8222090: Add Hygon Dhyana processor support
authordholmes
Thu, 11 Apr 2019 19:36:00 -0400
changeset 54519 a2795025f417
parent 54518 50b34791a1d2
child 54520 f48312257bc6
8222090: Add Hygon Dhyana processor support Reviewed-by: kvn, dholmes, coleenp, rwestberg Contributed-by: Jinke Fan <fanjinke51@yeah.net>
src/hotspot/cpu/x86/assembler_x86.cpp
src/hotspot/cpu/x86/vm_version_ext_x86.cpp
src/hotspot/cpu/x86/vm_version_x86.cpp
src/hotspot/cpu/x86/vm_version_x86.hpp
--- a/src/hotspot/cpu/x86/assembler_x86.cpp	Thu Apr 11 12:06:58 2019 -0700
+++ b/src/hotspot/cpu/x86/assembler_x86.cpp	Thu Apr 11 19:36:00 2019 -0400
@@ -3099,7 +3099,7 @@
     }
     return;
   }
-  if (UseAddressNop && VM_Version::is_amd()) {
+  if (UseAddressNop && VM_Version::is_amd_family()) {
     //
     // Using multi-bytes nops "0x0F 0x1F [address]" for AMD.
     //  1: 0x90
--- a/src/hotspot/cpu/x86/vm_version_ext_x86.cpp	Thu Apr 11 12:06:58 2019 -0700
+++ b/src/hotspot/cpu/x86/vm_version_ext_x86.cpp	Thu Apr 11 19:36:00 2019 -0400
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2013, 2018, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2013, 2019, Oracle and/or its affiliates. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
  * This code is free software; you can redistribute it and/or modify it
@@ -340,6 +340,10 @@
     return !is_amd_Barcelona();
   }
 
+  if (is_hygon()) {
+    return true;
+  }
+
   return false;
 }
 
@@ -407,6 +411,10 @@
     }
     return _family_id_intel[cpu_family_id];
   }
+  if (is_hygon()) {
+    return "Dhyana";
+  }
+
   return "Unknown x86";
 }
 
@@ -423,6 +431,9 @@
   } else if (is_amd()) {
     cpu_type = "AMD";
     x64 = cpu_is_em64t() ? " AMD64" : "";
+  } else if (is_hygon()) {
+    cpu_type = "Hygon";
+    x64 = cpu_is_em64t() ? " AMD64" : "";
   } else {
     cpu_type = "Unknown x86";
     x64 = cpu_is_em64t() ? " x86_64" : "";
--- a/src/hotspot/cpu/x86/vm_version_x86.cpp	Thu Apr 11 12:06:58 2019 -0700
+++ b/src/hotspot/cpu/x86/vm_version_x86.cpp	Thu Apr 11 19:36:00 2019 -0400
@@ -680,7 +680,7 @@
     _features &= ~CPU_HT;
   }
 
-  if( is_intel() ) { // Intel cpus specific settings
+  if (is_intel()) { // Intel cpus specific settings
     if (is_knights_family()) {
       _features &= ~CPU_VZEROUPPER;
     }
@@ -781,7 +781,7 @@
           FLAG_SET_DEFAULT(UseAESCTRIntrinsics, false);
         }
       } else {
-        if(supports_sse4_1()) {
+        if (supports_sse4_1()) {
           if (FLAG_IS_DEFAULT(UseAESCTRIntrinsics)) {
             FLAG_SET_DEFAULT(UseAESCTRIntrinsics, true);
           }
@@ -1001,7 +1001,7 @@
   } else if (UseAVX == 1 || UseAVX == 2) {
     // 32 bytes vectors (in YMM) are only supported with AVX+
     max_vector_size = 32;
-  } else if (UseAVX > 2 ) {
+  } else if (UseAVX > 2) {
     // 64 bytes vectors (in ZMM) are only supported with AVX 3
     max_vector_size = 64;
   }
@@ -1165,38 +1165,38 @@
     }
   }
 
