[𝘀𝗽𝗿] initial version

Created using spr 1.3.6

Author: qinkunbao

clang/docs/ReleaseNotes.rst

@@ -960,6 +960,7 @@ Sanitizers
 ----------
 
 - ``-fsanitize=vptr`` is no longer a part of ``-fsanitize=undefined``.
+- Sanitizer ignorelists now support the syntax ``src:*=sanitize``.
 
 Python Binding Changes
 ----------------------

clang/docs/SanitizerSpecialCaseList.rst

@@ -58,7 +58,7 @@ Usage with UndefinedBehaviorSanitizer
 ability to adjust instrumentation based on type.
 
 By default, supported sanitizers will have their instrumentation disabled for
-types specified within an ignorelist.
+entris specified within an ignorelist.
 
 .. code-block:: bash
 
@@ -77,10 +77,8 @@ For example, supplying the above ``ignorelist.txt`` to
 ``-fsanitize-ignorelist=ignorelist.txt`` disables overflow sanitizer
 instrumentation for arithmetic operations containing values of type ``int``.
 
-The ``=sanitize`` category is also supported. Any types assigned to the
-``sanitize`` category will have their sanitizer instrumentation remain. If the
-same type appears within or across ignorelists with different categories the
-``sanitize`` category takes precedence -- regardless of order.
+The ``=sanitize`` category is also supported. Any entries assigned to the
+``sanitize`` category will have their sanitizer instrumentation remain.
 
 With this, one may disable instrumentation for some or all types and
 specifically allow instrumentation for one or many types -- including types
@@ -103,6 +101,22 @@ supported sanitizers.
     char c = toobig; // also not instrumented
   }
 
+Conflicting entries are resolved by the latest entry, which takes precedence.
+
+.. code-block:: bash
+
+  $ cat ignorelist1.txt
+  # test.cc will be instrumented.
+  src:*
+  src:*/mylib/*=sanitize
+  src:*/mylib/test.cc
+
+  $ cat ignorelist2.txt
+  # test.cc will not be instrumented.
+  src:*
+  src:*/mylib/test.cc
+  src:*/mylib/*=sanitize
+
 Format
 ======
 

clang/include/clang/Basic/SanitizerSpecialCaseList.h

@@ -43,13 +43,18 @@ class SanitizerSpecialCaseList : public llvm::SpecialCaseList {
   bool inSection(SanitizerMask Mask, StringRef Prefix, StringRef Query,
                  StringRef Category = StringRef()) const;
 
+  // Query ignorelisted entries if any bit in Mask matches the entry's section.
+  // Return 0 if not found. If found, return the line number (starts with 1).
+  unsigned inSectionBlame(SanitizerMask Mask, StringRef Prefix, StringRef Query,
+                          StringRef Category = StringRef()) const;
+
 protected:
   // Initialize SanitizerSections.
   void createSanitizerSections();
 
   struct SanitizerSection {
     SanitizerSection(SanitizerMask SM, SectionEntries &E)
-        : Mask(SM), Entries(E){};
+        : Mask(SM), Entries(E) {};
 
     SanitizerMask Mask;
     SectionEntries &Entries;

clang/lib/Basic/NoSanitizeList.cpp

@@ -44,7 +44,15 @@ bool NoSanitizeList::containsFunction(SanitizerMask Mask,
 
 bool NoSanitizeList::containsFile(SanitizerMask Mask, StringRef FileName,
                                   StringRef Category) const {
-  return SSCL->inSection(Mask, "src", FileName, Category);
+  unsigned NoSanLine = SSCL->inSectionBlame(Mask, "src", FileName, Category);
+  if (NoSanLine == 0)
+    return false;
+  unsigned SanLine = SSCL->inSectionBlame(Mask, "src", FileName, "sanitize");
+  // If we have two cases such as `src:a.cpp=sanitize` and `src:a.cpp`, the
+  // current entry override the previous entry.
+  if (SanLine > 0)
+    return NoSanLine > SanLine;
+  return true;
 }
 
