[clang] Cleanup docs and code for legacy no_sanitize attributes (NFC). #161311
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Update generated docs for legacy attributes:
* no_sanitize_(address|thread|memory)
* no_address_safety_analysis
Those are older forms of no_sanitize("list", "of", "sanitizers")
attribute. They were previously as various spellings of the same
attribute, which made the auto-generated documentation confusing.
Fix this by explicitly making them three different attributes.
This would also allow to simplify the delegation to the new no_sanitize
form slightly, as we can instead rely on auto-generated code to check
that TSan and MSan can't be disabled for globals.
llvmbot
added
clang
|
@llvm/pr-subscribers-clang Author: Alexey Samsonov (vonosmas) ChangesUpdate generated docs for legacy attributes:
Those are older forms of no_sanitize("list", "of", "sanitizers") attribute. They were previously as various spellings of the same attribute, which made the auto-generated documentation confusing. Fix this by explicitly making them three different attributes. This would also allow to simplify the delegation to the new no_sanitize form slightly, as we can instead rely on auto-generated code to check that TSan and MSan can't be disabled for globals. See the HTML diff: Full diff: https://github.com/llvm/llvm-project/pull/161311.diff 4 Files Affected:
diff --git a/clang/include/clang/Basic/Attr.td b/clang/include/clang/Basic/Attr.td
index 2623f9ff6972f..de56bb38fd63e 100644
--- a/clang/include/clang/Basic/Attr.td
+++ b/clang/include/clang/Basic/Attr.td
@@ -3921,16 +3921,31 @@ def NoSanitize : InheritableAttr {
}];
}
-// Attributes to disable a specific sanitizer. No new sanitizers should be added
+// Attribute to disable AddressSanitizer. No new spellings should be added
// to this list; the no_sanitize attribute should be extended instead.
-def NoSanitizeSpecific : InheritableAttr {
+def NoSanitizeAddress : InheritableAttr {
let Spellings = [GCC<"no_address_safety_analysis">,
- GCC<"no_sanitize_address">,
- GCC<"no_sanitize_thread">,
- Clang<"no_sanitize_memory">];
+ GCC<"no_sanitize_address">];
let Subjects = SubjectList<[Function, GlobalVar], ErrorDiag>;
- let Documentation = [NoSanitizeAddressDocs, NoSanitizeThreadDocs,
- NoSanitizeMemoryDocs];
+ let Documentation = [NoSanitizeAddressDocs];
+ let ASTNode = 0;
+}
+
+// Attribute to disable ThreadSanitizer. No new spellings should be added
+// to this list; the no_sanitize attribute should be extended instead.
+def NoSanitizeThread : InheritableAttr {
+ let Spellings = [GCC<"no_sanitize_thread">];
+ let Subjects = SubjectList<[Function], ErrorDiag>;
+ let Documentation = [NoSanitizeThreadDocs];
+ let ASTNode = 0;
+}
+
+// Attribute to disable MemorySanitizer. No new spellings should be added
+// to this list; the no_sanitize attribute should be extended instead.
+def NoSanitizeMemory : InheritableAttr {
+ let Spellings = [Clang<"no_sanitize_memory">];
+ let Subjects = SubjectList<[Function], ErrorDiag>;
+ let Documentation = [NoSanitizeMemoryDocs];
let ASTNode = 0;
}
diff --git a/clang/include/clang/Basic/AttrDocs.td b/clang/include/clang/Basic/AttrDocs.td
index ee212a9b50f36..20a52b49a8f10 100644
--- a/clang/include/clang/Basic/AttrDocs.td
+++ b/clang/include/clang/Basic/AttrDocs.td
@@ -13,16 +13,16 @@
// version control.
//
// To run clang-tblgen to generate the .rst file:
-// clang-tblgen -gen-attr-docs -I <root>/llvm/tools/clang/include
-// <root>/llvm/tools/clang/include/clang/Basic/Attr.td -o
-// <root>/llvm/tools/clang/docs/AttributeReference.rst
+// clang-tblgen -gen-attr-docs -I <root>/clang/include
+// <root>/clang/include/clang/Basic/Attr.td -o
+// <root>/clang/docs/AttributeReference.rst
//
// To run sphinx to generate the .html files (note that sphinx-build must be
// available on the PATH):
// Windows (from within the clang\docs directory):
// make.bat html
-// Non-Windows (from within the clang\docs directory):
-// sphinx-build -b html _build/html
+// Non-Windows (from within the clang/docs directory):
+// sphinx-build -b html . _build/html
def GlobalDocumentation {
code Intro =[{..
