views/asm-docs-source/XSAVES.html - compiler-explorer - Rivoreo Source Code Repositories

 <!DOCTYPE html>

 <html>
 <head>
 <meta charset="UTF-8">
 <link href="style.css" type="text/css" rel="stylesheet">
 <title>XSAVES—Save Processor Extended States Supervisor </title></head>
 <body>
 <h1>XSAVES—Save Processor Extended States Supervisor</h1>
 <table>
 <tr>
 <th>Opcode</th>
 <th>Instruction</th>
 <th>Op/En</th>
 <th>64-Bit Mode</th>
 <th>Compat/Leg Mode</th>
 <th>Description</th></tr>
 <tr>
 <td>0F C7 /5</td>
 <td>XSAVES <em>mem</em></td>
 <td>M</td>
 <td>Valid</td>
 <td>Valid</td>
 <td>Save state components specified by EDX:EAX to <em>mem</em> with compaction, optimizing if possible.</td></tr>
 <tr>
 <td>REX.W+ 0F C7 /5</td>
 <td>XSAVES64 <em>mem</em></td>
 <td>M</td>
 <td>Valid</td>
 <td>N.E.</td>
 <td>Save state components specified by EDX:EAX to <em>mem</em> with compaction, optimizing if possible.</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>M</td>
 <td>ModRM:r/m (w)</td>
 <td>NA</td>
 <td>NA</td>
 <td>NA</td></tr></table>
 <h2>Description</h2>
 <p>Performs a full or partial save of processor state components to the XSAVE area located at the memory address specified by the destination operand. The implicit EDX:EAX register pair specifies a 64-bit instruction mask. The specific state components saved correspond to the bits set in the requested-feature bitmap (RFBM), the logical-AND of EDX:EAX and the logical-OR of XCR0 with the IA32_XSS MSR. XSAVES may be executed only if CPL = 0.</p>
 <p>The format of the XSAVE area is detailed in Section 13.4, “XSAVE Area,” of <em>Intel® 64 and IA-32 Architectures Soft-ware Developer’s Manual, Volume 1</em>.</p>
 <p>Section 13.10, “Operation of XSAVES,” of <em>Intel® 64 and IA-32 Architectures Software Developer’s Manual, Volume 1</em> provides a detailed description of the operation of the XSAVES instruction. The following items provide a high-level outline:</p>
 <p>Use of a destination operand not aligned to 64-byte boundary (in either 64-bit or 32-bit modes) results in a general-protection (#GP) exception. In 64-bit mode, the upper 32 bits of RDX and RAX are ignored.</p>
 <p>1.</p>
 <p>There is an exception for state component 1 (SSE). MXCSR is part of SSE state, but XINUSE[1] may be 0 even if MXCSR does not have its initial value of 1F80H. In this case, the init optimization does not apply and XSAVEC will save SSE state as long as RFBM[1] = 1 and the modified optimization is not being applied.</p>
 <p>2.</p>
 <p>There is an exception for state component 1 (SSE). MXCSR is part of SSE state, but XINUSE[1] may be 0 even if MXCSR does not have its initial value of 1F80H. In this case, XSAVES sets XSTATE_BV[1] to 1 as long as RFBM[1] = 1.</p>
 <h2>Operation</h2>
 <pre>RFBM ← XCR0 AND EDX:EAX;
                                     /* bitwise logical AND */
 IF in VMX non-root operation
     THEN VMXNR ← 1;
     ELSE VMXNR ← 0;
 FI;
 LAXA ← linear address of XSAVE area;
 COMPMASK ← RFBM OR 80000000_00000000H;
 IF XRSTOR_INFO = (cid:162)CPL,VMXNR,LAXA,COMPMASK(cid:178)
     THEN MODOPT ← 1;
     ELSE MODOPT ← 0;
 FI;
 IF RFBM[0] = 1 and XINUSE[0] = 1
     THEN store x87 state into legacy region of XSAVE area;
     /* might avoid saving if x87 state is not modified and MODOPT = 1 */
 FI;
 IF RFBM[1] = 1 and (XINUSE[1] = 1 or MXCSR ≠ 1F80H)
     THEN store SSE state into legacy region of XSAVE area;
     /* might avoid saving if SSE state is not modified and MODOPT = 1 */
 FI;
 IF RFBM[2] = 1 AND XINUSE[2] = 1
     THEN store AVX state into extended region of XSAVE area;
     /* might avoid saving if AVX state is not modified and MODOPT = 1 */
 FI;
 XSTATE_BV field in XSAVE header ← XINUSE AND RFBM;
 XCOMP_BV field in XSAVE header ← COMPMASK;</pre>
 <h2>Flags Affected</h2>
 <p>None.</p>
 <h2>Intel C/C++ Compiler Intrinsic Equivalent</h2>
