8367031: [backout] Change java.time month/day field types to 'byte'

Backport-of: 00be643fa3eff6fd66d39f5f5ea70ff347296318

.github/workflows/main.yml

@@ -310,7 +310,7 @@ jobs:
     uses: ./.github/workflows/build-windows.yml
     with:
       platform: windows-x64
-      msvc-toolset-version: '14.43'
+      msvc-toolset-version: '14.44'
       msvc-toolset-architecture: 'x86.x64'
       configure-arguments: ${{ github.event.inputs.configure-arguments }}
       make-arguments: ${{ github.event.inputs.make-arguments }}
@@ -322,7 +322,7 @@ jobs:
     uses: ./.github/workflows/build-windows.yml
     with:
       platform: windows-aarch64
-      msvc-toolset-version: '14.43'
+      msvc-toolset-version: '14.44'
       msvc-toolset-architecture: 'arm64'
       make-target: 'hotspot'
       extra-conf-options: '--openjdk-target=aarch64-unknown-cygwin'

.jcheck/conf

@@ -1,7 +1,7 @@
 [general]
-project=jdk
+project=jdk-updates
 jbs=JDK
-version=26
+version=25.0.1
 
 [checks]
 error=author,committer,reviewers,merge,issues,executable,symlink,message,hg-tag,whitespace,problemlists,copyright

make/autoconf/configure

@@ -1,6 +1,6 @@
 #!/bin/bash
 #
-# Copyright (c) 2012, 2023, Oracle and/or its affiliates. All rights reserved.
+# Copyright (c) 2012, 2025, 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
@@ -366,7 +366,7 @@ EOT
 
     # Print additional help, e.g. a list of toolchains and JVM features.
     # This must be done by the autoconf script.
-    ( CONFIGURE_PRINT_ADDITIONAL_HELP=true . $generated_script PRINTF=printf )
+    ( CONFIGURE_PRINT_ADDITIONAL_HELP=true . $generated_script PRINTF=printf ECHO=echo )
 
     cat <<EOT
 

make/autoconf/lib-tests.m4

@@ -28,7 +28,7 @@
 ################################################################################
 
 # Minimum supported versions
-JTREG_MINIMUM_VERSION=7.5.2
+JTREG_MINIMUM_VERSION=7.5.1
 GTEST_MINIMUM_VERSION=1.14.0
 
 ################################################################################

make/conf/github-actions.conf

@@ -26,7 +26,7 @@
 # Versions and download locations for dependencies used by GitHub Actions (GHA)
 
 GTEST_VERSION=1.14.0
-JTREG_VERSION=7.5.2+1
+JTREG_VERSION=7.5.1+1
 
 LINUX_X64_BOOT_JDK_EXT=tar.gz
 LINUX_X64_BOOT_JDK_URL=https://download.java.net/java/GA/jdk24/1f9ff9062db4449d8ca828c504ffae90/36/GPL/openjdk-24_linux-x64_bin.tar.gz

make/conf/jib-profiles.js

@@ -1174,9 +1174,9 @@ var getJibProfilesDependencies = function (input, common) {
         jtreg: {
             server: "jpg",
             product: "jtreg",
-            version: "7.5.2",
+            version: "7.5.1",
             build_number: "1",
-            file: "bundles/jtreg-7.5.2+1.zip",
+            file: "bundles/jtreg-7.5.1+1.zip",
             environment_name: "JT_HOME",
             environment_path: input.get("jtreg", "home_path") + "/bin",
             configure_args: "--with-jtreg=" + input.get("jtreg", "home_path"),
@@ -1192,8 +1192,8 @@ var getJibProfilesDependencies = function (input, common) {
             server: "jpg",
             product: "jcov",
             version: "3.0",
-            build_number: "3",
-            file: "bundles/jcov-3.0+3.zip",
+            build_number: "1",
+            file: "bundles/jcov-3.0+1.zip",
             environment_name: "JCOV_HOME",
         },
 

make/conf/version-numbers.conf

@@ -26,17 +26,17 @@
 # Default version, product, and vendor information to use,
 # unless overridden by configure
 
-DEFAULT_VERSION_FEATURE=26
+DEFAULT_VERSION_FEATURE=25
 DEFAULT_VERSION_INTERIM=0
-DEFAULT_VERSION_UPDATE=0
+DEFAULT_VERSION_UPDATE=1
 DEFAULT_VERSION_PATCH=0
 DEFAULT_VERSION_EXTRA1=0
 DEFAULT_VERSION_EXTRA2=0
 DEFAULT_VERSION_EXTRA3=0
-DEFAULT_VERSION_DATE=2026-03-17
-DEFAULT_VERSION_CLASSFILE_MAJOR=70  # "`$EXPR $DEFAULT_VERSION_FEATURE + 44`"
+DEFAULT_VERSION_DATE=2025-10-21
+DEFAULT_VERSION_CLASSFILE_MAJOR=69  # "`$EXPR $DEFAULT_VERSION_FEATURE + 44`"
 DEFAULT_VERSION_CLASSFILE_MINOR=0
 DEFAULT_VERSION_DOCS_API_SINCE=11
-DEFAULT_ACCEPTABLE_BOOT_VERSIONS="24 25 26"
-DEFAULT_JDK_SOURCE_TARGET_VERSION=26
-DEFAULT_PROMOTED_VERSION_PRE=ea
+DEFAULT_ACCEPTABLE_BOOT_VERSIONS="24 25"
+DEFAULT_JDK_SOURCE_TARGET_VERSION=25
+DEFAULT_PROMOTED_VERSION_PRE=

make/jdk/src/classes/build/tools/cldrconverter/Bundle.java

@@ -542,10 +542,10 @@ class Bundle {
                     if (pattern != null) {
                         // Perform date-time format pattern conversion which is
                         // applicable to both SimpleDateFormat and j.t.f.DateTimeFormatter.
-                        String transPattern = translateDateFormatLetters(calendarType, pattern, this::convertDateTimePatternLetter);
+                        String transPattern = translateDateFormatLetters(calendarType, key, pattern, this::convertDateTimePatternLetter);
                         dateTimePatterns.add(i, transPattern);
                         // Additionally, perform SDF specific date-time format pattern conversion
-                        sdfPatterns.add(i, translateDateFormatLetters(calendarType, transPattern, this::convertSDFLetter));
+                        sdfPatterns.add(i, translateDateFormatLetters(calendarType, key, transPattern, this::convertSDFLetter));
                     } else {
                         dateTimePatterns.add(i, null);
                         sdfPatterns.add(i, null);
@@ -568,7 +568,7 @@ class Bundle {
         }
     }
 
-    private String translateDateFormatLetters(CalendarType calendarType, String cldrFormat, ConvertDateTimeLetters converter) {
+    private String translateDateFormatLetters(CalendarType calendarType, String patternKey, String cldrFormat, ConvertDateTimeLetters converter) {
         String pattern = cldrFormat;
         int length = pattern.length();
         boolean inQuote = false;
@@ -587,7 +587,7 @@ class Bundle {
                     if (nextc == '\'') {
                         i++;
                         if (count != 0) {
-                            converter.convert(calendarType, lastLetter, count, jrePattern);
+                            converter.convert(calendarType, patternKey, lastLetter, count, jrePattern);
                             lastLetter = 0;
                             count = 0;
                         }
@@ -597,7 +597,7 @@ class Bundle {
                 }
                 if (!inQuote) {
                     if (count != 0) {
-                        converter.convert(calendarType, lastLetter, count, jrePattern);
+                        converter.convert(calendarType, patternKey, lastLetter, count, jrePattern);
                         lastLetter = 0;
                         count = 0;
                     }
@@ -614,7 +614,7 @@ class Bundle {
             }
             if (!(c >= 'a' && c <= 'z' || c >= 'A' && c <= 'Z')) {
                 if (count != 0) {
-                    converter.convert(calendarType, lastLetter, count, jrePattern);
+                    converter.convert(calendarType, patternKey, lastLetter, count, jrePattern);
                     lastLetter = 0;
                     count = 0;
                 }
@@ -627,7 +627,7 @@ class Bundle {
                 count++;
                 continue;
             }
-            converter.convert(calendarType, lastLetter, count, jrePattern);
+            converter.convert(calendarType, patternKey, lastLetter, count, jrePattern);
             lastLetter = c;
             count = 1;
         }
@@ -637,7 +637,7 @@ class Bundle {
         }
 
         if (count != 0) {
-            converter.convert(calendarType, lastLetter, count, jrePattern);
+            converter.convert(calendarType, patternKey, lastLetter, count, jrePattern);
         }
         if (cldrFormat.contentEquals(jrePattern)) {
             return cldrFormat;
@@ -661,7 +661,7 @@ class Bundle {
      * on the support given by the SimpleDateFormat and the j.t.f.DateTimeFormatter
      * for date-time formatting.
      */
-    private void convertDateTimePatternLetter(CalendarType calendarType, char cldrLetter, int count, StringBuilder sb) {
+    private void convertDateTimePatternLetter(CalendarType calendarType, String patternKey, char cldrLetter, int count, StringBuilder sb) {
         switch (cldrLetter) {
             case 'u':
             case 'U':
@@ -683,7 +683,7 @@ class Bundle {
      * Perform a conversion of CLDR date-time format pattern letter which is
      * specific to the SimpleDateFormat.
      */
-    private void convertSDFLetter(CalendarType calendarType, char cldrLetter, int count, StringBuilder sb) {
+    private void convertSDFLetter(CalendarType calendarType, String patternKey, char cldrLetter, int count, StringBuilder sb) {
         switch (cldrLetter) {
             case 'G':
                 if (calendarType != CalendarType.GREGORIAN) {
@@ -722,6 +722,17 @@ class Bundle {
                 appendN('z', count, sb);
                 break;
 
+            case 'y':
+                // If the style is FULL/LONG for a Japanese Calendar, make the
+                // count == 4 for Gan-nen
+                if (calendarType == CalendarType.JAPANESE &&
+                        (patternKey.contains("full-") ||
+                         patternKey.contains("long-"))) {
+                    count = 4;
+                }
+                appendN(cldrLetter, count, sb);
+                break;
+
             case 'Z':
                 if (count == 4 || count == 5) {
                     sb.append("XXX");
@@ -767,6 +778,7 @@ class Bundle {
             .collect(Collectors.toMap(
                 e -> calendarPrefix + e.getKey(),
                 e -> translateDateFormatLetters(calendarType,
+                        e.getKey(),
                         (String)e.getValue(),
                         this::convertDateTimePatternLetter)
             ))
@@ -775,7 +787,7 @@ class Bundle {
 
     @FunctionalInterface
     private interface ConvertDateTimeLetters {
-        void convert(CalendarType calendarType, char cldrLetter, int count, StringBuilder sb);
+        void convert(CalendarType calendarType, String patternKey, char cldrLetter, int count, StringBuilder sb);
     }
 
     /**

make/modules/java.base/Gensrc.gmk

@@ -46,8 +46,6 @@ CLDR_GEN_DONE := $(GENSRC_DIR)/_cldr-gensrc.marker
 TZ_DATA_DIR := $(MODULE_SRC)/share/data/tzdata
 ZONENAME_TEMPLATE := $(MODULE_SRC)/share/classes/java/time/format/ZoneName.java.template
 
-# The `-utf8` option is used even for US English, as some names
-# may contain non-ASCII characters, such as “Türkiye”.
 $(CLDR_GEN_DONE): $(wildcard $(CLDR_DATA_DIR)/dtd/*.dtd) \
     $(wildcard $(CLDR_DATA_DIR)/main/en*.xml) \
     $(wildcard $(CLDR_DATA_DIR)/supplemental/*.xml) \
@@ -63,8 +61,7 @@ $(CLDR_GEN_DONE): $(wildcard $(CLDR_DATA_DIR)/dtd/*.dtd) \
 	    -basemodule \
 	    -year $(COPYRIGHT_YEAR) \
 	    -zntempfile $(ZONENAME_TEMPLATE) \
-	    -tzdatadir $(TZ_DATA_DIR) \
-	    -utf8)
+	    -tzdatadir $(TZ_DATA_DIR))
 	$(TOUCH) $@
 
 TARGETS += $(CLDR_GEN_DONE)

make/modules/jdk.localedata/Gensrc.gmk

@@ -45,8 +45,7 @@ $(CLDR_GEN_DONE): $(wildcard $(CLDR_DATA_DIR)/dtd/*.dtd) \
 	    -baselocales "en-US" \
 	    -year $(COPYRIGHT_YEAR) \
 	    -o $(GENSRC_DIR) \
-	    -tzdatadir $(TZ_DATA_DIR) \
-	    -utf8)
+	    -tzdatadir $(TZ_DATA_DIR))
 	$(TOUCH) $@
 
 TARGETS += $(CLDR_GEN_DONE)

src/demo/share/jfc/Stylepad/HelloWorld.java

@@ -187,18 +187,22 @@ public class HelloWorld {
             new Run("none", "Hello from Cupertino")
         }),
         new Paragraph("title", new Run[] {
-            new Run("none", "台北問候您!")
+            new Run("none", "\u53F0\u5317\u554F\u5019\u60A8\u0021")
         }),
         new Paragraph("title", new Run[] {
-            new Run("none", "Αθηναι ασπαζονται υμας!") // Greek
+            new Run("none", "\u0391\u03B8\u03B7\u03BD\u03B1\u03B9\u0020" // Greek
+            + "\u03B1\u03C3\u03C0\u03B1\u03B6\u03BF\u03BD"
+            + "\u03C4\u03B1\u03B9\u0020\u03C5\u03BC\u03B1"
+            + "\u03C2\u0021")
         }),
         new Paragraph("title", new Run[] {
-            new Run("none", "東京から今日は")
+            new Run("none", "\u6771\u4eac\u304b\u3089\u4eca\u65e5\u306f")
         }),
         new Paragraph("title", new Run[] {
-            new Run("none", "שלום מירושלים")
+            new Run("none", "\u05e9\u05dc\u05d5\u05dd \u05de\u05d9\u05e8\u05d5"
+            + "\u05e9\u05dc\u05d9\u05dd")
         }),
         new Paragraph("title", new Run[] {
-            new Run("none", "سلام")
+            new Run("none", "\u0633\u0644\u0627\u0645")
         }), };
 }

src/demo/share/jfc/SwingSet2/resources/swingset_de.properties

@@ -456,13 +456,13 @@ SliderDemo.horizontal=Horizontal
 SliderDemo.vertical=Vertikal
 SliderDemo.plain=Einfach
 SliderDemo.a_plain_slider=Ein einfacher Schieberegler
-SliderDemo.majorticks=Grobteilungen
-SliderDemo.majorticksdescription=Ein Schieberegler mit Grobteilungsmarkierungen
-SliderDemo.ticks=Feinteilungen, Teilungen zum Einrasten und Labels
-SliderDemo.minorticks=Feinteilungen
-SliderDemo.minorticksdescription=Ein Schieberegler mit Grob- und Feinteilungen, mit Teilungen, in die der Schieberegler einrastet, wobei einige Teilungen mit einem sichtbaren Label versehen sind
+SliderDemo.majorticks=Hauptteilstriche
+SliderDemo.majorticksdescription=Ein Schieberegler mit Hauptteilstrichen
+SliderDemo.ticks=Hilfsteilstriche, zum Einrasten und Beschriften
+SliderDemo.minorticks=Hilfsteilstriche
+SliderDemo.minorticksdescription=Ein Schieberegler mit Haupt- und Hilfsteilstrichen, in die der Schieberegler einrastet, wobei einige Teilstriche mit einer sichtbaren Beschriftung versehen sind
 SliderDemo.disabled=Deaktiviert
-SliderDemo.disableddescription=Ein Schieberegler mit Grob- und Feinteilungen, der nicht aktiviert ist (kann nicht bearbeitet werden)
+SliderDemo.disableddescription=Ein Schieberegler mit Haupt- und Hilfsteilstrichen, der nicht aktiviert ist (kann nicht bearbeitet werden)
 
 ### SplitPane Demo ###
 

src/hotspot/cpu/aarch64/aarch64.ad

@@ -3921,10 +3921,6 @@ ins_attrib ins_alignment(4);    // Required alignment attribute (must
                                 // compute_padding() function must be
                                 // provided for the instruction
 
-// Whether this node is expanded during code emission into a sequence of
-// instructions and the first instruction can perform an implicit null check.
-ins_attrib ins_is_late_expanded_null_check_candidate(false);
-
 //----------OPERANDS-----------------------------------------------------------
 // Operand definitions must precede instruction definitions for correct parsing
 // in the ADLC because operands constitute user defined types which are used in

src/hotspot/cpu/aarch64/gc/shenandoah/shenandoahBarrierSetAssembler_aarch64.cpp

@@ -292,7 +292,8 @@ void ShenandoahBarrierSetAssembler::load_reference_barrier(MacroAssembler* masm,
   } else {
     assert(is_phantom, "only remaining strength");
     assert(!is_narrow, "phantom access cannot be narrow");
-    __ mov(lr, CAST_FROM_FN_PTR(address, ShenandoahRuntime::load_reference_barrier_phantom));
+    // AOT saved adapters need relocation for this call.
+    __ lea(lr, RuntimeAddress(CAST_FROM_FN_PTR(address, ShenandoahRuntime::load_reference_barrier_phantom)));
   }
   __ blr(lr);
   __ mov(rscratch1, r0);

src/hotspot/cpu/aarch64/gc/z/z_aarch64.ad

@@ -106,13 +106,6 @@ instruct zLoadP(iRegPNoSp dst, memory8 mem, rFlagsReg cr)
   match(Set dst (LoadP mem));
   predicate(UseZGC && !needs_acquiring_load(n) && n->as_Load()->barrier_data() != 0);
   effect(TEMP dst, KILL cr);
-  // The main load is a candidate to implement implicit null checks, as long as
-  // legitimize_address() does not require a preceding lea instruction to
-  // materialize the memory operand. The absence of a preceding lea instruction
-  // is guaranteed for immLoffset8 memory operands, because these do not lead to
-  // out-of-range offsets (see definition of immLoffset8). Fortunately,
-  // immLoffset8 memory operands are the most common ones in practice.
-  ins_is_late_expanded_null_check_candidate(opnd_array(1)->opcode() == INDOFFL8);
 
   ins_cost(4 * INSN_COST);
 
@@ -124,11 +117,7 @@ instruct zLoadP(iRegPNoSp dst, memory8 mem, rFlagsReg cr)
       // Fix up any out-of-range offsets.
       assert_different_registers(rscratch2, as_Register($mem$$base));
       assert_different_registers(rscratch2, $dst$$Register);
-      int size = 8;
-      assert(!this->is_late_expanded_null_check_candidate() ||
-             !MacroAssembler::legitimize_address_requires_lea(ref_addr, size),
-             "an instruction that can be used for implicit null checking should emit the candidate memory access first");
-      ref_addr = __ legitimize_address(ref_addr, size, rscratch2);
+      ref_addr = __ legitimize_address(ref_addr, 8, rscratch2);
     }
     __ ldr($dst$$Register, ref_addr);
     z_load_barrier(masm, this, ref_addr, $dst$$Register, rscratch1);

src/hotspot/cpu/aarch64/macroAssembler_aarch64.hpp

@@ -129,21 +129,16 @@ class MacroAssembler: public Assembler {
     a.lea(this, r);
   }
 
-  // Whether materializing the given address for a LDR/STR requires an
-  // additional lea instruction.
-  static bool legitimize_address_requires_lea(const Address &a, int size) {
-    return a.getMode() == Address::base_plus_offset &&
-           !Address::offset_ok_for_immed(a.offset(), exact_log2(size));
-  }
-
   /* Sometimes we get misaligned loads and stores, usually from Unsafe
      accesses, and these can exceed the offset range. */
   Address legitimize_address(const Address &a, int size, Register scratch) {
-    if (legitimize_address_requires_lea(a, size)) {
-      block_comment("legitimize_address {");
-      lea(scratch, a);
-      block_comment("} legitimize_address");
-      return Address(scratch);
+    if (a.getMode() == Address::base_plus_offset) {
+      if (! Address::offset_ok_for_immed(a.offset(), exact_log2(size))) {
+        block_comment("legitimize_address {");
+        lea(scratch, a);
+        block_comment("} legitimize_address");
+        return Address(scratch);
+      }
     }
     return a;
   }

src/hotspot/cpu/arm/arm.ad

@@ -8888,13 +8888,8 @@ instruct TailCalljmpInd(IPRegP jump_target, inline_cache_regP method_ptr) %{
   match(TailCall jump_target method_ptr);
 
   ins_cost(CALL_COST);
-  format %{ "MOV    Rexception_pc, LR\n\t"
-            "jump   $jump_target  \t! $method_ptr holds method" %}
+  format %{ "jump   $jump_target  \t! $method_ptr holds method" %}
   ins_encode %{
-    __ mov(Rexception_pc, LR);   // this is used only to call
-                                 // StubRoutines::forward_exception_entry()
-                                 // which expects PC of exception in
-                                 // R5. FIXME?
     __ jump($jump_target$$Register);
   %}
   ins_pipe(tail_call);
@@ -8939,8 +8934,10 @@ instruct ForwardExceptionjmp()
   match(ForwardException);
   ins_cost(CALL_COST);
 
-  format %{ "b    forward_exception_stub" %}
+  format %{ "MOV  Rexception_pc, LR\n\t"
+            "b    forward_exception_entry" %}
   ins_encode %{
+    __ mov(Rexception_pc, LR);
     // OK to trash Rtemp, because Rtemp is used by stub
     __ jump(StubRoutines::forward_exception_entry(), relocInfo::runtime_call_type, Rtemp);
   %}

src/hotspot/cpu/ppc/gc/z/z_ppc.ad

@@ -141,7 +141,6 @@ instruct zLoadP(iRegPdst dst, memoryAlg4 mem, flagsRegCR0 cr0)
 %{
   match(Set dst (LoadP mem));
   effect(TEMP_DEF dst, KILL cr0);
-  ins_is_late_expanded_null_check_candidate(true);
   ins_cost(MEMORY_REF_COST);
 
   predicate((UseZGC && n->as_Load()->barrier_data() != 0)
@@ -161,7 +160,6 @@ instruct zLoadP_acq(iRegPdst dst, memoryAlg4 mem, flagsRegCR0 cr0)
 %{
   match(Set dst (LoadP mem));
   effect(TEMP_DEF dst, KILL cr0);
-  ins_is_late_expanded_null_check_candidate(true);
   ins_cost(3 * MEMORY_REF_COST);
 
