bionic/tests/unistd_test.cpp
Derek Xue bc6447673f bionic/test: migrate sysconf tests from system/extras to bionic/tests
The old test are implemented in file:
  system/extras/tests/bionic/libc/other/test_sysconf.c
This change is to migrate them to bionic/tests with the gtest format.

and since the sysconf is defined in unistd.h, will put the test under
bionic/tests/unistd_test.cpp file as unistd.syscon test

Change-Id: Ie519147c1c86a6c4cefa8c88b18bf58bdfbffbdb
Signed-off-by: Yongqin Liu <yongqin.liu@linaro.org>
2014-11-15 09:48:36 +08:00

654 lines
18 KiB
C++

/*
* Copyright (C) 2012 The Android Open Source Project
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
#include <gtest/gtest.h>
#include "BionicDeathTest.h"
#include "ScopedSignalHandler.h"
#include "TemporaryFile.h"
#include <errno.h>
#include <fcntl.h>
#include <limits.h>
#include <stdint.h>
#include <sys/param.h>
#include <sys/syscall.h>
#include <sys/types.h>
#include <sys/utsname.h>
#include <sys/wait.h>
#include <unistd.h>
static void* get_brk() {
return sbrk(0);
}
static void* page_align(uintptr_t addr) {
uintptr_t mask = sysconf(_SC_PAGE_SIZE) - 1;
return reinterpret_cast<void*>((addr + mask) & ~mask);
}
TEST(unistd, brk) {
void* initial_break = get_brk();
// The kernel aligns the break to a page.
void* new_break = reinterpret_cast<void*>(reinterpret_cast<uintptr_t>(initial_break) + 1);
ASSERT_EQ(0, brk(new_break));
ASSERT_GE(get_brk(), new_break);
new_break = page_align(reinterpret_cast<uintptr_t>(initial_break) + sysconf(_SC_PAGE_SIZE));
ASSERT_EQ(0, brk(new_break));
ASSERT_EQ(get_brk(), new_break);
}
TEST(unistd, brk_ENOMEM) {
ASSERT_EQ(-1, brk(reinterpret_cast<void*>(-1)));
ASSERT_EQ(ENOMEM, errno);
}
#if defined(__GLIBC__)
#define SBRK_MIN INTPTR_MIN
#define SBRK_MAX INTPTR_MAX
#else
#define SBRK_MIN PTRDIFF_MIN
#define SBRK_MAX PTRDIFF_MAX
#endif
TEST(unistd, sbrk_ENOMEM) {
#if defined(__BIONIC__) && !defined(__LP64__)
// There is no way to guarantee that all overflow conditions can be tested
// without manipulating the underlying values of the current break.
extern void* __bionic_brk;
class ScopedBrk {
public:
ScopedBrk() : saved_brk_(__bionic_brk) {}
virtual ~ScopedBrk() { __bionic_brk = saved_brk_; }
private:
void* saved_brk_;
};
ScopedBrk scope_brk;
// Set the current break to a point that will cause an overflow.
__bionic_brk = reinterpret_cast<void*>(static_cast<uintptr_t>(PTRDIFF_MAX) + 2);
// Can't increase by so much that we'd overflow.
ASSERT_EQ(reinterpret_cast<void*>(-1), sbrk(PTRDIFF_MAX));
ASSERT_EQ(ENOMEM, errno);
// Set the current break to a point that will cause an overflow.
__bionic_brk = reinterpret_cast<void*>(static_cast<uintptr_t>(PTRDIFF_MAX));
ASSERT_EQ(reinterpret_cast<void*>(-1), sbrk(PTRDIFF_MIN));
ASSERT_EQ(ENOMEM, errno);
__bionic_brk = reinterpret_cast<void*>(static_cast<uintptr_t>(PTRDIFF_MAX) - 1);
ASSERT_EQ(reinterpret_cast<void*>(-1), sbrk(PTRDIFF_MIN + 1));
ASSERT_EQ(ENOMEM, errno);
#else
class ScopedBrk {
public:
ScopedBrk() : saved_brk_(get_brk()) {}
virtual ~ScopedBrk() { brk(saved_brk_); }
private:
void* saved_brk_;
};
ScopedBrk scope_brk;
uintptr_t cur_brk = reinterpret_cast<uintptr_t>(get_brk());
if (cur_brk < static_cast<uintptr_t>(-(SBRK_MIN+1))) {
// Do the overflow test for a max negative increment.