 <p>XSAVES:</p>
 <p>void _xsaves( void * , unsigned __int64);</p>
 <p>XSAVES64:</p>
 <p>void _xsaves64( void * , unsigned __int64);</p>
 <h2>Protected Mode Exceptions</h2>
 <table class="exception-table">
 <tr>
 <td>#GP(0)</td>
 <td>
 <p>If a memory operand effective address is outside the CS, DS, ES, FS, or GS segment limit.</p>
 <p>If a memory operand is not aligned on a 64-byte boundary, regardless of segment.</p></td></tr>
 <tr>
 <td>#SS(0)</td>
 <td>If a memory operand effective address is outside the SS segment limit.</td></tr>
 <tr>
 <td>#PF(fault-code)</td>
 <td>If a page fault occurs.</td></tr>
 <tr>
 <td>#NM</td>
 <td>If CR0.TS[bit 3] = 1.</td></tr>
 <tr>
 <td>#UD</td>
 <td>
 <p>If CPUID.01H:ECX.XSAVE[bit 26] = 0 or CPUID.(EAX=0DH,ECX=1):EAX.XSS[bit 3] = 0.</p>
 <p>If CR4.OSXSAVE[bit 18] = 0.</p>
 <p>If any of the LOCK, 66H, F3H or F2H prefixes is used.</p></td></tr>
 <tr>
 <td>#AC</td>
 <td>
 <p>If this exception is disabled a general protection exception (#GP) is signaled if the memory operand is not aligned on a 16-byte boundary, as described above. If the alignment check exception (#AC) is enabled (and the CPL is 3), signaling of #AC is not guaranteed and may vary with implementation, as follows. In all implementations where #AC is not signaled, a general protection exception is signaled in its place. In addition, the width of the alignment check may also vary with implementation. For instance, for a given implementation, an align-</p>
 <p>ment check exception might be signaled for a 2-byte misalignment, whereas a general protec-tion exception might be signaled for all other misalignments (4-, 8-, or 16-byte misalignments).</p></td></tr></table>
 <h2>Real-Address Mode Exceptions</h2>
 <table class="exception-table">
 <tr>
 <td>#GP</td>
 <td>
 <p>If a memory operand is not aligned on a 64-byte boundary, regardless of segment.</p>
 <p>If any part of the operand lies outside the effective address space from 0 to FFFFH.</p></td></tr>
 <tr>
 <td>#NM</td>
 <td>If CR0.TS[bit 3] = 1.</td></tr>
 <tr>
 <td>#UD</td>
 <td>
 <p>If CPUID.01H:ECX.XSAVE[bit 26] = 0 or CPUID.(EAX=0DH,ECX=1):EAX.XSS[bit 3] = 0.</p>
 <p>If CR4.OSXSAVE[bit 18] = 0.</p>
 <p>If any of the LOCK, 66H, F3H or F2H prefixes is used.</p></td></tr></table>
 <h2>Virtual-8086 Mode Exceptions</h2>
 <p>Same exceptions as in protected mode.</p>
 <h2>Compatibility Mode Exceptions</h2>
 <p>Same exceptions as in protected mode.</p>
 <h2>64-Bit Mode Exceptions</h2>
 <table class="exception-table">
 <tr>
 <td>#GP(0)</td>
 <td>
 <p>If the memory address is in a non-canonical form.</p>
 <p>If a memory operand is not aligned on a 64-byte boundary, regardless of segment.</p></td></tr>
 <tr>
 <td>#SS(0)</td>
 <td>If a memory address referencing the SS segment is in a non-canonical form.</td></tr>
 <tr>
 <td>#PF(fault-code)</td>
 <td>If a page fault occurs.</td></tr>
 <tr>
 <td>#NM</td>
 <td>If CR0.TS[bit 3] = 1.</td></tr>
 <tr>
 <td>#UD</td>
 <td>
 <p>If CPUID.01H:ECX.XSAVE[bit 26] = 0 or CPUID.(EAX=0DH,ECX=1):EAX.XSS[bit 3] = 0.</p>
 <p>If CR4.OSXSAVE[bit 18] = 0.</p>
 <p>If any of the LOCK, 66H, F3H or F2H prefixes is used.</p></td></tr>
 <tr>
 <td>#AC</td>
 <td>If this exception is disabled a general protection exception (#GP) is signaled if the memory operand is not aligned on a 16-byte boundary, as described above. If the alignment check exception (#AC) is enabled (and the CPL is 3), signaling of #AC is not guaranteed and may vary with implementation, as follows. In all implementations where #AC is not signaled, a general protection exception is signaled in its place. In addition, the width of the alignment check may also vary with implementation. For instance, for a given implementation, an align-ment check exception might be signaled for a 2-byte misalignment, whereas a general protec-tion exception might be signaled for all other misalignments (4-, 8-, or 16-byte misalignments).</td></tr></table></body></html>