   // Predicate on instruction order is implicitly present due to the predicate of the cheaper zLoadP operation

src/hotspot/cpu/ppc/macroAssembler_ppc.cpp

@@ -3928,8 +3928,10 @@ void MacroAssembler::kernel_crc32_vpmsum_aligned(Register crc, Register buf, Reg
   Label L_outer_loop, L_inner_loop, L_last;
 
   // Set DSCR pre-fetch to deepest.
-  load_const_optimized(t0, VM_Version::_dscr_val | 7);
-  mtdscr(t0);
+  if (VM_Version::has_mfdscr()) {
+    load_const_optimized(t0, VM_Version::_dscr_val | 7);
+    mtdscr(t0);
+  }
 
   mtvrwz(VCRC, crc); // crc lives in VCRC, now
 
@@ -4073,8 +4075,10 @@ void MacroAssembler::kernel_crc32_vpmsum_aligned(Register crc, Register buf, Reg
   // ********** Main loop end **********
 
   // Restore DSCR pre-fetch value.
-  load_const_optimized(t0, VM_Version::_dscr_val);
-  mtdscr(t0);
+  if (VM_Version::has_mfdscr()) {
+    load_const_optimized(t0, VM_Version::_dscr_val);
+    mtdscr(t0);
+  }
 
   // ********** Simple loop for remaining 16 byte blocks **********
   {

src/hotspot/cpu/ppc/ppc.ad

@@ -4036,10 +4036,6 @@ ins_attrib ins_field_cbuf_insts_offset(-1);
 ins_attrib ins_field_load_ic_hi_node(0);
 ins_attrib ins_field_load_ic_node(0);
 
-// Whether this node is expanded during code emission into a sequence of
-// instructions and the first instruction can perform an implicit null check.
-ins_attrib ins_is_late_expanded_null_check_candidate(false);
-
 //----------OPERANDS-----------------------------------------------------------
 // Operand definitions must precede instruction definitions for correct
 // parsing in the ADLC because operands constitute user defined types

src/hotspot/cpu/ppc/stubGenerator_ppc.cpp

@@ -952,8 +952,10 @@ class StubGenerator: public StubCodeGenerator {
     address start_pc = __ pc();
     Register tmp1 = R6_ARG4;
     // probably copy stub would have changed value reset it.
-    __ load_const_optimized(tmp1, VM_Version::_dscr_val);
-    __ mtdscr(tmp1);
+    if (VM_Version::has_mfdscr()) {
+      __ load_const_optimized(tmp1, VM_Version::_dscr_val);
+      __ mtdscr(tmp1);
+    }
     __ li(R3_RET, 0); // return 0
     __ blr();
     return start_pc;
@@ -1070,9 +1072,10 @@ class StubGenerator: public StubCodeGenerator {
         __ dcbt(R3_ARG1, 0);
 
         // If supported set DSCR pre-fetch to deepest.
-        __ load_const_optimized(tmp2, VM_Version::_dscr_val | 7);
-        __ mtdscr(tmp2);
-
+        if (VM_Version::has_mfdscr()) {
+          __ load_const_optimized(tmp2, VM_Version::_dscr_val | 7);
+          __ mtdscr(tmp2);
+        }
         __ li(tmp1, 16);
 
         // Backbranch target aligned to 32-byte. Not 16-byte align as
@@ -1092,8 +1095,10 @@ class StubGenerator: public StubCodeGenerator {
         __ bdnz(l_10);                       // Dec CTR and loop if not zero.
 
         // Restore DSCR pre-fetch value.
-        __ load_const_optimized(tmp2, VM_Version::_dscr_val);
-        __ mtdscr(tmp2);
+        if (VM_Version::has_mfdscr()) {
+          __ load_const_optimized(tmp2, VM_Version::_dscr_val);
+          __ mtdscr(tmp2);
+        }
 
      } // FasterArrayCopy
 
@@ -1344,8 +1349,10 @@ class StubGenerator: public StubCodeGenerator {
           __ dcbt(R3_ARG1, 0);
 
           // If supported set DSCR pre-fetch to deepest.
-          __ load_const_optimized(tmp2, VM_Version::_dscr_val | 7);
-          __ mtdscr(tmp2);
+          if (VM_Version::has_mfdscr()) {
+            __ load_const_optimized(tmp2, VM_Version::_dscr_val | 7);
+            __ mtdscr(tmp2);
+          }
           __ li(tmp1, 16);
 
           // Backbranch target aligned to 32-byte. It's not aligned 16-byte
@@ -1365,8 +1372,11 @@ class StubGenerator: public StubCodeGenerator {
           __ bdnz(l_9);                        // Dec CTR and loop if not zero.
 
           // Restore DSCR pre-fetch value.
-          __ load_const_optimized(tmp2, VM_Version::_dscr_val);
-          __ mtdscr(tmp2);
+          if (VM_Version::has_mfdscr()) {
+            __ load_const_optimized(tmp2, VM_Version::_dscr_val);
+            __ mtdscr(tmp2);
+          }
+
       } // FasterArrayCopy
       __ bind(l_6);
 
@@ -1527,9 +1537,10 @@ class StubGenerator: public StubCodeGenerator {
     __ dcbt(R3_ARG1, 0);
 
     // Set DSCR pre-fetch to deepest.
-    __ load_const_optimized(tmp2, VM_Version::_dscr_val | 7);
-    __ mtdscr(tmp2);
-
+    if (VM_Version::has_mfdscr()) {
+      __ load_const_optimized(tmp2, VM_Version::_dscr_val | 7);
+      __ mtdscr(tmp2);
+    }
     __ li(tmp1, 16);
 
     // Backbranch target aligned to 32-byte. Not 16-byte align as
@@ -1549,9 +1560,10 @@ class StubGenerator: public StubCodeGenerator {
     __ bdnz(l_7);                        // Dec CTR and loop if not zero.
 
     // Restore DSCR pre-fetch value.
-    __ load_const_optimized(tmp2, VM_Version::_dscr_val);
-    __ mtdscr(tmp2);
-
+    if (VM_Version::has_mfdscr()) {
+      __ load_const_optimized(tmp2, VM_Version::_dscr_val);
+      __ mtdscr(tmp2);
+    }
 
    } // FasterArrayCopy
 
@@ -1672,9 +1684,10 @@ class StubGenerator: public StubCodeGenerator {
       __ dcbt(R3_ARG1, 0);
 
       // Set DSCR pre-fetch to deepest.
-      __ load_const_optimized(tmp2, VM_Version::_dscr_val | 7);
-      __ mtdscr(tmp2);
-
+      if (VM_Version::has_mfdscr()) {
+        __ load_const_optimized(tmp2, VM_Version::_dscr_val | 7);
+        __ mtdscr(tmp2);
+      }
       __ li(tmp1, 16);
 
       // Backbranch target aligned to 32-byte. Not 16-byte align as
@@ -1694,8 +1707,10 @@ class StubGenerator: public StubCodeGenerator {
       __ bdnz(l_4);
 
       // Restore DSCR pre-fetch value.
-      __ load_const_optimized(tmp2, VM_Version::_dscr_val);
-      __ mtdscr(tmp2);
+      if (VM_Version::has_mfdscr()) {
+        __ load_const_optimized(tmp2, VM_Version::_dscr_val);
+        __ mtdscr(tmp2);
+      }
 
       __ cmpwi(CR0, R5_ARG3, 0);
       __ beq(CR0, l_6);
@@ -1788,9 +1803,10 @@ class StubGenerator: public StubCodeGenerator {
       __ dcbt(R3_ARG1, 0);
 
       // Set DSCR pre-fetch to deepest.
-      __ load_const_optimized(tmp2, VM_Version::_dscr_val | 7);
-      __ mtdscr(tmp2);
-
+      if (VM_Version::has_mfdscr()) {
+        __ load_const_optimized(tmp2, VM_Version::_dscr_val | 7);
+        __ mtdscr(tmp2);
+      }
       __ li(tmp1, 16);
 
       // Backbranch target aligned to 32-byte. Not 16-byte align as
@@ -1810,8 +1826,10 @@ class StubGenerator: public StubCodeGenerator {
       __ bdnz(l_5);                        // Dec CTR and loop if not zero.
 
       // Restore DSCR pre-fetch value.
-      __ load_const_optimized(tmp2, VM_Version::_dscr_val);
-      __ mtdscr(tmp2);
+      if (VM_Version::has_mfdscr()) {
+        __ load_const_optimized(tmp2, VM_Version::_dscr_val);
+        __ mtdscr(tmp2);
+      }
 
    } // FasterArrayCopy
 
@@ -1910,9 +1928,10 @@ class StubGenerator: public StubCodeGenerator {
       __ dcbt(R3_ARG1, 0);
 
       // Set DSCR pre-fetch to deepest.
-      __ load_const_optimized(tmp2, VM_Version::_dscr_val | 7);
-      __ mtdscr(tmp2);
-
+      if (VM_Version::has_mfdscr()) {
+        __ load_const_optimized(tmp2, VM_Version::_dscr_val | 7);
+        __ mtdscr(tmp2);
+      }
       __ li(tmp1, 16);
 
       // Backbranch target aligned to 32-byte. Not 16-byte align as
@@ -1932,8 +1951,10 @@ class StubGenerator: public StubCodeGenerator {
       __ bdnz(l_4);
 
       // Restore DSCR pre-fetch value.
-      __ load_const_optimized(tmp2, VM_Version::_dscr_val);
-      __ mtdscr(tmp2);
+      if (VM_Version::has_mfdscr()) {
+        __ load_const_optimized(tmp2, VM_Version::_dscr_val);
+        __ mtdscr(tmp2);
+      }
 
       __ cmpwi(CR0, R5_ARG3, 0);
       __ beq(CR0, l_1);

src/hotspot/cpu/ppc/vm_version_ppc.cpp

@@ -80,7 +80,9 @@ void VM_Version::initialize() {
             "%zu on this machine", PowerArchitecturePPC64);
 
   // Power 8: Configure Data Stream Control Register.
-  config_dscr();
+  if (VM_Version::has_mfdscr()) {
+    config_dscr();
+  }
 
   if (!UseSIGTRAP) {
     MSG(TrapBasedICMissChecks);
@@ -170,7 +172,8 @@ void VM_Version::initialize() {
   // Create and print feature-string.
   char buf[(num_features+1) * 16]; // Max 16 chars per feature.
   jio_snprintf(buf, sizeof(buf),
-               "ppc64 sha aes%s%s",
+               "ppc64 sha aes%s%s%s",
+               (has_mfdscr()  ? " mfdscr"  : ""),
                (has_darn()    ? " darn"    : ""),
                (has_brw()     ? " brw"     : "")
                // Make sure number of %s matches num_features!
@@ -488,6 +491,7 @@ void VM_Version::determine_features() {
   uint32_t *code = (uint32_t *)a->pc();
   // Keep R3_ARG1 unmodified, it contains &field (see below).
   // Keep R4_ARG2 unmodified, it contains offset = 0 (see below).
+  a->mfdscr(R0);
   a->darn(R7);
   a->brw(R5, R6);
   a->blr();
@@ -524,6 +528,7 @@ void VM_Version::determine_features() {
 
   // determine which instructions are legal.
   int feature_cntr = 0;
+  if (code[feature_cntr++]) features |= mfdscr_m;
   if (code[feature_cntr++]) features |= darn_m;
   if (code[feature_cntr++]) features |= brw_m;
 

src/hotspot/cpu/ppc/vm_version_ppc.hpp

@@ -32,12 +32,14 @@
 class VM_Version: public Abstract_VM_Version {
 protected:
   enum Feature_Flag {
+    mfdscr,
     darn,
     brw,
     num_features // last entry to count features
   };
   enum Feature_Flag_Set {
     unknown_m             = 0,
+    mfdscr_m              = (1 << mfdscr ),
     darn_m                = (1 << darn   ),
     brw_m                 = (1 << brw    ),
     all_features_m        = (unsigned long)-1
@@ -67,8 +69,9 @@ public:
 
   static bool is_determine_features_test_running() { return _is_determine_features_test_running; }
   // CPU instruction support
-  static bool has_darn()    { return (_features & darn_m) != 0; }
-  static bool has_brw()     { return (_features & brw_m) != 0; }
+  static bool has_mfdscr() { return (_features & mfdscr_m) != 0; } // Power8, but may be unavailable (QEMU)
+  static bool has_darn()   { return (_features & darn_m) != 0; }
+  static bool has_brw()    { return (_features & brw_m) != 0; }
 
   // Assembler testing
   static void allow_all();

src/hotspot/cpu/riscv/c2_MacroAssembler_riscv.cpp

@@ -2170,15 +2170,13 @@ void C2_MacroAssembler::enc_cmove_cmp_fp(int cmpFlag, FloatRegister op1, FloatRe
       cmov_cmp_fp_le(op1, op2, dst, src, is_single);
       break;
     case BoolTest::ge:
-      assert(false, "Should go to BoolTest::le case");
-      ShouldNotReachHere();
+      cmov_cmp_fp_ge(op1, op2, dst, src, is_single);
       break;
     case BoolTest::lt:
       cmov_cmp_fp_lt(op1, op2, dst, src, is_single);
       break;
     case BoolTest::gt:
-      assert(false, "Should go to BoolTest::lt case");
-      ShouldNotReachHere();
+      cmov_cmp_fp_gt(op1, op2, dst, src, is_single);
       break;
     default:
       assert(false, "unsupported compare condition");

src/hotspot/cpu/riscv/gc/z/z_riscv.ad

@@ -96,7 +96,6 @@ instruct zLoadP(iRegPNoSp dst, memory mem, iRegPNoSp tmp, rFlagsReg cr)
   match(Set dst (LoadP mem));
   predicate(UseZGC && n->as_Load()->barrier_data() != 0);
   effect(TEMP dst, TEMP tmp, KILL cr);
-  ins_is_late_expanded_null_check_candidate(true);
 
   ins_cost(4 * DEFAULT_COST);
 

src/hotspot/cpu/riscv/macroAssembler_riscv.cpp

@@ -1268,12 +1268,19 @@ void MacroAssembler::cmov_gtu(Register cmp1, Register cmp2, Register dst, Regist
 }
 
 // ----------- cmove, compare float -----------
+//
+// For CmpF/D + CMoveI/L, ordered ones are quite straight and simple,
+// so, just list behaviour of unordered ones as follow.
+//
+// Set dst (CMoveI (Binary cop (CmpF/D op1 op2)) (Binary dst src))
+// (If one or both inputs to the compare are NaN, then)
+//    1. (op1 lt op2) => true  => CMove: dst = src
+//    2. (op1 le op2) => true  => CMove: dst = src
+//    3. (op1 gt op2) => false => CMove: dst = dst
+//    4. (op1 ge op2) => false => CMove: dst = dst
+//    5. (op1 eq op2) => false => CMove: dst = dst
+//    6. (op1 ne op2) => true  => CMove: dst = src
 
-// Move src to dst only if cmp1 == cmp2,
-// otherwise leave dst unchanged, including the case where one of them is NaN.
-// Clarification:
-//   java code      :  cmp1 != cmp2 ? dst : src
-//   transformed to :  CMove dst, (cmp1 eq cmp2), dst, src
 void MacroAssembler::cmov_cmp_fp_eq(FloatRegister cmp1, FloatRegister cmp2, Register dst, Register src, bool is_single) {
   if (UseZicond) {
     if (is_single) {
@@ -1289,7 +1296,7 @@ void MacroAssembler::cmov_cmp_fp_eq(FloatRegister cmp1, FloatRegister cmp2, Regi
   Label no_set;
   if (is_single) {
     // jump if cmp1 != cmp2, including the case of NaN
-    // not jump (i.e. move src to dst) if cmp1 == cmp2
+    // fallthrough (i.e. move src to dst) if cmp1 == cmp2
     float_bne(cmp1, cmp2, no_set);
   } else {
     double_bne(cmp1, cmp2, no_set);
@@ -1298,11 +1305,6 @@ void MacroAssembler::cmov_cmp_fp_eq(FloatRegister cmp1, FloatRegister cmp2, Regi
   bind(no_set);
 }
 
-// Keep dst unchanged only if cmp1 == cmp2,
-// otherwise move src to dst, including the case where one of them is NaN.
-// Clarification:
-//   java code      :  cmp1 == cmp2 ? dst : src
-//   transformed to :  CMove dst, (cmp1 ne cmp2), dst, src
 void MacroAssembler::cmov_cmp_fp_ne(FloatRegister cmp1, FloatRegister cmp2, Register dst, Register src, bool is_single) {
   if (UseZicond) {
     if (is_single) {
@@ -1318,7 +1320,7 @@ void MacroAssembler::cmov_cmp_fp_ne(FloatRegister cmp1, FloatRegister cmp2, Regi
   Label no_set;
   if (is_single) {
     // jump if cmp1 == cmp2
-    // not jump (i.e. move src to dst) if cmp1 != cmp2, including the case of NaN
+    // fallthrough (i.e. move src to dst) if cmp1 != cmp2, including the case of NaN
     float_beq(cmp1, cmp2, no_set);
   } else {
     double_beq(cmp1, cmp2, no_set);
@@ -1327,14 +1329,6 @@ void MacroAssembler::cmov_cmp_fp_ne(FloatRegister cmp1, FloatRegister cmp2, Regi
   bind(no_set);
 }
 
-// When cmp1 <= cmp2 or any of them is NaN then dst = src, otherwise, dst = dst
-// Clarification
-//   scenario 1:
-//     java code      :  cmp2 < cmp1 ? dst : src
-//     transformed to :  CMove dst, (cmp1 le cmp2), dst, src
-//   scenario 2:
-//     java code      :  cmp1 > cmp2 ? dst : src
-//     transformed to :  CMove dst, (cmp1 le cmp2), dst, src
 void MacroAssembler::cmov_cmp_fp_le(FloatRegister cmp1, FloatRegister cmp2, Register dst, Register src, bool is_single) {
   if (UseZicond) {
     if (is_single) {
@@ -1350,7 +1344,7 @@ void MacroAssembler::cmov_cmp_fp_le(FloatRegister cmp1, FloatRegister cmp2, Regi
   Label no_set;
   if (is_single) {
     // jump if cmp1 > cmp2
-    // not jump (i.e. move src to dst) if cmp1 <= cmp2 or either is NaN
+    // fallthrough (i.e. move src to dst) if cmp1 <= cmp2 or either is NaN
     float_bgt(cmp1, cmp2, no_set);
   } else {
     double_bgt(cmp1, cmp2, no_set);
@@ -1359,14 +1353,30 @@ void MacroAssembler::cmov_cmp_fp_le(FloatRegister cmp1, FloatRegister cmp2, Regi
   bind(no_set);
 }
 
-// When cmp1 < cmp2 or any of them is NaN then dst = src, otherwise, dst = dst
-// Clarification
-//   scenario 1:
-//     java code      :  cmp2 <= cmp1 ? dst : src
-//     transformed to :  CMove dst, (cmp1 lt cmp2), dst, src
-//   scenario 2:
-//     java code      :  cmp1 >= cmp2 ? dst : src
-//     transformed to :  CMove dst, (cmp1 lt cmp2), dst, src
+void MacroAssembler::cmov_cmp_fp_ge(FloatRegister cmp1, FloatRegister cmp2, Register dst, Register src, bool is_single) {
+  if (UseZicond) {
+    if (is_single) {
+      fle_s(t0, cmp2, cmp1);
+    } else {
+      fle_d(t0, cmp2, cmp1);
+    }
+    czero_nez(dst, dst, t0);
+    czero_eqz(t0 , src, t0);
+    orr(dst, dst, t0);
+    return;
+  }
+  Label no_set;
+  if (is_single) {
+    // jump if cmp1 < cmp2 or either is NaN
+    // fallthrough (i.e. move src to dst) if cmp1 >= cmp2
+    float_blt(cmp1, cmp2, no_set, false, true);
+  } else {
+    double_blt(cmp1, cmp2, no_set, false, true);
+  }
+  mv(dst, src);
+  bind(no_set);
+}
+
 void MacroAssembler::cmov_cmp_fp_lt(FloatRegister cmp1, FloatRegister cmp2, Register dst, Register src, bool is_single) {
   if (UseZicond) {
     if (is_single) {
@@ -1382,7 +1392,7 @@ void MacroAssembler::cmov_cmp_fp_lt(FloatRegister cmp1, FloatRegister cmp2, Regi
   Label no_set;
   if (is_single) {
     // jump if cmp1 >= cmp2
-    // not jump (i.e. move src to dst) if cmp1 < cmp2 or either is NaN
+    // fallthrough (i.e. move src to dst) if cmp1 < cmp2 or either is NaN
     float_bge(cmp1, cmp2, no_set);
   } else {
     double_bge(cmp1, cmp2, no_set);
@@ -1391,6 +1401,30 @@ void MacroAssembler::cmov_cmp_fp_lt(FloatRegister cmp1, FloatRegister cmp2, Regi
   bind(no_set);
 }
 
+void MacroAssembler::cmov_cmp_fp_gt(FloatRegister cmp1, FloatRegister cmp2, Register dst, Register src, bool is_single) {
+  if (UseZicond) {
+    if (is_single) {
+      flt_s(t0, cmp2, cmp1);
+    } else {
+      flt_d(t0, cmp2, cmp1);
+    }
+    czero_nez(dst, dst, t0);
+    czero_eqz(t0 , src, t0);
+    orr(dst, dst, t0);
+    return;
+  }
+  Label no_set;
+  if (is_single) {
+    // jump if cmp1 <= cmp2 or either is NaN
+    // fallthrough (i.e. move src to dst) if cmp1 > cmp2
+    float_ble(cmp1, cmp2, no_set, false, true);
+  } else {
+    double_ble(cmp1, cmp2, no_set, false, true);
+  }
+  mv(dst, src);
+  bind(no_set);
+}
+
 // Float compare branch instructions
 
 #define INSN(NAME, FLOATCMP, BRANCH)                                                                                    \
@@ -5310,42 +5344,6 @@ void MacroAssembler::add2_with_carry(Register final_dest_hi, Register dest_hi, R
   add(final_dest_hi, dest_hi, carry);
 }
 
