[libc-commits] [libc] b6a20a4 - [libc] Add memchr implementation.
via libc-commits
libc-commits at lists.llvm.org
Tue Jul 7 14:09:07 PDT 2020
Author: cgyurgyik
Date: 2020-07-07T17:08:50-04:00
New Revision: b6a20a4970813e0c64e2799f83ec246a82f70438
URL: https://github.com/llvm/llvm-project/commit/b6a20a4970813e0c64e2799f83ec246a82f70438
DIFF: https://github.com/llvm/llvm-project/commit/b6a20a4970813e0c64e2799f83ec246a82f70438.diff
LOG: [libc] Add memchr implementation.
Added:
libc/src/string/memchr.cpp
libc/src/string/memchr.h
libc/test/src/string/memchr_test.cpp
Modified:
libc/config/linux/aarch64/entrypoints.txt
libc/config/linux/x86_64/entrypoints.txt
libc/src/string/CMakeLists.txt
libc/test/src/CMakeLists.txt
libc/test/src/string/CMakeLists.txt
Removed:
################################################################################
diff --git a/libc/config/linux/aarch64/entrypoints.txt b/libc/config/linux/aarch64/entrypoints.txt
index af6c16bed19c..1eb9e8a99516 100644
--- a/libc/config/linux/aarch64/entrypoints.txt
+++ b/libc/config/linux/aarch64/entrypoints.txt
@@ -9,6 +9,7 @@ set(TARGET_LIBC_ENTRYPOINTS
libc.src.string.strcpy
libc.src.string.strcat
libc.src.string.strlen
+ libc.src.string.memchr
)
set(TARGET_LIBM_ENTRYPOINTS
diff --git a/libc/config/linux/x86_64/entrypoints.txt b/libc/config/linux/x86_64/entrypoints.txt
index 4ba34feeb7bf..322bf050c716 100644
--- a/libc/config/linux/x86_64/entrypoints.txt
+++ b/libc/config/linux/x86_64/entrypoints.txt
@@ -27,6 +27,7 @@ set(TARGET_LIBC_ENTRYPOINTS
libc.src.string.strcat
libc.src.string.strlen
libc.src.string.strcmp
+ libc.src.string.memchr
# sys/mman.h entrypoints
libc.src.sys.mman.mmap
diff --git a/libc/src/string/CMakeLists.txt b/libc/src/string/CMakeLists.txt
index 51456e8a94c8..05deee10d714 100644
--- a/libc/src/string/CMakeLists.txt
+++ b/libc/src/string/CMakeLists.txt
@@ -44,6 +44,15 @@ add_entrypoint_object(
libc.include.string
)
+add_entrypoint_object(
+ memchr
+ SRCS
+ memchr.cpp
+ HDRS
+ memchr.h
+)
+
+
# Helper to define a function with multiple implementations
# - Computes flags to satisfy required/rejected features and arch,
# - Declares an entry point,
diff --git a/libc/src/string/memchr.cpp b/libc/src/string/memchr.cpp
new file mode 100644
index 000000000000..eee3d4971edc
--- /dev/null
+++ b/libc/src/string/memchr.cpp
@@ -0,0 +1,23 @@
+//===-- Implementation of memchr ------------------------------------------===//
+//
+// 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/memchr.h"
+#include "src/__support/common.h"
+#include <stddef.h>
+
+namespace __llvm_libc {
+
+// TODO: Look at performance benefits of comparing words.
+void *LLVM_LIBC_ENTRYPOINT(memchr)(const void *src, int c, size_t n) {
+ const unsigned char *str = reinterpret_cast<const unsigned char *>(src);
+ for (; n && *str != c; --n, ++str)
+ ;
+ return n ? const_cast<unsigned char *>(str) : nullptr;
+}
+
+} // namespace __llvm_libc
diff --git a/libc/src/string/memchr.h b/libc/src/string/memchr.h
new file mode 100644
index 000000000000..369475305236
--- /dev/null
+++ b/libc/src/string/memchr.h
@@ -0,0 +1,20 @@
+//===-- Implementation header for memchr ------------------------*- 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_MEMCHR_H
+#define LLVM_LIBC_SRC_STRING_MEMCHR_H
+
+#include <stddef.h>
+
+namespace __llvm_libc {
+
+void *memchr(const void *src, int c, size_t n);
+
+} // namespace __llvm_libc
+
+#endif // LLVM_LIBC_SRC_STRING_MEMCHR_H
diff --git a/libc/test/src/CMakeLists.txt b/libc/test/src/CMakeLists.txt
index 432b6190e7bb..3e7bad72482b 100644
--- a/libc/test/src/CMakeLists.txt
+++ b/libc/test/src/CMakeLists.txt
@@ -91,4 +91,4 @@ endif()
target_link_libraries(libc-integration-test
PRIVATE
${library_files}
-)
+)
\ No newline at end of file
diff --git a/libc/test/src/string/CMakeLists.txt b/libc/test/src/string/CMakeLists.txt
index 992cebe5067c..0055591be5c7 100644
--- a/libc/test/src/string/CMakeLists.txt
+++ b/libc/test/src/string/CMakeLists.txt
@@ -42,6 +42,16 @@ add_libc_unittest(
libc.src.string.strcmp
)
+add_libc_unittest(
+ memchr_test
+ SUITE
+ libc_string_unittests
+ SRCS
+ memchr_test.cpp
+ DEPENDS
+ libc.src.string.memchr
+)