	<!DOCTYPE html>

	<html>
	<head>
	<meta charset="UTF-8">
	<link href="style.css" type="text/css" rel="stylesheet">
	<title>XSAVES—Save Processor Extended States Supervisor </title></head>
	<body>
	<h1>XSAVES—Save Processor Extended States Supervisor</h1>
	<table>
	<tr>
	<th>Opcode</th>
	<th>Instruction</th>
	<th>Op/En</th>
	<th>64-Bit Mode</th>
	<th>Compat/Leg Mode</th>
	<th>Description</th></tr>
	<tr>
	<td>0F C7 /5</td>
	<td>XSAVES <em>mem</em></td>
	<td>M</td>
	<td>Valid</td>
	<td>Valid</td>
	<td>Save state components specified by EDX:EAX to <em>mem</em> with compaction, optimizing if possible.</td></tr>
	<tr>
	<td>REX.W+ 0F C7 /5</td>
	<td>XSAVES64 <em>mem</em></td>
	<td>M</td>
	<td>Valid</td>
	<td>N.E.</td>
	<td>Save state components specified by EDX:EAX to <em>mem</em> with compaction, optimizing if possible.</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>M</td>
	<td>ModRM:r/m (w)</td>
	<td>NA</td>
	<td>NA</td>
	<td>NA</td></tr></table>
	<h2>Description</h2>
	<p>Performs a full or partial save of processor state components to the XSAVE area located at the memory address specified by the destination operand. The implicit EDX:EAX register pair specifies a 64-bit instruction mask. The specific state components saved correspond to the bits set in the requested-feature bitmap (RFBM), the logical-AND of EDX:EAX and the logical-OR of XCR0 with the IA32_XSS MSR. XSAVES may be executed only if CPL = 0.</p>
	<p>The format of the XSAVE area is detailed in Section 13.4, “XSAVE Area,” of <em>Intel® 64 and IA-32 Architectures Soft-ware Developer’s Manual, Volume 1</em>.</p>
	<p>Section 13.10, “Operation of XSAVES,” of <em>Intel® 64 and IA-32 Architectures Software Developer’s Manual, Volume 1</em> provides a detailed description of the operation of the XSAVES instruction. The following items provide a high-level outline:</p>
	<p>Use of a destination operand not aligned to 64-byte boundary (in either 64-bit or 32-bit modes) results in a general-protection (#GP) exception. In 64-bit mode, the upper 32 bits of RDX and RAX are ignored.</p>
	<p>1.</p>
	<p>There is an exception for state component 1 (SSE). MXCSR is part of SSE state, but XINUSE[1] may be 0 even if MXCSR does not have its initial value of 1F80H. In this case, the init optimization does not apply and XSAVEC will save SSE state as long as RFBM[1] = 1 and the modified optimization is not being applied.</p>
	<p>2.</p>
	<p>There is an exception for state component 1 (SSE). MXCSR is part of SSE state, but XINUSE[1] may be 0 even if MXCSR does not have its initial value of 1F80H. In this case, XSAVES sets XSTATE_BV[1] to 1 as long as RFBM[1] = 1.</p>
	<h2>Operation</h2>
	<pre>RFBM ← XCR0 AND EDX:EAX;
	/* bitwise logical AND */
	IF in VMX non-root operation
	THEN VMXNR ← 1;
	ELSE VMXNR ← 0;
	FI;
	LAXA ← linear address of XSAVE area;
	COMPMASK ← RFBM OR 80000000_00000000H;
	IF XRSTOR_INFO = (cid:162)CPL,VMXNR,LAXA,COMPMASK(cid:178)
	THEN MODOPT ← 1;
	ELSE MODOPT ← 0;
	FI;
	IF RFBM[0] = 1 and XINUSE[0] = 1
	THEN store x87 state into legacy region of XSAVE area;
	/* might avoid saving if x87 state is not modified and MODOPT = 1 */
	FI;
	IF RFBM[1] = 1 and (XINUSE[1] = 1 or MXCSR ≠ 1F80H)
	THEN store SSE state into legacy region of XSAVE area;
	/* might avoid saving if SSE state is not modified and MODOPT = 1 */
	FI;
	IF RFBM[2] = 1 AND XINUSE[2] = 1
	THEN store AVX state into extended region of XSAVE area;
	/* might avoid saving if AVX state is not modified and MODOPT = 1 */
	FI;
	XSTATE_BV field in XSAVE header ← XINUSE AND RFBM;
	XCOMP_BV field in XSAVE header ← COMPMASK;</pre>
	<h2>Flags Affected</h2>
	<p>None.</p>
	<h2>Intel C/C++ Compiler Intrinsic Equivalent</h2>
	<p>XSAVES:</p>
	<p>void _xsaves( void * , unsigned __int64);</p>
	<p>XSAVES64:</p>