-/**
- * Multiply 32 bit by 32 bit first loop.
- */
-void MacroAssembler::multiply_32_x_32_loop(Register x, Register xstart, Register x_xstart,
-                                           Register y, Register y_idx, Register z,
-                                           Register carry, Register product,
-                                           Register idx, Register kdx) {
-  // jlong carry, x[], y[], z[];
-  // for (int idx=ystart, kdx=ystart+1+xstart; idx >= 0; idx--, kdx--) {
-  //     long product = y[idx] * x[xstart] + carry;
-  //     z[kdx] = (int)product;
-  //     carry = product >>> 32;
-  // }
-  // z[xstart] = (int)carry;
-
-  Label L_first_loop, L_first_loop_exit;
-  blez(idx, L_first_loop_exit);
-
-  shadd(t0, xstart, x, t0, LogBytesPerInt);
-  lwu(x_xstart, Address(t0, 0));
-
-  bind(L_first_loop);
-  subiw(idx, idx, 1);
-  shadd(t0, idx, y, t0, LogBytesPerInt);
-  lwu(y_idx, Address(t0, 0));
-  mul(product, x_xstart, y_idx);
-  add(product, product, carry);
-  srli(carry, product, 32);
-  subiw(kdx, kdx, 1);
-  shadd(t0, kdx, z, t0, LogBytesPerInt);
-  sw(product, Address(t0, 0));
-  bgtz(idx, L_first_loop);
-
-  bind(L_first_loop_exit);
-}
-
 /**
  * Multiply 64 bit by 64 bit first loop.
  */
@@ -5562,77 +5560,16 @@ void MacroAssembler::multiply_to_len(Register x, Register xlen, Register y, Regi
   const Register carry = tmp5;
   const Register product = xlen;
   const Register x_xstart = tmp0;
+  const Register jdx = tmp1;
 
   mv(idx, ylen);         // idx = ylen;
   addw(kdx, xlen, ylen); // kdx = xlen+ylen;
   mv(carry, zr);         // carry = 0;
 
-  Label L_multiply_64_x_64_loop, L_done;
-
+  Label L_done;
   subiw(xstart, xlen, 1);
   bltz(xstart, L_done);
 
-  const Register jdx = tmp1;
-
-  if (AvoidUnalignedAccesses) {
-    int base_offset = arrayOopDesc::base_offset_in_bytes(T_INT);
-    assert((base_offset % (UseCompactObjectHeaders ? 4 :
-                           (UseCompressedClassPointers ? 8 : 4))) == 0, "Must be");
-
-    if ((base_offset % 8) == 0) {
-      // multiply_64_x_64_loop emits 8-byte load/store to access two elements
-      // at a time from int arrays x and y. When base_offset is 8 bytes, these
-      // accesses are naturally aligned if both xlen and ylen are even numbers.
-      orr(t0, xlen, ylen);
-      test_bit(t0, t0, 0);
-      beqz(t0, L_multiply_64_x_64_loop);
-    }
-
-    Label L_second_loop_unaligned, L_third_loop, L_third_loop_exit;
-
-    multiply_32_x_32_loop(x, xstart, x_xstart, y, y_idx, z, carry, product, idx, kdx);
-    shadd(t0, xstart, z, t0, LogBytesPerInt);
-    sw(carry, Address(t0, 0));
-
-    bind(L_second_loop_unaligned);
-    mv(carry, zr);
-    mv(jdx, ylen);
-    subiw(xstart, xstart, 1);
-    bltz(xstart, L_done);
-
-    subi(sp, sp, 2 * wordSize);
-    sd(z, Address(sp, 0));
-    sd(zr, Address(sp, wordSize));
-    shadd(t0, xstart, z, t0, LogBytesPerInt);
-    addi(z, t0, 4);
-    shadd(t0, xstart, x, t0, LogBytesPerInt);
-    lwu(product, Address(t0, 0));
-
-    blez(jdx, L_third_loop_exit);
-
-    bind(L_third_loop);
-    subiw(jdx, jdx, 1);
-    shadd(t0, jdx, y, t0, LogBytesPerInt);
-    lwu(t0, Address(t0, 0));
-    mul(t1, t0, product);
-    add(t0, t1, carry);
-    shadd(tmp6, jdx, z, t1, LogBytesPerInt);
-    lwu(t1, Address(tmp6, 0));
-    add(t0, t0, t1);
-    sw(t0, Address(tmp6, 0));
-    srli(carry, t0, 32);
-    bgtz(jdx, L_third_loop);
-
-    bind(L_third_loop_exit);
-    ld(z, Address(sp, 0));
-    addi(sp, sp, 2 * wordSize);
-    shadd(t0, xstart, z, t0, LogBytesPerInt);
-    sw(carry, Address(t0, 0));
-
-    j(L_second_loop_unaligned);
-  }
-
-  bind(L_multiply_64_x_64_loop);
   multiply_64_x_64_loop(x, xstart, x_xstart, y, y_idx, z, carry, product, idx, kdx);
 
   Label L_second_loop_aligned;

src/hotspot/cpu/riscv/macroAssembler_riscv.hpp

@@ -660,7 +660,9 @@ class MacroAssembler: public Assembler {
   void cmov_cmp_fp_eq(FloatRegister cmp1, FloatRegister cmp2, Register dst, Register src, bool is_single);
   void cmov_cmp_fp_ne(FloatRegister cmp1, FloatRegister cmp2, Register dst, Register src, bool is_single);
   void cmov_cmp_fp_le(FloatRegister cmp1, FloatRegister cmp2, Register dst, Register src, bool is_single);
+  void cmov_cmp_fp_ge(FloatRegister cmp1, FloatRegister cmp2, Register dst, Register src, bool is_single);
   void cmov_cmp_fp_lt(FloatRegister cmp1, FloatRegister cmp2, Register dst, Register src, bool is_single);
+  void cmov_cmp_fp_gt(FloatRegister cmp1, FloatRegister cmp2, Register dst, Register src, bool is_single);
 
  public:
   // We try to follow risc-v asm menomics.
@@ -1382,10 +1384,6 @@ public:
   void adc(Register dst, Register src1, Register src2, Register carry);
   void add2_with_carry(Register final_dest_hi, Register dest_hi, Register dest_lo,
                        Register src1, Register src2, Register carry);
-  void multiply_32_x_32_loop(Register x, Register xstart, Register x_xstart,
-                             Register y, Register y_idx, Register z,
-                             Register carry, Register product,
-                             Register idx, Register kdx);
   void multiply_64_x_64_loop(Register x, Register xstart, Register x_xstart,
                              Register y, Register y_idx, Register z,
                              Register carry, Register product,

src/hotspot/cpu/riscv/riscv.ad

@@ -2619,10 +2619,6 @@ ins_attrib ins_alignment(4);    // Required alignment attribute (must
                                 // compute_padding() function must be
                                 // provided for the instruction
 
-// Whether this node is expanded during code emission into a sequence of
-// instructions and the first instruction can perform an implicit null check.
-ins_attrib ins_is_late_expanded_null_check_candidate(false);
-
 //----------OPERANDS-----------------------------------------------------------
 // Operand definitions must precede instruction definitions for correct parsing
 // in the ADLC because operands constitute user defined types which are used in
@@ -8435,6 +8431,17 @@ instruct castVV(vReg dst)
   ins_pipe(pipe_class_empty);
 %}
 
+instruct castVVMask(vRegMask dst)
+%{
+  match(Set dst (CastVV dst));
+
+  size(0);
+  format %{ "# castVV of $dst" %}
+  ins_encode(/* empty encoding */);
+  ins_cost(0);
+  ins_pipe(pipe_class_empty);
+%}
+
 // ============================================================================
 // Convert Instructions
 

src/hotspot/cpu/riscv/vm_version_riscv.cpp

@@ -203,15 +203,15 @@ void VM_Version::common_initialize() {
     }
   }
 
-  // Misc Intrinsics could depend on RVV
+  // Misc Intrinsics that could depend on RVV.
 
-  if (UseZba || UseRVV) {
+  if (!AvoidUnalignedAccesses && (UseZba || UseRVV)) {
     if (FLAG_IS_DEFAULT(UseCRC32Intrinsics)) {
       FLAG_SET_DEFAULT(UseCRC32Intrinsics, true);
     }
   } else {
     if (!FLAG_IS_DEFAULT(UseCRC32Intrinsics)) {
-      warning("CRC32 intrinsic requires Zba or RVV instructions (not available on this CPU)");
+      warning("CRC32 intrinsic are not available on this CPU.");
     }
     FLAG_SET_DEFAULT(UseCRC32Intrinsics, false);
   }
@@ -325,20 +325,40 @@ void VM_Version::c2_initialize() {
     FLAG_SET_DEFAULT(UseMulAddIntrinsic, true);
   }
 
-  if (FLAG_IS_DEFAULT(UseMultiplyToLenIntrinsic)) {
-    FLAG_SET_DEFAULT(UseMultiplyToLenIntrinsic, true);
+  if (!AvoidUnalignedAccesses) {
+    if (FLAG_IS_DEFAULT(UseMultiplyToLenIntrinsic)) {
+      FLAG_SET_DEFAULT(UseMultiplyToLenIntrinsic, true);
+    }
+  } else if (UseMultiplyToLenIntrinsic) {
+    warning("Intrinsics for BigInteger.multiplyToLen() not available on this CPU.");
+    FLAG_SET_DEFAULT(UseMultiplyToLenIntrinsic, false);
   }
 
-  if (FLAG_IS_DEFAULT(UseSquareToLenIntrinsic)) {
-    FLAG_SET_DEFAULT(UseSquareToLenIntrinsic, true);
+  if (!AvoidUnalignedAccesses) {
+    if (FLAG_IS_DEFAULT(UseSquareToLenIntrinsic)) {
+      FLAG_SET_DEFAULT(UseSquareToLenIntrinsic, true);
+    }
+  } else if (UseSquareToLenIntrinsic) {
+    warning("Intrinsics for BigInteger.squareToLen() not available on this CPU.");
+    FLAG_SET_DEFAULT(UseSquareToLenIntrinsic, false);
   }
 
-  if (FLAG_IS_DEFAULT(UseMontgomeryMultiplyIntrinsic)) {
-    FLAG_SET_DEFAULT(UseMontgomeryMultiplyIntrinsic, true);
+  if (!AvoidUnalignedAccesses) {
+    if (FLAG_IS_DEFAULT(UseMontgomeryMultiplyIntrinsic)) {
+      FLAG_SET_DEFAULT(UseMontgomeryMultiplyIntrinsic, true);
+    }
+  } else if (UseMontgomeryMultiplyIntrinsic) {
+    warning("Intrinsics for BigInteger.montgomeryMultiply() not available on this CPU.");
+    FLAG_SET_DEFAULT(UseMontgomeryMultiplyIntrinsic, false);
   }
 
-  if (FLAG_IS_DEFAULT(UseMontgomerySquareIntrinsic)) {
-    FLAG_SET_DEFAULT(UseMontgomerySquareIntrinsic, true);
+  if (!AvoidUnalignedAccesses) {
+    if (FLAG_IS_DEFAULT(UseMontgomerySquareIntrinsic)) {
+      FLAG_SET_DEFAULT(UseMontgomerySquareIntrinsic, true);
+    }
+  } else if (UseMontgomerySquareIntrinsic) {
+    warning("Intrinsics for BigInteger.montgomerySquare() not available on this CPU.");
+    FLAG_SET_DEFAULT(UseMontgomerySquareIntrinsic, false);
   }
 
   // Adler32

src/hotspot/cpu/s390/frame_s390.hpp

@@ -410,7 +410,7 @@
 
   // C2I adapter frames:
   //
-  // STACK (interpreted called from compiled, on entry to template interpreter):
+  // STACK (interpreted called from compiled, on entry to frame manager):
   //
   //       [TOP_C2I_FRAME]
   //       [JIT_FRAME]

src/hotspot/cpu/s390/register_s390.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2016, 2025, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2016, 2024, Oracle and/or its affiliates. All rights reserved.
  * Copyright (c) 2016, 2023 SAP SE. All rights reserved.
  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  *
@@ -414,7 +414,7 @@ constexpr FloatRegister  Z_FARG2   = Z_F2;
 constexpr FloatRegister  Z_FARG3   = Z_F4;
 constexpr FloatRegister  Z_FARG4   = Z_F6;
 
-// Register declarations to be used in template interpreter assembly code.
+// Register declarations to be used in frame manager assembly code.
 // Use only non-volatile registers in order to keep values across C-calls.
 
 // Register to cache the integer value on top of the operand stack.
@@ -439,7 +439,7 @@ constexpr Register      Z_bcp          = Z_R13;
 // Bytecode which is dispatched (short lived!).
 constexpr Register      Z_bytecode     = Z_R14;
 
-// Temporary registers to be used within template interpreter. We can use
+// Temporary registers to be used within frame manager. We can use
 // the nonvolatile ones because the call stub has saved them.
 // Use only non-volatile registers in order to keep values across C-calls.
 constexpr Register Z_tmp_1 =  Z_R10;

src/hotspot/cpu/s390/runtime_s390.cpp

@@ -118,7 +118,7 @@ ExceptionBlob* OptoRuntime::generate_exception_blob() {
   __ z_lgr(Z_SP, saved_sp);
 
   // [Z_RET] isn't null was possible in hotspot5 but not in sapjvm6.
-  // C2I adapter extensions are now removed by a resize in the template interpreter
+  // C2I adapter extensions are now removed by a resize in the frame manager
   // (unwind_initial_activation_pending_exception).
 #ifdef ASSERT
   __ z_ltgr(handle_exception, handle_exception);

src/hotspot/cpu/s390/sharedRuntime_s390.cpp

@@ -2139,7 +2139,7 @@ static address gen_c2i_adapter(MacroAssembler  *masm,
   Register  value       = Z_R12;
 
   // Remember the senderSP so we can pop the interpreter arguments off of the stack.
-  // In addition, template interpreter expects initial_caller_sp in Z_R10.
+  // In addition, frame manager expects initial_caller_sp in Z_R10.
   __ z_lgr(sender_SP, Z_SP);
 
   // This should always fit in 14 bit immediate.

src/hotspot/cpu/s390/stubGenerator_s390.cpp

@@ -115,7 +115,7 @@ class StubGenerator: public StubCodeGenerator {
   //   [SP+176]  - thread                   : Thread*
   //
   address generate_call_stub(address& return_address) {
-    // Set up a new C frame, copy Java arguments, call template interpreter
+    // Set up a new C frame, copy Java arguments, call frame manager
     // or native_entry, and process result.
 
     StubGenStubId stub_id = StubGenStubId::call_stub_id;
@@ -272,10 +272,10 @@ class StubGenerator: public StubCodeGenerator {
 
     BLOCK_COMMENT("call {");
     {
-      // Call template interpreter or native entry.
+      // Call frame manager or native entry.
 
       //
-      // Register state on entry to template interpreter / native entry:
+      // Register state on entry to frame manager / native entry:
       //
       //   Z_ARG1 = r_top_of_arguments_addr  - intptr_t *sender tos (prepushed)
       //                                       Lesp = (SP) + copied_arguments_offset - 8
@@ -290,7 +290,7 @@ class StubGenerator: public StubCodeGenerator {
       __ z_lgr(Z_esp, r_top_of_arguments_addr);
 
       //
-      // Stack on entry to template interpreter / native entry:
+      // Stack on entry to frame manager / native entry:
       //
       //     F0      [TOP_IJAVA_FRAME_ABI]
       //             [outgoing Java arguments]
@@ -300,7 +300,7 @@ class StubGenerator: public StubCodeGenerator {
       //
 
       // Do a light-weight C-call here, r_new_arg_entry holds the address
-      // of the interpreter entry point (template interpreter or native entry)
+      // of the interpreter entry point (frame manager or native entry)
       // and save runtime-value of return_pc in return_address
       // (call by reference argument).
       return_address = __ call_stub(r_new_arg_entry);
@@ -309,11 +309,11 @@ class StubGenerator: public StubCodeGenerator {
 
     {
       BLOCK_COMMENT("restore registers {");
-      // Returned from template interpreter or native entry.
+      // Returned from frame manager or native entry.
       // Now pop frame, process result, and return to caller.
 
       //
-      // Stack on exit from template interpreter / native entry:
+      // Stack on exit from frame manager / native entry:
       //
       //     F0      [ABI]
       //             ...
@@ -330,7 +330,7 @@ class StubGenerator: public StubCodeGenerator {
       __ pop_frame();
 
       // Reload some volatile registers which we've spilled before the call
-      // to template interpreter / native entry.
+      // to frame manager / native entry.
       // Access all locals via frame pointer, because we know nothing about
       // the topmost frame's size.
       __ z_lg(r_arg_result_addr, result_address_offset, r_entryframe_fp);

src/hotspot/cpu/s390/templateInterpreterGenerator_s390.cpp

@@ -1217,7 +1217,7 @@ void TemplateInterpreterGenerator::generate_fixed_frame(bool native_call) {
 
 // Various method entries
 
-// Math function, template interpreter must set up an interpreter state, etc.
+// Math function, frame manager must set up an interpreter state, etc.
 address TemplateInterpreterGenerator::generate_math_entry(AbstractInterpreter::MethodKind kind) {
 
   // Decide what to do: Use same platform specific instructions and runtime calls as compilers.

src/hotspot/cpu/x86/assembler_x86.cpp

@@ -15681,6 +15681,8 @@ void Assembler::pusha_uncached() { // 64bit
     // Push pair of original stack pointer along with remaining registers
     // at 16B aligned boundary.
     push2p(rax, r31);
+    // Restore the original contents of RAX register.
+    movq(rax, Address(rax));
     push2p(r30, r29);
     push2p(r28, r27);
     push2p(r26, r25);

src/hotspot/cpu/x86/c2_MacroAssembler_x86.cpp

@@ -4655,6 +4655,7 @@ static void convertF2I_slowpath(C2_MacroAssembler& masm, C2GeneralStub<Register,
   __ subptr(rsp, 8);
   __ movdbl(Address(rsp), src);
   __ call(RuntimeAddress(target));
+  // APX REX2 encoding for pop(dst) increases the stub size by 1 byte.
   __ pop(dst);
   __ jmp(stub.continuation());
 #undef __
@@ -4687,7 +4688,9 @@ void C2_MacroAssembler::convertF2I(BasicType dst_bt, BasicType src_bt, Register
     }
   }
 
-  auto stub = C2CodeStub::make<Register, XMMRegister, address>(dst, src, slowpath_target, 23, convertF2I_slowpath);
+  // Using the APX extended general purpose registers increases the instruction encoding size by 1 byte.
+  int max_size = 23 + (UseAPX ? 1 : 0);
+  auto stub = C2CodeStub::make<Register, XMMRegister, address>(dst, src, slowpath_target, max_size, convertF2I_slowpath);
   jcc(Assembler::equal, stub->entry());
   bind(stub->continuation());
 }

src/hotspot/cpu/x86/gc/shenandoah/shenandoahBarrierSetAssembler_x86.cpp

@@ -353,7 +353,7 @@ void ShenandoahBarrierSetAssembler::load_reference_barrier(MacroAssembler* masm,
 
   // The rest is saved with the optimized path
 
-  uint num_saved_regs = 4 + (dst != rax ? 1 : 0) + 4;
+  uint num_saved_regs = 4 + (dst != rax ? 1 : 0) + 4 + (UseAPX ? 16 : 0);
   __ subptr(rsp, num_saved_regs * wordSize);
   uint slot = num_saved_regs;
   if (dst != rax) {
@@ -367,6 +367,25 @@ void ShenandoahBarrierSetAssembler::load_reference_barrier(MacroAssembler* masm,
   __ movptr(Address(rsp, (--slot) * wordSize), r9);
   __ movptr(Address(rsp, (--slot) * wordSize), r10);
   __ movptr(Address(rsp, (--slot) * wordSize), r11);
+  // Save APX extended registers r16–r31 if enabled
+  if (UseAPX) {
+    __ movptr(Address(rsp, (--slot) * wordSize), r16);
+    __ movptr(Address(rsp, (--slot) * wordSize), r17);
+    __ movptr(Address(rsp, (--slot) * wordSize), r18);
+    __ movptr(Address(rsp, (--slot) * wordSize), r19);
+    __ movptr(Address(rsp, (--slot) * wordSize), r20);
+    __ movptr(Address(rsp, (--slot) * wordSize), r21);
+    __ movptr(Address(rsp, (--slot) * wordSize), r22);
+    __ movptr(Address(rsp, (--slot) * wordSize), r23);
+    __ movptr(Address(rsp, (--slot) * wordSize), r24);
+    __ movptr(Address(rsp, (--slot) * wordSize), r25);
+    __ movptr(Address(rsp, (--slot) * wordSize), r26);
+    __ movptr(Address(rsp, (--slot) * wordSize), r27);
+    __ movptr(Address(rsp, (--slot) * wordSize), r28);
+    __ movptr(Address(rsp, (--slot) * wordSize), r29);
+    __ movptr(Address(rsp, (--slot) * wordSize), r30);
+    __ movptr(Address(rsp, (--slot) * wordSize), r31);
+  }
   // r12-r15 are callee saved in all calling conventions
   assert(slot == 0, "must use all slots");
 
@@ -398,6 +417,25 @@ void ShenandoahBarrierSetAssembler::load_reference_barrier(MacroAssembler* masm,
     __ super_call_VM_leaf(CAST_FROM_FN_PTR(address, ShenandoahRuntime::load_reference_barrier_phantom), arg0, arg1);
   }
 