@@ -3629,6 +3629,7 @@ instrumentations should not be applied.
The attribute takes a list of string literals with the following accepted
values:
+
* all values accepted by ``-fno-sanitize=``;
* ``coverage``, to disable SanitizerCoverage instrumentation.
diff --git a/clang/lib/Sema/SemaDeclAttr.cpp b/clang/lib/Sema/SemaDeclAttr.cpp
index a8dfa4d7df2d5..fceb8f0eda289 100644
--- a/clang/lib/Sema/SemaDeclAttr.cpp
+++ b/clang/lib/Sema/SemaDeclAttr.cpp
@@ -6361,19 +6361,8 @@ static void handleNoSanitizeAttr(Sema &S, Decl *D, const ParsedAttr &AL) {
Sanitizers.size()));
}
-static void handleNoSanitizeSpecificAttr(Sema &S, Decl *D,
- const ParsedAttr &AL) {
- StringRef AttrName = AL.getAttrName()->getName();
- normalizeName(AttrName);
- StringRef SanitizerName = llvm::StringSwitch<StringRef>(AttrName)
- .Case("no_address_safety_analysis", "address")
- .Case("no_sanitize_address", "address")
- .Case("no_sanitize_thread", "thread")
- .Case("no_sanitize_memory", "memory");
- if (isGlobalVar(D) && SanitizerName != "address")
- S.Diag(D->getLocation(), diag::err_attribute_wrong_decl_type)
- << AL << AL.isRegularKeywordAttribute() << ExpectedFunction;
-
+static AttributeCommonInfo
+getNoSanitizeAttrInfo(const ParsedAttr &NoSanitizeSpecificAttr) {
// FIXME: Rather than create a NoSanitizeSpecificAttr, this creates a
// NoSanitizeAttr object; but we need to calculate the correct spelling list
// index rather than incorrectly assume the index for NoSanitizeSpecificAttr
@@ -6383,11 +6372,34 @@ static void handleNoSanitizeSpecificAttr(Sema &S, Decl *D,
// getSpelling() or prettyPrint() on the resulting semantic attribute object
// without failing assertions.
unsigned TranslatedSpellingIndex = 0;
- if (AL.isStandardAttributeSyntax())
+ if (NoSanitizeSpecificAttr.isStandardAttributeSyntax())
TranslatedSpellingIndex = 1;
- AttributeCommonInfo Info = AL;
+ AttributeCommonInfo Info = NoSanitizeSpecificAttr;
Info.setAttributeSpellingListIndex(TranslatedSpellingIndex);
+ return Info;
+}
+
+static void handleNoSanitizeAddressAttr(Sema &S, Decl *D,
+ const ParsedAttr &AL) {
+ StringRef SanitizerName = "address";
+ AttributeCommonInfo Info = getNoSanitizeAttrInfo(AL);
+ D->addAttr(::new (S.Context)
+ NoSanitizeAttr(S.Context, Info, &SanitizerName, 1));
+}
+
+static void handleNoSanitizeThreadAttr(Sema &S, Decl *D,
+ const ParsedAttr &AL) {
+ StringRef SanitizerName = "thread";
+ AttributeCommonInfo Info = getNoSanitizeAttrInfo(AL);
+ D->addAttr(::new (S.Context)
+ NoSanitizeAttr(S.Context, Info, &SanitizerName, 1));
+}
+
+static void handleNoSanitizeMemoryAttr(Sema &S, Decl *D,
+ const ParsedAttr &AL) {
+ StringRef SanitizerName = "memory";
+ AttributeCommonInfo Info = getNoSanitizeAttrInfo(AL);
D->addAttr(::new (S.Context)
NoSanitizeAttr(S.Context, Info, &SanitizerName, 1));
}
@@ -7513,8 +7525,14 @@ ProcessDeclAttribute(Sema &S, Scope *scope, Decl *D, const ParsedAttr &AL,
case ParsedAttr::AT_NoSanitize:
handleNoSanitizeAttr(S, D, AL);
break;
- case ParsedAttr::AT_NoSanitizeSpecific:
- handleNoSanitizeSpecificAttr(S, D, AL);
+ case ParsedAttr::AT_NoSanitizeAddress:
+ handleNoSanitizeAddressAttr(S, D, AL);
+ break;
+ case ParsedAttr::AT_NoSanitizeThread:
+ handleNoSanitizeThreadAttr(S, D, AL);
+ break;
+ case ParsedAttr::AT_NoSanitizeMemory:
+ handleNoSanitizeMemoryAttr(S, D, AL);
break;
