blob: 10586a64b617fb6637e06812852a9ebb9214a580 [file] [log] [blame] [raw]
<!DOCTYPE html>
<html>
<head>
<meta charset="UTF-8">
<link href="style.css" type="text/css" rel="stylesheet">
<title>PSHUFD—Shuffle Packed Doublewords </title></head>
<body>
<h1>PSHUFD—Shuffle Packed Doublewords</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>
<p>66 0F 70 /<em>r</em> ib</p>
<p>PSHUFD <em>xmm1</em>, <em>xmm2/m128</em>, <em>imm8</em></p></td>
<td>RMI</td>
<td>V/V</td>
<td> SSE2</td>
<td>Shuffle the doublewords in <em>xmm2/m128 </em>based on the encoding in <em>imm8</em> and store the result in <em>xmm1</em>.</td></tr>
<tr>
<td>
<p>VEX.128.66.0F.WIG 70 /r ib</p>
<p>VPSHUFD <em>xmm1, xmm2/m128, imm8</em></p></td>
<td>RMI</td>
<td>V/V</td>
<td>AVX</td>
<td>Shuffle the doublewords in <em>xmm2/m128 </em>based on the encoding in <em>imm8</em> and store the result in <em>xmm1</em>.</td></tr>
<tr>
<td>
<p>VEX.256.66.0F.WIG 70 /r ib</p>
<p>VPSHUFD <em>ymm1, ymm2/m256, imm8</em></p></td>
<td>RMI</td>
<td>V/V</td>
<td>AVX2</td>
<td>Shuffle the doublewords in <em>ymm2/m256 </em>based on the encoding in <em>imm8</em> and store the result in <em>ymm1</em>.</td></tr></table>
<h3>Instruction Operand Encoding</h3>
<table>
<tr>
<td>Op/En</td>
<td>Operand 1</td>
<td>Operand 2</td>
<td>Operand 3</td>
<td>Operand 4</td></tr>
<tr>
<td>RMI</td>
<td>ModRM:reg (w)</td>
<td>ModRM:r/m (r)</td>
<td>imm8</td>
<td>NA</td></tr></table>
<h2>Description</h2>
<p>Copies doublewords from source operand (second operand) and inserts them in the destination operand (first operand) at the locations selected with the order operand (third operand). Figure 4-12 shows the operation of the 256-bit VPSHUFD instruction and the encoding of the order operand. Each 2-bit field in the order operand selects the contents of one doubleword location within a 128-bit lane and copy to the target element in the destination operand. For example, bits 0 and 1 of the order operand targets the first doubleword element in the low and high 128-bit lane of the destination operand for 256-bit VPSHUFD. The encoded value of bits 1:0 of the order operand (see the field encoding in Figure 4-12) determines which doubleword element (from the respective 128-bit lane) of the source operand will be copied to doubleword 0 of the destination operand.</p>
<p>For 128-bit operation, only the low 128-bit lane are operative. The source operand can be an XMM register or a 128-bit memory location. The destination operand is an XMM register. The order operand is an 8-bit immediate. Note that this instruction permits a doubleword in the source operand to be copied to more than one doubleword location in the destination operand.</p>
<p>SRC</p>
<p>X7</p>
<p>X6</p>
<p>X5</p>
<p>X4</p>
<p>X3</p>
<p>X2</p>
<p>X1</p>
<p>X0</p>
<p>DEST</p>
<p>Y7</p>
<p>Y6</p>
<p>Y5</p>
<p>Y4</p>
<p>Y3</p>
<p>Y2</p>
<p>Y1</p>
<p>Y0</p>
<p>00B - X4</p>
<p>Encoding</p>
<p>00B - X0</p>
<p>Encoding</p>
<p>01B - X5</p>
<p>of Fields in</p>