+  // Restore APX extended registers r31–r16 if previously saved
+  if (UseAPX) {
+    __ movptr(r31, Address(rsp, (slot++) * wordSize));
+    __ movptr(r30, Address(rsp, (slot++) * wordSize));
+    __ movptr(r29, Address(rsp, (slot++) * wordSize));
+    __ movptr(r28, Address(rsp, (slot++) * wordSize));
+    __ movptr(r27, Address(rsp, (slot++) * wordSize));
+    __ movptr(r26, Address(rsp, (slot++) * wordSize));
+    __ movptr(r25, Address(rsp, (slot++) * wordSize));
+    __ movptr(r24, Address(rsp, (slot++) * wordSize));
+    __ movptr(r23, Address(rsp, (slot++) * wordSize));
+    __ movptr(r22, Address(rsp, (slot++) * wordSize));
+    __ movptr(r21, Address(rsp, (slot++) * wordSize));
+    __ movptr(r20, Address(rsp, (slot++) * wordSize));
+    __ movptr(r19, Address(rsp, (slot++) * wordSize));
+    __ movptr(r18, Address(rsp, (slot++) * wordSize));
+    __ movptr(r17, Address(rsp, (slot++) * wordSize));
+    __ movptr(r16, Address(rsp, (slot++) * wordSize));
+  }
   __ movptr(r11, Address(rsp, (slot++) * wordSize));
   __ movptr(r10, Address(rsp, (slot++) * wordSize));
   __ movptr(r9,  Address(rsp, (slot++) * wordSize));

src/hotspot/cpu/x86/gc/z/z_x86_64.ad

@@ -118,10 +118,6 @@ instruct zLoadP(rRegP dst, memory mem, rFlagsReg cr)
   predicate(UseZGC && n->as_Load()->barrier_data() != 0);
   match(Set dst (LoadP mem));
   effect(TEMP dst, KILL cr);
-  // The main load is a candidate to implement implicit null checks. The
-  // barrier's slow path includes an identical reload, which does not need to be
-  // registered in the exception table because it is dominated by the main one.
-  ins_is_late_expanded_null_check_candidate(true);
 
   ins_cost(125);
 

src/hotspot/cpu/x86/stubDeclarations_x86.hpp

@@ -30,7 +30,7 @@
                                       do_arch_blob,                     \
                                       do_arch_entry,                    \
                                       do_arch_entry_init)               \
-  do_arch_blob(initial, 20000 WINDOWS_ONLY(+1000))                      \
+  do_arch_blob(initial, PRODUCT_ONLY(20000) NOT_PRODUCT(21000) WINDOWS_ONLY(+1000))                      \
   do_stub(initial, verify_mxcsr)                                        \
   do_arch_entry(x86, initial, verify_mxcsr, verify_mxcsr_entry,         \
                 verify_mxcsr_entry)                                     \
@@ -239,7 +239,7 @@
                                     do_arch_blob,                       \
                                     do_arch_entry,                      \
                                     do_arch_entry_init)                 \
-  do_arch_blob(final, 31000                                             \
+  do_arch_blob(final, 33000                                             \
                WINDOWS_ONLY(+22000) ZGC_ONLY(+20000))                   \
 
 #endif // CPU_X86_STUBDECLARATIONS_HPP

src/hotspot/cpu/x86/stubGenerator_x86_64_cbrt.cpp

@@ -46,6 +46,12 @@
 //
 /******************************************************************************/
 
+/* Represents 0x7FFFFFFFFFFFFFFF double precision in lower 64 bits*/
+ATTRIBUTE_ALIGNED(16) static const juint _ABS_MASK[] =
+{
+    4294967295, 2147483647, 0, 0
+};
+
 ATTRIBUTE_ALIGNED(4) static const juint _SIG_MASK[] =
 {
     0, 1032192
@@ -188,10 +194,10 @@ address StubGenerator::generate_libmCbrt() {
   StubCodeMark mark(this, stub_id);
   address start = __ pc();
 
-  Label L_2TAG_PACKET_0_0_1, L_2TAG_PACKET_1_0_1, L_2TAG_PACKET_2_0_1, L_2TAG_PACKET_3_0_1;
-  Label L_2TAG_PACKET_4_0_1, L_2TAG_PACKET_5_0_1, L_2TAG_PACKET_6_0_1;
+  Label L_2TAG_PACKET_0_0_1, L_2TAG_PACKET_1_0_1, L_2TAG_PACKET_2_0_1;
   Label B1_1, B1_2, B1_4;
 
+  address ABS_MASK        = (address)_ABS_MASK;
   address SIG_MASK        = (address)_SIG_MASK;
   address EXP_MASK        = (address)_EXP_MASK;
   address EXP_MSK2        = (address)_EXP_MSK2;
@@ -208,8 +214,12 @@ address StubGenerator::generate_libmCbrt() {
   __ enter(); // required for proper stackwalking of RuntimeStub frame
 
   __ bind(B1_1);
-  __ subq(rsp, 24);
-  __ movsd(Address(rsp), xmm0);
+  __ ucomisd(xmm0, ExternalAddress(ZERON), r11 /*rscratch*/);
+  __ jcc(Assembler::equal, L_2TAG_PACKET_1_0_1); // Branch only if x is +/- zero or NaN
+  __ movq(xmm1, xmm0);
+  __ andpd(xmm1, ExternalAddress(ABS_MASK), r11 /*rscratch*/);
+  __ ucomisd(xmm1, ExternalAddress(INF), r11 /*rscratch*/);
+  __ jcc(Assembler::equal, B1_4); // Branch only if x is +/- INF
 
   __ bind(B1_2);
   __ movq(xmm7, xmm0);
@@ -228,8 +238,6 @@ address StubGenerator::generate_libmCbrt() {
   __ andl(rdx, rax);
   __ cmpl(rdx, 0);
   __ jcc(Assembler::equal, L_2TAG_PACKET_0_0_1); // Branch only if |x| is denormalized
-  __ cmpl(rdx, 524032);
-  __ jcc(Assembler::equal, L_2TAG_PACKET_1_0_1); // Branch only if |x| is INF or NaN
   __ shrl(rdx, 8);
   __ shrq(r9, 8);
   __ andpd(xmm2, xmm0);
@@ -297,8 +305,6 @@ address StubGenerator::generate_libmCbrt() {
   __ andl(rdx, rax);
   __ shrl(rdx, 8);
   __ shrq(r9, 8);
-  __ cmpl(rdx, 0);
-  __ jcc(Assembler::equal, L_2TAG_PACKET_3_0_1); // Branch only if |x| is zero
   __ andpd(xmm2, xmm0);
   __ andpd(xmm0, xmm5);
   __ orpd(xmm3, xmm2);
@@ -322,41 +328,10 @@ address StubGenerator::generate_libmCbrt() {
   __ psllq(xmm7, 52);
   __ jmp(L_2TAG_PACKET_2_0_1);
 
-  __ bind(L_2TAG_PACKET_3_0_1);
-  __ cmpq(r9, 0);
-  __ jcc(Assembler::notEqual, L_2TAG_PACKET_4_0_1); // Branch only if x is negative zero
-  __ xorpd(xmm0, xmm0);
-  __ jmp(B1_4);
-
-  __ bind(L_2TAG_PACKET_4_0_1);
-  __ movsd(xmm0, ExternalAddress(ZERON), r11 /*rscratch*/);
-  __ jmp(B1_4);
-
   __ bind(L_2TAG_PACKET_1_0_1);
-  __ movl(rax, Address(rsp, 4));
-  __ movl(rdx, Address(rsp));
-  __ movl(rcx, rax);
-  __ andl(rcx, 2147483647);
-  __ cmpl(rcx, 2146435072);
-  __ jcc(Assembler::above, L_2TAG_PACKET_5_0_1); // Branch only if |x| is NaN
-  __ cmpl(rdx, 0);
-  __ jcc(Assembler::notEqual, L_2TAG_PACKET_5_0_1); // Branch only if |x| is NaN
-  __ cmpl(rax, 2146435072);
-  __ jcc(Assembler::notEqual, L_2TAG_PACKET_6_0_1); // Branch only if x is negative INF
-  __ movsd(xmm0, ExternalAddress(INF), r11 /*rscratch*/);
-  __ jmp(B1_4);
-
-  __ bind(L_2TAG_PACKET_6_0_1);
-  __ movsd(xmm0, ExternalAddress(NEG_INF), r11 /*rscratch*/);
-  __ jmp(B1_4);
-
-  __ bind(L_2TAG_PACKET_5_0_1);
-  __ movsd(xmm0, Address(rsp));
   __ addsd(xmm0, xmm0);
-  __ movq(Address(rsp, 8), xmm0);
 
   __ bind(B1_4);
-  __ addq(rsp, 24);
   __ leave(); // required for proper stackwalking of RuntimeStub frame
   __ ret(0);
 

src/hotspot/cpu/x86/templateInterpreterGenerator_x86_64.cpp

@@ -465,19 +465,13 @@ address TemplateInterpreterGenerator::generate_math_entry(AbstractInterpreter::M
       __ call_VM_leaf0(CAST_FROM_FN_PTR(address, SharedRuntime::dtan));
     }
   } else if (kind == Interpreter::java_lang_math_tanh) {
-    if (StubRoutines::dtanh() != nullptr) {
+      assert(StubRoutines::dtanh() != nullptr, "not initialized");
       __ movdbl(xmm0, Address(rsp, wordSize));
       __ call(RuntimeAddress(CAST_FROM_FN_PTR(address, StubRoutines::dtanh())));
-    } else {
-      return nullptr; // Fallback to default implementation
-    }
   } else if (kind == Interpreter::java_lang_math_cbrt) {
-    if (StubRoutines::dcbrt() != nullptr) {
-      __ movdbl(xmm0, Address(rsp, wordSize));
-      __ call(RuntimeAddress(CAST_FROM_FN_PTR(address, StubRoutines::dcbrt())));
-    } else {
-      return nullptr; // Fallback to default implementation
-    }
+    assert(StubRoutines::dcbrt() != nullptr, "not initialized");
+    __ movdbl(xmm0, Address(rsp, wordSize));
+    __ call(RuntimeAddress(CAST_FROM_FN_PTR(address, StubRoutines::dcbrt())));
   } else if (kind == Interpreter::java_lang_math_abs) {
     assert(StubRoutines::x86::double_sign_mask() != nullptr, "not initialized");
     __ movdbl(xmm0, Address(rsp, wordSize));

src/hotspot/cpu/x86/vm_version_x86.cpp

@@ -440,7 +440,6 @@ class VM_Version_StubGenerator: public StubCodeGenerator {
     __ andl(rax, Address(rbp, in_bytes(VM_Version::xem_xcr0_offset()))); // xcr0 bits apx_f
     __ jcc(Assembler::equal, vector_save_restore);
 
-#ifndef PRODUCT
     bool save_apx = UseAPX;
     VM_Version::set_apx_cpuFeatures();
     UseAPX = true;
@@ -457,7 +456,6 @@ class VM_Version_StubGenerator: public StubCodeGenerator {
     __ movq(Address(rsi, 8), r31);
 
     UseAPX = save_apx;
-#endif
     __ bind(vector_save_restore);
     //
     // Check if OS has enabled XGETBV instruction to access XCR0
@@ -1022,8 +1020,6 @@ void VM_Version::get_processor_features() {
   if (UseAPX && !apx_supported) {
     warning("UseAPX is not supported on this CPU, setting it to false");
     FLAG_SET_DEFAULT(UseAPX, false);
-  } else if (FLAG_IS_DEFAULT(UseAPX)) {
-    FLAG_SET_DEFAULT(UseAPX, apx_supported ? true : false);
   }
 
   if (!UseAPX) {
@@ -2111,7 +2107,7 @@ bool VM_Version::is_intel_cascade_lake() {
 // has improved implementation of 64-byte load/stores and so the default
 // threshold is set to 0 for these platforms.
 int VM_Version::avx3_threshold() {
-  return (is_intel_family_core() &&
+  return (is_intel_server_family() &&
           supports_serialize() &&
           FLAG_IS_DEFAULT(AVX3Threshold)) ? 0 : AVX3Threshold;
 }
@@ -3151,17 +3147,11 @@ bool VM_Version::os_supports_apx_egprs() {
   if (!supports_apx_f()) {
     return false;
   }
-  // Enable APX support for product builds after
-  // completion of planned features listed in JDK-8329030.
-#if !defined(PRODUCT)
   if (_cpuid_info.apx_save[0] != egpr_test_value() ||
       _cpuid_info.apx_save[1] != egpr_test_value()) {
     return false;
   }
   return true;
-#else
-  return false;
-#endif
 }
 
 uint VM_Version::cores_per_cpu() {

src/hotspot/cpu/x86/x86_64.ad

@@ -2055,10 +2055,6 @@ ins_attrib ins_alignment(1);    // Required alignment attribute (must
                                 // compute_padding() function must be
                                 // provided for the instruction
 
-// Whether this node is expanded during code emission into a sequence of
-// instructions and the first instruction can perform an implicit null check.
-ins_attrib ins_is_late_expanded_null_check_candidate(false);
-
 //----------OPERANDS-----------------------------------------------------------
 // Operand definitions must precede instruction definitions for correct parsing
 // in the ADLC because operands constitute user defined types which are used in

src/hotspot/os/aix/os_aix.cpp

@@ -1261,6 +1261,69 @@ void os::pd_print_cpu_info(outputStream* st, char* buf, size_t buflen) {
   // Nothing to do beyond of what os::print_cpu_info() does.
 }
 
+static char saved_jvm_path[MAXPATHLEN] = {0};
+
+// Find the full path to the current module, libjvm.so.
+void os::jvm_path(char *buf, jint buflen) {
+  // Error checking.
+  if (buflen < MAXPATHLEN) {
+    assert(false, "must use a large-enough buffer");
+    buf[0] = '\0';
+    return;
+  }
+  // Lazy resolve the path to current module.
+  if (saved_jvm_path[0] != 0) {
+    strcpy(buf, saved_jvm_path);
+    return;
+  }
+
+  Dl_info dlinfo;
+  int ret = dladdr(CAST_FROM_FN_PTR(void *, os::jvm_path), &dlinfo);
+  assert(ret != 0, "cannot locate libjvm");
+  char* rp = os::realpath((char *)dlinfo.dli_fname, buf, buflen);
+  assert(rp != nullptr, "error in realpath(): maybe the 'path' argument is too long?");
+
+  // If executing unit tests we require JAVA_HOME to point to the real JDK.
+  if (Arguments::executing_unit_tests()) {
+    // Look for JAVA_HOME in the environment.
+    char* java_home_var = ::getenv("JAVA_HOME");
+    if (java_home_var != nullptr && java_home_var[0] != 0) {
+
+      // Check the current module name "libjvm.so".
+      const char* p = strrchr(buf, '/');
+      if (p == nullptr) {
+        return;
+      }
+      assert(strstr(p, "/libjvm") == p, "invalid library name");
+
+      stringStream ss(buf, buflen);
+      rp = os::realpath(java_home_var, buf, buflen);
+      if (rp == nullptr) {
+        return;
+      }
+
+      assert((int)strlen(buf) < buflen, "Ran out of buffer room");
+      ss.print("%s/lib", buf);
+
+      if (0 == access(buf, F_OK)) {
+        // Use current module name "libjvm.so"
+        ss.print("/%s/libjvm%s", Abstract_VM_Version::vm_variant(), JNI_LIB_SUFFIX);
+        assert(strcmp(buf + strlen(buf) - strlen(JNI_LIB_SUFFIX), JNI_LIB_SUFFIX) == 0,
+               "buf has been truncated");
+      } else {
+        // Go back to path of .so
+        rp = os::realpath((char *)dlinfo.dli_fname, buf, buflen);
+        if (rp == nullptr) {
+          return;
+        }
+      }
+    }
+  }
+
+  strncpy(saved_jvm_path, buf, sizeof(saved_jvm_path));
+  saved_jvm_path[sizeof(saved_jvm_path) - 1] = '\0';
+}
+
 ////////////////////////////////////////////////////////////////////////////////
 // Virtual Memory
 

src/hotspot/os/bsd/os_bsd.cpp

@@ -154,8 +154,7 @@ julong os::Bsd::available_memory() {
   assert(kerr == KERN_SUCCESS,
          "host_statistics64 failed - check mach_host_self() and count");
   if (kerr == KERN_SUCCESS) {
-    // free_count is just a lowerbound, other page categories can be freed too and make memory available
-    available = (vmstat.free_count + vmstat.inactive_count + vmstat.purgeable_count) * os::vm_page_size();
+    available = vmstat.free_count * os::vm_page_size();
   }
 #endif
   return available;
@@ -1483,6 +1482,83 @@ void os::print_memory_info(outputStream* st) {
   st->cr();
 }
 
+static char saved_jvm_path[MAXPATHLEN] = {0};
+
+// Find the full path to the current module, libjvm
+void os::jvm_path(char *buf, jint buflen) {
+  // Error checking.
+  if (buflen < MAXPATHLEN) {
+    assert(false, "must use a large-enough buffer");
+    buf[0] = '\0';
+    return;
+  }
+  // Lazy resolve the path to current module.
+  if (saved_jvm_path[0] != 0) {
+    strcpy(buf, saved_jvm_path);
+    return;
+  }
+
+  char dli_fname[MAXPATHLEN];
+  dli_fname[0] = '\0';
+  bool ret = dll_address_to_library_name(
+                                         CAST_FROM_FN_PTR(address, os::jvm_path),
+                                         dli_fname, sizeof(dli_fname), nullptr);
+  assert(ret, "cannot locate libjvm");
+  char *rp = nullptr;
+  if (ret && dli_fname[0] != '\0') {
+    rp = os::realpath(dli_fname, buf, buflen);
+  }
+  if (rp == nullptr) {
+    return;
+  }
+
+  // If executing unit tests we require JAVA_HOME to point to the real JDK.
+  if (Arguments::executing_unit_tests()) {
+    // Look for JAVA_HOME in the environment.
+    char* java_home_var = ::getenv("JAVA_HOME");
+    if (java_home_var != nullptr && java_home_var[0] != 0) {
+
+      // Check the current module name "libjvm"
+      const char* p = strrchr(buf, '/');
+      assert(strstr(p, "/libjvm") == p, "invalid library name");
+
+      stringStream ss(buf, buflen);
+      rp = os::realpath(java_home_var, buf, buflen);
+      if (rp == nullptr) {
+        return;
+      }
+
+      assert((int)strlen(buf) < buflen, "Ran out of buffer space");
+      // Add the appropriate library and JVM variant subdirs
+      ss.print("%s/lib/%s", buf, Abstract_VM_Version::vm_variant());
+
+      if (0 != access(buf, F_OK)) {
+        ss.reset();
+        ss.print("%s/lib", buf);
+      }
+
+      // If the path exists within JAVA_HOME, add the JVM library name
+      // to complete the path to JVM being overridden.  Otherwise fallback
+      // to the path to the current library.
+      if (0 == access(buf, F_OK)) {
+        // Use current module name "libjvm"
+        ss.print("/libjvm%s", JNI_LIB_SUFFIX);
+        assert(strcmp(buf + strlen(buf) - strlen(JNI_LIB_SUFFIX), JNI_LIB_SUFFIX) == 0,
+               "buf has been truncated");
+      } else {
+        // Fall back to path of current library
+        rp = os::realpath(dli_fname, buf, buflen);
+        if (rp == nullptr) {
+          return;
+        }
+      }
+    }
+  }
+
+  strncpy(saved_jvm_path, buf, MAXPATHLEN);
+  saved_jvm_path[MAXPATHLEN - 1] = '\0';
+}
+
 ////////////////////////////////////////////////////////////////////////////////
 // Virtual Memory
 

src/hotspot/os/linux/globals_linux.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2005, 2025, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 2005, 2024, 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
@@ -35,6 +35,9 @@
                          range,                                         \
                          constraint)                                    \
                                                                         \
+  product(bool, UseOprofile, false,                                     \
+        "(Deprecated) enable support for Oprofile profiler")            \
+                                                                        \
   product(bool, UseTransparentHugePages, false,                         \
           "Use MADV_HUGEPAGE for large pages")                          \
                                                                         \

src/hotspot/os/linux/os_linux.cpp

@@ -2746,9 +2746,118 @@ void os::get_summary_cpu_info(char* cpuinfo, size_t length) {
 #endif
 }
 
+static char saved_jvm_path[MAXPATHLEN] = {0};
+
+// Find the full path to the current module, libjvm.so
+void os::jvm_path(char *buf, jint buflen) {
+  // Error checking.
+  if (buflen < MAXPATHLEN) {
+    assert(false, "must use a large-enough buffer");
+    buf[0] = '\0';
+    return;
+  }
+  // Lazy resolve the path to current module.
+  if (saved_jvm_path[0] != 0) {
+    strcpy(buf, saved_jvm_path);
+    return;
+  }
+
+  char dli_fname[MAXPATHLEN];
+  dli_fname[0] = '\0';
+  bool ret = dll_address_to_library_name(
+                                         CAST_FROM_FN_PTR(address, os::jvm_path),
+                                         dli_fname, sizeof(dli_fname), nullptr);
+  assert(ret, "cannot locate libjvm");
+  char *rp = nullptr;
+  if (ret && dli_fname[0] != '\0') {
+    rp = os::realpath(dli_fname, buf, buflen);
+  }
+  if (rp == nullptr) {
+    return;
+  }
+
+  // If executing unit tests we require JAVA_HOME to point to the real JDK.
+  if (Arguments::executing_unit_tests()) {
+    // Look for JAVA_HOME in the environment.
+    char* java_home_var = ::getenv("JAVA_HOME");
+    if (java_home_var != nullptr && java_home_var[0] != 0) {
+
+      // Check the current module name "libjvm.so".
+      const char* p = strrchr(buf, '/');
+      if (p == nullptr) {
+        return;
+      }
+      assert(strstr(p, "/libjvm") == p, "invalid library name");
+
+      stringStream ss(buf, buflen);
+      rp = os::realpath(java_home_var, buf, buflen);
+      if (rp == nullptr) {
+        return;
+      }
+
+      assert((int)strlen(buf) < buflen, "Ran out of buffer room");
+      ss.print("%s/lib", buf);
+
+      if (0 == access(buf, F_OK)) {
+        // Use current module name "libjvm.so"
+        ss.print("/%s/libjvm%s", Abstract_VM_Version::vm_variant(), JNI_LIB_SUFFIX);
+        assert(strcmp(buf + strlen(buf) - strlen(JNI_LIB_SUFFIX), JNI_LIB_SUFFIX) == 0,
+               "buf has been truncated");
+      } else {
+        // Go back to path of .so
+        rp = os::realpath(dli_fname, buf, buflen);
+        if (rp == nullptr) {
+          return;
+        }
+      }
+    }
+  }
+
+  strncpy(saved_jvm_path, buf, MAXPATHLEN);
+  saved_jvm_path[MAXPATHLEN - 1] = '\0';
+}
+
 ////////////////////////////////////////////////////////////////////////////////
 // Virtual Memory
 
