| <!DOCTYPE html> |
| |
| <html> |
| <head> |
| <meta charset="UTF-8"> |
| <link href="style.css" type="text/css" rel="stylesheet"> |
| <title>VPBROADCAST—Broadcast Integer Data </title></head> |
| <body> |
| <h1>VPBROADCAST—Broadcast Integer Data</h1> |
| <table> |
| <tr> |
| <th>Opcode/Instruction</th> |
| <th>Op/En</th> |
| <th>64/32 -bit Mode</th> |
| <th>CPUID Feature Flag</th> |
| <th>Description</th></tr> |
| <tr> |
| <td> |
| <p>VEX.128.66.0F38.W0 78 /r</p> |
| <p>VPBROADCASTB <em>xmm1, xmm2/m8</em></p></td> |
| <td>RM</td> |
| <td>V/V</td> |
| <td>AVX2</td> |
| <td>Broadcast a byte integer in the source operand to six-teen locations in <em>xmm1</em>.</td></tr> |
| <tr> |
| <td> |
| <p>VEX.256.66.0F38.W0 78 /r</p> |
| <p>VPBROADCASTB <em>ymm1, xmm2/m8</em></p></td> |
| <td>RM</td> |
| <td>V/V</td> |
| <td>AVX2</td> |
| <td>Broadcast a byte integer in the source operand to thirty-two locations in <em>ymm1</em>.</td></tr> |
| <tr> |
| <td> |
| <p>VEX.128.66.0F38.W0 79 /r</p> |
| <p>VPBROADCASTW <em>xmm1, xmm2/m16</em></p></td> |
| <td>RM</td> |
| <td>V/V</td> |
| <td>AVX2</td> |
| <td>Broadcast a word integer in the source operand to eight locations in <em>xmm1</em>.</td></tr> |
| <tr> |
| <td> |
| <p>VEX.256.66.0F38.W0 79 /r</p> |
| <p>VPBROADCASTW <em>ymm1, xmm2/m16</em></p></td> |
| <td>RM</td> |
| <td>V/V</td> |
| <td>AVX2</td> |
| <td>Broadcast a word integer in the source operand to six-teen locations in <em>ymm1</em>.</td></tr> |
| <tr> |
| <td> |
| <p>VEX.128.66.0F38.W0 58 /r</p> |
| <p>VPBROADCASTD <em>xmm1, xmm2/m32</em></p></td> |
| <td>RM</td> |
| <td>V/V</td> |
| <td>AVX2</td> |
| <td>Broadcast a dword integer in the source operand to four locations in <em>xmm1</em>.</td></tr> |
| <tr> |
| <td> |
| <p>VEX.256.66.0F38.W0 58 /r</p> |
| <p>VPBROADCASTD <em>ymm1, xmm2/m32</em></p></td> |
| <td>RM</td> |
| <td>V/V</td> |
| <td>AVX2</td> |
| <td>Broadcast a dword integer in the source operand to eight locations in <em>ymm1</em>.</td></tr> |
| <tr> |
| <td> |
| <p>VEX.128.66.0F38.W0 59 /r</p> |
| <p>VPBROADCASTQ <em>xmm1, xmm2/m64</em></p></td> |
| <td>RM</td> |
| <td>V/V</td> |
| <td>AVX2</td> |
| <td>Broadcast a qword element in mem to two locations in <em>xmm1</em>.</td></tr> |
| <tr> |
| <td> |
| <p>VEX.256.66.0F38.W0 59 /r</p> |
| <p>VPBROADCASTQ <em>ymm1, xmm2/m64</em></p></td> |
| <td>RM</td> |
| <td>V/V</td> |
| <td>AVX2</td> |
| <td>Broadcast a qword element in mem to four locations in <em>ymm1</em>.</td></tr> |
| <tr> |
| <td> |
| <p>VEX.256.66.0F38.W0 5A /r</p> |
| <p>VBROADCASTI128 <em>ymm1, m128</em></p></td> |
| <td>RM</td> |
| <td>V/V</td> |
| <td>AVX2</td> |
| <td>Broadcast 128 bits of integer data in mem to low and high 128-bits 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>RM</td> |
| <td>ModRM:reg (w)</td> |
| <td>ModRM:r/m (r)</td> |
| <td>NA</td> |
| <td>NA</td></tr></table> |
| <h2>Description</h2> |
| <p>Load integer data from the source operand (second operand) and broadcast to all elements of the destination operand (first operand).</p> |
| <p>The destination operand is a YMM register. The source operand is 8-bit, 16-bit 32-bit, 64-bit memory location or the low 8-bit, 16-bit 32-bit, 64-bit data in an XMM register. VPBROADCASTB/D/W/Q also support XMM register as the source operand.</p> |
| <p>VBROADCASTI128: The destination operand is a YMM register. The source operand is 128-bit memory location. Register source encodings for VBROADCASTI128 are reserved and will #UD.</p> |
| <p>VPBROADCASTB/W/D/Q is supported in both 128-bit and 256-bit wide versions.</p> |
| <p>VBROADCASTI128 is only supported as a 256-bit wide version.</p> |
| <p>Note: In VEX-encoded versions, VEX.vvvv is reserved and must be 1111b otherwise instructions will #UD. Attempts to execute any VPBROADCAST* instruction with VEX.W = 1 will cause #UD. If VBROADCASTI128 is encoded with VEX.L= 0, an attempt to execute the instruction encoded with VEX.L= 0 will cause an #UD exception.</p> |