 bool NoSanitizeList::containsMainFile(SanitizerMask Mask, StringRef FileName,

clang/lib/Basic/SanitizerSpecialCaseList.cpp

@@ -56,10 +56,20 @@ void SanitizerSpecialCaseList::createSanitizerSections() {
 bool SanitizerSpecialCaseList::inSection(SanitizerMask Mask, StringRef Prefix,
                                          StringRef Query,
                                          StringRef Category) const {
-  for (auto &S : SanitizerSections)
-    if ((S.Mask & Mask) &&
-        SpecialCaseList::inSectionBlame(S.Entries, Prefix, Query, Category))
-      return true;
+  return inSectionBlame(Mask, Prefix, Query, Category) > 0;
+}
 
-  return false;
+unsigned SanitizerSpecialCaseList::inSectionBlame(SanitizerMask Mask,
+                                                  StringRef Prefix,
+                                                  StringRef Query,
+                                                  StringRef Category) const {
+  for (auto &S : SanitizerSections) {
+    if (S.Mask & Mask) {
+      unsigned lineNum =
+          SpecialCaseList::inSectionBlame(S.Entries, Prefix, Query, Category);
+      if (lineNum > 0)
+        return lineNum;
+    }
+  }
+  return 0;
 }

clang/test/CodeGen/ubsan-src-ignorelist-category.test

@@ -3,11 +3,11 @@
 
 // RUN: %clang_cc1 -triple x86_64-linux-gnu -fsanitize=signed-integer-overflow -fsanitize-ignorelist=%t/src.ignorelist -emit-llvm %t/test2.c -o -             | FileCheck %s --check-prefixes=CHECK2
 
-// RUN: %clang_cc1 -triple x86_64-linux-gnu -fsanitize=signed-integer-overflow -fsanitize-ignorelist=%t/src.ignorelist -emit-llvm %t/test1.c -o -             | FileCheck %s --check-prefixes=CHECK1,IGNORE
+// RUN: %clang_cc1 -triple x86_64-linux-gnu -fsanitize=signed-integer-overflow -fsanitize-ignorelist=%t/src.ignorelist -emit-llvm %t/test1.c -o -             | FileCheck %s --check-prefixes=CHECK1,SANITIZE
 // RUN: %clang_cc1 -triple x86_64-linux-gnu -fsanitize=signed-integer-overflow -fsanitize-ignorelist=%t/src.ignorelist.contradict1 -emit-llvm %t/test1.c -o - | FileCheck %s --check-prefixes=CHECK1,IGNORE
-// RUN: %clang_cc1 -triple x86_64-linux-gnu -fsanitize=signed-integer-overflow -fsanitize-ignorelist=%t/src.ignorelist.contradict2 -emit-llvm %t/test1.c -o - | FileCheck %s --check-prefixes=CHECK1,IGNORE
+// RUN: %clang_cc1 -triple x86_64-linux-gnu -fsanitize=signed-integer-overflow -fsanitize-ignorelist=%t/src.ignorelist.contradict2 -emit-llvm %t/test1.c -o - | FileCheck %s --check-prefixes=CHECK1,SANITIZE
 // RUN: %clang_cc1 -triple x86_64-linux-gnu -fsanitize=signed-integer-overflow -fsanitize-ignorelist=%t/src.ignorelist.contradict3 -emit-llvm %t/test1.c -o - | FileCheck %s --check-prefixes=CHECK1,IGNORE
-// RUN: %clang_cc1 -triple x86_64-linux-gnu -fsanitize=signed-integer-overflow -fsanitize-ignorelist=%t/src.ignorelist.contradict4 -emit-llvm %t/test1.c -o - | FileCheck %s --check-prefixes=CHECK1,IGNORE
+// RUN: %clang_cc1 -triple x86_64-linux-gnu -fsanitize=signed-integer-overflow -fsanitize-ignorelist=%t/src.ignorelist.contradict4 -emit-llvm %t/test1.c -o - | FileCheck %s --check-prefixes=CHECK1,SANITIZE
 