	<p>void _xsaves64( void * , unsigned __int64);</p>
	<h2>Protected Mode Exceptions</h2>
	<table class="exception-table">
	<tr>
	<td>#GP(0)</td>
	<td>
	<p>If a memory operand effective address is outside the CS, DS, ES, FS, or GS segment limit.</p>
	<p>If a memory operand is not aligned on a 64-byte boundary, regardless of segment.</p></td></tr>
	<tr>
	<td>#SS(0)</td>
	<td>If a memory operand effective address is outside the SS segment limit.</td></tr>
	<tr>
	<td>#PF(fault-code)</td>
	<td>If a page fault occurs.</td></tr>
	<tr>
	<td>#NM</td>
	<td>If CR0.TS[bit 3] = 1.</td></tr>
	<tr>
	<td>#UD</td>
	<td>
	<p>If CPUID.01H:ECX.XSAVE[bit 26] = 0 or CPUID.(EAX=0DH,ECX=1):EAX.XSS[bit 3] = 0.</p>
	<p>If CR4.OSXSAVE[bit 18] = 0.</p>
	<p>If any of the LOCK, 66H, F3H or F2H prefixes is used.</p></td></tr>
	<tr>
	<td>#AC</td>
	<td>
	<p>If this exception is disabled a general protection exception (#GP) is signaled if the memory operand is not aligned on a 16-byte boundary, as described above. If the alignment check exception (#AC) is enabled (and the CPL is 3), signaling of #AC is not guaranteed and may vary with implementation, as follows. In all implementations where #AC is not signaled, a general protection exception is signaled in its place. In addition, the width of the alignment check may also vary with implementation. For instance, for a given implementation, an align-</p>
	<p>ment check exception might be signaled for a 2-byte misalignment, whereas a general protec-tion exception might be signaled for all other misalignments (4-, 8-, or 16-byte misalignments).</p></td></tr></table>
	<h2>Real-Address Mode Exceptions</h2>
	<table class="exception-table">
	<tr>
	<td>#GP</td>
	<td>
	<p>If a memory operand is not aligned on a 64-byte boundary, regardless of segment.</p>
	<p>If any part of the operand lies outside the effective address space from 0 to FFFFH.</p></td></tr>
	<tr>
	<td>#NM</td>
	<td>If CR0.TS[bit 3] = 1.</td></tr>
	<tr>
	<td>#UD</td>
	<td>
	<p>If CPUID.01H:ECX.XSAVE[bit 26] = 0 or CPUID.(EAX=0DH,ECX=1):EAX.XSS[bit 3] = 0.</p>
	<p>If CR4.OSXSAVE[bit 18] = 0.</p>
	<p>If any of the LOCK, 66H, F3H or F2H prefixes is used.</p></td></tr></table>
	<h2>Virtual-8086 Mode Exceptions</h2>
	<p>Same exceptions as in protected mode.</p>
	<h2>Compatibility Mode Exceptions</h2>
	<p>Same exceptions as in protected mode.</p>
	<h2>64-Bit Mode Exceptions</h2>
	<table class="exception-table">
	<tr>
	<td>#GP(0)</td>
	<td>
	<p>If the memory address is in a non-canonical form.</p>
	<p>If a memory operand is not aligned on a 64-byte boundary, regardless of segment.</p></td></tr>
	<tr>
	<td>#SS(0)</td>
	<td>If a memory address referencing the SS segment is in a non-canonical form.</td></tr>
	<tr>
	<td>#PF(fault-code)</td>
	<td>If a page fault occurs.</td></tr>
	<tr>
	<td>#NM</td>
	<td>If CR0.TS[bit 3] = 1.</td></tr>
	<tr>
	<td>#UD</td>
	<td>
	<p>If CPUID.01H:ECX.XSAVE[bit 26] = 0 or CPUID.(EAX=0DH,ECX=1):EAX.XSS[bit 3] = 0.</p>
	<p>If CR4.OSXSAVE[bit 18] = 0.</p>
	<p>If any of the LOCK, 66H, F3H or F2H prefixes is used.</p></td></tr>
	<tr>
	<td>#AC</td>
	<td>If this exception is disabled a general protection exception (#GP) is signaled if the memory operand is not aligned on a 16-byte boundary, as described above. If the alignment check exception (#AC) is enabled (and the CPL is 3), signaling of #AC is not guaranteed and may vary with implementation, as follows. In all implementations where #AC is not signaled, a general protection exception is signaled in its place. In addition, the width of the alignment check may also vary with implementation. For instance, for a given implementation, an align-ment check exception might be signaled for a 2-byte misalignment, whereas a general protec-tion exception might be signaled for all other misalignments (4-, 8-, or 16-byte misalignments).</td></tr></table></body></html>