drginfo/ia32-64
1
0
Fork 0
forked from NRZCode/ia32-64
Code Pull requests Activity
ia32-64/x86/insertps.html
NRZ Code 92622c37d6 Initial commit
2025-07-08 02:23:29 -03:00

169 lines
7.7 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>INSERTPS
— Insert Scalar Single Precision Floating-Point Value</title></head><body><header><nav><ul><li><a href='index.html'>Index</a></li><li>December 2023</li></ul></nav></header><h1>INSERTPS
— Insert Scalar Single Precision Floating-Point Value</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>66 0F 3A 21 /r ib INSERTPS xmm1, xmm2/m32, imm8</td>
<td>A</td>
<td>V/V</td>
<td>SSE4_1</td>
<td>Insert a single precision floating-point value selected by imm8 from xmm2/m32 into xmm1 at the specified destination element specified by imm8 and zero out destination elements in xmm1 as indicated in imm8.</td></tr>
<tr>
<td>VEX.128.66.0F3A.WIG 21 /r ib VINSERTPS xmm1, xmm2, xmm3/m32, imm8</td>
<td>B</td>
<td>V/V</td>
<td>AVX</td>
<td>Insert a single precision floating-point value selected by imm8 from xmm3/m32 and merge with values in xmm2 at the specified destination element specified by imm8 and write out the result and zero out destination elements in xmm1 as indicated in imm8.</td></tr>
<tr>
<td>EVEX.128.66.0F3A.W0 21 /r ib VINSERTPS xmm1, xmm2, xmm3/m32, imm8</td>
<td>C</td>
<td>V/V</td>
<td>AVX512F</td>
<td>Insert a single precision floating-point value selected by imm8 from xmm3/m32 and merge with values in xmm2 at the specified destination element specified by imm8 and write out the result and zero out destination elements in xmm1 as indicated in imm8.</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>N/A</td>
<td>ModRM:reg (r, w)</td>
<td>ModRM:r/m (r)</td>
<td>imm8</td>
<td>N/A</td></tr>
<tr>
<td>B</td>
<td>N/A</td>
<td>ModRM:reg (w)</td>
<td>VEX.vvvv (r)</td>
<td>ModRM:r/m (r)</td>
<td>imm8</td></tr>
<tr>
<td>C</td>
<td>Tuple1 Scalar</td>
<td>ModRM:reg (w)</td>
<td>EVEX.vvvv (r)</td>
<td>ModRM:r/m (r)</td>
<td>imm8</td></tr></table>
<h2 id="description">Description<a class="anchor" href="#description">
</a></h2>
<p>(register source form)</p>
<p>Copy a single precision scalar floating-point element into a 128-bit vector register. The immediate operand has three fields, where the ZMask bits specify which elements of the destination will be set to zero, the Count_D bits specify which element of the destination will be overwritten with the scalar value, and for vector register sources the Count_S bits specify which element of the source will be copied. When the scalar source is a memory operand the Count_S bits are ignored.</p>
<p>(memory source form)</p>
<p>Load a floating-point element from a 32-bit memory location and destination operand it into the first source at the location indicated by the Count_D bits of the immediate operand. Store in the destination and zero out destination elements based on the ZMask bits of the immediate operand.</p>
<p>128-bit Legacy SSE version: The first source register is an XMM register. The second source operand is either an XMM register or a 32-bit memory location. The destination is not distinct from the first source XMM register and the upper bits (MAXVL-1:128) of the corresponding register destination are unmodified.</p>
<p>VEX.128 and EVEX encoded version: The destination and first source register is an XMM register. The second source operand is either an XMM register or a 32-bit memory location. The upper bits (MAXVL-1:128) of the corresponding register destination are zeroed.</p>