ASSERT_EQ(reinterpret_cast<void*>(-1), sbrk(SBRK_MIN));
#if defined(__BIONIC__)
// GLIBC does not set errno in overflow case.
ASSERT_EQ(ENOMEM, errno);
#endif
}
uintptr_t overflow_brk = static_cast<uintptr_t>(SBRK_MAX) + 2;
if (cur_brk < overflow_brk) {
// Try and move the value to PTRDIFF_MAX + 2.
cur_brk = reinterpret_cast<uintptr_t>(sbrk(overflow_brk));
}
if (cur_brk >= overflow_brk) {
ASSERT_EQ(reinterpret_cast<void*>(-1), sbrk(SBRK_MAX));
#if defined(__BIONIC__)
// GLIBC does not set errno in overflow case.
ASSERT_EQ(ENOMEM, errno);
#endif
}
#endif
}
TEST(unistd, truncate) {
TemporaryFile tf;
ASSERT_EQ(0, close(tf.fd));
ASSERT_EQ(0, truncate(tf.filename, 123));
struct stat sb;
ASSERT_EQ(0, stat(tf.filename, &sb));
ASSERT_EQ(123, sb.st_size);
}
TEST(unistd, truncate64) {
TemporaryFile tf;
ASSERT_EQ(0, close(tf.fd));
ASSERT_EQ(0, truncate64(tf.filename, 123));
struct stat sb;
ASSERT_EQ(0, stat(tf.filename, &sb));
ASSERT_EQ(123, sb.st_size);
}
TEST(unistd, ftruncate) {
TemporaryFile tf;
ASSERT_EQ(0, ftruncate(tf.fd, 123));
ASSERT_EQ(0, close(tf.fd));
struct stat sb;
ASSERT_EQ(0, stat(tf.filename, &sb));
ASSERT_EQ(123, sb.st_size);
}
TEST(unistd, ftruncate64) {
TemporaryFile tf;
ASSERT_EQ(0, ftruncate64(tf.fd, 123));
ASSERT_EQ(0, close(tf.fd));
struct stat sb;
ASSERT_EQ(0, stat(tf.filename, &sb));
ASSERT_EQ(123, sb.st_size);
}
static bool g_pause_test_flag = false;
static void PauseTestSignalHandler(int) {
g_pause_test_flag = true;
}
TEST(unistd, pause) {
ScopedSignalHandler handler(SIGALRM, PauseTestSignalHandler);
alarm(1);
ASSERT_FALSE(g_pause_test_flag);
ASSERT_EQ(-1, pause());
ASSERT_TRUE(g_pause_test_flag);
}
TEST(unistd, read) {
int fd = open("/proc/version", O_RDONLY);
ASSERT_TRUE(fd != -1);
char buf[5];
ASSERT_EQ(5, read(fd, buf, 5));
ASSERT_EQ(buf[0], 'L');
ASSERT_EQ(buf[1], 'i');
ASSERT_EQ(buf[2], 'n');
ASSERT_EQ(buf[3], 'u');
ASSERT_EQ(buf[4], 'x');
close(fd);
}
TEST(unistd, read_EBADF) {
// read returns ssize_t which is 64-bits on LP64, so it's worth explicitly checking that
// our syscall stubs correctly return a 64-bit -1.
char buf[1];
ASSERT_EQ(-1, read(-1, buf, sizeof(buf)));
ASSERT_EQ(EBADF, errno);
}
TEST(unistd, syscall_long) {
// Check that syscall(3) correctly returns long results.
// https://code.google.com/p/android/issues/detail?id=73952
// We assume that the break is > 4GiB, but this is potentially flaky.