+// Rationale behind this function:
+//  current (Mon Apr 25 20:12:18 MSD 2005) oprofile drops samples without executable
+//  mapping for address (see lookup_dcookie() in the kernel module), thus we cannot get
+//  samples for JITted code. Here we create private executable mapping over the code cache
+//  and then we can use standard (well, almost, as mapping can change) way to provide
+//  info for the reporting script by storing timestamp and location of symbol
+void linux_wrap_code(char* base, size_t size) {
+  static volatile jint cnt = 0;
+
+  static_assert(sizeof(off_t) == 8, "Expected Large File Support in this file");
+
+  if (!UseOprofile) {
+    return;
+  }
+
+  char buf[PATH_MAX+1];
+  int num = Atomic::add(&cnt, 1);
+
+  snprintf(buf, sizeof(buf), "%s/hs-vm-%d-%d",
+           os::get_temp_directory(), os::current_process_id(), num);
+  unlink(buf);
+
+  int fd = ::open(buf, O_CREAT | O_RDWR, S_IRWXU);
+
+  if (fd != -1) {
+    off_t rv = ::lseek(fd, size-2, SEEK_SET);
+    if (rv != (off_t)-1) {
+      if (::write(fd, "", 1) == 1) {
+        mmap(base, size,
+             PROT_READ|PROT_WRITE|PROT_EXEC,
+             MAP_PRIVATE|MAP_FIXED|MAP_NORESERVE, fd, 0);
+      }
+    }
+    ::close(fd);
+    unlink(buf);
+  }
+}
+
 static bool recoverable_mmap_error(int err) {
   // See if the error is one we can let the caller handle. This
   // list of errno values comes from JBS-6843484. I can't find a

src/hotspot/os/posix/os_posix.cpp

@@ -59,7 +59,6 @@
 #ifdef AIX
 #include "loadlib_aix.hpp"
 #include "os_aix.hpp"
-#include "porting_aix.hpp"
 #endif
 #ifdef LINUX
 #include "os_linux.hpp"
@@ -1061,95 +1060,6 @@ bool os::same_files(const char* file1, const char* file2) {
   return is_same;
 }
 
-static char saved_jvm_path[MAXPATHLEN] = {0};
-
-// Find the full path to the current module, libjvm.so
-void os::jvm_path(char *buf, jint buflen) {
-  // Error checking.
-  if (buflen < MAXPATHLEN) {
-    assert(false, "must use a large-enough buffer");
-    buf[0] = '\0';
-    return;
-  }
-  // Lazy resolve the path to current module.
-  if (saved_jvm_path[0] != 0) {
-    strcpy(buf, saved_jvm_path);
-    return;
-  }
-
-  const char* fname;
-#ifdef AIX
-  Dl_info dlinfo;
-  int ret = dladdr(CAST_FROM_FN_PTR(void *, os::jvm_path), &dlinfo);
-  assert(ret != 0, "cannot locate libjvm");
-  if (ret == 0) {
-    return;
-  }
-  fname = dlinfo.dli_fname;
-#else
-  char dli_fname[MAXPATHLEN];
-  dli_fname[0] = '\0';
-  bool ret = dll_address_to_library_name(
-                                         CAST_FROM_FN_PTR(address, os::jvm_path),
-                                         dli_fname, sizeof(dli_fname), nullptr);
-  assert(ret, "cannot locate libjvm");
-  if (!ret) {
-    return;
-  }
-  fname = dli_fname;
-#endif // AIX
-  char* rp = nullptr;
-  if (fname[0] != '\0') {
-    rp = os::realpath(fname, buf, buflen);
-  }
-  if (rp == nullptr) {
-    return;
-  }
-
-  // If executing unit tests we require JAVA_HOME to point to the real JDK.
-  if (Arguments::executing_unit_tests()) {
-    // Look for JAVA_HOME in the environment.
-    char* java_home_var = ::getenv("JAVA_HOME");
-    if (java_home_var != nullptr && java_home_var[0] != 0) {
-
-      // Check the current module name "libjvm.so".
-      const char* p = strrchr(buf, '/');
-      if (p == nullptr) {
-        return;
-      }
-      assert(strstr(p, "/libjvm") == p, "invalid library name");
-
-      stringStream ss(buf, buflen);
-      rp = os::realpath(java_home_var, buf, buflen);
-      if (rp == nullptr) {
-        return;
-      }
-
-      assert((int)strlen(buf) < buflen, "Ran out of buffer room");
-      ss.print("%s/lib", buf);
-
-      // If the path exists within JAVA_HOME, add the VM variant directory and JVM
-      // library name to complete the path to JVM being overridden.  Otherwise fallback
-      // to the path to the current library.
-      if (0 == access(buf, F_OK)) {
-        // Use current module name "libjvm.so"
-        ss.print("/%s/libjvm%s", Abstract_VM_Version::vm_variant(), JNI_LIB_SUFFIX);
-        assert(strcmp(buf + strlen(buf) - strlen(JNI_LIB_SUFFIX), JNI_LIB_SUFFIX) == 0,
-               "buf has been truncated");
-      } else {
-        // Go back to path of .so
-        rp = os::realpath(fname, buf, buflen);
-        if (rp == nullptr) {
-          return;
-        }
-      }
-    }
-  }
-
-  strncpy(saved_jvm_path, buf, MAXPATHLEN);
-  saved_jvm_path[MAXPATHLEN - 1] = '\0';
-}
-
 // Called when creating the thread.  The minimum stack sizes have already been calculated
 size_t os::Posix::get_initial_stack_size(ThreadType thr_type, size_t req_stack_size) {
   size_t stack_size;

src/hotspot/os/windows/os_windows.cpp

@@ -2623,7 +2623,6 @@ LONG WINAPI topLevelExceptionFilter(struct _EXCEPTION_POINTERS* exceptionInfo) {
     return Handle_Exception(exceptionInfo, VM_Version::cpuinfo_cont_addr());
   }
 
-#if !defined(PRODUCT)
   if ((exception_code == EXCEPTION_ACCESS_VIOLATION) &&
       VM_Version::is_cpuinfo_segv_addr_apx(pc)) {
     // Verify that OS save/restore APX registers.
@@ -2631,7 +2630,6 @@ LONG WINAPI topLevelExceptionFilter(struct _EXCEPTION_POINTERS* exceptionInfo) {
     return Handle_Exception(exceptionInfo, VM_Version::cpuinfo_cont_addr_apx());
   }
 #endif
-#endif
 
 #ifdef CAN_SHOW_REGISTERS_ON_ASSERT
   if (VMError::was_assert_poison_crash(exception_record)) {

src/hotspot/os_cpu/bsd_aarch64/os_bsd_aarch64.cpp

@@ -81,12 +81,14 @@
 #endif
 
 #define SPELL_REG_SP "sp"
+#define SPELL_REG_FP "fp"
 
 #ifdef __APPLE__
 // see darwin-xnu/osfmk/mach/arm/_structs.h
 
 // 10.5 UNIX03 member name prefixes
 #define DU3_PREFIX(s, m) __ ## s.__ ## m
+#endif
 
 #define context_x    uc_mcontext->DU3_PREFIX(ss,x)
 #define context_fp   uc_mcontext->DU3_PREFIX(ss,fp)
@@ -95,31 +97,6 @@
 #define context_pc   uc_mcontext->DU3_PREFIX(ss,pc)
 #define context_cpsr uc_mcontext->DU3_PREFIX(ss,cpsr)
 #define context_esr  uc_mcontext->DU3_PREFIX(es,esr)
-#endif
-
-#ifdef __FreeBSD__
-# define context_x  uc_mcontext.mc_gpregs.gp_x
-# define context_fp context_x[REG_FP]
-# define context_lr uc_mcontext.mc_gpregs.gp_lr
-# define context_sp uc_mcontext.mc_gpregs.gp_sp
-# define context_pc uc_mcontext.mc_gpregs.gp_elr
-#endif
-
-#ifdef __NetBSD__
-# define context_x  uc_mcontext.__gregs
-# define context_fp uc_mcontext.__gregs[_REG_FP]
-# define context_lr uc_mcontext.__gregs[_REG_LR]
-# define context_sp uc_mcontext.__gregs[_REG_SP]
-# define context_pc uc_mcontext.__gregs[_REG_ELR]
-#endif
-
-#ifdef __OpenBSD__
-# define context_x  sc_x
-# define context_fp sc_x[REG_FP]
-# define context_lr sc_lr
-# define context_sp sc_sp
-# define context_pc sc_elr
-#endif
 
 #define REG_BCP context_x[22]
 
@@ -520,11 +497,9 @@ int os::extra_bang_size_in_bytes() {
   return 0;
 }
 
-#ifdef __APPLE__
 void os::current_thread_enable_wx(WXMode mode) {
   pthread_jit_write_protect_np(mode == WXExec);
 }
-#endif
 
 static inline void atomic_copy64(const volatile void *src, volatile void *dst) {
   *(jlong *) dst = *(const jlong *) src;

src/hotspot/os_cpu/bsd_x86/os_bsd_x86.cpp

@@ -429,13 +429,11 @@ bool PosixSignals::pd_hotspot_signal_handler(int sig, siginfo_t* info,
       stub = VM_Version::cpuinfo_cont_addr();
     }
 
-#if !defined(PRODUCT) && defined(_LP64)
     if ((sig == SIGSEGV || sig == SIGBUS) && VM_Version::is_cpuinfo_segv_addr_apx(pc)) {
       // Verify that OS save/restore APX registers.
       stub = VM_Version::cpuinfo_cont_addr_apx();
       VM_Version::clear_apx_test_state();
     }
-#endif
 
     // We test if stub is already set (by the stack overflow code
     // above) so it is not overwritten by the code that follows. This

src/hotspot/os_cpu/linux_x86/os_linux_x86.cpp

@@ -255,13 +255,11 @@ bool PosixSignals::pd_hotspot_signal_handler(int sig, siginfo_t* info,
       stub = VM_Version::cpuinfo_cont_addr();
     }
 
-#if !defined(PRODUCT) && defined(_LP64)
     if ((sig == SIGSEGV) && VM_Version::is_cpuinfo_segv_addr_apx(pc)) {
       // Verify that OS save/restore APX registers.
       stub = VM_Version::cpuinfo_cont_addr_apx();
       VM_Version::clear_apx_test_state();
     }
-#endif
 
     if (thread->thread_state() == _thread_in_Java) {
       // Java thread running in Java code => find exception handler if any

src/hotspot/share/adlc/main.cpp

@@ -481,3 +481,7 @@ int get_legal_text(FileBuff &fbuf, char **legal_text)
   *legal_text = legal_start;
   return (int) (legal_end - legal_start);
 }
+
+void *operator new( size_t size, int, const char *, int ) throw() {
+  return ::operator new( size );
+}

src/hotspot/share/adlc/output_h.cpp

@@ -1626,8 +1626,6 @@ void ArchDesc::declareClasses(FILE *fp) {
     while (attr != nullptr) {
       if (strcmp (attr->_ident, "ins_is_TrapBasedCheckNode") == 0) {
         fprintf(fp, "  virtual bool           is_TrapBasedCheckNode() const { return %s; }\n", attr->_val);
-      } else if (strcmp (attr->_ident, "ins_is_late_expanded_null_check_candidate") == 0) {
-        fprintf(fp, "  virtual bool           is_late_expanded_null_check_candidate() const { return %s; }\n", attr->_val);
       } else if (strcmp (attr->_ident, "ins_cost") != 0 &&
           strncmp(attr->_ident, "ins_field_", 10) != 0 &&
           // Must match function in node.hpp: return type bool, no prefix "ins_".

src/hotspot/share/c1/c1_Runtime1.cpp

@@ -818,7 +818,7 @@ JRT_ENTRY(void, Runtime1::deoptimize(JavaThread* current, jint trap_request))
   Deoptimization::DeoptReason reason = Deoptimization::trap_request_reason(trap_request);
 
   if (action == Deoptimization::Action_make_not_entrant) {
-    if (nm->make_not_entrant(nmethod::ChangeReason::C1_deoptimize)) {
+    if (nm->make_not_entrant("C1 deoptimize")) {
       if (reason == Deoptimization::Reason_tenured) {
         MethodData* trap_mdo = Deoptimization::get_method_data(current, method, true /*create_if_missing*/);
         if (trap_mdo != nullptr) {
@@ -1110,7 +1110,7 @@ JRT_ENTRY(void, Runtime1::patch_code(JavaThread* current, C1StubId stub_id ))
     // safepoint, but if it's still alive then make it not_entrant.
     nmethod* nm = CodeCache::find_nmethod(caller_frame.pc());
     if (nm != nullptr) {
-      nm->make_not_entrant(nmethod::ChangeReason::C1_codepatch);
+      nm->make_not_entrant("C1 code patch");
     }
 
     Deoptimization::deoptimize_frame(current, caller_frame.id());
@@ -1358,7 +1358,7 @@ void Runtime1::patch_code(JavaThread* current, C1StubId stub_id) {
     // Make sure the nmethod is invalidated, i.e. made not entrant.
     nmethod* nm = CodeCache::find_nmethod(caller_frame.pc());
     if (nm != nullptr) {
-      nm->make_not_entrant(nmethod::ChangeReason::C1_deoptimize_for_patching);
+      nm->make_not_entrant("C1 deoptimize for patching");
     }
   }
 
@@ -1486,7 +1486,7 @@ JRT_ENTRY(void, Runtime1::predicate_failed_trap(JavaThread* current))
 
   nmethod* nm = CodeCache::find_nmethod(caller_frame.pc());
   assert (nm != nullptr, "no more nmethod?");
-  nm->make_not_entrant(nmethod::ChangeReason::C1_predicate_failed_trap);
+  nm->make_not_entrant("C1 predicate failed trap");
 
   methodHandle m(current, nm->method());
   MethodData* mdo = m->method_data();

src/hotspot/share/c1/c1_ValueMap.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1999, 2023, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 1999, 2025, 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
@@ -187,7 +187,13 @@ class ValueNumberingVisitor: public InstructionVisitor {
   void do_Convert        (Convert*         x) { /* nothing to do */ }
   void do_NullCheck      (NullCheck*       x) { /* nothing to do */ }
   void do_TypeCast       (TypeCast*        x) { /* nothing to do */ }
-  void do_NewInstance    (NewInstance*     x) { /* nothing to do */ }
+  void do_NewInstance    (NewInstance*     x) {
+    ciInstanceKlass* c = x->klass();
+    if (c != nullptr && !c->is_initialized() &&
+        (!c->is_loaded() || c->has_class_initializer())) {
+      kill_memory();
+    }
+  }
   void do_NewTypeArray   (NewTypeArray*    x) { /* nothing to do */ }
   void do_NewObjectArray (NewObjectArray*  x) { /* nothing to do */ }
   void do_NewMultiArray  (NewMultiArray*   x) { /* nothing to do */ }

src/hotspot/share/cds/cdsConfig.cpp

@@ -110,24 +110,12 @@ const char* CDSConfig::default_archive_path() {
   // before CDSConfig::ergo_initialize() is called.
   assert(_cds_ergo_initialize_started, "sanity");
   if (_default_archive_path == nullptr) {
+    char jvm_path[JVM_MAXPATHLEN];
+    os::jvm_path(jvm_path, sizeof(jvm_path));
+    char *end = strrchr(jvm_path, *os::file_separator());
+    if (end != nullptr) *end = '\0';
     stringStream tmp;
-    if (is_vm_statically_linked()) {
-      // It's easier to form the path using JAVA_HOME as os::jvm_path
-      // gives the path to the launcher executable on static JDK.
-      const char* subdir = WINDOWS_ONLY("bin") NOT_WINDOWS("lib");
-      tmp.print("%s%s%s%s%s%sclasses",
-                Arguments::get_java_home(), os::file_separator(),
-                subdir, os::file_separator(),
-                Abstract_VM_Version::vm_variant(), os::file_separator());
-    } else {
-      // Assume .jsa is in the same directory where libjvm resides on
-      // non-static JDK.
-      char jvm_path[JVM_MAXPATHLEN];
-      os::jvm_path(jvm_path, sizeof(jvm_path));
-      char *end = strrchr(jvm_path, *os::file_separator());
-      if (end != nullptr) *end = '\0';
-      tmp.print("%s%sclasses", jvm_path, os::file_separator());
-    }
+    tmp.print("%s%sclasses", jvm_path, os::file_separator());
 #ifdef _LP64
     if (!UseCompressedOops) {
       tmp.print_raw("_nocoops");

src/hotspot/share/cds/cds_globals.hpp

@@ -147,7 +147,7 @@
   product(bool, AOTVerifyTrainingData, trueInDebug, DIAGNOSTIC,             \
           "Verify archived training data")                                  \
                                                                             \
-  product(bool, AOTCompileEagerly, false, DIAGNOSTIC,                       \
+  product(bool, AOTCompileEagerly, false, EXPERIMENTAL,                     \
           "Compile methods as soon as possible")                            \
                                                                             \
   /* AOT Code flags */                                                      \

src/hotspot/share/cds/metaspaceShared.cpp

@@ -837,11 +837,10 @@ void MetaspaceShared::preload_and_dump(TRAPS) {
       struct stat st;
       if (os::stat(AOTCache, &st) != 0) {
         tty->print_cr("AOTCache creation failed: %s", AOTCache);
-        vm_exit(0);
       } else {
         tty->print_cr("AOTCache creation is complete: %s " INT64_FORMAT " bytes", AOTCache, (int64_t)(st.st_size));
-        vm_exit(0);
       }
+      vm_direct_exit(0);
     }
   }
 }

src/hotspot/share/ci/ciInstanceKlass.cpp

@@ -549,6 +549,11 @@ bool ciInstanceKlass::compute_has_trusted_loader() {
   return java_lang_ClassLoader::is_trusted_loader(loader_oop);
 }
 
+bool ciInstanceKlass::has_class_initializer() {
+  VM_ENTRY_MARK;
+  return get_instanceKlass()->class_initializer() != nullptr;
+}
+
 // ------------------------------------------------------------------
 // ciInstanceKlass::find_method
 //

src/hotspot/share/ci/ciInstanceKlass.hpp

@@ -231,6 +231,8 @@ public:
   ciInstanceKlass* unique_concrete_subklass();
   bool has_finalizable_subclass();
 
+  bool has_class_initializer();
+
   bool contains_field_offset(int offset);
 
   // Get the instance of java.lang.Class corresponding to

src/hotspot/share/ci/ciReplay.cpp

@@ -802,7 +802,7 @@ class CompileReplay : public StackObj {
     // Make sure the existence of a prior compile doesn't stop this one
     nmethod* nm = (entry_bci != InvocationEntryBci) ? method->lookup_osr_nmethod_for(entry_bci, comp_level, true) : method->code();
     if (nm != nullptr) {
-      nm->make_not_entrant(nmethod::ChangeReason::CI_replay);
+      nm->make_not_entrant("CI replay");
     }
     replay_state = this;
     CompileBroker::compile_method(methodHandle(THREAD, method), entry_bci, comp_level,

src/hotspot/share/classfile/classFileParser.cpp

@@ -154,8 +154,6 @@
 
 #define JAVA_25_VERSION                   69
 
-#define JAVA_26_VERSION                   70
-
 void ClassFileParser::set_class_bad_constant_seen(short bad_constant) {
   assert((bad_constant == JVM_CONSTANT_Module ||
           bad_constant == JVM_CONSTANT_Package) && _major_version >= JAVA_9_VERSION,
@@ -3740,6 +3738,7 @@ void ClassFileParser::apply_parsed_class_metadata(
   _cp->set_pool_holder(this_klass);
   this_klass->set_constants(_cp);
   this_klass->set_fieldinfo_stream(_fieldinfo_stream);
+  this_klass->set_fieldinfo_search_table(_fieldinfo_search_table);
   this_klass->set_fields_status(_fields_status);
   this_klass->set_methods(_methods);
   this_klass->set_inner_classes(_inner_classes);
@@ -3749,6 +3748,8 @@ void ClassFileParser::apply_parsed_class_metadata(
   this_klass->set_permitted_subclasses(_permitted_subclasses);
   this_klass->set_record_components(_record_components);
 
