SSSE3 + MMX |
PABSD xmm, m128
PABSD_XMMdq_MEMdq | 66 0F 38 1E /r | SSSE3 |
PABSD xmm, xmm
PABSD_XMMdq_XMMdq | 66 0F 38 1E /r | SSSE3 |
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.
Every doubleword works alike: 2 and 4 doublewords in the mm and xmm forms.
Result doubleword i comes from doubleword i of xmm2. Result doublewords 3 to 0: |a3|, |a2|, |a1|, |a0|.
Operation
What the forms do, in pseudocode that our tests run against results recorded on real processors. How to read.
PABSD mm, mm/m64for i in range(2):
mm1.d[i] = abs(src2.sd[i])
# x87: TOP = 0, all tags valid, bits 79:64 of mm1's register all onesRun on AMD Zen 5 (EPYC 9645): 2 cases.
PABSD xmm, xmm/m128for i in range(4):
xmm1.d[i] = abs(src2.sd[i])
# bits 128 and up of the register are unchangedRun on AMD Zen 5 (EPYC 9645): 2 cases.
Operands
| Operand | Kind | Action |
|---|---|---|
PABSD mm, mm/m64 | ||
mm | explicit | write |
mm/m64 | explicit | read |
PABSD xmm, xmm/m128 | ||
xmm | explicit | write |
xmm/m128 | 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.
Attributes
What else XED records about the forms. How to read.
PABSD mm, mm/m64- may abort a TSX transaction
PABSD xmm, m128- aligned memory operand
Intrinsics
The C intrinsics of Clang 23.1.2 that compile to PABSD, 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.
| Intrinsic | Compiles to |
|---|---|
_mm_abs_epi32
__m128i _mm_abs_epi32(__m128i __a) |
|
_mm_abs_pi32
__m64 _mm_abs_pi32(__m64 __a) |
|
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.
More SSSE3 instructions
Sources
- Intel XED v2026.08.23 (commit
0bcb6237345c): forms, encodings, operands, flags, exception classes, attributes, ISA sets, CPUID bits and chip model. - The CPU data of x86-64.net: Zen 3 to Zen 6, from GCC’s
-marchfeature sets and AMD’s Processor Programming References, which the page of each processor cites. - 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. - Intel 64 and IA-32 Architectures Software Developer's Manual, Volume 2 (325383-093, September 2026): Volume 2B, page 4-177 (PABSB/
PABSW/ PABSD/ PABSQ) - Software Optimization Guide for the AMD Zen5 Microarchitecture (58455, rev. 1.00, August 2024), Wayback Machine copy of AMD's ZIP
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.