uintptr_t p = reinterpret_cast<uintptr_t>(sbrk(0));
ASSERT_EQ(p, static_cast<uintptr_t>(syscall(__NR_brk, 0)));
}
TEST(unistd, alarm) {
ASSERT_EQ(0U, alarm(0));
}
TEST(unistd, _exit) {
int pid = fork();
ASSERT_NE(-1, pid) << strerror(errno);
if (pid == 0) {
_exit(99);
}
int status;
ASSERT_EQ(pid, waitpid(pid, &status, 0));
ASSERT_TRUE(WIFEXITED(status));
ASSERT_EQ(99, WEXITSTATUS(status));
}
TEST(unistd, getenv_unsetenv) {
ASSERT_EQ(0, setenv("test-variable", "hello", 1));
ASSERT_STREQ("hello", getenv("test-variable"));
ASSERT_EQ(0, unsetenv("test-variable"));
ASSERT_TRUE(getenv("test-variable") == NULL);
}
TEST(unistd, unsetenv_EINVAL) {
EXPECT_EQ(-1, unsetenv(""));
EXPECT_EQ(EINVAL, errno);
EXPECT_EQ(-1, unsetenv("a=b"));
EXPECT_EQ(EINVAL, errno);
}
TEST(unistd, setenv_EINVAL) {
EXPECT_EQ(-1, setenv(NULL, "value", 0));
EXPECT_EQ(EINVAL, errno);
EXPECT_EQ(-1, setenv(NULL, "value", 1));
EXPECT_EQ(EINVAL, errno);
EXPECT_EQ(-1, setenv("", "value", 0));
EXPECT_EQ(EINVAL, errno);
EXPECT_EQ(-1, setenv("", "value", 1));
EXPECT_EQ(EINVAL, errno);
EXPECT_EQ(-1, setenv("a=b", "value", 0));
EXPECT_EQ(EINVAL, errno);
EXPECT_EQ(-1, setenv("a=b", "value", 1));
EXPECT_EQ(EINVAL, errno);
}
TEST(unistd, setenv) {
ASSERT_EQ(0, unsetenv("test-variable"));
char a[] = "a";
char b[] = "b";
char c[] = "c";
// New value.
EXPECT_EQ(0, setenv("test-variable", a, 0));
EXPECT_STREQ(a, getenv("test-variable"));
// Existing value, no overwrite.
EXPECT_EQ(0, setenv("test-variable", b, 0));
EXPECT_STREQ(a, getenv("test-variable"));
// Existing value, overwrite.
EXPECT_EQ(0, setenv("test-variable", c, 1));
EXPECT_STREQ(c, getenv("test-variable"));
// But the arrays backing the values are unchanged.
EXPECT_EQ('a', a[0]);
EXPECT_EQ('b', b[0]);
EXPECT_EQ('c', c[0]);
ASSERT_EQ(0, unsetenv("test-variable"));
}
TEST(unistd, putenv) {
ASSERT_EQ(0, unsetenv("a"));
char* s1 = strdup("a=b");
ASSERT_EQ(0, putenv(s1));
ASSERT_STREQ("b", getenv("a"));
s1[2] = 'c';
ASSERT_STREQ("c", getenv("a"));
char* s2 = strdup("a=b");
ASSERT_EQ(0, putenv(s2));
ASSERT_STREQ("b", getenv("a"));
ASSERT_EQ('c', s1[2]);
ASSERT_EQ(0, unsetenv("a"));
free(s1);
free(s2);
}
TEST(unistd, clearenv) {
extern char** environ;
// Guarantee that environ is not initially empty...
ASSERT_EQ(0, setenv("test-variable", "a", 1));
// Stash a copy.
std::vector<char*> old_environ;
for (size_t i = 0; environ[i] != NULL; ++i) {
old_environ.push_back(strdup(environ[i]));
}
ASSERT_EQ(0, clearenv());
EXPECT_TRUE(environ == NULL || environ[0] == NULL);
EXPECT_EQ(NULL, getenv("test-variable"));
EXPECT_EQ(0, setenv("test-variable", "post-clear", 1));
EXPECT_STREQ("post-clear", getenv("test-variable"));
// Put the old environment back.
for (size_t i = 0; i < old_environ.size(); ++i) {
EXPECT_EQ(0, putenv(old_environ[i]));
}
// Check it wasn't overwritten.
EXPECT_STREQ("a", getenv("test-variable"));
EXPECT_EQ(0, unsetenv("test-variable"));
}
static void TestFsyncFunction(int (*fn)(int)) {
int fd;
// Can't sync an invalid fd.
errno = 0;
EXPECT_EQ(-1, fn(-1));
EXPECT_EQ(EBADF, errno);
// It doesn't matter whether you've opened a file for write or not.
TemporaryFile tf;
ASSERT_NE(-1, tf.fd);
EXPECT_EQ(0, fn(tf.fd));
ASSERT_NE(-1, fd = open(tf.filename, O_RDONLY));
EXPECT_EQ(0, fn(fd));
close(fd);
ASSERT_NE(-1, fd = open(tf.filename, O_RDWR));
EXPECT_EQ(0, fn(fd));
close(fd);
// The fd can even be a directory.