<p>If VINSERTPS is encoded with VEX.L= 1, an attempt to execute the instruction encoded with VEX.L= 1 will cause an #UD exception.</p>
<h2 id="operation">Operation<a class="anchor" href="#operation">
</a></h2>
<h3 id="vinsertps--vex-128-and-evex-encoded-version-">VINSERTPS (VEX.128 and EVEX Encoded Version)<a class="anchor" href="#vinsertps--vex-128-and-evex-encoded-version-">
</a></h3>
<pre>IF (SRC = REG) THEN COUNT_S := imm8[7:6]
ELSE COUNT_S := 0
COUNT_D := imm8[5:4]
ZMASK := imm8[3:0]
CASE (COUNT_S) OF
0: TMP := SRC2[31:0]
1: TMP := SRC2[63:32]
2: TMP := SRC2[95:64]
3: TMP := SRC2[127:96]
ESAC;
CASE (COUNT_D) OF
0: TMP2[31:0] := TMP
TMP2[127:32] := SRC1[127:32]
1: TMP2[63:32] := TMP
TMP2[31:0] := SRC1[31:0]
TMP2[127:64] := SRC1[127:64]
2: TMP2[95:64] := TMP
TMP2[63:0] := SRC1[63:0]
TMP2[127:96] := SRC1[127:96]
3: TMP2[127:96] := TMP
TMP2[95:0] := SRC1[95:0]
ESAC;
IF (ZMASK[0] = 1) THEN DEST[31:0] := 00000000H
ELSE DEST[31:0] := TMP2[31:0]
IF (ZMASK[1] = 1) THEN DEST[63:32] := 00000000H
ELSE DEST[63:32] := TMP2[63:32]
IF (ZMASK[2] = 1) THEN DEST[95:64] := 00000000H
ELSE DEST[95:64] := TMP2[95:64]
IF (ZMASK[3] = 1) THEN DEST[127:96] := 00000000H
ELSE DEST[127:96] := TMP2[127:96]
DEST[MAXVL-1:128] := 0
</pre>
<h3 id="insertps--128-bit-legacy-sse-version-">INSERTPS (128-bit Legacy SSE Version)<a class="anchor" href="#insertps--128-bit-legacy-sse-version-">
</a></h3>
<pre>IF (SRC = REG) THEN COUNT_S :=imm8[7:6]
ELSE COUNT_S :=0
COUNT_D := imm8[5:4]
ZMASK := imm8[3:0]
CASE (COUNT_S) OF
0: TMP := SRC[31:0]
1: TMP := SRC[63:32]
2: TMP := SRC[95:64]
3: TMP := SRC[127:96]
ESAC;
CASE (COUNT_D) OF
0: TMP2[31:0] := TMP
TMP2[127:32] := DEST[127:32]
1: TMP2[63:32] := TMP
TMP2[31:0] := DEST[31:0]
TMP2[127:64] := DEST[127:64]
2: TMP2[95:64] := TMP
TMP2[63:0] := DEST[63:0]
TMP2[127:96] := DEST[127:96]
3: TMP2[127:96] := TMP
TMP2[95:0] := DEST[95:0]
ESAC;
IF (ZMASK[0] = 1) THEN DEST[31:0] := 00000000H
ELSE DEST[31:0] := TMP2[31:0]
IF (ZMASK[1] = 1) THEN DEST[63:32] := 00000000H
ELSE DEST[63:32] := TMP2[63:32]
IF (ZMASK[2] = 1) THEN DEST[95:64] := 00000000H
ELSE DEST[95:64] := TMP2[95:64]
IF (ZMASK[3] = 1) THEN DEST[127:96] := 00000000H
ELSE DEST[127:96] := TMP2[127:96]
DEST[MAXVL-1:128] (Unmodified)
</pre>
<h2 id="intel-c-c++-compiler-intrinsic-equivalent">Intel C/C++ Compiler Intrinsic Equivalent<a class="anchor" href="#intel-c-c++-compiler-intrinsic-equivalent">
</a></h2>
<pre>VINSERTPS __m128 _mm_insert_ps(__m128 dst, __m128 src, const int nidx);
</pre>
<pre>INSETRTPS __m128 _mm_insert_ps(__m128 dst, __m128 src, const int nidx);
</pre>
<h2 class="exceptions" id="simd-floating-point-exceptions">SIMD Floating-Point Exceptions<a class="anchor" href="#simd-floating-point-exceptions">
</a></h2>
<p>None.</p>
<h2 class="exceptions" id="other-exceptions">Other Exceptions<a class="anchor" href="#other-exceptions">
</a></h2>
<p>Non-EVEX-encoded instruction, see <span class="not-imported">Table 2-22</span>, “Type 5 Class Exception Conditions,” additionally:</p>
<table>
<tr>
<td>#UD</td>
<td>If VEX.L = 0.</td></tr></table>
<p>EVEX-encoded instruction, see <span class="not-imported">Table 2-57</span>, “Type E9NF Class Exception Conditions.”</p><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 Developer’s Manual</a> for anything serious.
</p></footer></body></html>
View git blame Copy permalink