+
# Tests all implementations that can run on the host.
function(add_libc_multi_impl_test name)
get_property(fq_implementations GLOBAL PROPERTY ${name}_implementations)
@@ -66,4 +76,3 @@ endfunction()
add_libc_multi_impl_test(memcpy SRCS memcpy_test.cpp)
add_libc_multi_impl_test(memset SRCS memset_test.cpp)
add_libc_multi_impl_test(bzero SRCS bzero_test.cpp)
-
diff --git a/libc/test/src/string/memchr_test.cpp b/libc/test/src/string/memchr_test.cpp
new file mode 100644
index 000000000000..370aa2fa9f13
--- /dev/null
+++ b/libc/test/src/string/memchr_test.cpp
@@ -0,0 +1,113 @@
+//===-- Unittests for memchr ----------------------------------------------===//
+//
+// 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/memchr.h"
+#include "utils/UnitTest/Test.h"
+#include <stddef.h>
+
+// A helper function that calls memchr and abstracts away the explicit cast for
+// readability purposes.
+const char *call_memchr(const void *src, int c, size_t size) {
+ return reinterpret_cast<const char *>(__llvm_libc::memchr(src, c, size));
+}
+
+TEST(MemChrTest, FindsCharacterAfterNullTerminator) {
+ // memchr should continue searching after a null terminator.
+ const size_t size = 5;
+ const unsigned char src[size] = {'a', '\0', 'b', 'c', '\0'};
+ // Should return 'b', 'c', '\0' even when after null terminator.
+ ASSERT_STREQ(call_memchr(src, 'b', size), "bc");
+}
+
+TEST(MemChrTest, FindsCharacterInNonNullTerminatedCollection) {
+ const size_t size = 3;
+ const unsigned char src[size] = {'a', 'b', 'c'};
+ // Should return 'b', 'c'.
+ const char *ret = call_memchr(src, 'b', size);
+ ASSERT_EQ(ret[0], 'b');
+ ASSERT_EQ(ret[1], 'c');
+}
+
+TEST(MemChrTest, 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_memchr(src, 'a', size), "abcde");
+}
+
+TEST(MemChrTest, 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_memchr(src, 'c', size), "cde");
+}
+
+TEST(MemChrTest, 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_memchr(src, 'e', size), "e");
+}
+
+TEST(MemChrTest, FindsNullTerminator) {
+ const size_t size = 6;
+ const unsigned char src[size] = {'a', 'b', 'c', 'd', 'e', '\0'};
+ // Should return null terminator.
+ ASSERT_STREQ(call_memchr(src, '\0', size), "");
+}
+
+TEST(MemChrTest, 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_memchr(src, 'z', size), nullptr);
+}
+
+TEST(MemChrTest, CharacterNotWithinSizeShouldReturnNullptr) {
+ const unsigned char src[5] = {'1', '2', '3', '4', '\0'};
+ // Since '4' is not the first or second character, this should return nullptr.
+ const size_t size = 2;
+ ASSERT_STREQ(call_memchr(src, '4', size), nullptr);
+}
+
+TEST(MemChrTest, TheSourceShouldNotChange) {
+ const size_t size = 6;
+ const unsigned char src[size] = {'a', 'b', 'c', 'd', 'e', '\0'};
+ const char *src_copy = reinterpret_cast<const char *>(src);
+ // When the character is found, the source string should not change.
+ __llvm_libc::memchr(src, 'd', size);
+ ASSERT_STREQ(reinterpret_cast<const char *>(src), src_copy);
+ // Same case for when the character is not found.
+ __llvm_libc::memchr(src, 'z', size);
+ ASSERT_STREQ(reinterpret_cast<const char *>(src), src_copy);
+}
+
+TEST(MemChrTest, ShouldFindFirstOfDuplicates) {
+ const size_t size = 12; // 11 characters + null terminator.
+ const char *dups = "abc1def1ghi";
+ // 1 is duplicated in 'dups', but it should find the first copy.
+ ASSERT_STREQ(call_memchr(dups, '1', size), "1def1ghi");
+}
+
+TEST(MemChrTest, EmptyStringShouldOnlyMatchNullTerminator) {
+ const size_t size = 1; // Null terminator.
+ const char *empty_string = "";
+ // Null terminator should match.
+ ASSERT_STREQ(call_memchr(empty_string, '\0', size), "");
+ // All other characters should not match.
+ ASSERT_STREQ(call_memchr(empty_string, 'A', size), nullptr);
+ ASSERT_STREQ(call_memchr(empty_string, '9', size), nullptr);
+ ASSERT_STREQ(call_memchr(empty_string, '?', size), nullptr);
+}
+
+TEST(MemChrTest, SingleRepeatedCharacterShouldReturnFirst) {
+ const char *dups = "XXXXX";
+ const size_t size = 6; // 5 characters + null terminator.
+ // Should return original string since X is first character.
+ ASSERT_STREQ(call_memchr(dups, 'X', size), dups);
+}
More information about the libc-commits
mailing list