ASSERT_NE(-1, fd = open("/", O_RDONLY));
EXPECT_EQ(0, fn(fd));
close(fd);
// But some file systems may choose to be fussy...
errno = 0;
ASSERT_NE(-1, fd = open("/proc/version", O_RDONLY));
EXPECT_EQ(-1, fn(fd));
EXPECT_EQ(EINVAL, errno);
close(fd);
}
TEST(unistd, fdatasync) {
TestFsyncFunction(fdatasync);
}
TEST(unistd, fsync) {
TestFsyncFunction(fsync);
}
static void AssertGetPidCorrect() {
// The loop is just to make manual testing/debugging with strace easier.
pid_t getpid_syscall_result = syscall(__NR_getpid);
for (size_t i = 0; i < 128; ++i) {
ASSERT_EQ(getpid_syscall_result, getpid());
}
}
TEST(unistd, getpid_caching_and_fork) {
pid_t parent_pid = getpid();
ASSERT_EQ(syscall(__NR_getpid), parent_pid);
pid_t fork_result = fork();
ASSERT_NE(fork_result, -1);
if (fork_result == 0) {
// We're the child.
AssertGetPidCorrect();
ASSERT_EQ(parent_pid, getppid());
_exit(123);
} else {
// We're the parent.
ASSERT_EQ(parent_pid, getpid());
int status;
ASSERT_EQ(fork_result, waitpid(fork_result, &status, 0));
ASSERT_TRUE(WIFEXITED(status));
ASSERT_EQ(123, WEXITSTATUS(status));
}
}
static int GetPidCachingCloneStartRoutine(void*) {
AssertGetPidCorrect();
return 123;
}
TEST(unistd, getpid_caching_and_clone) {
pid_t parent_pid = getpid();
ASSERT_EQ(syscall(__NR_getpid), parent_pid);
void* child_stack[1024];
int clone_result = clone(GetPidCachingCloneStartRoutine, &child_stack[1024], CLONE_NEWNS | SIGCHLD, NULL);
if (clone_result == -1 && errno == EPERM && getuid() != 0) {
GTEST_LOG_(INFO) << "This test only works if you have permission to CLONE_NEWNS; try running as root.\n";
return;
}
ASSERT_NE(clone_result, -1);
ASSERT_EQ(parent_pid, getpid());
int status;
ASSERT_EQ(clone_result, waitpid(clone_result, &status, 0));
ASSERT_TRUE(WIFEXITED(status));
ASSERT_EQ(123, WEXITSTATUS(status));
}
static void* GetPidCachingPthreadStartRoutine(void*) {
AssertGetPidCorrect();
return NULL;
}
TEST(unistd, getpid_caching_and_pthread_create) {
pid_t parent_pid = getpid();
pthread_t t;
ASSERT_EQ(0, pthread_create(&t, NULL, GetPidCachingPthreadStartRoutine, NULL));
ASSERT_EQ(parent_pid, getpid());
void* result;
ASSERT_EQ(0, pthread_join(t, &result));
ASSERT_EQ(NULL, result);
}
class unistd_DeathTest : public BionicDeathTest {};
TEST_F(unistd_DeathTest, abort) {
ASSERT_EXIT(abort(), testing::KilledBySignal(SIGABRT), "");
}
TEST(unistd, sethostname) {
// The permissions check happens before the argument check, so this will
// fail for a different reason if you're running as root than if you're
// not, but it'll fail either way. Checking that we have the symbol is about
// all we can do for sethostname(2).
ASSERT_EQ(-1, sethostname("", -1));
}
TEST(unistd, gethostname) {
char hostname[HOST_NAME_MAX + 1];
memset(hostname, 0, sizeof(hostname));
// Can we get the hostname with a big buffer?