+  DEBUG_ONLY(FieldInfoStream::validate_search_table(_cp, _fieldinfo_stream, _fieldinfo_search_table));
+
   // Delay the setting of _local_interfaces and _transitive_interfaces until after
   // initialize_supers() in fill_instance_klass(). It is because the _local_interfaces could
   // be shared with _transitive_interfaces and _transitive_interfaces may be shared with
@@ -5056,6 +5057,7 @@ void ClassFileParser::fill_instance_klass(InstanceKlass* ik,
   // note that is not safe to use the fields in the parser from this point on
   assert(nullptr == _cp, "invariant");
   assert(nullptr == _fieldinfo_stream, "invariant");
+  assert(nullptr == _fieldinfo_search_table, "invariant");
   assert(nullptr == _fields_status, "invariant");
   assert(nullptr == _methods, "invariant");
   assert(nullptr == _inner_classes, "invariant");
@@ -5276,6 +5278,7 @@ ClassFileParser::ClassFileParser(ClassFileStream* stream,
   _super_klass(),
   _cp(nullptr),
   _fieldinfo_stream(nullptr),
+  _fieldinfo_search_table(nullptr),
   _fields_status(nullptr),
   _methods(nullptr),
   _inner_classes(nullptr),
@@ -5352,6 +5355,7 @@ void ClassFileParser::clear_class_metadata() {
   // deallocated if classfile parsing returns an error.
   _cp = nullptr;
   _fieldinfo_stream = nullptr;
+  _fieldinfo_search_table = nullptr;
   _fields_status = nullptr;
   _methods = nullptr;
   _inner_classes = nullptr;
@@ -5374,6 +5378,7 @@ ClassFileParser::~ClassFileParser() {
   if (_fieldinfo_stream != nullptr) {
     MetadataFactory::free_array<u1>(_loader_data, _fieldinfo_stream);
   }
+  MetadataFactory::free_array<u1>(_loader_data, _fieldinfo_search_table);
 
   if (_fields_status != nullptr) {
     MetadataFactory::free_array<FieldStatus>(_loader_data, _fields_status);
@@ -5774,6 +5779,7 @@ void ClassFileParser::post_process_parsed_stream(const ClassFileStream* const st
   _fieldinfo_stream =
     FieldInfoStream::create_FieldInfoStream(_temp_field_info, _java_fields_count,
                                             injected_fields_count, loader_data(), CHECK);
+  _fieldinfo_search_table = FieldInfoStream::create_search_table(_cp, _fieldinfo_stream, _loader_data, CHECK);
   _fields_status =
     MetadataFactory::new_array<FieldStatus>(_loader_data, _temp_field_info->length(),
                                             FieldStatus(0), CHECK);

src/hotspot/share/classfile/classFileParser.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1997, 2024, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 1997, 2025, 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
@@ -123,6 +123,7 @@ class ClassFileParser {
   const InstanceKlass* _super_klass;
   ConstantPool* _cp;
   Array<u1>* _fieldinfo_stream;
+  Array<u1>* _fieldinfo_search_table;
   Array<FieldStatus>* _fields_status;
   Array<Method*>* _methods;
   Array<u2>* _inner_classes;

src/hotspot/share/classfile/fieldLayoutBuilder.cpp

@@ -301,7 +301,7 @@ void FieldLayout::reconstruct_layout(const InstanceKlass* ik, bool& has_instance
   BasicType last_type;
   int last_offset = -1;
   while (ik != nullptr) {
-    for (AllFieldStream fs(ik->fieldinfo_stream(), ik->constants()); !fs.done(); fs.next()) {
+    for (AllFieldStream fs(ik); !fs.done(); fs.next()) {
       BasicType type = Signature::basic_type(fs.signature());
       // distinction between static and non-static fields is missing
       if (fs.access_flags().is_static()) continue;
@@ -461,7 +461,7 @@ void FieldLayout::print(outputStream* output, bool is_static, const InstanceKlas
         bool found = false;
         const InstanceKlass* ik = super;
         while (!found && ik != nullptr) {
-          for (AllFieldStream fs(ik->fieldinfo_stream(), ik->constants()); !fs.done(); fs.next()) {
+          for (AllFieldStream fs(ik); !fs.done(); fs.next()) {
             if (fs.offset() == b->offset()) {
               output->print_cr(" @%d \"%s\" %s %d/%d %s",
                   b->offset(),

src/hotspot/share/classfile/javaClasses.cpp

@@ -967,6 +967,13 @@ void java_lang_Class::fixup_mirror(Klass* k, TRAPS) {
       Array<u1>* new_fis = FieldInfoStream::create_FieldInfoStream(fields, java_fields, injected_fields, k->class_loader_data(), CHECK);
       ik->set_fieldinfo_stream(new_fis);
       MetadataFactory::free_array<u1>(k->class_loader_data(), old_stream);
+
+      Array<u1>* old_table = ik->fieldinfo_search_table();
+      Array<u1>* search_table = FieldInfoStream::create_search_table(ik->constants(), new_fis, k->class_loader_data(), CHECK);
+      ik->set_fieldinfo_search_table(search_table);
+      MetadataFactory::free_array<u1>(k->class_loader_data(), old_table);
+
+      DEBUG_ONLY(FieldInfoStream::validate_search_table(ik->constants(), new_fis, search_table));
     }
   }
 

src/hotspot/share/classfile/stackMapTable.cpp

@@ -132,8 +132,16 @@ bool StackMapTable::match_stackmap(
 }
 
 void StackMapTable::check_jump_target(
-    StackMapFrame* frame, int32_t target, TRAPS) const {
+    StackMapFrame* frame, int bci, int offset, TRAPS) const {
   ErrorContext ctx;
+  // Jump targets must be within the method and the method size is limited. See JVMS 4.11
+  int min_offset = -1 * max_method_code_size;
+  if (offset < min_offset || offset > max_method_code_size) {
+    frame->verifier()->verify_error(ErrorContext::bad_stackmap(bci, frame),
+        "Illegal target of jump or branch (bci %d + offset %d)", bci, offset);
+    return;
+  }
+  int target = bci + offset;
   bool match = match_stackmap(
     frame, target, true, false, &ctx, CHECK_VERIFY(frame->verifier()));
   if (!match || (target < 0 || target >= _code_length)) {

src/hotspot/share/classfile/stackMapTable.hpp

@@ -67,7 +67,7 @@ class StackMapTable : public StackObj {
 
   // Check jump instructions. Make sure there are no uninitialized
   // instances on backward branch.
-  void check_jump_target(StackMapFrame* frame, int32_t target, TRAPS) const;
+  void check_jump_target(StackMapFrame* frame, int bci, int offset, TRAPS) const;
 
   // The following methods are only used inside this class.
 

src/hotspot/share/classfile/stringTable.cpp

@@ -32,6 +32,7 @@
 #include "classfile/javaClasses.inline.hpp"
 #include "classfile/stringTable.hpp"
 #include "classfile/vmClasses.hpp"
+#include "compiler/compileBroker.hpp"
 #include "gc/shared/collectedHeap.hpp"
 #include "gc/shared/oopStorage.inline.hpp"
 #include "gc/shared/oopStorageSet.hpp"
@@ -115,6 +116,7 @@ OopStorage*   StringTable::_oop_storage;
 
 static size_t _current_size = 0;
 static volatile size_t _items_count = 0;
+DEBUG_ONLY(static bool _disable_interning_during_cds_dump = false);
 
 volatile bool _alt_hash = false;
 
@@ -346,6 +348,10 @@ bool StringTable::has_work() {
   return Atomic::load_acquire(&_has_work);
 }
 
+size_t StringTable::items_count_acquire() {
+  return Atomic::load_acquire(&_items_count);
+}
+
 void StringTable::trigger_concurrent_work() {
   // Avoid churn on ServiceThread
   if (!has_work()) {
@@ -504,6 +510,9 @@ oop StringTable::intern(const char* utf8_string, TRAPS) {
 }
 
 oop StringTable::intern(const StringWrapper& name, TRAPS) {
+  assert(!Atomic::load_acquire(&_disable_interning_during_cds_dump),
+         "All threads that may intern strings should have been stopped before CDS starts copying the interned string table");
+
   // shared table always uses java_lang_String::hash_code
   unsigned int hash = hash_wrapped_string(name);
   oop found_string = lookup_shared(name, hash);
@@ -793,7 +802,7 @@ void StringTable::verify() {
 }
 
 // Verification and comp
-class VerifyCompStrings : StackObj {
+class StringTable::VerifyCompStrings : StackObj {
   static unsigned string_hash(oop const& str) {
     return java_lang_String::hash_code_noupdate(str);
   }
@@ -805,7 +814,7 @@ class VerifyCompStrings : StackObj {
                               string_hash, string_equals> _table;
  public:
   size_t _errors;
-  VerifyCompStrings() : _table(unsigned(_items_count / 8) + 1, 0 /* do not resize */), _errors(0) {}
+  VerifyCompStrings() : _table(unsigned(items_count_acquire() / 8) + 1, 0 /* do not resize */), _errors(0) {}
   bool operator()(WeakHandle* val) {
     oop s = val->resolve();
     if (s == nullptr) {
@@ -939,20 +948,31 @@ oop StringTable::lookup_shared(const jchar* name, int len) {
   return _shared_table.lookup(wrapped_name, java_lang_String::hash_code(name, len), 0);
 }
 
-// This is called BEFORE we enter the CDS safepoint. We can allocate heap objects.
-// This should be called when we know no more strings will be added (which will be easy
-// to guarantee because CDS runs with a single Java thread. See JDK-8253495.)
+// This is called BEFORE we enter the CDS safepoint. We can still allocate Java object arrays to
+// be used by the shared strings table.
 void StringTable::allocate_shared_strings_array(TRAPS) {
   if (!CDSConfig::is_dumping_heap()) {
     return;
   }
-  assert(CDSConfig::allow_only_single_java_thread(), "No more interned strings can be added");
 
-  if (_items_count > (size_t)max_jint) {
-    fatal("Too many strings to be archived: %zu", _items_count);
+  CompileBroker::wait_for_no_active_tasks();
+
+  precond(CDSConfig::allow_only_single_java_thread());
+
+  // At this point, no more strings will be added:
+  // - There's only a single Java thread (this thread). It no longer executes Java bytecodes
+  //   so JIT compilation will eventually stop.
+  // - CompileBroker has no more active tasks, so all JIT requests have been processed.
+
+  // This flag will be cleared after intern table dumping has completed, so we can run the
+  // compiler again (for future AOT method compilation, etc).
+  DEBUG_ONLY(Atomic::release_store(&_disable_interning_during_cds_dump, true));
+
+  if (items_count_acquire() > (size_t)max_jint) {
+    fatal("Too many strings to be archived: %zu", items_count_acquire());
   }
 
-  int total = (int)_items_count;
+  int total = (int)items_count_acquire();
   size_t single_array_size = objArrayOopDesc::object_size(total);
 
   log_info(aot)("allocated string table for %d strings", total);
@@ -972,7 +992,7 @@ void StringTable::allocate_shared_strings_array(TRAPS) {
       // This can only happen if you have an extremely large number of classes that
       // refer to more than 16384 * 16384 = 26M interned strings! Not a practical concern
       // but bail out for safety.
-      log_error(aot)("Too many strings to be archived: %zu", _items_count);
+      log_error(aot)("Too many strings to be archived: %zu", items_count_acquire());
       MetaspaceShared::unrecoverable_writing_error();
     }
 
@@ -1070,7 +1090,7 @@ oop StringTable::init_shared_strings_array() {
 
 void StringTable::write_shared_table() {
   _shared_table.reset();
-  CompactHashtableWriter writer((int)_items_count, ArchiveBuilder::string_stats());
+  CompactHashtableWriter writer((int)items_count_acquire(), ArchiveBuilder::string_stats());
 
   int index = 0;
   auto copy_into_shared_table = [&] (WeakHandle* val) {
@@ -1084,6 +1104,8 @@ void StringTable::write_shared_table() {
   };
   _local_table->do_safepoint_scan(copy_into_shared_table);
   writer.dump(&_shared_table, "string");
+
+  DEBUG_ONLY(Atomic::release_store(&_disable_interning_during_cds_dump, false));
 }
 
 void StringTable::set_shared_strings_array_index(int root_index) {

src/hotspot/share/classfile/stringTable.hpp

@@ -40,7 +40,7 @@ class StringTableConfig;
 
 class StringTable : AllStatic {
   friend class StringTableConfig;
-
+  class VerifyCompStrings;
   static volatile bool _has_work;
 
   // Set if one bucket is out of balance due to hash algorithm deficiency
@@ -74,6 +74,7 @@ private:
 
   static void item_added();
   static void item_removed();
+  static size_t items_count_acquire();
 
   static oop intern(const StringWrapper& name, TRAPS);
   static oop do_intern(const StringWrapper& name, uintx hash, TRAPS);

src/hotspot/share/classfile/verifier.cpp

@@ -781,7 +781,6 @@ void ClassVerifier::verify_method(const methodHandle& m, TRAPS) {
 
     // Merge with the next instruction
     {
-      int target;
       VerificationType type, type2;
       VerificationType atype;
 
@@ -1606,9 +1605,8 @@ void ClassVerifier::verify_method(const methodHandle& m, TRAPS) {
         case Bytecodes::_ifle:
           current_frame.pop_stack(
             VerificationType::integer_type(), CHECK_VERIFY(this));
-          target = bcs.dest();
           stackmap_table.check_jump_target(
-            &current_frame, target, CHECK_VERIFY(this));
+            &current_frame, bcs.bci(), bcs.get_offset_s2(), CHECK_VERIFY(this));
           no_control_flow = false; break;
         case Bytecodes::_if_acmpeq :
         case Bytecodes::_if_acmpne :
@@ -1619,19 +1617,16 @@ void ClassVerifier::verify_method(const methodHandle& m, TRAPS) {
         case Bytecodes::_ifnonnull :
           current_frame.pop_stack(
             VerificationType::reference_check(), CHECK_VERIFY(this));
-          target = bcs.dest();
           stackmap_table.check_jump_target
-            (&current_frame, target, CHECK_VERIFY(this));
+            (&current_frame, bcs.bci(), bcs.get_offset_s2(), CHECK_VERIFY(this));
           no_control_flow = false; break;
         case Bytecodes::_goto :
-          target = bcs.dest();
           stackmap_table.check_jump_target(
-            &current_frame, target, CHECK_VERIFY(this));
+            &current_frame, bcs.bci(), bcs.get_offset_s2(), CHECK_VERIFY(this));
           no_control_flow = true; break;
         case Bytecodes::_goto_w :
-          target = bcs.dest_w();
           stackmap_table.check_jump_target(
-            &current_frame, target, CHECK_VERIFY(this));
+            &current_frame, bcs.bci(), bcs.get_offset_s4(), CHECK_VERIFY(this));
           no_control_flow = true; break;
         case Bytecodes::_tableswitch :
         case Bytecodes::_lookupswitch :
@@ -2280,15 +2275,14 @@ void ClassVerifier::verify_switch(
       }
     }
   }
-  int target = bci + default_offset;
-  stackmap_table->check_jump_target(current_frame, target, CHECK_VERIFY(this));
+  stackmap_table->check_jump_target(current_frame, bci, default_offset, CHECK_VERIFY(this));
   for (int i = 0; i < keys; i++) {
     // Because check_jump_target() may safepoint, the bytecode could have
     // moved, which means 'aligned_bcp' is no good and needs to be recalculated.
     aligned_bcp = align_up(bcs->bcp() + 1, jintSize);
-    target = bci + (jint)Bytes::get_Java_u4(aligned_bcp+(3+i*delta)*jintSize);
+    int offset = (jint)Bytes::get_Java_u4(aligned_bcp+(3+i*delta)*jintSize);
     stackmap_table->check_jump_target(
-      current_frame, target, CHECK_VERIFY(this));
+      current_frame, bci, offset, CHECK_VERIFY(this));
   }
   NOT_PRODUCT(aligned_bcp = nullptr);  // no longer valid at this point
 }
@@ -2549,7 +2543,12 @@ bool ClassVerifier::ends_in_athrow(u4 start_bc_offset) {
 
       case Bytecodes::_goto:
       case Bytecodes::_goto_w: {
-        int target = (opcode == Bytecodes::_goto ? bcs.dest() : bcs.dest_w());
+        int offset = (opcode == Bytecodes::_goto ? bcs.get_offset_s2() : bcs.get_offset_s4());
+        int min_offset = -1 * max_method_code_size;
+        // Check offset for overflow
+        if (offset < min_offset || offset > max_method_code_size) return false;
+
+        int target = bci + offset;
         if (visited_branches->contains(bci)) {
           if (bci_stack->is_empty()) {
             if (handler_stack->is_empty()) {
@@ -2607,7 +2606,10 @@ bool ClassVerifier::ends_in_athrow(u4 start_bc_offset) {
 
           // Push the switch alternatives onto the stack.
           for (int i = 0; i < keys; i++) {
-            int target = bci + (jint)Bytes::get_Java_u4(aligned_bcp+(3+i*delta)*jintSize);
+            int min_offset = -1 * max_method_code_size;
+            int offset = (jint)Bytes::get_Java_u4(aligned_bcp+(3+i*delta)*jintSize);
+            if (offset < min_offset || offset > max_method_code_size) return false;
+            int target = bci + offset;
             if (target > code_length) return false;
             bci_stack->push(target);
           }

src/hotspot/share/classfile/vmIntrinsics.cpp

@@ -289,6 +289,8 @@ bool vmIntrinsics::disabled_by_jvm_flags(vmIntrinsics::ID id) {
   case vmIntrinsics::_dsin:
   case vmIntrinsics::_dcos:
   case vmIntrinsics::_dtan:
+  case vmIntrinsics::_dtanh:
+  case vmIntrinsics::_dcbrt:
   case vmIntrinsics::_dlog:
   case vmIntrinsics::_dexp:
   case vmIntrinsics::_dpow:
@@ -314,13 +316,6 @@ bool vmIntrinsics::disabled_by_jvm_flags(vmIntrinsics::ID id) {
   case vmIntrinsics::_fmaF:
     if (!InlineMathNatives || !UseFMA) return true;
     break;
-  case vmIntrinsics::_dtanh:
-  case vmIntrinsics::_dcbrt:
-    if (!InlineMathNatives || !InlineIntrinsics) return true;
-#if defined(AMD64) && (defined(COMPILER1) || defined(COMPILER2))
-    if (!UseLibmIntrinsic) return true;
-#endif
-    break;
   case vmIntrinsics::_floatToFloat16:
   case vmIntrinsics::_float16ToFloat:
     if (!InlineIntrinsics) return true;

src/hotspot/share/code/aotCodeCache.cpp

@@ -344,6 +344,7 @@ AOTCodeCache::~AOTCodeCache() {
     _store_buffer = nullptr;
   }
   if (_table != nullptr) {
+    MutexLocker ml(AOTCodeCStrings_lock, Mutex::_no_safepoint_check_flag);
     delete _table;
     _table = nullptr;
   }
@@ -774,6 +775,9 @@ bool AOTCodeCache::store_code_blob(CodeBlob& blob, AOTCodeEntry::Kind entry_kind
   // we need to take a lock to prevent race between compiler threads generating AOT code
   // and the main thread generating adapter
   MutexLocker ml(Compile_lock);
+  if (!is_on()) {
+    return false; // AOT code cache was already dumped and closed.
+  }
   if (!cache->align_write()) {
     return false;
   }
@@ -1434,6 +1438,9 @@ AOTCodeAddressTable::~AOTCodeAddressTable() {
   if (_extrs_addr != nullptr) {
     FREE_C_HEAP_ARRAY(address, _extrs_addr);
   }
+  if (_stubs_addr != nullptr) {
+    FREE_C_HEAP_ARRAY(address, _stubs_addr);
+  }
   if (_shared_blobs_addr != nullptr) {
     FREE_C_HEAP_ARRAY(address, _shared_blobs_addr);
   }
@@ -1485,6 +1492,7 @@ void AOTCodeCache::load_strings() {
 
 int AOTCodeCache::store_strings() {
   if (_C_strings_used > 0) {
+    MutexLocker ml(AOTCodeCStrings_lock, Mutex::_no_safepoint_check_flag);
     uint offset = _write_position;
     uint length = 0;
     uint* lengths = (uint *)reserve_bytes(sizeof(uint) * _C_strings_used);
@@ -1510,15 +1518,17 @@ int AOTCodeCache::store_strings() {
 
 const char* AOTCodeCache::add_C_string(const char* str) {
   if (is_on_for_dump() && str != nullptr) {
-    return _cache->_table->add_C_string(str);
+    MutexLocker ml(AOTCodeCStrings_lock, Mutex::_no_safepoint_check_flag);
+    AOTCodeAddressTable* table = addr_table();
+    if (table != nullptr) {
+      return table->add_C_string(str);
+    }
   }
   return str;
 }
 
 const char* AOTCodeAddressTable::add_C_string(const char* str) {
   if (_extrs_complete) {
-    LogStreamHandle(Trace, aot, codecache, stringtable) log; // ctor outside lock
-    MutexLocker ml(AOTCodeCStrings_lock, Mutex::_no_safepoint_check_flag);
     // Check previous strings address
     for (int i = 0; i < _C_strings_count; i++) {
       if (_C_strings_in[i] == str) {
@@ -1535,9 +1545,7 @@ const char* AOTCodeAddressTable::add_C_string(const char* str) {
       _C_strings_in[_C_strings_count] = str;
       const char* dup = os::strdup(str);
       _C_strings[_C_strings_count++] = dup;
-      if (log.is_enabled()) {
-        log.print_cr("add_C_string: [%d] " INTPTR_FORMAT " '%s'", _C_strings_count, p2i(dup), dup);
-      }
+      log_trace(aot, codecache, stringtable)("add_C_string: [%d] " INTPTR_FORMAT " '%s'", _C_strings_count, p2i(dup), dup);
       return dup;
     } else {
       assert(false, "Number of C strings >= MAX_STR_COUNT");

src/hotspot/share/code/aotCodeCache.hpp

@@ -136,6 +136,7 @@ private:
 public:
   AOTCodeAddressTable() :
     _extrs_addr(nullptr),
+    _stubs_addr(nullptr),
     _shared_blobs_addr(nullptr),
     _C1_blobs_addr(nullptr),
     _extrs_length(0),

src/hotspot/share/code/codeBlob.cpp

@@ -160,7 +160,7 @@ CodeBlob::CodeBlob(const char* name, CodeBlobKind kind, CodeBuffer* cb, int size
     }
   } else {
     // We need unique and valid not null address
-    assert(_mutable_data = blob_end(), "sanity");
+    assert(_mutable_data == blob_end(), "sanity");
   }
 
   set_oop_maps(oop_maps);
@@ -177,6 +177,7 @@ CodeBlob::CodeBlob(const char* name, CodeBlobKind kind, int size, uint16_t heade
   _code_offset(_content_offset),
   _data_offset(size),
   _frame_size(0),
+  _mutable_data_size(0),
   S390_ONLY(_ctable_offset(0) COMMA)
   _header_size(header_size),
   _frame_complete_offset(CodeOffsets::frame_never_safe),
@@ -185,7 +186,7 @@ CodeBlob::CodeBlob(const char* name, CodeBlobKind kind, int size, uint16_t heade
 {
   assert(is_aligned(size,            oopSize), "unaligned size");
   assert(is_aligned(header_size,     oopSize), "unaligned size");
-  assert(_mutable_data = blob_end(), "sanity");
+  assert(_mutable_data == blob_end(), "sanity");
 }
 
 void CodeBlob::restore_mutable_data(address reloc_data) {
@@ -195,8 +196,11 @@ void CodeBlob::restore_mutable_data(address reloc_data) {
     if (_mutable_data == nullptr) {
       vm_exit_out_of_memory(_mutable_data_size, OOM_MALLOC_ERROR, "codebuffer: no space for mutable data");
     }
+  } else {
+    _mutable_data = blob_end(); // default value
   }
   if (_relocation_size > 0) {
+    assert(_mutable_data_size > 0, "relocation is part of mutable data section");
     memcpy((address)relocation_begin(), reloc_data, relocation_size());
   }
 }
@@ -206,6 +210,8 @@ void CodeBlob::purge() {
   if (_mutable_data != blob_end()) {
     os::free(_mutable_data);
     _mutable_data = blob_end(); // Valid not null address
+    _mutable_data_size = 0;
+    _relocation_size = 0;
   }
   if (_oop_maps != nullptr) {
     delete _oop_maps;

src/hotspot/share/code/codeBlob.hpp

@@ -247,7 +247,7 @@ public:
   // Sizes
   int size() const               { return _size; }
   int header_size() const        { return _header_size; }
-  int relocation_size() const    { return pointer_delta_as_int((address) relocation_end(), (address) relocation_begin()); }
+  int relocation_size() const    { return _relocation_size; }
   int content_size() const       { return pointer_delta_as_int(content_end(), content_begin()); }
   int code_size() const          { return pointer_delta_as_int(code_end(), code_begin()); }
 

src/hotspot/share/code/codeCache.cpp

@@ -1361,7 +1361,7 @@ void CodeCache::make_marked_nmethods_deoptimized() {
   while(iter.next()) {
     nmethod* nm = iter.method();
     if (nm->is_marked_for_deoptimization() && !nm->has_been_deoptimized() && nm->can_be_deoptimized()) {
-      nm->make_not_entrant(nmethod::ChangeReason::marked_for_deoptimization);
+      nm->make_not_entrant("marked for deoptimization");
       nm->make_deoptimized();
     }
   }

src/hotspot/share/code/nmethod.cpp

@@ -28,7 +28,6 @@
 #include "code/dependencies.hpp"
 #include "code/nativeInst.hpp"
 #include "code/nmethod.inline.hpp"
-#include "code/relocInfo.hpp"
 #include "code/scopeDesc.hpp"
 #include "compiler/abstractCompiler.hpp"
 #include "compiler/compilationLog.hpp"
@@ -1653,10 +1652,6 @@ void nmethod::maybe_print_nmethod(const DirectiveSet* directive) {
 }
 
 void nmethod::print_nmethod(bool printmethod) {
-  // Enter a critical section to prevent a race with deopts that patch code and updates the relocation info.
-  // Unfortunately, we have to lock the NMethodState_lock before the tty lock due to the deadlock rules and
-  // cannot lock in a more finely grained manner.
-  ConditionalMutexLocker ml(NMethodState_lock, !NMethodState_lock->owned_by_self(), Mutex::_no_safepoint_check_flag);
   ttyLocker ttyl;  // keep the following output all in one block
   if (xtty != nullptr) {
     xtty->begin_head("print_nmethod");
@@ -1975,12 +1970,14 @@ void nmethod::invalidate_osr_method() {
   }
 }
 
