forked from NRZCode/ia32-64
217 lines
9.7 KiB
HTML
217 lines
9.7 KiB
HTML
|
<!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>PHSUBW/PHSUBD
|
|||
|
|
— Packed Horizontal Subtract</title></head><body><header><nav><ul><li><a href='index.html'>Index</a></li><li>December 2023</li></ul></nav></header><h1>PHSUBW/PHSUBD
|
|||
|
|
— Packed Horizontal Subtract</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>NP 0F 38 05 /r<sup>1</sup> PHSUBW mm1, mm2/m64</td>
|
|||
|
|
<td>RM</td>
|
|||
|
|
<td>V/V</td>
|
|||
|
|
<td>SSSE3</td>
|
|||
|
|
<td>Subtract 16-bit signed integers horizontally, pack to mm1.</td></tr>
|
|||
|
|
<tr>
|
|||
|
|
<td>66 0F 38 05 /r PHSUBW xmm1, xmm2/m128</td>
|
|||
|
|
<td>RM</td>
|
|||
|
|
<td>V/V</td>
|
|||
|
|
<td>SSSE3</td>
|
|||
|
|
<td>Subtract 16-bit signed integers horizontally, pack to xmm1.</td></tr>
|
|||
|
|
<tr>
|
|||
|
|
<td>NP 0F 38 06 /r PHSUBD mm1, mm2/m64</td>
|
|||
|
|
<td>RM</td>
|
|||
|
|
<td>V/V</td>
|
|||
|
|
<td>SSSE3</td>
|
|||
|
|
<td>Subtract 32-bit signed integers horizontally, pack to mm1.</td></tr>
|
|||
|
|
<tr>
|
|||
|
|
<td>66 0F 38 06 /r PHSUBD xmm1, xmm2/m128</td>
|
|||
|
|
<td>RM</td>
|
|||
|
|
<td>V/V</td>
|
|||
|
|
<td>SSSE3</td>
|
|||
|
|
<td>Subtract 32-bit signed integers horizontally, pack to xmm1.</td></tr>
|
|||
|
|
<tr>
|
|||
|
|
<td>VEX.128.66.0F38.WIG 05 /r VPHSUBW xmm1, xmm2, xmm3/m128</td>
|
|||
|
|
<td>RVM</td>
|
|||
|
|
<td>V/V</td>
|
|||
|
|
<td>AVX</td>
|
|||
|
|
<td>Subtract 16-bit signed integers horizontally, pack to xmm1.</td></tr>
|
|||
|
|
<tr>
|
|||
|
|
<td>VEX.128.66.0F38.WIG 06 /r VPHSUBD xmm1, xmm2, xmm3/m128</td>
|
|||
|
|
<td>RVM</td>
|
|||
|
|
<td>V/V</td>
|
|||
|
|
<td>AVX</td>
|
|||
|
|
<td>Subtract 32-bit signed integers horizontally, pack to xmm1.</td></tr>
|
|||
|
|
<tr>
|
|||
|
|
<td>VEX.256.66.0F38.WIG 05 /r VPHSUBW ymm1, ymm2, ymm3/m256</td>
|
|||
|
|
<td>RVM</td>
|
|||
|
|
<td>V/V</td>
|
|||
|
|
<td>AVX2</td>
|
|||
|
|
<td>Subtract 16-bit signed integers horizontally, pack to ymm1.</td></tr>
|
|||
|
|
<tr>
|
|||
|
|
<td>VEX.256.66.0F38.WIG 06 /r VPHSUBD ymm1, ymm2, ymm3/m256</td>
|
|||
|
|
<td>RVM</td>
|
|||
|
|
<td>V/V</td>
|
|||
|
|
<td>AVX2</td>
|
|||
|
|
<td>Subtract 32-bit signed integers horizontally, pack to ymm1.</td></tr></table>
|
|||
|
|
<blockquote>
|
|||
|
|
<p>1. See note in Section 2.5, “Intel® AVX and Intel® SSE Instruction Exception Classification,” in the Intel<sup>®</sup> 64 and IA-32 Architectures Software Developer’s Manual, Volume 2A, and Section 23.25.3, “Exception Conditions of Legacy SIMD Instructions Operating on MMX Registers,” in the Intel<sup>®</sup> 64 and IA-32 Architectures Software Developer’s Manual, Volume 3B.</p></blockquote>
|
|||
|
|