| <p>X0</p> |
| <p>m32</p> |
| <p>DEST</p> |
| <p>X0</p> |
| <p>X0</p> |
| <p>X0</p> |
| <p>X0</p> |
| <p>X0</p> |
| <p>X0</p> |
| <p>X0</p> |
| <p>X0</p> |
| <h3>Figure 4-33. VPBROADCASTD Operation (VEX.256 encoded version)</h3> |
| <p>X0</p> |
| <p>m32</p> |
| <p>DEST</p> |
| <p>0</p> |
| <p>0</p> |
| <p>0</p> |
| <p>0</p> |
| <p>X0</p> |
| <p>X0</p> |
| <p>X0</p> |
| <p>X0</p> |
| <h3>Figure 4-34. VPBROADCASTD Operation (128-bit version)</h3> |
| <p>m64</p> |
| <p>X0</p> |
| <p>DEST</p> |
| <p>X0</p> |
| <p>X0</p> |
| <p>X0</p> |
| <p>X0</p> |
| <h3>Figure 4-35. VPBROADCASTQ Operation</h3> |
| <p>m128i</p> |
| <p>X0</p> |
| <svg width="445.499985" viewBox="145.860000 876124.379980 296.999990 13.500030" height="20.250045"> |
| <rect y="876124.38" x="145.86" style="fill:rgba(0,0,0,0);stroke:rgb(0,0,0);stroke-width:1pt;" height="13.5" width="148.5"></rect> |
| <text y="876134.1135" x="206.64" style="font-size:7.500000pt" lengthAdjust="spacingAndGlyphs" textLength="9.15">X0</text></svg> |
| <p>DEST</p> |
| <p>X0</p> |
| <h3>Figure 4-36. VBROADCASTI128 Operation</h3> |
| <h2>Operation</h2> |
| <p><strong>VPBROADCASTB (VEX.128 encoded version)</strong></p> |
| <pre>temp ← SRC[7:0] |
| FOR j ← 0 TO 15 |
| DEST[7+j*8: j*8] ← temp |
| ENDFOR |
| DEST[VLMAX-1:128] ← 0</pre> |
| <p><strong>VPBROADCASTB (VEX.256 encoded version)</strong></p> |
| <pre>temp ← SRC[7:0] |
| FOR j ← 0 TO 31 |
| DEST[7+j*8: j*8] ← temp |
| ENDFOR</pre> |
| <p><strong>VPBROADCASTW (VEX.128 encoded version)</strong></p> |
| <pre>temp ← SRC[15:0] |
| FOR j ← 0 TO 7 |
| DEST[15+j*16: j*16] ← temp |
| ENDFOR |
| DEST[VLMAX-1:128] ← 0</pre> |
| <p><strong>VPBROADCASTW (VEX.256 encoded version)</strong></p> |
| <pre>temp ← SRC[15:0] |
| FOR j ← 0 TO 15 |
| DEST[15+j*16: j*16] ← temp |
| ENDFOR</pre> |
| <p><strong>VPBROADCASTD (128 bit version)</strong></p> |
| <pre>temp ← SRC[31:0] |
| FOR j ← 0 TO 3 |
| DEST[31+j*32: j*32] ← temp |
| ENDFOR |
| DEST[VLMAX-1:128] ← 0</pre> |
| <p><strong>VPBROADCASTD (VEX.256 encoded version)</strong></p> |
| <pre>temp ← SRC[31:0] |
| FOR j ← 0 TO 7 |
| DEST[31+j*32: j*32] ← temp |
| ENDFOR</pre> |
| <p><strong>VPBROADCASTQ (VEX.128 encoded version)</strong></p> |
| <pre>temp ← SRC[63:0] |
| DEST[63:0] ← temp |
| DEST[127:64] ← temp |
| DEST[VLMAX-1:128] ← 0</pre> |
| <p><strong>VPBROADCASTQ (VEX.256 encoded version)</strong></p> |
| <pre>temp ← SRC[63:0] |
| DEST[63:0] ← temp |
| DEST[127:64] ← temp |
| DEST[191:128] ← temp |
| DEST[255:192] ← temp</pre> |
| <p><strong>VBROADCASTI128</strong></p> |
| <pre>temp ← SRC[127:0] |
| DEST[127:0] ← temp |
| DEST[VLMAX-1:128] ← temp</pre> |
| <h2>Intel C/C++ Compiler Intrinsic Equivalent</h2> |
| <p>VPBROADCASTB:</p> |
| <p>__m256i _mm256_broadcastb_epi8(__m128i );</p> |
| <p>VPBROADCASTW:</p> |
| <p>__m256i _mm256_broadcastw_epi16(__m128i );</p> |
| <p>VPBROADCASTD:</p> |
| <p>__m256i _mm256_broadcastd_epi32(__m128i );</p> |
| <p>VPBROADCASTQ:</p> |
| <p>__m256i _mm256_broadcastq_epi64(__m128i );</p> |
| <p>VPBROADCASTB:</p> |
| <p>__m128i _mm_broadcastb_epi8(__m128i );</p> |
| <p>VPBROADCASTW:</p> |
| <p>__m128i _mm_broadcastw_epi16(__m128i );</p> |
| <p>VPBROADCASTD:</p> |
| <p>__m128i _mm_broadcastd_epi32(__m128i );</p> |
| <p>VPBROADCASTQ:</p> |
| <p>__m128i _mm_broadcastq_epi64(__m128i );</p> |
| <p>VBROADCASTI128:</p> |
| <p> __m256i _mm256_broadcastsi128_si256(__m128i );</p> |
| <h2>SIMD Floating-Point Exceptions</h2> |
| <p>None</p> |
| <h2>Other Exceptions</h2> |
| <p>See Exceptions Type 6; additionally</p> |
| <table class="exception-table"> |
| <tr> |
| <td>#UD</td> |
| <td> |
| <p>If VEX.W = 1,</p> |
| <p>If VEX.L = 0 for VBROADCASTI128.</p></td></tr></table></body></html> |