case ParsedAttr::AT_GuardedBy:
handleGuardedByAttr(S, D, AL);
diff --git a/clang/test/Misc/pragma-attribute-supported-attributes-list.test b/clang/test/Misc/pragma-attribute-supported-attributes-list.test
index 37ff33e5a1523..73d4cb1769ed5 100644
--- a/clang/test/Misc/pragma-attribute-supported-attributes-list.test
+++ b/clang/test/Misc/pragma-attribute-supported-attributes-list.test
@@ -126,7 +126,9 @@
// CHECK-NEXT: NoProfileFunction (SubjectMatchRule_function)
// CHECK-NEXT: NoRandomizeLayout (SubjectMatchRule_record)
// CHECK-NEXT: NoSanitize (SubjectMatchRule_function, SubjectMatchRule_objc_method, SubjectMatchRule_variable_is_global)
-// CHECK-NEXT: NoSanitizeSpecific (SubjectMatchRule_function, SubjectMatchRule_variable_is_global)
+// CHECK-NEXT: NoSanitizeAddress (SubjectMatchRule_function, SubjectMatchRule_variable_is_global)
+// CHECK-NEXT: NoSanitizeMemory (SubjectMatchRule_function)
+// CHECK-NEXT: NoSanitizeThread (SubjectMatchRule_function)
// CHECK-NEXT: NoSpeculativeLoadHardening (SubjectMatchRule_function, SubjectMatchRule_objc_method)
// CHECK-NEXT: NoSplitStack (SubjectMatchRule_function)
// CHECK-NEXT: NoStackProtector (SubjectMatchRule_function)
|
|
✅ With the latest revision this PR passed the C/C++ code formatter. |
vitalybuka
approved these changes
Sep 30, 2025
erichkeane
approved these changes
Sep 30, 2025
|
A release note is a good idea, just 'improved the heck outa sanitize docs' kinda thing. |
Done |
erichkeane
reviewed
Sep 30, 2025
Co-authored-by: Erich Keane <ekeane@nvidia.com>
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/27/builds/16833 Here is the relevant piece of the build log for the reference |
mahesh-attarde
pushed a commit
to mahesh-attarde/llvm-project
that referenced
this pull request
Oct 3, 2025
llvm#161311) Update generated docs for legacy attributes: * no_sanitize_(address|thread|memory) * no_address_safety_analysis Those are older forms of no_sanitize("list", "of", "sanitizers") attribute. They were previously as various spellings of the same attribute, which made the auto-generated documentation confusing. Fix this by explicitly making them three different attributes. This would also allow to simplify the delegation to the new no_sanitize form slightly, as we can instead rely on auto-generated code to check that TSan and MSan can't be disabled for globals. **HTML docs before:** <img width="1004" height="1175" alt="rendered-docs-before" src="https://github.com/user-attachments/assets/407b5fc1-799c-4882-8ff8-44a5ef3cf4f1" /> **HTML docs after:** <img width="1098" height="1118" alt="rendered-docs-after" src="https://github.com/user-attachments/assets/236ca93f-25f8-4d58-95ac-ede95ce18d01" /> --------- Co-authored-by: Erich Keane <ekeane@nvidia.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Update generated docs for legacy attributes:
Those are older forms of no_sanitize("list", "of", "sanitizers") attribute. They were previously as various spellings of the same attribute, which made the auto-generated documentation confusing.
Fix this by explicitly making them three different attributes. This would also allow to simplify the delegation to the new no_sanitize form slightly, as we can instead rely on auto-generated code to check that TSan and MSan can't be disabled for globals.
HTML docs before:
HTML docs after:
