[libc] Adds implementation for memrchr.

Reviewed By: sivachandra
Differential Revision: https://reviews.llvm.org/D84469

Author: cgyurgyik

libc/config/linux/aarch64/entrypoints.txt

@@ -13,6 +13,7 @@ set(TARGET_LIBC_ENTRYPOINTS
     libc.src.string.strchr
     libc.src.string.strstr
     libc.src.string.strnlen
+    libc.src.string.memrchr
 )
 
 set(TARGET_LIBM_ENTRYPOINTS

libc/config/linux/api.td

@@ -216,7 +216,8 @@ def StringAPI : PublicAPI<"string.h"> {
     "strtok",
     "strerror",
     "strlen",
-    "strnlen"
+    "strnlen",
+    "memrchr"
   ];
 
   let TypeDeclarations = [

libc/config/linux/x86_64/entrypoints.txt

@@ -31,6 +31,7 @@ set(TARGET_LIBC_ENTRYPOINTS
     libc.src.string.strchr
     libc.src.string.strstr
     libc.src.string.strnlen
+    libc.src.string.memrchr
 
     # sys/mman.h entrypoints
     libc.src.sys.mman.mmap

libc/spec/gnu_ext.td

@@ -12,8 +12,22 @@ def GnuExtensions : StandardSpec<"GNUExtensions"> {
         >,
       ]
   >;
+  
+  HeaderSpec String = HeaderSpec<
+      "string.h",
+      [], // Macros
+      [], // Types
+      [], // Enumerations
+      [ 
+        FunctionSpec<
+            "memrchr",
+            RetValSpec<VoidPtr>,
+            [ArgSpec<VoidPtr>, ArgSpec<IntType>, ArgSpec<SizeTType>]
+        >,
+      ]
+  >;
 
   let Headers = [
-    Math,
+    Math, String,
   ];
 }

libc/src/string/CMakeLists.txt

@@ -78,6 +78,14 @@ add_entrypoint_object(
     .memchr
 )
 
+add_entrypoint_object(
+  memrchr
+  SRCS
+    memrchr.cpp
+  HDRS
+    memrchr.h
+)
+
 # Helper to define a function with multiple implementations
 # - Computes flags to satisfy required/rejected features and arch,
 # - Declares an entry point,

libc/src/string/memrchr.cpp

@@ -0,0 +1,26 @@
+//===-- Implementation of memrchr -----------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "src/string/memrchr.h"
+#include "src/__support/common.h"
+#include <stddef.h>
+
+namespace __llvm_libc {
+
+void *LLVM_LIBC_ENTRYPOINT(memrchr)(const void *src, int c, size_t n) {
+  const unsigned char *str = reinterpret_cast<const unsigned char *>(src);
+  const unsigned char ch = c;
+  for (; n != 0; --n) {
+    const unsigned char *s = str + n - 1;
+    if (*s == ch)
+      return const_cast<unsigned char *>(s);
+  }
+  return nullptr;
+}
+
+} // namespace __llvm_libc

libc/src/string/memrchr.h

@@ -0,0 +1,20 @@
+//===-- Implementation header for memrchr -----------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_LIBC_SRC_STRING_MEMRCHR_H
+#define LLVM_LIBC_SRC_STRING_MEMRCHR_H
+
+#include <stddef.h>
+
+namespace __llvm_libc {
+
+void *memrchr(const void *src, int c, size_t n);
+
+} // namespace __llvm_libc
+
+#endif // LLVM_LIBC_SRC_STRING_MEMRCHR_H

libc/test/src/string/CMakeLists.txt

@@ -82,6 +82,16 @@ add_libc_unittest(
     libc.src.string.strnlen
 )
 
+add_libc_unittest(
+  memrchr_test
+  SUITE
+    libc_string_unittests
+  SRCS
+    memrchr_test.cpp
+  DEPENDS
+    libc.src.string.memrchr
+)
+
 # Tests all implementations that can run on the host.
 function(add_libc_multi_impl_test name)
   get_property(fq_implementations GLOBAL PROPERTY ${name}_implementations)

