ia32-64/x86/vcvtuqq2ps.html
2025-07-08 02:23:29 -03:00

154 lines
6.6 KiB
HTML
Raw Permalink Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

<!DOCTYPE html>
<html xmlns="http://www.w3.org/1999/xhtml" xmlns:svg="http://www.w3.org/2000/svg" xmlns:x86="http://www.felixcloutier.com/x86"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8"><link rel="stylesheet" type="text/css" href="style.css"></link><title>VCVTUQQ2PS
— Convert Packed Unsigned Quadword Integers to Packed Single PrecisionFloating-Point Values</title></head><body><header><nav><ul><li><a href='index.html'>Index</a></li><li>December 2023</li></ul></nav></header><h1>VCVTUQQ2PS
— Convert Packed Unsigned Quadword Integers to Packed Single PrecisionFloating-Point Values</h1>
<table>
<tr>
<th>Opcode/Instruction</th>
<th>Op/En</th>
<th>64/32 Bit Mode Support</th>
<th>CPUID Feature Flag</th>
<th>Description</th></tr>
<tr>
<td>EVEX.128.F2.0F.W1 7A /r VCVTUQQ2PS xmm1 {k1}{z}, xmm2/m128/m64bcst</td>
<td>A</td>
<td>V/V</td>
<td>AVX512VL AVX512DQ</td>
<td>Convert two packed unsigned quadword integers from xmm2/m128/m64bcst to packed single precision floating-point values in zmm1 with writemask k1.</td></tr>
<tr>
<td>EVEX.256.F2.0F.W1 7A /r VCVTUQQ2PS xmm1 {k1}{z}, ymm2/m256/m64bcst</td>
<td>A</td>
<td>V/V</td>
<td>AVX512VL AVX512DQ</td>
<td>Convert four packed unsigned quadword integers from ymm2/m256/m64bcst to packed single precision floating-point values in xmm1 with writemask k1.</td></tr>
<tr>
<td>EVEX.512.F2.0F.W1 7A /r VCVTUQQ2PS ymm1 {k1}{z}, zmm2/m512/m64bcst{er}</td>
<td>A</td>
<td>V/V</td>
<td>AVX512DQ</td>
<td>Convert eight packed unsigned quadword integers from zmm2/m512/m64bcst to eight packed single precision floating-point values in zmm1 with writemask k1.</td></tr></table>
<h2 id="instruction-operand-encoding">Instruction Operand Encoding<a class="anchor" href="#instruction-operand-encoding">
</a></h2>
<table>
<tr>
<th>Op/En</th>
<th>Tuple Type</th>
<th>Operand 1</th>
<th>Operand 2</th>
<th>Operand 3</th>
<th>Operand 4</th></tr>
<tr>
<td>A</td>
<td>Full</td>
<td>ModRM:reg (w)</td>
<td>ModRM:r/m (r)</td>
<td>N/A</td>
<td>N/A</td></tr></table>
<h3 id="description">Description<a class="anchor" href="#description">
</a></h3>
<p>Converts packed unsigned quadword integers in the source operand (second operand) to single precision floating-point values in the destination operand (first operand).</p>
<p>EVEX encoded versions: The source operand is a ZMM/YMM/XMM register or a 512/256/128-bit memory location. The destination operand is a YMM/XMM/XMM (low 64 bits) register conditionally updated with writemask k1.</p>
<p>Note: EVEX.vvvv is reserved and must be 1111b, otherwise instructions will #UD.</p>
<h3 id="operation">Operation<a class="anchor" href="#operation">
</a></h3>
<h4 id="vcvtuqq2ps--evex-encoded-version--when-src-operand-is-a-register">VCVTUQQ2PS (EVEX Encoded Version) When SRC Operand is a Register<a class="anchor" href="#vcvtuqq2ps--evex-encoded-version--when-src-operand-is-a-register">
</a></h4>
<pre>(KL, VL) = (2, 128), (4, 256), (8, 512)
IF (VL = 512) AND (EVEX.b = 1)
THEN
SET_ROUNDING_MODE_FOR_THIS_INSTRUCTION(EVEX.RC);
ELSE
SET_ROUNDING_MODE_FOR_THIS_INSTRUCTION(MXCSR.RC);