<h2 id="instruction-operand-encoding">Instruction Operand Encoding<a class="anchor" href="#instruction-operand-encoding">
|
|||
|
|
¶
|
|||
|
|
</a></h2>
|
|||
|
|
<table>
|
|||
|
|
<tr>
|
|||
|
|
<th>Op/En</th>
|
|||
|
|
<th>Operand 1</th>
|
|||
|
|
<th>Operand 2</th>
|
|||
|
|
<th>Operand 3</th>
|
|||
|
|
<th>Operand 4</th></tr>
|
|||
|
|
<tr>
|
|||
|
|
<td>RM</td>
|
|||
|
|
<td>ModRM:reg (r, w)</td>
|
|||
|
|
<td>ModRM:r/m (r)</td>
|
|||
|
|
<td>N/A</td>
|
|||
|
|
<td>N/A</td></tr>
|
|||
|
|
<tr>
|
|||
|
|
<td>RVM</td>
|
|||
|
|
<td>ModRM:reg (r, w)</td>
|
|||
|
|
<td>VEX.vvvv (r)</td>
|
|||
|
|
<td>ModRM:r/m (r)</td>
|
|||
|
|
<td>N/A</td></tr></table>
|
|||
|
|
<h2 id="description">Description<a class="anchor" href="#description">
|
|||
|
|
¶
|
|||
|
|
</a></h2>
|
|||
|
|
<p>(V)PHSUBW performs horizontal subtraction on each adjacent pair of 16-bit signed integers by subtracting the most significant word from the least significant word of each pair in the source and destination operands, and packs the signed 16-bit results to the destination operand (first operand). (V)PHSUBD performs horizontal subtraction on each adjacent pair of 32-bit signed integers by subtracting the most significant doubleword from the least significant doubleword of each pair, and packs the signed 32-bit result to the destination operand. When the source operand is a 128-bit memory operand, the operand must be aligned on a 16-byte boundary or a general-protection exception (#GP) will be generated.</p>
|
|||
|
|
<p>Legacy SSE version: Both operands can be MMX registers. The second source operand can be an MMX register or a 64-bit memory location.</p>
|
|||
|
|
<p>128-bit Legacy SSE version: The first source and destination operands are XMM registers. The second source operand is an XMM register or a 128-bit memory location. Bits (MAXVL-1:128) of the corresponding YMM destination register remain unchanged.</p>
|
|||
|
|
<p>In 64-bit mode, use the REX prefix to access additional registers.</p>
|
|||
|
|
<p>VEX.128 encoded version: The first source and destination operands are XMM registers. The second source operand is an XMM register or a 128-bit memory location. Bits (MAXVL-1:128) of the destination YMM register are zeroed.</p>
|
|||
|
|
<p>VEX.256 encoded version: The first source and destination operands are YMM registers. The second source operand can be an YMM register or a 256-bit memory location.</p>
|
|||
|
|
<h2 id="operation">Operation<a class="anchor" href="#operation">
|
|||
|
|
¶
|
|||
|
|
</a></h2>
|
|||
|
|
<h3 id="phsubw--with-64-bit-operands-">PHSUBW (With 64-bit Operands)<a class="anchor" href="#phsubw--with-64-bit-operands-">
|
|||
|
|
¶
|
|||
|
|
</a></h3>
|
|||
|
|