libc/test/src/string/memrchr_test.cpp

@@ -0,0 +1,114 @@
+//===-- Unittests for memrchr ---------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "src/string/memrchr.h"
+#include "utils/UnitTest/Test.h"
+#include <stddef.h>
+
+// A helper function that calls memrchr and abstracts away the explicit cast for
+// readability purposes.
+const char *call_memrchr(const void *src, int c, size_t size) {
+  return reinterpret_cast<const char *>(__llvm_libc::memrchr(src, c, size));
+}
+
+TEST(MemRChrTest, FindsCharacterAfterNullTerminator) {
+  // memrchr should continue searching after a null terminator.
+  const size_t size = 6;
+  const unsigned char src[size] = {'a', '\0', 'b', 'c', 'd', '\0'};
+  // Should return 'b', 'c', 'd', '\0' even when after null terminator.
+  ASSERT_STREQ(call_memrchr(src, 'b', size), "bcd");
+}
+
+TEST(MemRChrTest, FindsCharacterInNonNullTerminatedCollection) {
+  const size_t size = 3;
+  const unsigned char src[size] = {'a', 'b', 'c'};
+  // Should return 'b', 'c'.
+  const char *ret = call_memrchr(src, 'b', size);
+  ASSERT_EQ(ret[0], 'b');
+  ASSERT_EQ(ret[1], 'c');
+}
+
+TEST(MemRChrTest, FindsFirstCharacter) {
+  const size_t size = 6;
+  const unsigned char src[size] = {'a', 'b', 'c', 'd', 'e', '\0'};
+  // Should return original array since 'a' is the first character.
+  ASSERT_STREQ(call_memrchr(src, 'a', size), "abcde");
+}
+
+TEST(MemRChrTest, FindsMiddleCharacter) {
+  const size_t size = 6;
+  const unsigned char src[size] = {'a', 'b', 'c', 'd', 'e', '\0'};
+  // Should return characters after (and including) 'c'.
+  ASSERT_STREQ(call_memrchr(src, 'c', size), "cde");
+}
+
+TEST(MemRChrTest, FindsLastCharacterThatIsNotNullTerminator) {
+  const size_t size = 6;
+  const unsigned char src[size] = {'a', 'b', 'c', 'd', 'e', '\0'};
+  // Should return 'e' and null-terminator.
+  ASSERT_STREQ(call_memrchr(src, 'e', size), "e");
+}
+
+TEST(MemRChrTest, FindsNullTerminator) {
+  const size_t size = 6;
+  const unsigned char src[size] = {'a', 'b', 'c', 'd', 'e', '\0'};
+  // Should return null terminator.
+  ASSERT_STREQ(call_memrchr(src, '\0', size), "");
+}
+
+TEST(MemRChrTest, CharacterNotWithinStringShouldReturnNullptr) {
+  const size_t size = 4;
+  const unsigned char src[size] = {'1', '2', '3', '?'};
+  // Since 'z' is not within 'characters', should return nullptr.
+  ASSERT_STREQ(call_memrchr(src, 'z', size), nullptr);
+}
+
+TEST(MemRChrTest, CharacterNotWithinSizeShouldReturnNullptr) {
+  const unsigned char src[5] = {'1', '2', '3', '4', '\0'};
+  // Since '4' is not within the first 2 characters, this should return nullptr.
+  const size_t size = 2;
+  ASSERT_STREQ(call_memrchr(src, '4', size), nullptr);
+}
+
+TEST(MemRChrTest, ShouldFindLastOfDuplicates) {
+  size_t size = 12; // 11 characters + null terminator.
+  const char *dups = "abc1def1ghi";
+  // 1 is duplicated in 'dups', but it should find the last copy.
+  ASSERT_STREQ(call_memrchr(dups, '1', size), "1ghi");
+
+  const char *repeated = "XXXXX";
+  size = 6; // 5 characters + null terminator.
+  // Should return the last X with the null terminator.
+  ASSERT_STREQ(call_memrchr(repeated, 'X', size), "X");
+}
+
+TEST(MemRChrTest, EmptyStringShouldOnlyMatchNullTerminator) {
+  const size_t size = 1; // Null terminator.
+  const char *empty_string = "";
+  // Null terminator should match.
+  ASSERT_STREQ(call_memrchr(empty_string, '\0', size), "");
+  // All other characters should not match.
+  ASSERT_STREQ(call_memrchr(empty_string, 'A', size), nullptr);
+  ASSERT_STREQ(call_memrchr(empty_string, '9', size), nullptr);
+  ASSERT_STREQ(call_memrchr(empty_string, '?', size), nullptr);
+}
+
+TEST(MemRChrTest, SignedCharacterFound) {
+  char c = -1;
+  const size_t size = 1;
+  char src[size] = {c};
+  const char *actual = call_memrchr(src, c, size);
+  // Should find the last character 'c'.
+  ASSERT_EQ(actual[0], c);
+}
+
+TEST(MemRChrTest, ZeroLengthShouldReturnNullptr) {
+  const unsigned char src[4] = {'a', 'b', 'c', '\0'};
+  // This will iterate over exactly zero characters, so should return nullptr.
+  ASSERT_STREQ(call_memrchr(src, 'd', 0), nullptr);
+}