FI;
FOR j := 0 TO KL-1
i := j * 32
k := j * 64
IF k1[j] OR *no writemask*
THEN DEST[i+31:i] :=
Convert_UQuadInteger_To_Single_Precision_Floating_Point(SRC[k+63:k])
ELSE
IF *merging-masking* ; merging-masking
THEN *DEST[i+31:i] remains unchanged*
ELSE ; zeroing-masking
DEST[i+31:i] := 0
FI
FI;
ENDFOR
DEST[MAXVL-1:VL/2] := 0
</pre>
<h4 id="vcvtuqq2ps--evex-encoded-version--when-src-operand-is-a-memory-source">VCVTUQQ2PS (EVEX Encoded Version) When SRC Operand is a Memory Source<a class="anchor" href="#vcvtuqq2ps--evex-encoded-version--when-src-operand-is-a-memory-source">
</a></h4>
<pre>(KL, VL) = (2, 128), (4, 256), (8, 512)
FOR j := 0 TO KL-1
i := j * 32
k := j * 64
IF k1[j] OR *no writemask*
THEN
IF (EVEX.b = 1)
THEN
DEST[i+31:i] :=
Convert_UQuadInteger_To_Single_Precision_Floating_Point(SRC[63:0])
ELSE
DEST[i+31:i] :=
Convert_UQuadInteger_To_Single_Precision_Floating_Point(SRC[k+63:k])
FI;
ELSE
IF *merging-masking* ; merging-masking
THEN *DEST[i+31:i] remains unchanged*
ELSE ; zeroing-masking
DEST[i+31:i] := 0
FI
FI;
ENDFOR
DEST[MAXVL-1:VL/2] := 0
</pre>
<h3 id="intel-c-c++-compiler-intrinsic-equivalent">Intel C/C++ Compiler Intrinsic Equivalent<a class="anchor" href="#intel-c-c++-compiler-intrinsic-equivalent">
</a></h3>
<pre>VCVTUQQ2PS __m256 _mm512_cvtepu64_ps( __m512i a);
</pre>
<pre>VCVTUQQ2PS __m256 _mm512_mask_cvtepu64_ps( __m256 s, __mmask8 k, __m512i a);
</pre>
<pre>VCVTUQQ2PS __m256 _mm512_maskz_cvtepu64_ps( __mmask8 k, __m512i a);
</pre>
<pre>VCVTUQQ2PS __m256 _mm512_cvt_roundepu64_ps( __m512i a, int r);
</pre>
<pre>VCVTUQQ2PS __m256 _mm512_mask_cvt_roundepu64_ps( __m256 s, __mmask8 k, __m512i a, int r);
</pre>
<pre>VCVTUQQ2PS __m256 _mm512_maskz_cvt_roundepu64_ps( __mmask8 k, __m512i a, int r);
</pre>
<pre>VCVTUQQ2PS __m128 _mm256_cvtepu64_ps( __m256i a);
</pre>
<pre>VCVTUQQ2PS __m128 _mm256_mask_cvtepu64_ps( __m128 s, __mmask8 k, __m256i a);
</pre>
<pre>VCVTUQQ2PS __m128 _mm256_maskz_cvtepu64_ps( __mmask8 k, __m256i a);
</pre>
<pre>VCVTUQQ2PS __m128 _mm_cvtepu64_ps( __m128i a);
</pre>
<pre>VCVTUQQ2PS __m128 _mm_mask_cvtepu64_ps( __m128 s, __mmask8 k, __m128i a);
</pre>
<pre>VCVTUQQ2PS __m128 _mm_maskz_cvtepu64_ps( __mmask8 k, __m128i a);
</pre>
<h3 class="exceptions" id="simd-floating-point-exceptions">SIMD Floating-Point Exceptions<a class="anchor" href="#simd-floating-point-exceptions">
</a></h3>
<p>Precision.</p>
<h3 class="exceptions" id="other-exceptions">Other Exceptions<a class="anchor" href="#other-exceptions">
</a></h3>
<p>EVEX-encoded instructions, see <span class="not-imported">Table 2-46</span>, “Type E2 Class Exception Conditions.”</p>
<p>Additionally:</p>
<table>
<tr>
<td>#UD</td>
<td>If EVEX.vvvv != 1111B.</td></tr></table><footer><p>
This UNOFFICIAL, mechanically-separated, non-verified reference is provided for convenience, but it may be
inc<span style="opacity: 0.2">omp</span>lete or b<sub>r</sub>oke<sub>n</sub> in various obvious or non-obvious
ways. Refer to <a href="https://software.intel.com/en-us/download/intel-64-and-ia-32-architectures-sdm-combined-volumes-1-2a-2b-2c-2d-3a-3b-3c-3d-and-4">Intel® 64 and IA-32 Architectures Software Developers Manual</a> for anything serious.
</p></footer></body></html>