-void nmethod::log_state_change(ChangeReason change_reason) const {
+void nmethod::log_state_change(const char* reason) const {
+  assert(reason != nullptr, "Must provide a reason");
+
   if (LogCompilation) {
     if (xtty != nullptr) {
       ttyLocker ttyl;  // keep the following output all in one block
       xtty->begin_elem("make_not_entrant thread='%zu' reason='%s'",
-                       os::current_thread_id(), change_reason_to_string(change_reason));
+                       os::current_thread_id(), reason);
       log_identity(xtty);
       xtty->stamp();
       xtty->end_elem();
@@ -1989,7 +1986,7 @@ void nmethod::log_state_change(ChangeReason change_reason) const {
 
   ResourceMark rm;
   stringStream ss(NEW_RESOURCE_ARRAY(char, 256), 256);
-  ss.print("made not entrant: %s", change_reason_to_string(change_reason));
+  ss.print("made not entrant: %s", reason);
 
   CompileTask::print_ul(this, ss.freeze());
   if (PrintCompilation) {
@@ -2004,7 +2001,9 @@ void nmethod::unlink_from_method() {
 }
 
 // Invalidate code
-bool nmethod::make_not_entrant(ChangeReason change_reason) {
+bool nmethod::make_not_entrant(const char* reason) {
+  assert(reason != nullptr, "Must provide a reason");
+
   // This can be called while the system is already at a safepoint which is ok
   NoSafepointVerifier nsv;
 
@@ -2042,17 +2041,6 @@ bool nmethod::make_not_entrant(ChangeReason change_reason) {
       // cache call.
       NativeJump::patch_verified_entry(entry_point(), verified_entry_point(),
                                        SharedRuntime::get_handle_wrong_method_stub());
-
-      // Update the relocation info for the patched entry.
-      // First, get the old relocation info...
-      RelocIterator iter(this, verified_entry_point(), verified_entry_point() + 8);
-      if (iter.next() && iter.addr() == verified_entry_point()) {
-        Relocation* old_reloc = iter.reloc();
-        // ...then reset the iterator to update it.
-        RelocIterator iter(this, verified_entry_point(), verified_entry_point() + 8);
-        relocInfo::change_reloc_info_for_address(&iter, verified_entry_point(), old_reloc->type(),
-                                                 relocInfo::relocType::runtime_call_type);
-      }
     }
 
     if (update_recompile_counts()) {
@@ -2073,7 +2061,7 @@ bool nmethod::make_not_entrant(ChangeReason change_reason) {
     assert(success, "Transition can't fail");
 
     // Log the transition once
-    log_state_change(change_reason);
+    log_state_change(reason);
 
     // Remove nmethod from method.
     unlink_from_method();
@@ -2178,6 +2166,7 @@ void nmethod::purge(bool unregister_nmethod) {
   }
   CodeCache::unregister_old_nmethod(this);
 
+  JVMCI_ONLY( _metadata_size = 0; )
   CodeBlob::purge();
 }
 

src/hotspot/share/code/nmethod.hpp

@@ -471,85 +471,6 @@ class nmethod : public CodeBlob {
   void oops_do_set_strong_done(nmethod* old_head);
 
 public:
-  enum class ChangeReason : u1 {
-    C1_codepatch,
-    C1_deoptimize,
-    C1_deoptimize_for_patching,
-    C1_predicate_failed_trap,
-    CI_replay,
-    JVMCI_invalidate_nmethod,
-    JVMCI_invalidate_nmethod_mirror,
-    JVMCI_materialize_virtual_object,
-    JVMCI_new_installation,
-    JVMCI_register_method,
-    JVMCI_replacing_with_new_code,
-    JVMCI_reprofile,
-    marked_for_deoptimization,
-    missing_exception_handler,
-    not_used,
-    OSR_invalidation_back_branch,
-    OSR_invalidation_for_compiling_with_C1,
-    OSR_invalidation_of_lower_level,
-    set_native_function,
-    uncommon_trap,
-    whitebox_deoptimization,
-    zombie,
-  };
-
-
-  static const char* change_reason_to_string(ChangeReason change_reason) {
-    switch (change_reason) {
-      case ChangeReason::C1_codepatch:
-        return "C1 code patch";
-      case ChangeReason::C1_deoptimize:
-        return "C1 deoptimized";
-      case ChangeReason::C1_deoptimize_for_patching:
-        return "C1 deoptimize for patching";
-      case ChangeReason::C1_predicate_failed_trap:
-        return "C1 predicate failed trap";
-      case ChangeReason::CI_replay:
-        return "CI replay";
-      case ChangeReason::JVMCI_invalidate_nmethod:
-        return "JVMCI invalidate nmethod";
-      case ChangeReason::JVMCI_invalidate_nmethod_mirror:
-        return "JVMCI invalidate nmethod mirror";
-      case ChangeReason::JVMCI_materialize_virtual_object:
-        return "JVMCI materialize virtual object";
-      case ChangeReason::JVMCI_new_installation:
-        return "JVMCI new installation";
-      case ChangeReason::JVMCI_register_method:
-        return "JVMCI register method";
-      case ChangeReason::JVMCI_replacing_with_new_code:
-        return "JVMCI replacing with new code";
-      case ChangeReason::JVMCI_reprofile:
-        return "JVMCI reprofile";
-      case ChangeReason::marked_for_deoptimization:
-        return "marked for deoptimization";
-      case ChangeReason::missing_exception_handler:
-        return "missing exception handler";
-      case ChangeReason::not_used:
-        return "not used";
-      case ChangeReason::OSR_invalidation_back_branch:
-        return "OSR invalidation back branch";
-      case ChangeReason::OSR_invalidation_for_compiling_with_C1:
-        return "OSR invalidation for compiling with C1";
-      case ChangeReason::OSR_invalidation_of_lower_level:
-        return "OSR invalidation of lower level";
-      case ChangeReason::set_native_function:
-        return "set native function";
-      case ChangeReason::uncommon_trap:
-        return "uncommon trap";
-      case ChangeReason::whitebox_deoptimization:
-        return "whitebox deoptimization";
-      case ChangeReason::zombie:
-        return "zombie";
-      default: {
-        assert(false, "Unhandled reason");
-        return "Unknown";
-      }
-    }
-  }
-
   // create nmethod with entry_bci
   static nmethod* new_nmethod(const methodHandle& method,
                               int compile_id,
@@ -712,8 +633,8 @@ public:
   // alive.  It is used when an uncommon trap happens.  Returns true
   // if this thread changed the state of the nmethod or false if
   // another thread performed the transition.
-  bool  make_not_entrant(ChangeReason change_reason);
-  bool  make_not_used() { return make_not_entrant(ChangeReason::not_used); }
+  bool  make_not_entrant(const char* reason);
+  bool  make_not_used()    { return make_not_entrant("not used"); }
 
   bool  is_marked_for_deoptimization() const { return deoptimization_status() != not_marked; }
   bool  has_been_deoptimized() const { return deoptimization_status() == deoptimize_done; }
@@ -1026,7 +947,7 @@ public:
   // Logging
   void log_identity(xmlStream* log) const;
   void log_new_nmethod() const;
-  void log_state_change(ChangeReason change_reason) const;
+  void log_state_change(const char* reason) const;
 
   // Prints block-level comments, including nmethod specific block labels:
   void print_nmethod_labels(outputStream* stream, address block_begin, bool print_section_labels=true) const;

src/hotspot/share/compiler/compilationPolicy.cpp

@@ -924,7 +924,7 @@ void CompilationPolicy::compile(const methodHandle& mh, int bci, CompLevel level
         nmethod* osr_nm = mh->lookup_osr_nmethod_for(bci, CompLevel_simple, false);
         if (osr_nm != nullptr && osr_nm->comp_level() > CompLevel_simple) {
           // Invalidate the existing OSR nmethod so that a compile at CompLevel_simple is permitted.
-          osr_nm->make_not_entrant(nmethod::ChangeReason::OSR_invalidation_for_compiling_with_C1);
+          osr_nm->make_not_entrant("OSR invalidation for compiling with C1");
         }
         compile(mh, bci, CompLevel_simple, THREAD);
       }
@@ -1516,7 +1516,7 @@ void CompilationPolicy::method_back_branch_event(const methodHandle& mh, const m
               int osr_bci = nm->is_osr_method() ? nm->osr_entry_bci() : InvocationEntryBci;
               print_event(MAKE_NOT_ENTRANT, mh(), mh(), osr_bci, level);
             }
-            nm->make_not_entrant(nmethod::ChangeReason::OSR_invalidation_back_branch);
+            nm->make_not_entrant("OSR invalidation, back branch");
           }
         }
         // Fix up next_level if necessary to avoid deopts

src/hotspot/share/compiler/compileBroker.cpp

@@ -1750,6 +1750,10 @@ void CompileBroker::wait_for_completion(CompileTask* task) {
   }
 }
 
+void CompileBroker::wait_for_no_active_tasks() {
+  CompileTask::wait_for_no_active_tasks();
+}
+
 /**
  * Initialize compiler thread(s) + compiler object(s). The postcondition
  * of this function is that the compiler runtimes are initialized and that

src/hotspot/share/compiler/compileBroker.hpp

@@ -383,6 +383,9 @@ public:
   static bool is_compilation_disabled_forever() {
     return _should_compile_new_jobs == shutdown_compilation;
   }
+
+  static void wait_for_no_active_tasks();
+
   static void handle_full_code_cache(CodeBlobType code_blob_type);
   // Ensures that warning is only printed once.
   static bool should_print_compiler_warning() {

src/hotspot/share/compiler/compileTask.cpp

@@ -37,12 +37,13 @@
 #include "runtime/mutexLocker.hpp"
 
 CompileTask*  CompileTask::_task_free_list = nullptr;
+int CompileTask::_active_tasks = 0;
 
 /**
  * Allocate a CompileTask, from the free list if possible.
  */
 CompileTask* CompileTask::allocate() {
-  MutexLocker locker(CompileTaskAlloc_lock);
+  MonitorLocker locker(CompileTaskAlloc_lock);
   CompileTask* task = nullptr;
 
   if (_task_free_list != nullptr) {
@@ -56,6 +57,7 @@ CompileTask* CompileTask::allocate() {
   }
   assert(task->is_free(), "Task must be free.");
   task->set_is_free(false);
+  _active_tasks++;
   return task;
 }
 
@@ -63,7 +65,7 @@ CompileTask* CompileTask::allocate() {
 * Add a task to the free list.
 */
 void CompileTask::free(CompileTask* task) {
-  MutexLocker locker(CompileTaskAlloc_lock);
+  MonitorLocker locker(CompileTaskAlloc_lock);
   if (!task->is_free()) {
     if ((task->_method_holder != nullptr && JNIHandles::is_weak_global_handle(task->_method_holder))) {
       JNIHandles::destroy_weak_global(task->_method_holder);
@@ -79,6 +81,17 @@ void CompileTask::free(CompileTask* task) {
     task->set_is_free(true);
     task->set_next(_task_free_list);
     _task_free_list = task;
+    _active_tasks--;
+    if (_active_tasks == 0) {
+      locker.notify_all();
+    }
+  }
+}
+
+void CompileTask::wait_for_no_active_tasks() {
+  MonitorLocker locker(CompileTaskAlloc_lock);
+  while (_active_tasks > 0) {
+    locker.wait();
   }
 }
 

src/hotspot/share/compiler/compileTask.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 1998, 2024, Oracle and/or its affiliates. All rights reserved.
+ * Copyright (c) 1998, 2025, 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
@@ -83,6 +83,7 @@ class CompileTask : public CHeapObj<mtCompiler> {
 
  private:
   static CompileTask*  _task_free_list;
+  static int           _active_tasks;
   int                  _compile_id;
   Method*              _method;
   jobject              _method_holder;
@@ -123,6 +124,7 @@ class CompileTask : public CHeapObj<mtCompiler> {
 
   static CompileTask* allocate();
   static void         free(CompileTask* task);
+  static void         wait_for_no_active_tasks();
 
   int          compile_id() const                { return _compile_id; }
   Method*      method() const                    { return _method; }

src/hotspot/share/gc/parallel/parallelArguments.cpp

@@ -98,15 +98,15 @@ void ParallelArguments::initialize() {
   FullGCForwarding::initialize_flags(heap_reserved_size_bytes());
 }
 
-// The alignment used for spaces in young gen and old gen
-static size_t default_space_alignment() {
+// The alignment used for boundary between young gen and old gen
+static size_t default_gen_alignment() {
   return 64 * K * HeapWordSize;
 }
 
 void ParallelArguments::initialize_alignments() {
   // Initialize card size before initializing alignments
   CardTable::initialize_card_size();
-  SpaceAlignment = default_space_alignment();
+  SpaceAlignment = GenAlignment = default_gen_alignment();
   HeapAlignment = compute_heap_alignment();
 }
 
@@ -123,8 +123,9 @@ void ParallelArguments::initialize_heap_flags_and_sizes() {
 
   // Can a page size be something else than a power of two?
   assert(is_power_of_2((intptr_t)page_sz), "must be a power of 2");
-  size_t new_alignment = align_up(page_sz, SpaceAlignment);
-  if (new_alignment != SpaceAlignment) {
+  size_t new_alignment = align_up(page_sz, GenAlignment);
+  if (new_alignment != GenAlignment) {
+    GenAlignment = new_alignment;
     SpaceAlignment = new_alignment;
     // Redo everything from the start
     initialize_heap_flags_and_sizes_one_pass();

src/hotspot/share/gc/parallel/parallelInitLogger.cpp

@@ -29,8 +29,10 @@
 void ParallelInitLogger::print_heap() {
   log_info_p(gc, init)("Alignments:"
                        " Space " EXACTFMT ","
+                       " Generation " EXACTFMT ","
                        " Heap " EXACTFMT,
                        EXACTFMTARGS(SpaceAlignment),
+                       EXACTFMTARGS(GenAlignment),
                        EXACTFMTARGS(HeapAlignment));
   GCInitLogger::print_heap();
 }

src/hotspot/share/gc/parallel/parallelScavengeHeap.cpp

@@ -69,8 +69,8 @@ jint ParallelScavengeHeap::initialize() {
 
   initialize_reserved_region(heap_rs);
   // Layout the reserved space for the generations.
-  ReservedSpace old_rs   = heap_rs.first_part(MaxOldSize, SpaceAlignment);
-  ReservedSpace young_rs = heap_rs.last_part(MaxOldSize, SpaceAlignment);
+  ReservedSpace old_rs   = heap_rs.first_part(MaxOldSize, GenAlignment);
+  ReservedSpace young_rs = heap_rs.last_part(MaxOldSize, GenAlignment);
   assert(young_rs.size() == MaxNewSize, "Didn't reserve all of the heap");
 
   PSCardTable* card_table = new PSCardTable(_reserved);
@@ -107,7 +107,7 @@ jint ParallelScavengeHeap::initialize() {
     new PSAdaptiveSizePolicy(eden_capacity,
                              initial_promo_size,
                              young_gen()->to_space()->capacity_in_bytes(),
-                             SpaceAlignment,
+                             GenAlignment,
                              max_gc_pause_sec,
                              GCTimeRatio
                              );

src/hotspot/share/gc/parallel/psOldGen.cpp

@@ -41,7 +41,7 @@ PSOldGen::PSOldGen(ReservedSpace rs, size_t initial_size, size_t min_size,
   _min_gen_size(min_size),
   _max_gen_size(max_size)
 {
-  initialize(rs, initial_size, SpaceAlignment);
+  initialize(rs, initial_size, GenAlignment);
 }
 
 void PSOldGen::initialize(ReservedSpace rs, size_t initial_size, size_t alignment) {

src/hotspot/share/gc/parallel/psYoungGen.cpp

@@ -47,7 +47,7 @@ PSYoungGen::PSYoungGen(ReservedSpace rs, size_t initial_size, size_t min_size, s
   _from_counters(nullptr),
   _to_counters(nullptr)
 {
-  initialize(rs, initial_size, SpaceAlignment);
+  initialize(rs, initial_size, GenAlignment);
 }
 
 void PSYoungGen::initialize_virtual_space(ReservedSpace rs,
@@ -746,7 +746,7 @@ size_t PSYoungGen::available_to_live() {
   }
 
   size_t delta_in_bytes = unused_committed + delta_in_survivor;
-  delta_in_bytes = align_down(delta_in_bytes, SpaceAlignment);
+  delta_in_bytes = align_down(delta_in_bytes, GenAlignment);
   return delta_in_bytes;
 }
 

src/hotspot/share/gc/serial/serialHeap.cpp

@@ -188,8 +188,8 @@ jint SerialHeap::initialize() {
 
   initialize_reserved_region(heap_rs);
 
-  ReservedSpace young_rs = heap_rs.first_part(MaxNewSize, SpaceAlignment);
-  ReservedSpace old_rs = heap_rs.last_part(MaxNewSize, SpaceAlignment);
+  ReservedSpace young_rs = heap_rs.first_part(MaxNewSize, GenAlignment);
+  ReservedSpace old_rs = heap_rs.last_part(MaxNewSize, GenAlignment);
 
   _rem_set = new CardTableRS(_reserved);
   _rem_set->initialize(young_rs.base(), old_rs.base());

src/hotspot/share/gc/shared/gcArguments.hpp

@@ -35,7 +35,7 @@ extern size_t SpaceAlignment;
 
 class GCArguments {
 protected:
-  // Initialize HeapAlignment, SpaceAlignment
+  // Initialize HeapAlignment, SpaceAlignment, and extra alignments (E.g. GenAlignment)
   virtual void initialize_alignments() = 0;
   virtual void initialize_heap_flags_and_sizes();
   virtual void initialize_size_info();

src/hotspot/share/gc/shared/genArguments.cpp

@@ -42,15 +42,17 @@ size_t MaxOldSize = 0;
 // See more in JDK-8346005
 size_t OldSize = ScaleForWordSize(4*M);
 
+size_t GenAlignment = 0;
+
 size_t GenArguments::conservative_max_heap_alignment() { return (size_t)Generation::GenGrain; }
 
 static size_t young_gen_size_lower_bound() {
   // The young generation must be aligned and have room for eden + two survivors
-  return 3 * SpaceAlignment;
+  return align_up(3 * SpaceAlignment, GenAlignment);
 }
 
 static size_t old_gen_size_lower_bound() {
-  return SpaceAlignment;
+  return align_up(SpaceAlignment, GenAlignment);
 }
 
 size_t GenArguments::scale_by_NewRatio_aligned(size_t base_size, size_t alignment) {
@@ -67,20 +69,23 @@ static size_t bound_minus_alignment(size_t desired_size,
 void GenArguments::initialize_alignments() {
   // Initialize card size before initializing alignments
   CardTable::initialize_card_size();
-  SpaceAlignment = (size_t)Generation::GenGrain;
+  SpaceAlignment = GenAlignment = (size_t)Generation::GenGrain;
   HeapAlignment = compute_heap_alignment();
 }
 
 void GenArguments::initialize_heap_flags_and_sizes() {
   GCArguments::initialize_heap_flags_and_sizes();
 
-  assert(SpaceAlignment != 0, "Generation alignment not set up properly");
-  assert(HeapAlignment >= SpaceAlignment,
-         "HeapAlignment: %zu less than SpaceAlignment: %zu",
-         HeapAlignment, SpaceAlignment);
-  assert(HeapAlignment % SpaceAlignment == 0,
-         "HeapAlignment: %zu not aligned by SpaceAlignment: %zu",
-         HeapAlignment, SpaceAlignment);
+  assert(GenAlignment != 0, "Generation alignment not set up properly");
+  assert(HeapAlignment >= GenAlignment,
+         "HeapAlignment: %zu less than GenAlignment: %zu",
+         HeapAlignment, GenAlignment);
+  assert(GenAlignment % SpaceAlignment == 0,
+         "GenAlignment: %zu not aligned by SpaceAlignment: %zu",
+         GenAlignment, SpaceAlignment);
+  assert(HeapAlignment % GenAlignment == 0,
+         "HeapAlignment: %zu not aligned by GenAlignment: %zu",
+         HeapAlignment, GenAlignment);
 
   // All generational heaps have a young gen; handle those flags here
 
@@ -101,7 +106,7 @@ void GenArguments::initialize_heap_flags_and_sizes() {
 
   // Make sure NewSize allows an old generation to fit even if set on the command line
   if (FLAG_IS_CMDLINE(NewSize) && NewSize >= InitialHeapSize) {
-    size_t revised_new_size = bound_minus_alignment(NewSize, InitialHeapSize, SpaceAlignment);
+    size_t revised_new_size = bound_minus_alignment(NewSize, InitialHeapSize, GenAlignment);
     log_warning(gc, ergo)("NewSize (%zuk) is equal to or greater than initial heap size (%zuk).  A new "
                           "NewSize of %zuk will be used to accomodate an old generation.",
                           NewSize/K, InitialHeapSize/K, revised_new_size/K);
@@ -110,8 +115,8 @@ void GenArguments::initialize_heap_flags_and_sizes() {
 
