[llvm] [AllocToken] Introduce sanitize_alloc_token attribute and alloc_token metadata (PR #160131)

via llvm-commits llvm-commits at lists.llvm.org
Mon Sep 22 08:37:45 PDT 2025


llvmbot wrote:


<!--LLVM PR SUMMARY COMMENT-->

@llvm/pr-subscribers-llvm-ir

Author: Marco Elver (melver)

<details>
<summary>Changes</summary>

In preparation of adding the "AllocToken" pass, add the pre-requisite
`sanitize_alloc_token` function attribute and `alloc_token` metadata.

---

This change is part of the following series:
  1. https://github.com/llvm/llvm-project/pull/160131
  2. https://github.com/llvm/llvm-project/pull/156838
  3. https://github.com/llvm/llvm-project/pull/156839
  4. https://github.com/llvm/llvm-project/pull/156840
  5. https://github.com/llvm/llvm-project/pull/156841
  6. https://github.com/llvm/llvm-project/pull/156842

---
Full diff: https://github.com/llvm/llvm-project/pull/160131.diff


15 Files Affected:

- (modified) llvm/docs/LangRef.rst (+10) 
- (modified) llvm/include/llvm/Bitcode/LLVMBitCodes.h (+1) 
- (modified) llvm/include/llvm/IR/Attributes.td (+3) 
- (modified) llvm/include/llvm/IR/FixedMetadataKinds.def (+1) 
- (modified) llvm/lib/Bitcode/Reader/BitcodeReader.cpp (+2) 
- (modified) llvm/lib/Bitcode/Writer/BitcodeWriter.cpp (+2) 
- (modified) llvm/lib/Transforms/Utils/CodeExtractor.cpp (+1) 
- (modified) llvm/lib/Transforms/Utils/Local.cpp (+4) 
- (modified) llvm/test/Bitcode/attributes.ll (+6) 
- (modified) llvm/test/Bitcode/compatibility.ll (+6-2) 
- (modified) llvm/test/Transforms/Inline/attributes.ll (+42) 
- (modified) llvm/utils/emacs/llvm-mode.el (+1-1) 
- (modified) llvm/utils/llvm.grm (+1) 
- (modified) llvm/utils/vim/syntax/llvm.vim (+1) 
- (modified) llvm/utils/vscode/llvm/syntaxes/ll.tmLanguage.yaml (+1) 


``````````diff
diff --git a/llvm/docs/LangRef.rst b/llvm/docs/LangRef.rst
index e64b9343b7622..1cf7ccc463b38 100644
--- a/llvm/docs/LangRef.rst
+++ b/llvm/docs/LangRef.rst
@@ -2427,6 +2427,9 @@ For example:
     if the attributed function is called during invocation of a function
     attributed with ``sanitize_realtime``.
     This attribute is incompatible with the ``sanitize_realtime`` attribute.
+``sanitize_alloc_token``
+    This attribute indicates that implicit allocation token instrumentation
+    is enabled for this function.
 ``speculative_load_hardening``
     This attribute indicates that
     `Speculative Load Hardening <https://llvm.org/docs/SpeculativeLoadHardening.html>`_
@@ -8389,6 +8392,13 @@ Example:
 The ``nofree`` metadata indicates the memory pointed by the pointer will not be
 freed after the attached instruction.
 
+'``alloc_token``' Metadata
+^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+The ``alloc_token`` metadata may be attached to calls to memory allocation
+functions, and contains richer semantic information about the type of the
+allocation. This information is consumed by the ``alloc-token`` pass to
+instrument such calls with allocation token IDs.
 
 Module Flags Metadata
 =====================
diff --git a/llvm/include/llvm/Bitcode/LLVMBitCodes.h b/llvm/include/llvm/Bitcode/LLVMBitCodes.h
index 1c7d3462b6bae..464f475098ec5 100644
--- a/llvm/include/llvm/Bitcode/LLVMBitCodes.h
+++ b/llvm/include/llvm/Bitcode/LLVMBitCodes.h
@@ -800,6 +800,7 @@ enum AttributeKindCodes {
   ATTR_KIND_SANITIZE_TYPE = 101,
   ATTR_KIND_CAPTURES = 102,
   ATTR_KIND_DEAD_ON_RETURN = 103,
+  ATTR_KIND_SANITIZE_ALLOC_TOKEN = 104,
 };
 
 enum ComdatSelectionKindCodes {
diff --git a/llvm/include/llvm/IR/Attributes.td b/llvm/include/llvm/IR/Attributes.td
index ef816fb86ed1d..8e7d9dcebfe2a 100644
--- a/llvm/include/llvm/IR/Attributes.td
+++ b/llvm/include/llvm/IR/Attributes.td
@@ -342,6 +342,9 @@ def SanitizeRealtime : EnumAttr<"sanitize_realtime", IntersectPreserve, [FnAttr]
 /// during a real-time sanitized function (see `sanitize_realtime`).
 def SanitizeRealtimeBlocking : EnumAttr<"sanitize_realtime_blocking", IntersectPreserve, [FnAttr]>;
 
+/// Allocation token instrumentation is on.
+def SanitizeAllocToken : EnumAttr<"sanitize_alloc_token", IntersectPreserve, [FnAttr]>;
+
 /// Speculative Load Hardening is enabled.
 ///
 /// Note that this uses the default compatibility (always compatible during
diff --git a/llvm/include/llvm/IR/FixedMetadataKinds.def b/llvm/include/llvm/IR/FixedMetadataKinds.def
index d09cc15d65ff6..86a4d79afc2de 100644
--- a/llvm/include/llvm/IR/FixedMetadataKinds.def
+++ b/llvm/include/llvm/IR/FixedMetadataKinds.def
@@ -55,3 +55,4 @@ LLVM_FIXED_MD_KIND(MD_mmra, "mmra", 40)
 LLVM_FIXED_MD_KIND(MD_noalias_addrspace, "noalias.addrspace", 41)
 LLVM_FIXED_MD_KIND(MD_callee_type, "callee_type", 42)
 LLVM_FIXED_MD_KIND(MD_nofree, "nofree", 43)
+LLVM_FIXED_MD_KIND(MD_alloc_token, "alloc_token", 44)
diff --git a/llvm/lib/Bitcode/Reader/BitcodeReader.cpp b/llvm/lib/Bitcode/Reader/BitcodeReader.cpp
index 22a0d0ffdbaab..67ad4a2655ecd 100644
--- a/llvm/lib/Bitcode/Reader/BitcodeReader.cpp
+++ b/llvm/lib/Bitcode/Reader/BitcodeReader.cpp
@@ -2203,6 +2203,8 @@ static Attribute::AttrKind getAttrFromCode(uint64_t Code) {
     return Attribute::SanitizeRealtime;
   case bitc::ATTR_KIND_SANITIZE_REALTIME_BLOCKING:
     return Attribute::SanitizeRealtimeBlocking;
+  case bitc::ATTR_KIND_SANITIZE_ALLOC_TOKEN:
+    return Attribute::SanitizeAllocToken;
   case bitc::ATTR_KIND_SPECULATIVE_LOAD_HARDENING:
     return Attribute::SpeculativeLoadHardening;
   case bitc::ATTR_KIND_SWIFT_ERROR:
diff --git a/llvm/lib/Bitcode/Writer/BitcodeWriter.cpp b/llvm/lib/Bitcode/Writer/BitcodeWriter.cpp
index a1d5b36bde64d..742c1836c32cd 100644
--- a/llvm/lib/Bitcode/Writer/BitcodeWriter.cpp
+++ b/llvm/lib/Bitcode/Writer/BitcodeWriter.cpp
@@ -883,6 +883,8 @@ static uint64_t getAttrKindEncoding(Attribute::AttrKind Kind) {
     return bitc::ATTR_KIND_STRUCT_RET;
   case Attribute::SanitizeAddress:
     return bitc::ATTR_KIND_SANITIZE_ADDRESS;
+  case Attribute::SanitizeAllocToken:
+    return bitc::ATTR_KIND_SANITIZE_ALLOC_TOKEN;
   case Attribute::SanitizeHWAddress:
     return bitc::ATTR_KIND_SANITIZE_HWADDRESS;
   case Attribute::SanitizeThread:
diff --git a/llvm/lib/Transforms/Utils/CodeExtractor.cpp b/llvm/lib/Transforms/Utils/CodeExtractor.cpp
index bbd1ed6a3ab2d..5ba6f95f5fae8 100644
--- a/llvm/lib/Transforms/Utils/CodeExtractor.cpp
+++ b/llvm/lib/Transforms/Utils/CodeExtractor.cpp
@@ -970,6 +970,7 @@ Function *CodeExtractor::constructFunctionDeclaration(
       case Attribute::SanitizeMemTag:
       case Attribute::SanitizeRealtime:
       case Attribute::SanitizeRealtimeBlocking:
+      case Attribute::SanitizeAllocToken:
       case Attribute::SpeculativeLoadHardening:
       case Attribute::StackProtect:
       case Attribute::StackProtectReq:
diff --git a/llvm/lib/Transforms/Utils/Local.cpp b/llvm/lib/Transforms/Utils/Local.cpp
index b94ed7db91580..0649e70f4f5c8 100644
--- a/llvm/lib/Transforms/Utils/Local.cpp
+++ b/llvm/lib/Transforms/Utils/Local.cpp
@@ -3025,6 +3025,10 @@ static void combineMetadata(Instruction *K, const Instruction *J,
         // Preserve !nosanitize if both K and J have it.
         K->setMetadata(Kind, JMD);
         break;
+      case LLVMContext::MD_alloc_token:
+        // Preserve !alloc_token if both K and J have it.
+        K->setMetadata(Kind, JMD);
+        break;
       }
   }
   // Set !invariant.group from J if J has it. If both instructions have it
diff --git a/llvm/test/Bitcode/attributes.ll b/llvm/test/Bitcode/attributes.ll
index 8c1a76365e1b4..aef7810fe2c3b 100644
--- a/llvm/test/Bitcode/attributes.ll
+++ b/llvm/test/Bitcode/attributes.ll
@@ -516,6 +516,11 @@ define void @f93() sanitize_realtime_blocking {
         ret void;
 }
 
+; CHECK: define void @f_sanitize_alloc_token() #55
+define void @f_sanitize_alloc_token() sanitize_alloc_token {
+        ret void;
+}
+
 ; CHECK: define void @f87() [[FNRETTHUNKEXTERN:#[0-9]+]]
 define void @f87() fn_ret_thunk_extern { ret void }
 
@@ -627,6 +632,7 @@ define void @dead_on_return(ptr dead_on_return %p) {
 ; CHECK: attributes #52 = { nosanitize_bounds }
 ; CHECK: attributes #53 = { sanitize_realtime }
 ; CHECK: attributes #54 = { sanitize_realtime_blocking }
+; CHECK: attributes #55 = { sanitize_alloc_token }
 ; CHECK: attributes [[FNRETTHUNKEXTERN]] = { fn_ret_thunk_extern }
 ; CHECK: attributes [[SKIPPROFILE]] = { skipprofile }
 ; CHECK: attributes [[OPTDEBUG]] = { optdebug }
diff --git a/llvm/test/Bitcode/compatibility.ll b/llvm/test/Bitcode/compatibility.ll
index 0b5ce08c00a23..e21786e5ee330 100644
--- a/llvm/test/Bitcode/compatibility.ll
+++ b/llvm/test/Bitcode/compatibility.ll
@@ -1718,7 +1718,7 @@ exit:
   ; CHECK: select <2 x i1> <i1 true, i1 false>, <2 x i8> <i8 2, i8 3>, <2 x i8> <i8 3, i8 2>
 
   call void @f.nobuiltin() builtin
-  ; CHECK: call void @f.nobuiltin() #54
+  ; CHECK: call void @f.nobuiltin() #55
 
   call fastcc noalias ptr @f.noalias() noinline
   ; CHECK: call fastcc noalias ptr @f.noalias() #12
@@ -2151,6 +2151,9 @@ declare void @f.sanitize_realtime() sanitize_realtime
 declare void @f.sanitize_realtime_blocking() sanitize_realtime_blocking
 ; CHECK: declare void @f.sanitize_realtime_blocking() #53
 
+declare void @f.sanitize_alloc_token() sanitize_alloc_token