-  if( is_amd() ) { // AMD cpus specific settings
-    if( supports_sse2() && FLAG_IS_DEFAULT(UseAddressNop) ) {
+  if (is_amd_family()) { // AMD cpus specific settings
+    if (supports_sse2() && FLAG_IS_DEFAULT(UseAddressNop)) {
       // Use it on new AMD cpus starting from Opteron.
       UseAddressNop = true;
     }
-    if( supports_sse2() && FLAG_IS_DEFAULT(UseNewLongLShift) ) {
+    if (supports_sse2() && FLAG_IS_DEFAULT(UseNewLongLShift)) {
       // Use it on new AMD cpus starting from Opteron.
       UseNewLongLShift = true;
     }
-    if( FLAG_IS_DEFAULT(UseXmmLoadAndClearUpper) ) {
+    if (FLAG_IS_DEFAULT(UseXmmLoadAndClearUpper)) {
       if (supports_sse4a()) {
         UseXmmLoadAndClearUpper = true; // use movsd only on '10h' Opteron
       } else {
         UseXmmLoadAndClearUpper = false;
       }
     }
-    if( FLAG_IS_DEFAULT(UseXmmRegToRegMoveAll) ) {
-      if( supports_sse4a() ) {
+    if (FLAG_IS_DEFAULT(UseXmmRegToRegMoveAll)) {
+      if (supports_sse4a()) {
         UseXmmRegToRegMoveAll = true; // use movaps, movapd only on '10h'
       } else {
         UseXmmRegToRegMoveAll = false;
       }
     }
-    if( FLAG_IS_DEFAULT(UseXmmI2F) ) {
-      if( supports_sse4a() ) {
+    if (FLAG_IS_DEFAULT(UseXmmI2F)) {
+      if (supports_sse4a()) {
         UseXmmI2F = true;
       } else {
         UseXmmI2F = false;
       }
     }
-    if( FLAG_IS_DEFAULT(UseXmmI2D) ) {
-      if( supports_sse4a() ) {
+    if (FLAG_IS_DEFAULT(UseXmmI2D)) {
+      if (supports_sse4a()) {
         UseXmmI2D = true;
       } else {
         UseXmmI2D = false;
@@ -1214,7 +1214,7 @@
     }
 
     // some defaults for AMD family 15h
-    if ( cpu_family() == 0x15 ) {
+    if (cpu_family() == 0x15) {
       // On family 15h processors default is no sw prefetch
       if (FLAG_IS_DEFAULT(AllocatePrefetchStyle)) {
         FLAG_SET_DEFAULT(AllocatePrefetchStyle, 0);
@@ -1239,8 +1239,8 @@
     }
 #endif // COMPILER2
 
-    // Some defaults for AMD family 17h
-    if ( cpu_family() == 0x17 ) {
+    // Some defaults for AMD family 17h || Hygon family 18h
+    if (cpu_family() == 0x17 || cpu_family() == 0x18) {
       // On family 17h processors use XMM and UnalignedLoadStores for Array Copy
       if (supports_sse2() && FLAG_IS_DEFAULT(UseXMMForArrayCopy)) {
         FLAG_SET_DEFAULT(UseXMMForArrayCopy, true);
@@ -1256,29 +1256,29 @@
     }
   }
 