   // Now take the actual NewSize into account. We will silently increase NewSize
   // if the user specified a smaller or unaligned value.
-  size_t bounded_new_size = bound_minus_alignment(NewSize, MaxHeapSize, SpaceAlignment);
-  bounded_new_size = MAX2(smallest_new_size, align_down(bounded_new_size, SpaceAlignment));
+  size_t bounded_new_size = bound_minus_alignment(NewSize, MaxHeapSize, GenAlignment);
+  bounded_new_size = MAX2(smallest_new_size, align_down(bounded_new_size, GenAlignment));
   if (bounded_new_size != NewSize) {
     FLAG_SET_ERGO(NewSize, bounded_new_size);
   }
@@ -120,7 +125,7 @@ void GenArguments::initialize_heap_flags_and_sizes() {
   if (!FLAG_IS_DEFAULT(MaxNewSize)) {
     if (MaxNewSize >= MaxHeapSize) {
       // Make sure there is room for an old generation
-      size_t smaller_max_new_size = MaxHeapSize - SpaceAlignment;
+      size_t smaller_max_new_size = MaxHeapSize - GenAlignment;
       if (FLAG_IS_CMDLINE(MaxNewSize)) {
         log_warning(gc, ergo)("MaxNewSize (%zuk) is equal to or greater than the entire "
                               "heap (%zuk).  A new max generation size of %zuk will be used.",
@@ -132,8 +137,8 @@ void GenArguments::initialize_heap_flags_and_sizes() {
       }
     } else if (MaxNewSize < NewSize) {
       FLAG_SET_ERGO(MaxNewSize, NewSize);
-    } else if (!is_aligned(MaxNewSize, SpaceAlignment)) {
-      FLAG_SET_ERGO(MaxNewSize, align_down(MaxNewSize, SpaceAlignment));
+    } else if (!is_aligned(MaxNewSize, GenAlignment)) {
+      FLAG_SET_ERGO(MaxNewSize, align_down(MaxNewSize, GenAlignment));
     }
   }
 
@@ -161,13 +166,13 @@ void GenArguments::initialize_heap_flags_and_sizes() {
       // exceed it. Adjust New/OldSize as necessary.
       size_t calculated_size = NewSize + OldSize;
       double shrink_factor = (double) MaxHeapSize / calculated_size;
-      size_t smaller_new_size = align_down((size_t)(NewSize * shrink_factor), SpaceAlignment);
+      size_t smaller_new_size = align_down((size_t)(NewSize * shrink_factor), GenAlignment);
       FLAG_SET_ERGO(NewSize, MAX2(young_gen_size_lower_bound(), smaller_new_size));
 
       // OldSize is already aligned because above we aligned MaxHeapSize to
       // HeapAlignment, and we just made sure that NewSize is aligned to
-      // SpaceAlignment. In initialize_flags() we verified that HeapAlignment
-      // is a multiple of SpaceAlignment.
+      // GenAlignment. In initialize_flags() we verified that HeapAlignment
+      // is a multiple of GenAlignment.
       OldSize = MaxHeapSize - NewSize;
     } else {
       FLAG_SET_ERGO(MaxHeapSize, align_up(NewSize + OldSize, HeapAlignment));
@@ -195,7 +200,7 @@ void GenArguments::initialize_size_info() {
   // Determine maximum size of the young generation.
 
   if (FLAG_IS_DEFAULT(MaxNewSize)) {
-    max_young_size = scale_by_NewRatio_aligned(MaxHeapSize, SpaceAlignment);
+    max_young_size = scale_by_NewRatio_aligned(MaxHeapSize, GenAlignment);
     // Bound the maximum size by NewSize below (since it historically
     // would have been NewSize and because the NewRatio calculation could
     // yield a size that is too small) and bound it by MaxNewSize above.
@@ -224,18 +229,18 @@ void GenArguments::initialize_size_info() {
       // If NewSize is set on the command line, we should use it as
       // the initial size, but make sure it is within the heap bounds.
       initial_young_size =
-        MIN2(max_young_size, bound_minus_alignment(NewSize, InitialHeapSize, SpaceAlignment));
-      MinNewSize = bound_minus_alignment(initial_young_size, MinHeapSize, SpaceAlignment);
+        MIN2(max_young_size, bound_minus_alignment(NewSize, InitialHeapSize, GenAlignment));
+      MinNewSize = bound_minus_alignment(initial_young_size, MinHeapSize, GenAlignment);
     } else {
       // For the case where NewSize is not set on the command line, use
       // NewRatio to size the initial generation size. Use the current
       // NewSize as the floor, because if NewRatio is overly large, the resulting
       // size can be too small.
       initial_young_size =
-        clamp(scale_by_NewRatio_aligned(InitialHeapSize, SpaceAlignment), NewSize, max_young_size);
+        clamp(scale_by_NewRatio_aligned(InitialHeapSize, GenAlignment), NewSize, max_young_size);
 
       // Derive MinNewSize from MinHeapSize
-      MinNewSize = MIN2(scale_by_NewRatio_aligned(MinHeapSize, SpaceAlignment), initial_young_size);
+      MinNewSize = MIN2(scale_by_NewRatio_aligned(MinHeapSize, GenAlignment), initial_young_size);
     }
   }
 
@@ -247,7 +252,7 @@ void GenArguments::initialize_size_info() {
   // The maximum old size can be determined from the maximum young
   // and maximum heap size since no explicit flags exist
   // for setting the old generation maximum.
-  MaxOldSize = MAX2(MaxHeapSize - max_young_size, SpaceAlignment);
+  MaxOldSize = MAX2(MaxHeapSize - max_young_size, GenAlignment);
   MinOldSize = MIN3(MaxOldSize,
                     InitialHeapSize - initial_young_size,
                     MinHeapSize - MinNewSize);
@@ -310,10 +315,10 @@ void GenArguments::assert_flags() {
   assert(NewSize >= MinNewSize, "Ergonomics decided on a too small young gen size");
   assert(NewSize <= MaxNewSize, "Ergonomics decided on incompatible initial and maximum young gen sizes");
   assert(FLAG_IS_DEFAULT(MaxNewSize) || MaxNewSize < MaxHeapSize, "Ergonomics decided on incompatible maximum young gen and heap sizes");
-  assert(NewSize % SpaceAlignment == 0, "NewSize alignment");
-  assert(FLAG_IS_DEFAULT(MaxNewSize) || MaxNewSize % SpaceAlignment == 0, "MaxNewSize alignment");
+  assert(NewSize % GenAlignment == 0, "NewSize alignment");
+  assert(FLAG_IS_DEFAULT(MaxNewSize) || MaxNewSize % GenAlignment == 0, "MaxNewSize alignment");
   assert(OldSize + NewSize <= MaxHeapSize, "Ergonomics decided on incompatible generation and heap sizes");
-  assert(OldSize % SpaceAlignment == 0, "OldSize alignment");
+  assert(OldSize % GenAlignment == 0, "OldSize alignment");
 }
 
 void GenArguments::assert_size_info() {
@@ -322,19 +327,19 @@ void GenArguments::assert_size_info() {
   assert(MaxNewSize < MaxHeapSize, "Ergonomics decided on incompatible maximum young and heap sizes");
   assert(MinNewSize <= NewSize, "Ergonomics decided on incompatible minimum and initial young gen sizes");
   assert(NewSize <= MaxNewSize, "Ergonomics decided on incompatible initial and maximum young gen sizes");
-  assert(MinNewSize % SpaceAlignment == 0, "_min_young_size alignment");
-  assert(NewSize % SpaceAlignment == 0, "_initial_young_size alignment");
-  assert(MaxNewSize % SpaceAlignment == 0, "MaxNewSize alignment");
-  assert(MinNewSize <= bound_minus_alignment(MinNewSize, MinHeapSize, SpaceAlignment),
+  assert(MinNewSize % GenAlignment == 0, "_min_young_size alignment");
+  assert(NewSize % GenAlignment == 0, "_initial_young_size alignment");
+  assert(MaxNewSize % GenAlignment == 0, "MaxNewSize alignment");
+  assert(MinNewSize <= bound_minus_alignment(MinNewSize, MinHeapSize, GenAlignment),
       "Ergonomics made minimum young generation larger than minimum heap");
-  assert(NewSize <=  bound_minus_alignment(NewSize, InitialHeapSize, SpaceAlignment),
+  assert(NewSize <=  bound_minus_alignment(NewSize, InitialHeapSize, GenAlignment),
       "Ergonomics made initial young generation larger than initial heap");
-  assert(MaxNewSize <= bound_minus_alignment(MaxNewSize, MaxHeapSize, SpaceAlignment),
+  assert(MaxNewSize <= bound_minus_alignment(MaxNewSize, MaxHeapSize, GenAlignment),
       "Ergonomics made maximum young generation lager than maximum heap");
   assert(MinOldSize <= OldSize, "Ergonomics decided on incompatible minimum and initial old gen sizes");
   assert(OldSize <= MaxOldSize, "Ergonomics decided on incompatible initial and maximum old gen sizes");
-  assert(MaxOldSize % SpaceAlignment == 0, "MaxOldSize alignment");
-  assert(OldSize % SpaceAlignment == 0, "OldSize alignment");
+  assert(MaxOldSize % GenAlignment == 0, "MaxOldSize alignment");
+  assert(OldSize % GenAlignment == 0, "OldSize alignment");
   assert(MaxHeapSize <= (MaxNewSize + MaxOldSize), "Total maximum heap sizes must be sum of generation maximum sizes");
   assert(MinNewSize + MinOldSize <= MinHeapSize, "Minimum generation sizes exceed minimum heap size");
   assert(NewSize + OldSize == InitialHeapSize, "Initial generation sizes should match initial heap size");

src/hotspot/share/gc/shared/genArguments.hpp

@@ -35,6 +35,8 @@ extern size_t MaxOldSize;
 
 extern size_t OldSize;
 
+extern size_t GenAlignment;
+
 class GenArguments : public GCArguments {
   friend class TestGenCollectorPolicy; // Testing
 private:

src/hotspot/share/gc/shenandoah/c2/shenandoahSupport.cpp

@@ -625,6 +625,34 @@ void ShenandoahBarrierC2Support::verify(RootNode* root) {
 }
 #endif
 
+bool ShenandoahBarrierC2Support::is_anti_dependent_load_at_control(PhaseIdealLoop* phase, Node* maybe_load, Node* store,
+                                                                   Node* control) {
+  return maybe_load->is_Load() && phase->C->can_alias(store->adr_type(), phase->C->get_alias_index(maybe_load->adr_type())) &&
+         phase->ctrl_or_self(maybe_load) == control;
+}
+
+void ShenandoahBarrierC2Support::maybe_push_anti_dependent_loads(PhaseIdealLoop* phase, Node* maybe_store, Node* control, Unique_Node_List &wq) {
+  if (!maybe_store->is_Store() && !maybe_store->is_LoadStore()) {
+    return;
+  }
+  Node* mem = maybe_store->in(MemNode::Memory);
+  for (DUIterator_Fast imax, i = mem->fast_outs(imax); i < imax; i++) {
+    Node* u = mem->fast_out(i);
+    if (is_anti_dependent_load_at_control(phase, u, maybe_store, control)) {
+      wq.push(u);
+    }
+  }
+}
+
+void ShenandoahBarrierC2Support::push_data_inputs_at_control(PhaseIdealLoop* phase, Node* n, Node* ctrl, Unique_Node_List &wq) {
+  for (uint i = 0; i < n->req(); i++) {
+    Node* in = n->in(i);
+    if (in != nullptr && phase->has_ctrl(in) && phase->get_ctrl(in) == ctrl) {
+      wq.push(in);
+    }
+  }
+}
+
 bool ShenandoahBarrierC2Support::is_dominator_same_ctrl(Node* c, Node* d, Node* n, PhaseIdealLoop* phase) {
   // That both nodes have the same control is not sufficient to prove
   // domination, verify that there's no path from d to n
@@ -639,22 +667,9 @@ bool ShenandoahBarrierC2Support::is_dominator_same_ctrl(Node* c, Node* d, Node*
     if (m->is_Phi() && m->in(0)->is_Loop()) {
       assert(phase->ctrl_or_self(m->in(LoopNode::EntryControl)) != c, "following loop entry should lead to new control");
     } else {
-      if (m->is_Store() || m->is_LoadStore()) {
-        // Take anti-dependencies into account
-        Node* mem = m->in(MemNode::Memory);
-        for (DUIterator_Fast imax, i = mem->fast_outs(imax); i < imax; i++) {
-          Node* u = mem->fast_out(i);
-          if (u->is_Load() && phase->C->can_alias(m->adr_type(), phase->C->get_alias_index(u->adr_type())) &&
-              phase->ctrl_or_self(u) == c) {
-            wq.push(u);
-          }
-        }
-      }
-      for (uint i = 0; i < m->req(); i++) {
-        if (m->in(i) != nullptr && phase->ctrl_or_self(m->in(i)) == c) {
-          wq.push(m->in(i));
-        }
-      }
+      // Take anti-dependencies into account
+      maybe_push_anti_dependent_loads(phase, m, c, wq);
+      push_data_inputs_at_control(phase, m, c, wq);
     }
   }
   return true;
@@ -1006,7 +1021,20 @@ void ShenandoahBarrierC2Support::call_lrb_stub(Node*& ctrl, Node*& val, Node* lo
   phase->register_new_node(val, ctrl);
 }
 
-void ShenandoahBarrierC2Support::fix_ctrl(Node* barrier, Node* region, const MemoryGraphFixer& fixer, Unique_Node_List& uses, Unique_Node_List& uses_to_ignore, uint last, PhaseIdealLoop* phase) {
+void ShenandoahBarrierC2Support::collect_nodes_above_barrier(Unique_Node_List &nodes_above_barrier, PhaseIdealLoop* phase, Node* ctrl, Node* init_raw_mem) {
+  nodes_above_barrier.clear();
+  if (phase->has_ctrl(init_raw_mem) && phase->get_ctrl(init_raw_mem) == ctrl && !init_raw_mem->is_Phi()) {
+    nodes_above_barrier.push(init_raw_mem);
+  }
+  for (uint next = 0; next < nodes_above_barrier.size(); next++) {
+    Node* n = nodes_above_barrier.at(next);
+    // Take anti-dependencies into account
+    maybe_push_anti_dependent_loads(phase, n, ctrl, nodes_above_barrier);
+    push_data_inputs_at_control(phase, n, ctrl, nodes_above_barrier);
+  }
+}
+
+void ShenandoahBarrierC2Support::fix_ctrl(Node* barrier, Node* region, const MemoryGraphFixer& fixer, Unique_Node_List& uses, Unique_Node_List& nodes_above_barrier, uint last, PhaseIdealLoop* phase) {
   Node* ctrl = phase->get_ctrl(barrier);
   Node* init_raw_mem = fixer.find_mem(ctrl, barrier);
 
@@ -1017,30 +1045,17 @@ void ShenandoahBarrierC2Support::fix_ctrl(Node* barrier, Node* region, const Mem
   // control will be after the expanded barrier. The raw memory (if
   // its memory is control dependent on the barrier's input control)
   // must stay above the barrier.
-  uses_to_ignore.clear();
-  if (phase->has_ctrl(init_raw_mem) && phase->get_ctrl(init_raw_mem) == ctrl && !init_raw_mem->is_Phi()) {
-    uses_to_ignore.push(init_raw_mem);
-  }
-  for (uint next = 0; next < uses_to_ignore.size(); next++) {
-    Node *n = uses_to_ignore.at(next);
-    for (uint i = 0; i < n->req(); i++) {
-      Node* in = n->in(i);
-      if (in != nullptr && phase->has_ctrl(in) && phase->get_ctrl(in) == ctrl) {
-        uses_to_ignore.push(in);
-      }
-    }
-  }
+  collect_nodes_above_barrier(nodes_above_barrier, phase, ctrl, init_raw_mem);
   for (DUIterator_Fast imax, i = ctrl->fast_outs(imax); i < imax; i++) {
     Node* u = ctrl->fast_out(i);
     if (u->_idx < last &&
         u != barrier &&
         !u->depends_only_on_test() && // preserve dependency on test
-        !uses_to_ignore.member(u) &&
+        !nodes_above_barrier.member(u) &&
         (u->in(0) != ctrl || (!u->is_Region() && !u->is_Phi())) &&
         (ctrl->Opcode() != Op_CatchProj || u->Opcode() != Op_CreateEx)) {
       Node* old_c = phase->ctrl_or_self(u);
-      Node* c = old_c;
-      if (c != ctrl ||
+      if (old_c != ctrl ||
           is_dominator_same_ctrl(old_c, barrier, u, phase) ||
           ShenandoahBarrierSetC2::is_shenandoah_state_load(u)) {
         phase->igvn().rehash_node_delayed(u);
@@ -1315,7 +1330,7 @@ void ShenandoahBarrierC2Support::pin_and_expand(PhaseIdealLoop* phase) {
 
   // Expand load-reference-barriers
   MemoryGraphFixer fixer(Compile::AliasIdxRaw, true, phase);
-  Unique_Node_List uses_to_ignore;
+  Unique_Node_List nodes_above_barriers;
   for (int i = state->load_reference_barriers_count() - 1; i >= 0; i--) {
     ShenandoahLoadReferenceBarrierNode* lrb = state->load_reference_barrier(i);
     uint last = phase->C->unique();
@@ -1410,7 +1425,7 @@ void ShenandoahBarrierC2Support::pin_and_expand(PhaseIdealLoop* phase) {
     Node* out_val = val_phi;
     phase->register_new_node(val_phi, region);
 
-    fix_ctrl(lrb, region, fixer, uses, uses_to_ignore, last, phase);
+    fix_ctrl(lrb, region, fixer, uses, nodes_above_barriers, last, phase);
 
     ctrl = orig_ctrl;
 

src/hotspot/share/gc/shenandoah/c2/shenandoahSupport.hpp

@@ -62,8 +62,12 @@ private:
                             PhaseIdealLoop* phase, int flags);
   static void call_lrb_stub(Node*& ctrl, Node*& val, Node* load_addr,
                             DecoratorSet decorators, PhaseIdealLoop* phase);
+
+  static void collect_nodes_above_barrier(Unique_Node_List &nodes_above_barrier, PhaseIdealLoop* phase, Node* ctrl,
+                                          Node* init_raw_mem);
+
   static void test_in_cset(Node*& ctrl, Node*& not_cset_ctrl, Node* val, Node* raw_mem, PhaseIdealLoop* phase);
-  static void fix_ctrl(Node* barrier, Node* region, const MemoryGraphFixer& fixer, Unique_Node_List& uses, Unique_Node_List& uses_to_ignore, uint last, PhaseIdealLoop* phase);
+  static void fix_ctrl(Node* barrier, Node* region, const MemoryGraphFixer& fixer, Unique_Node_List& uses, Unique_Node_List& nodes_above_barrier, uint last, PhaseIdealLoop* phase);
 
   static Node* get_load_addr(PhaseIdealLoop* phase, VectorSet& visited, Node* lrb);
 public:
@@ -76,6 +80,11 @@ public:
   static bool expand(Compile* C, PhaseIterGVN& igvn);
   static void pin_and_expand(PhaseIdealLoop* phase);
 
+  static void push_data_inputs_at_control(PhaseIdealLoop* phase, Node* n, Node* ctrl,
+                                          Unique_Node_List &wq);
+  static bool is_anti_dependent_load_at_control(PhaseIdealLoop* phase, Node* maybe_load, Node* store, Node* control);
+
+  static void maybe_push_anti_dependent_loads(PhaseIdealLoop* phase, Node* maybe_store, Node* control, Unique_Node_List &wq);
 #ifdef ASSERT
   static void verify(RootNode* root);
 #endif

src/hotspot/share/gc/shenandoah/shenandoahConcurrentGC.cpp

@@ -415,10 +415,6 @@ void ShenandoahConcurrentGC::entry_reset() {
                                 msg);
     op_reset();
   }
-
-  if (heap->mode()->is_generational()) {
-    heap->old_generation()->card_scan()->mark_read_table_as_clean();
-  }
 }
 
 void ShenandoahConcurrentGC::entry_scan_remembered_set() {
@@ -644,6 +640,10 @@ void ShenandoahConcurrentGC::op_reset() {
   } else {
     _generation->prepare_gc();
   }
+
+  if (heap->mode()->is_generational()) {
+    heap->old_generation()->card_scan()->mark_read_table_as_clean();
+  }
 }
 
 class ShenandoahInitMarkUpdateRegionStateClosure : public ShenandoahHeapRegionClosure {

src/hotspot/share/gc/shenandoah/shenandoahDegeneratedGC.cpp

@@ -136,9 +136,15 @@ void ShenandoahDegenGC::op_degenerated() {
       heap->set_unload_classes(_generation->heuristics()->can_unload_classes() &&
                                 (!heap->mode()->is_generational() || _generation->is_global()));
 
-      if (heap->mode()->is_generational() && _generation->is_young()) {
-        // Swap remembered sets for young
-        _generation->swap_card_tables();
+      if (heap->mode()->is_generational()) {
+        // Clean the read table before swapping it. The end goal here is to have a clean
+        // write table, and to have the read table updated with the previous write table.
+        heap->old_generation()->card_scan()->mark_read_table_as_clean();
+
+        if (_generation->is_young()) {
+          // Swap remembered sets for young
+          _generation->swap_card_tables();
+        }
       }
 
     case _degenerated_roots:

src/hotspot/share/gc/shenandoah/shenandoahGenerationalHeap.cpp

@@ -183,6 +183,29 @@ void ShenandoahGenerationalHeap::stop() {
   regulator_thread()->stop();
 }
 
+bool ShenandoahGenerationalHeap::requires_barriers(stackChunkOop obj) const {
+  if (is_idle()) {
+    return false;
+  }
+
+  if (is_concurrent_young_mark_in_progress() && is_in_young(obj) && !marking_context()->allocated_after_mark_start(obj)) {
+    // We are marking young, this object is in young, and it is below the TAMS
+    return true;
+  }
+
+  if (is_in_old(obj)) {
+    // Card marking barriers are required for objects in the old generation
+    return true;
+  }
+
+  if (has_forwarded_objects()) {
+    // Object may have pointers that need to be updated
+    return true;
+  }
+
+  return false;
+}
+
 void ShenandoahGenerationalHeap::evacuate_collection_set(bool concurrent) {
   ShenandoahRegionIterator regions;
   ShenandoahGenerationalEvacuationTask task(this, &regions, concurrent, false /* only promote regions */);

src/hotspot/share/gc/shenandoah/shenandoahGenerationalHeap.hpp

@@ -128,6 +128,8 @@ public:
 
   void stop() override;
 
+  bool requires_barriers(stackChunkOop obj) const override;
+
   // Used for logging the result of a region transfer outside the heap lock
   struct TransferResult {
     bool success;