+; CHECK: declare void @f.sanitize_alloc_token() #54
+
 ; CHECK: declare nofpclass(snan) float @nofpclass_snan(float nofpclass(snan))
 declare nofpclass(snan) float @nofpclass_snan(float nofpclass(snan))
 
@@ -2284,7 +2287,8 @@ define float @nofpclass_callsites(float %arg, { float } %arg1) {
 ; CHECK: attributes #51 = { sanitize_numerical_stability }
 ; CHECK: attributes #52 = { sanitize_realtime }
 ; CHECK: attributes #53 = { sanitize_realtime_blocking }
-; CHECK: attributes #54 = { builtin }
+; CHECK: attributes #54 = { sanitize_alloc_token }
+; CHECK: attributes #55 = { builtin }
 
 ;; Metadata
 
diff --git a/llvm/test/Transforms/Inline/attributes.ll b/llvm/test/Transforms/Inline/attributes.ll
index 42b1a3a29aec4..55ab430f201d6 100644
--- a/llvm/test/Transforms/Inline/attributes.ll
+++ b/llvm/test/Transforms/Inline/attributes.ll
@@ -26,6 +26,10 @@ define i32 @sanitize_memtag_callee(i32 %i) sanitize_memtag {
   ret i32 %i
 }
 
+define i32 @sanitize_alloc_token_callee(i32 %i) sanitize_alloc_token {
+  ret i32 %i
+}
+
 define i32 @safestack_callee(i32 %i) safestack {
   ret i32 %i
 }
@@ -58,6 +62,10 @@ define i32 @alwaysinline_sanitize_memtag_callee(i32 %i) alwaysinline sanitize_me
   ret i32 %i
 }
 
+define i32 @alwaysinline_sanitize_alloc_token_callee(i32 %i) alwaysinline sanitize_alloc_token {
+  ret i32 %i
+}
+
 define i32 @alwaysinline_safestack_callee(i32 %i) alwaysinline safestack {
   ret i32 %i
 }
@@ -184,6 +192,39 @@ define i32 @test_sanitize_memtag(i32 %arg) sanitize_memtag {
 ; CHECK-NEXT: ret i32
 }
 
+; ---------------------------------------------------------------------------- ;
+
+; Can inline sanitize_alloc_token functions into a noattr function. The
+; attribute is *not* viral, otherwise may break code.
+define i32 @test_no_sanitize_alloc_token(i32 %arg) {
+; CHECK-LABEL: @test_no_sanitize_alloc_token(
+; CHECK-SAME: ) {
+; CHECK-NOT: call
+; CHECK: ret i32
+entry:
+  %x1 = call i32 @noattr_callee(i32 %arg)
+  %x2 = call i32 @sanitize_alloc_token_callee(i32 %x1)
+  %x3 = call i32 @alwaysinline_callee(i32 %x2)
+  %x4 = call i32 @alwaysinline_sanitize_alloc_token_callee(i32 %x3)
+  ret i32 %x4
+}
+
+; Can inline noattr functions into a sanitize_alloc_token function. If
+; inlinable noattr functions cannot be instrumented, they should be marked with
+; explicit noinline.
+define i32 @test_sanitize_alloc_token(i32 %arg) sanitize_alloc_token {
+; CHECK-LABEL: @test_sanitize_alloc_token(
+; CHECK-SAME: ) [[SANITIZE_ALLOC_TOKEN:.*]] {
+; CHECK-NOT: call
+; CHECK: ret i32
+entry:
+  %x1 = call i32 @noattr_callee(i32 %arg)
+  %x2 = call i32 @sanitize_alloc_token_callee(i32 %x1)
+  %x3 = call i32 @alwaysinline_callee(i32 %x2)
+  %x4 = call i32 @alwaysinline_sanitize_alloc_token_callee(i32 %x3)
+  ret i32 %x4
+}
+
 define i32 @test_safestack(i32 %arg) safestack {
   %x1 = call i32 @noattr_callee(i32 %arg)
   %x2 = call i32 @safestack_callee(i32 %x1)
@@ -639,6 +680,7 @@ define i32 @loader_replaceable_caller() {
   ret i32 %1
 }
 