 
 // Verify ubsan only emits checks for files in the allowlist

llvm/include/llvm/Support/SpecialCaseList.h

@@ -138,6 +138,7 @@ class SpecialCaseList {
     std::unique_ptr<Matcher> SectionMatcher;
     SectionEntries Entries;
     std::string SectionStr;
+    unsigned LineNo;
   };
 
   std::vector<Section> Sections;

llvm/lib/Support/SpecialCaseList.cpp

@@ -63,6 +63,11 @@ Error SpecialCaseList::Matcher::insert(StringRef Pattern, unsigned LineNumber,
                        .moveInto(Pair.first))
       return Err;
     Pair.second = LineNumber;
+  } else {
+    // We should update the new line number if an entry with the same pattern
+    // repeats.
+    auto &Pair = It->getValue();
+    Pair.second = LineNumber;
   }
   return Error::success();
 }
@@ -132,18 +137,25 @@ bool SpecialCaseList::createInternal(const MemoryBuffer *MB,
 Expected<SpecialCaseList::Section *>
 SpecialCaseList::addSection(StringRef SectionStr, unsigned LineNo,
                             bool UseGlobs) {
-  Sections.emplace_back();
-  auto &Section = Sections.back();
-  Section.SectionStr = SectionStr;
-
-  if (auto Err = Section.SectionMatcher->insert(SectionStr, LineNo, UseGlobs)) {
+  auto it =
+      std::find_if(Sections.begin(), Sections.end(), [&](const Section &s) {
+        return s.SectionStr == SectionStr && s.LineNo == LineNo;
+      });
+  if (it == Sections.end()) {
+    Sections.emplace_back();
+    auto &sec = Sections.back();
+    sec.SectionStr = SectionStr;
+    sec.LineNo = LineNo;
+    it = std::prev(Sections.end());
+  }
+  if (auto Err = it->SectionMatcher->insert(SectionStr, LineNo, UseGlobs)) {
     return createStringError(errc::invalid_argument,
                              "malformed section at line " + Twine(LineNo) +
                                  ": '" + SectionStr +
                                  "': " + toString(std::move(Err)));
   }
 
-  return &Section;
+  return &(*it);
 }
 
 bool SpecialCaseList::parse(const MemoryBuffer *MB, std::string &Error) {

llvm/unittests/Support/SpecialCaseListTest.cpp

@@ -306,4 +306,26 @@ TEST_F(SpecialCaseListTest, Version2) {
   EXPECT_TRUE(SCL->inSection("sect2", "fun", "bar"));
   EXPECT_FALSE(SCL->inSection("sect3", "fun", "bar"));
 }
+
+TEST_F(SpecialCaseListTest, Version3) {
+  std::unique_ptr<SpecialCaseList> SCL = makeSpecialCaseList("[sect1]\n"
+                                                             "src:foo*\n"
+                                                             "[sect1]\n"
+                                                             "src:bar*\n"
+                                                             "src:def\n"
+                                                             "[sect2]\n"
+                                                             "src:def\n"
+                                                             "src:def\n");
+  EXPECT_TRUE(SCL->inSection("sect1", "src", "fooz"));
+  EXPECT_TRUE(SCL->inSection("sect1", "src", "barz"));
+  EXPECT_FALSE(SCL->inSection("sect2", "src", "fooz"));
+
+  EXPECT_TRUE(SCL->inSection("sect2", "src", "def"));
+  EXPECT_TRUE(SCL->inSection("sect1", "src", "def"));
+
+  EXPECT_EQ(2u, SCL->inSectionBlame("sect1", "src", "fooz"));
+  EXPECT_EQ(4u, SCL->inSectionBlame("sect1", "src", "barz"));
+  EXPECT_EQ(5u, SCL->inSectionBlame("sect1", "src", "def"));
+  EXPECT_EQ(8u, SCL->inSectionBlame("sect2", "src", "def"));
+}
 }