<p>ORDER</p>
<p>01B - X1</p>
<p>of Fields in</p>
<p>10B - X6</p>
<p>ORDER</p>
<p>10B - X2</p>
<p>ORDER</p>
<p>11B - X7</p>
<p>Operand</p>
<p>7</p>
<p>56</p>
<p>4</p>
<p>3</p>
<p>12</p>
<p>0</p>
<p>11B - X3</p>
<p>Operand</p>
<h3>Figure 4-12. 256-bit VPSHUFD Instruction Operation</h3>
<p>The source operand can be an XMM register or a 128-bit memory location. The destination operand is an XMM register. The order operand is an 8-bit immediate. Note that this instruction permits a doubleword in the source operand to be copied to more than one doubleword location in the destination operand.</p>
<p>Legacy SSE instructions: In 64-bit mode using a REX prefix in the form of REX.R permits this instruction to access additional registers (XMM8-XMM15).</p>
<p>128-bit Legacy SSE version: Bits (VLMAX-1:128) of the corresponding YMM destination register remain unchanged.</p>
<p>VEX.128 encoded version: Bits (VLMAX-1:128) of the destination YMM register are zeroed.</p>
<p>VEX.256 encoded version: Bits (255:128) of the destination stores the shuffled results of the upper 16 bytes of the source operand using the immediate byte as the order operand.</p>
<p>Note: VEX.vvvv is reserved and must be 1111b, VEX.L must be 0, otherwise the instruction will #UD.</p>
<h2>Operation</h2>
<p><strong>PSHUFD (128-bit Legacy SSE version)</strong></p>
<pre>DEST[31:0] ← (SRC &gt;&gt; (ORDER[1:0] * 32))[31:0];
DEST[63:32] ← (SRC &gt;&gt; (ORDER[3:2] * 32))[31:0];
DEST[95:64] ← (SRC &gt;&gt; (ORDER[5:4] * 32))[31:0];
DEST[127:96] ← (SRC &gt;&gt; (ORDER[7:6] * 32))[31:0];
DEST[VLMAX-1:128] (Unmodified)</pre>
<p><strong>VPSHUFD (VEX.128 encoded version)</strong></p>
<pre>DEST[31:0] ← (SRC &gt;&gt; (ORDER[1:0] * 32))[31:0];
DEST[63:32] ← (SRC &gt;&gt; (ORDER[3:2] * 32))[31:0];
DEST[95:64] ← (SRC &gt;&gt; (ORDER[5:4] * 32))[31:0];
DEST[127:96] ← (SRC &gt;&gt; (ORDER[7:6] * 32))[31:0];
DEST[VLMAX-1:128] ← 0</pre>
<p><strong>VPSHUFD (VEX.256 encoded version)</strong></p>
<pre>DEST[31:0] ← (SRC[127:0] &gt;&gt; (ORDER[1:0] * 32))[31:0];
DEST[63:32] ← (SRC[127:0] &gt;&gt; (ORDER[3:2] * 32))[31:0];
DEST[95:64] ← (SRC[127:0] &gt;&gt; (ORDER[5:4] * 32))[31:0];
DEST[127:96] ← (SRC[127:0] &gt;&gt; (ORDER[7:6] * 32))[31:0];
DEST[159:128] ← (SRC[255:128] &gt;&gt; (ORDER[1:0] * 32))[31:0];
DEST[191:160] ← (SRC[255:128] &gt;&gt; (ORDER[3:2] * 32))[31:0];
DEST[223:192] ← (SRC[255:128] &gt;&gt; (ORDER[5:4] * 32))[31:0];
DEST[255:224] ← (SRC[255:128] &gt;&gt; (ORDER[7:6] * 32))[31:0];</pre>
<h2>Intel C/C++ Compiler Intrinsic Equivalent</h2>
<p>(V)PSHUFD:</p>
<p>__m128i _mm_shuffle_epi32(__m128i a, int n)</p>
<p>VPSHUFD:</p>
<p>__m256i _mm256_shuffle_epi32(__m256i a, const int n)</p>
<h2>Flags Affected</h2>
<p>None.</p>
<h2>SIMD Floating-Point Exceptions</h2>
<p>None.</p>
<h2>Other Exceptions</h2>
<p>See Exceptions Type 4; additionally</p>
<table class="exception-table">
<tr>
<td>#UD</td>
<td>
<p>If VEX.L = 1.</p>
<p>If VEX.vvvv ≠ 1111B.</p></td></tr></table></body></html>