-  if( is_intel() ) { // Intel cpus specific settings
-    if( FLAG_IS_DEFAULT(UseStoreImmI16) ) {
+  if (is_intel()) { // Intel cpus specific settings
+    if (FLAG_IS_DEFAULT(UseStoreImmI16)) {
       UseStoreImmI16 = false; // don't use it on Intel cpus
     }
-    if( cpu_family() == 6 || cpu_family() == 15 ) {
-      if( FLAG_IS_DEFAULT(UseAddressNop) ) {
+    if (cpu_family() == 6 || cpu_family() == 15) {
+      if (FLAG_IS_DEFAULT(UseAddressNop)) {
         // Use it on all Intel cpus starting from PentiumPro
         UseAddressNop = true;
       }
     }
-    if( FLAG_IS_DEFAULT(UseXmmLoadAndClearUpper) ) {
+    if (FLAG_IS_DEFAULT(UseXmmLoadAndClearUpper)) {
       UseXmmLoadAndClearUpper = true; // use movsd on all Intel cpus
     }
-    if( FLAG_IS_DEFAULT(UseXmmRegToRegMoveAll) ) {
-      if( supports_sse3() ) {
+    if (FLAG_IS_DEFAULT(UseXmmRegToRegMoveAll)) {
+      if (supports_sse3()) {
         UseXmmRegToRegMoveAll = true; // use movaps, movapd on new Intel cpus
       } else {
         UseXmmRegToRegMoveAll = false;
       }
     }
-    if( cpu_family() == 6 && supports_sse3() ) { // New Intel cpus
+    if (cpu_family() == 6 && supports_sse3()) { // New Intel cpus
 #ifdef COMPILER2
-      if( FLAG_IS_DEFAULT(MaxLoopPad) ) {
+      if (FLAG_IS_DEFAULT(MaxLoopPad)) {
         // For new Intel cpus do the next optimization:
         // don't align the beginning of a loop if there are enough instructions
         // left (NumberOfLoopInstrToAlign defined in c2_globals.hpp)
@@ -1324,7 +1324,7 @@
         FLAG_SET_DEFAULT(UseIncDec, false);
       }
     }
-    if(FLAG_IS_DEFAULT(AllocatePrefetchInstr) && supports_3dnow_prefetch()) {
+    if (FLAG_IS_DEFAULT(AllocatePrefetchInstr) && supports_3dnow_prefetch()) {
       FLAG_SET_DEFAULT(AllocatePrefetchInstr, 3);
     }
   }
--- a/src/hotspot/cpu/x86/vm_version_x86.hpp	Thu Apr 11 12:06:58 2019 -0700
+++ b/src/hotspot/cpu/x86/vm_version_x86.hpp	Thu Apr 11 19:36:00 2019 -0400
@@ -495,13 +495,13 @@
       result |= CPU_CX8;
     if (_cpuid_info.std_cpuid1_edx.bits.cmov != 0)
       result |= CPU_CMOV;
-    if (_cpuid_info.std_cpuid1_edx.bits.fxsr != 0 || (is_amd() &&
+    if (_cpuid_info.std_cpuid1_edx.bits.fxsr != 0 || (is_amd_family() &&
         _cpuid_info.ext_cpuid1_edx.bits.fxsr != 0))
       result |= CPU_FXSR;
     // HT flag is set for multi-core processors also.
     if (threads_per_core() > 1)
       result |= CPU_HT;
-    if (_cpuid_info.std_cpuid1_edx.bits.mmx != 0 || (is_amd() &&
+    if (_cpuid_info.std_cpuid1_edx.bits.mmx != 0 || (is_amd_family() &&
         _cpuid_info.ext_cpuid1_edx.bits.mmx != 0))
       result |= CPU_MMX;
     if (_cpuid_info.std_cpuid1_edx.bits.sse != 0)
@@ -553,7 +553,7 @@
           result |= CPU_VNNI;
       }
     }
-    if(_cpuid_info.sef_cpuid7_ebx.bits.bmi1 != 0)
+    if (_cpuid_info.sef_cpuid7_ebx.bits.bmi1 != 0)
       result |= CPU_BMI1;
     if (_cpuid_info.std_cpuid1_edx.bits.tsc != 0)
       result |= CPU_TSC;
@@ -567,17 +567,17 @@
       result |= CPU_CLMUL;
     if (_cpuid_info.sef_cpuid7_ebx.bits.rtm != 0)
       result |= CPU_RTM;
-    if(_cpuid_info.sef_cpuid7_ebx.bits.adx != 0)
+    if (_cpuid_info.sef_cpuid7_ebx.bits.adx != 0)
        result |= CPU_ADX;
-    if(_cpuid_info.sef_cpuid7_ebx.bits.bmi2 != 0)
+    if (_cpuid_info.sef_cpuid7_ebx.bits.bmi2 != 0)
       result |= CPU_BMI2;
     if (_cpuid_info.sef_cpuid7_ebx.bits.sha != 0)
       result |= CPU_SHA;
     if (_cpuid_info.std_cpuid1_ecx.bits.fma != 0)
       result |= CPU_FMA;
 