<pre>mm1[15-0] = mm1[15-0] - mm1[31-16];
|
|||
|
|
mm1[31-16] = mm1[47-32] - mm1[63-48];
|
|||
|
|
mm1[47-32] = mm2/m64[15-0] - mm2/m64[31-16];
|
|||
|
|
mm1[63-48] = mm2/m64[47-32] - mm2/m64[63-48];
|
|||
|
|
</pre>
|
|||
|
|
<h3 id="phsubw--with-128-bit-operands-">PHSUBW (With 128-bit Operands)<a class="anchor" href="#phsubw--with-128-bit-operands-">
|
|||
|
|
¶
|
|||
|
|
</a></h3>
|
|||
|
|
<pre>xmm1[15-0] = xmm1[15-0] - xmm1[31-16];
|
|||
|
|
xmm1[31-16] = xmm1[47-32] - xmm1[63-48];
|
|||
|
|
xmm1[47-32] = xmm1[79-64] - xmm1[95-80];
|
|||
|
|
xmm1[63-48] = xmm1[111-96] - xmm1[127-112];
|
|||
|
|
xmm1[79-64] = xmm2/m128[15-0] - xmm2/m128[31-16];
|
|||
|
|
xmm1[95-80] = xmm2/m128[47-32] - xmm2/m128[63-48];
|
|||
|
|
xmm1[111-96] = xmm2/m128[79-64] - xmm2/m128[95-80];
|
|||
|
|
xmm1[127-112] = xmm2/m128[111-96] - xmm2/m128[127-112];
|
|||
|
|
</pre>
|
|||
|
|
<h3 id="vphsubw--vex-128-encoded-version-">VPHSUBW (VEX.128 Encoded Version)<a class="anchor" href="#vphsubw--vex-128-encoded-version-">
|
|||
|
|
¶
|
|||
|
|
</a></h3>
|
|||
|
|
<pre>DEST[15:0] := SRC1[15:0] - SRC1[31:16]
|
|||
|
|
DEST[31:16] := SRC1[47:32] - SRC1[63:48]
|
|||
|
|
DEST[47:32] := SRC1[79:64] - SRC1[95:80]
|
|||
|
|
DEST[63:48] := SRC1[111:96] - SRC1[127:112]
|
|||
|
|
DEST[79:64] := SRC2[15:0] - SRC2[31:16]
|
|||
|
|
DEST[95:80] := SRC2[47:32] - SRC2[63:48]
|
|||
|
|
DEST[111:96] := SRC2[79:64] - SRC2[95:80]
|
|||
|
|
DEST[127:112] := SRC2[111:96] - SRC2[127:112]
|
|||
|
|
DEST[MAXVL-1:128] := 0
|
|||
|
|
</pre>
|
|||
|
|
<h3 id="vphsubw--vex-256-encoded-version-">VPHSUBW (VEX.256 Encoded Version)<a class="anchor" href="#vphsubw--vex-256-encoded-version-">
|
|||
|
|
¶
|
|||
|
|
</a></h3>
|
|||
|
|
<pre>DEST[15:0] := SRC1[15:0] - SRC1[31:16]
|
|||
|
|
DEST[31:16] := SRC1[47:32] - SRC1[63:48]
|
|||
|
|
DEST[47:32] := SRC1[79:64] - SRC1[95:80]
|
|||
|
|
DEST[63:48] := SRC1[111:96] - SRC1[127:112]
|
|||
|
|
DEST[79:64] := SRC2[15:0] - SRC2[31:16]
|
|||
|
|
DEST[95:80] := SRC2[47:32] - SRC2[63:48]
|
|||
|
|
DEST[111:96] := SRC2[79:64] - SRC2[95:80]
|
|||
|
|
DEST[127:112] := SRC2[111:96] - SRC2[127:112]
|
|||
|
|
DEST[143:128] := SRC1[143:128] - SRC1[159:144]
|
|||
|
|
DEST[159:144] := SRC1[175:160] - SRC1[191:176]
|
|||
|
|
DEST[175:160] := SRC1[207:192] - SRC1[223:208]
|
|||
|
|
DEST[191:176] := SRC1[239:224] - SRC1[255:240]
|
|||
|
|
DEST[207:192] := SRC2[143:128] - SRC2[159:144]
|
|||
|
|
DEST[223:208] := SRC2[175:160] - SRC2[191:176]
|
|||
|
|
DEST[239:224] := SRC2[207:192] - SRC2[223:208]
|
|||
|
|
DEST[255:240] := SRC2[239:224] - SRC2[255:240]
|
|||
|
|
</pre>
|
|||
|
|
<h3 id="phsubd--with-64-bit-operands-">PHSUBD (With 64-bit Operands)<a class="anchor" href="#phsubd--with-64-bit-operands-">
|
|||
|
|
¶
|