+; CHECK: attributes [[SANITIZE_ALLOC_TOKEN]] = { sanitize_alloc_token }
 ; CHECK: attributes [[SLH]] = { speculative_load_hardening }
 ; CHECK: attributes [[FPMAD_FALSE]] = { "less-precise-fpmad"="false" }
 ; CHECK: attributes [[FPMAD_TRUE]] = { "less-precise-fpmad"="true" }
diff --git a/llvm/utils/emacs/llvm-mode.el b/llvm/utils/emacs/llvm-mode.el
index 660d0718f098c..240c13319f634 100644
--- a/llvm/utils/emacs/llvm-mode.el
+++ b/llvm/utils/emacs/llvm-mode.el
@@ -34,7 +34,7 @@
          "inaccessiblemem_or_argmemonly" "inalloca" "inlinehint" "jumptable" "minsize" "mustprogress" "naked" "nobuiltin" "nonnull" "nocapture"
          "nocallback" "nocf_check" "noduplicate" "noext" "nofree" "noimplicitfloat" "noinline" "nomerge" "nonlazybind" "noprofile" "noredzone" "noreturn"
          "norecurse" "nosync" "noundef" "nounwind" "nosanitize_bounds" "nosanitize_coverage" "null_pointer_is_valid" "optdebug" "optforfuzzing" "optnone" "optsize" "preallocated" "readnone" "readonly" "returned" "returns_twice"
-         "shadowcallstack" "signext" "speculatable" "speculative_load_hardening" "ssp" "sspreq" "sspstrong" "safestack" "sanitize_address" "sanitize_hwaddress" "sanitize_memtag"
+         "shadowcallstack" "signext" "speculatable" "speculative_load_hardening" "ssp" "sspreq" "sspstrong" "safestack" "sanitize_address" "sanitize_alloc_token" "sanitize_hwaddress" "sanitize_memtag"
          "sanitize_thread" "sanitize_memory" "strictfp" "swifterror" "uwtable" "vscale_range" "willreturn" "writeonly" "zeroext") 'symbols) . font-lock-constant-face)
    ;; Variables
    '("%[-a-zA-Z$._][-a-zA-Z$._0-9]*" . font-lock-variable-name-face)
diff --git a/llvm/utils/llvm.grm b/llvm/utils/llvm.grm
index 411323178bde1..dddfe3c301b65 100644
--- a/llvm/utils/llvm.grm
+++ b/llvm/utils/llvm.grm
@@ -173,6 +173,7 @@ FuncAttr      ::= noreturn
  | returns_twice
  | nonlazybind
  | sanitize_address
+ | sanitize_alloc_token
  | sanitize_thread
  | sanitize_memory
  | mustprogress
diff --git a/llvm/utils/vim/syntax/llvm.vim b/llvm/utils/vim/syntax/llvm.vim
index e3b8ff8629559..e048caa20406a 100644
--- a/llvm/utils/vim/syntax/llvm.vim
+++ b/llvm/utils/vim/syntax/llvm.vim
@@ -163,6 +163,7 @@ syn keyword llvmKeyword
       \ returns_twice
       \ safestack
       \ sanitize_address
+      \ sanitize_alloc_token
       \ sanitize_hwaddress
       \ sanitize_memory
       \ sanitize_memtag
diff --git a/llvm/utils/vscode/llvm/syntaxes/ll.tmLanguage.yaml b/llvm/utils/vscode/llvm/syntaxes/ll.tmLanguage.yaml
index b64482336f404..1faaf6b26f301 100644
--- a/llvm/utils/vscode/llvm/syntaxes/ll.tmLanguage.yaml
+++ b/llvm/utils/vscode/llvm/syntaxes/ll.tmLanguage.yaml
@@ -258,6 +258,7 @@ patterns:
             \\breturns_twice\\b|\
             \\bsafestack\\b|\
             \\bsanitize_address\\b|\
+            \\bsanitize_alloc_token\\b|\
             \\bsanitize_hwaddress\\b|\
             \\bsanitize_memory\\b|\
             \\bsanitize_memtag\\b|\

``````````

</details>


https://github.com/llvm/llvm-project/pull/160131


More information about the llvm-commits mailing list