-    // AMD features.
-    if (is_amd()) {
+    // AMD|Hygon features.
+    if (is_amd_family()) {
       if ((_cpuid_info.ext_cpuid1_edx.bits.tdnow != 0) ||
           (_cpuid_info.ext_cpuid1_ecx.bits.prefetchw != 0))
         result |= CPU_3DNOW_PREFETCH;
@@ -587,8 +587,8 @@
         result |= CPU_SSE4A;
     }
     // Intel features.
-    if(is_intel()) {
-      if(_cpuid_info.ext_cpuid1_ecx.bits.lzcnt_intel != 0)
+    if (is_intel()) {
+      if (_cpuid_info.ext_cpuid1_ecx.bits.lzcnt_intel != 0)
         result |= CPU_LZCNT;
       // for Intel, ecx.bits.misalignsse bit (bit 8) indicates support for prefetchw
       if (_cpuid_info.ext_cpuid1_ecx.bits.misalignsse != 0) {
@@ -714,6 +714,8 @@
   static int  cpu_family()        { return _cpu;}
   static bool is_P6()             { return cpu_family() >= 6; }
   static bool is_amd()            { assert_is_initialized(); return _cpuid_info.std_vendor_name_0 == 0x68747541; } // 'htuA'
+  static bool is_hygon()          { assert_is_initialized(); return _cpuid_info.std_vendor_name_0 == 0x6F677948; } // 'ogyH'
+  static bool is_amd_family()     { return is_amd() || is_hygon(); }
   static bool is_intel()          { assert_is_initialized(); return _cpuid_info.std_vendor_name_0 == 0x756e6547; } // 'uneG'
   static bool is_zx()             { assert_is_initialized(); return (_cpuid_info.std_vendor_name_0 == 0x746e6543) || (_cpuid_info.std_vendor_name_0 == 0x68532020); } // 'tneC'||'hS  '
   static bool is_atom_family()    { return ((cpu_family() == 0x06) && ((extended_cpu_model() == 0x36) || (extended_cpu_model() == 0x37) || (extended_cpu_model() == 0x4D))); } //Silvermont and Centerton
@@ -737,7 +739,7 @@
       if (!supports_topology || result == 0) {
         result = (_cpuid_info.dcp_cpuid4_eax.bits.cores_per_cpu + 1);
       }
-    } else if (is_amd()) {
+    } else if (is_amd_family()) {
       result = (_cpuid_info.ext_cpuid8_ecx.bits.cores_per_cpu + 1);
     } else if (is_zx()) {
       bool supports_topology = supports_processor_topology();
@@ -773,7 +775,7 @@
     intx result = 0;
     if (is_intel()) {
       result = (_cpuid_info.dcp_cpuid4_ebx.bits.L1_line_size + 1);
-    } else if (is_amd()) {
+    } else if (is_amd_family()) {
       result = _cpuid_info.ext_cpuid5_ecx.bits.L1_line_size;
     } else if (is_zx()) {
       result = (_cpuid_info.dcp_cpuid4_ebx.bits.L1_line_size + 1);
@@ -860,7 +862,7 @@
 
   // AMD features
   static bool supports_3dnow_prefetch()    { return (_features & CPU_3DNOW_PREFETCH) != 0; }
-  static bool supports_mmx_ext()  { return is_amd() && _cpuid_info.ext_cpuid1_edx.bits.mmx_amd != 0; }
+  static bool supports_mmx_ext()  { return is_amd_family() && _cpuid_info.ext_cpuid1_edx.bits.mmx_amd != 0; }
   static bool supports_lzcnt()    { return (_features & CPU_LZCNT) != 0; }
   static bool supports_sse4a()    { return (_features & CPU_SSE4A) != 0; }
 
@@ -873,8 +875,8 @@
   }
   static bool supports_tscinv() {
     return supports_tscinv_bit() &&
-           ( (is_amd() && !is_amd_Barcelona()) ||
-             is_intel_tsc_synched_at_init() );
+      ((is_amd_family() && !is_amd_Barcelona()) ||
+        is_intel_tsc_synched_at_init());
   }
 
   // Intel Core and newer cpus have fast IDIV instruction (excluding Atom).
@@ -899,7 +901,7 @@
     // Core      - 256 / prefetchnta
     // It will be used only when AllocatePrefetchStyle > 0
 
-    if (is_amd()) { // AMD
+    if (is_amd_family()) { // AMD | Hygon
       if (supports_sse2()) {
         return 256; // Opteron
       } else {