|||
|
|
</a></h3>
|
|||
|
|
<pre>mm1[31-0] = mm1[31-0] - mm1[63-32];
|
|||
|
|
mm1[63-32] = mm2/m64[31-0] - mm2/m64[63-32];
|
|||
|
|
</pre>
|
|||
|
|
<h3 id="phsubd--with-128-bit-operands-">PHSUBD (With 128-bit Operands)<a class="anchor" href="#phsubd--with-128-bit-operands-">
|
|||
|
|
¶
|
|||
|
|
</a></h3>
|
|||
|
|
<pre>xmm1[31-0] = xmm1[31-0] - xmm1[63-32];
|
|||
|
|
xmm1[63-32] = xmm1[95-64] - xmm1[127-96];
|
|||
|
|
xmm1[95-64] = xmm2/m128[31-0] - xmm2/m128[63-32];
|
|||
|
|
xmm1[127-96] = xmm2/m128[95-64] - xmm2/m128[127-96];
|
|||
|
|
</pre>
|
|||
|
|
<h3 id="vphsubd--vex-128-encoded-version-">VPHSUBD (VEX.128 Encoded Version)<a class="anchor" href="#vphsubd--vex-128-encoded-version-">
|
|||
|
|
¶
|
|||
|
|
</a></h3>
|
|||
|
|
<pre>DEST[31-0] := SRC1[31-0] - SRC1[63-32]
|
|||
|
|
DEST[63-32] := SRC1[95-64] - SRC1[127-96]
|
|||
|
|
DEST[95-64] := SRC2[31-0] - SRC2[63-32]
|
|||
|
|
DEST[127-96] := SRC2[95-64] - SRC2[127-96]
|
|||
|
|
DEST[MAXVL-1:128] := 0
|
|||
|
|
</pre>
|
|||
|
|
<h3 id="vphsubd--vex-256-encoded-version-">VPHSUBD (VEX.256 Encoded Version)<a class="anchor" href="#vphsubd--vex-256-encoded-version-">
|
|||
|
|
¶
|
|||
|
|
</a></h3>
|
|||
|
|
<pre>DEST[31:0] := SRC1[31:0] - SRC1[63:32]
|
|||
|
|
DEST[63:32] := SRC1[95:64] - SRC1[127:96]
|
|||
|
|
DEST[95:64] := SRC2[31:0] - SRC2[63:32]
|
|||
|
|
DEST[127:96] := SRC2[95:64] - SRC2[127:96]
|
|||
|
|
DEST[159:128] := SRC1[159:128] - SRC1[191:160]
|
|||
|
|
DEST[191:160] := SRC1[223:192] - SRC1[255:224]
|
|||
|
|
DEST[223:192] := SRC2[159:128] - SRC2[191:160]
|
|||
|
|
DEST[255:224] := SRC2[223:192] - SRC2[255:224]
|
|||
|
|
</pre>
|
|||
|
|
<h2 id="intel-c-c++-compiler-intrinsic-equivalents">Intel C/C++ Compiler Intrinsic Equivalents<a class="anchor" href="#intel-c-c++-compiler-intrinsic-equivalents">
|
|||
|
|
¶
|
|||
|
|
</a></h2>
|
|||
|
|
<pre>PHSUBW __m64 _mm_hsub_pi16 (__m64 a, __m64 b)
|
|||
|
|
</pre>
|
|||
|
|
<pre>PHSUBD __m64 _mm_hsub_pi32 (__m64 a, __m64 b)
|
|||
|
|
</pre>
|
|||
|
|
<pre>(V)PHSUBW __m128i _mm_hsub_epi16 (__m128i a, __m128i b)
|
|||
|
|
</pre>
|
|||
|
|
<pre>(V)PHSUBD __m128i _mm_hsub_epi32 (__m128i a, __m128i b)
|
|||
|
|
</pre>
|
|||
|
|
<pre>VPHSUBW __m256i _mm256_hsub_epi16 (__m256i a, __m256i b)
|
|||
|
|
</pre>
|
|||
|
|
<pre>VPHSUBD __m256i _mm256_hsub_epi32 (__m256i a, __m256i b)
|
|||
|
|
</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>See <span class="not-imported">Table 2-21</span>, “Type 4 Class Exception Conditions,” additionally:</p>
|
|||
|
|
<table>
|
|||
|
|
<tr>
|
|||
|
|
<td>#UD</td>
|
|||
|
|
<td>If VEX.L = 1.</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 Developer’s Manual</a> for anything serious.
|
|||
|
|
</p></footer></body></html>
|