ASSERT_EQ(0, gethostname(hostname, HOST_NAME_MAX));
// Can we get the hostname with a right-sized buffer?
errno = 0;
ASSERT_EQ(0, gethostname(hostname, strlen(hostname) + 1));
// Does uname(2) agree?
utsname buf;
ASSERT_EQ(0, uname(&buf));
ASSERT_EQ(0, strncmp(hostname, buf.nodename, SYS_NMLN));
ASSERT_GT(strlen(hostname), 0U);
// Do we correctly detect truncation?
errno = 0;
ASSERT_EQ(-1, gethostname(hostname, strlen(hostname)));
ASSERT_EQ(ENAMETOOLONG, errno);
}
TEST(unistd, pathconf_fpathconf) {
TemporaryFile tf;
long rc = 0L;
// As a file system's block size is always power of 2, the configure values
// for ALLOC and XFER should be power of 2 as well.
rc = pathconf(tf.filename, _PC_ALLOC_SIZE_MIN);
ASSERT_TRUE(rc > 0 && powerof2(rc));
rc = pathconf(tf.filename, _PC_REC_MIN_XFER_SIZE);
ASSERT_TRUE(rc > 0 && powerof2(rc));
rc = pathconf(tf.filename, _PC_REC_XFER_ALIGN);
ASSERT_TRUE(rc > 0 && powerof2(rc));
rc = fpathconf(tf.fd, _PC_ALLOC_SIZE_MIN);
ASSERT_TRUE(rc > 0 && powerof2(rc));
rc = fpathconf(tf.fd, _PC_REC_MIN_XFER_SIZE);
ASSERT_TRUE(rc > 0 && powerof2(rc));
rc = fpathconf(tf.fd, _PC_REC_XFER_ALIGN);
ASSERT_TRUE(rc > 0 && powerof2(rc));
}
#define verifySysconf(name, ret) \
{\
errno = 0;\
ret = sysconf(name);\
ASSERT_TRUE((0 == errno) && (-1 != ret)) << "name=" << #name << ", ret=" << ret << ", Error Message: " << strerror(errno);\
}\
TEST(unistd, sysconf) {
long ret;
verifySysconf(_SC_MONOTONIC_CLOCK, ret);
ASSERT_GT(ret, 0);
verifySysconf(_SC_ARG_MAX, ret);
ASSERT_GT(ret, 0);
verifySysconf(_SC_CHILD_MAX, ret);
ASSERT_GT(ret, 0);
verifySysconf(_SC_CLK_TCK, ret);
ASSERT_GT(ret, 0);
verifySysconf(_SC_LINE_MAX, ret);
ASSERT_GT(ret, 1);
verifySysconf(_SC_NGROUPS_MAX, ret);
ASSERT_GT(ret, 0);
verifySysconf(_SC_OPEN_MAX, ret);
ASSERT_GT(ret, 1);
verifySysconf(_SC_2_C_BIND, ret);
ASSERT_GT(ret, 0);
verifySysconf(_SC_2_C_VERSION, ret);
verifySysconf(_SC_JOB_CONTROL, ret);
ASSERT_EQ(1, ret);
verifySysconf(_SC_SAVED_IDS, ret);
ASSERT_EQ(1, ret);
verifySysconf(_SC_VERSION, ret);
verifySysconf(_SC_RE_DUP_MAX, ret);
verifySysconf(_SC_STREAM_MAX, ret);
ASSERT_GT(ret, 0);
verifySysconf(_SC_TZNAME_MAX, ret);
verifySysconf(_SC_XOPEN_VERSION, ret);
verifySysconf(_SC_ATEXIT_MAX, ret);
ASSERT_GT(ret, 1);
verifySysconf(_SC_IOV_MAX, ret);
ASSERT_GT(ret, 0);
verifySysconf(_SC_PAGESIZE, ret);
ASSERT_GE(ret, 1);
verifySysconf(_SC_PAGE_SIZE, ret);
ASSERT_GE(ret, 1);
verifySysconf(_SC_XOPEN_UNIX, ret);
verifySysconf(_SC_DELAYTIMER_MAX, ret);
ASSERT_GT(ret, 0);
verifySysconf(_SC_MQ_OPEN_MAX, ret);
verifySysconf(_SC_MQ_PRIO_MAX, ret);
ASSERT_GT(ret ,0);
verifySysconf(_SC_RTSIG_MAX, ret);
verifySysconf(_SC_SEM_NSEMS_MAX, ret);
verifySysconf(_SC_SEM_VALUE_MAX, ret);
verifySysconf(_SC_SIGQUEUE_MAX, ret);
ASSERT_GT(ret, 0);
verifySysconf(_SC_TIMER_MAX, ret);
verifySysconf(_SC_FSYNC, ret);
verifySysconf(_SC_MAPPED_FILES, ret);
verifySysconf(_SC_PRIORITY_SCHEDULING, ret);
verifySysconf(_SC_SEMAPHORES, ret);
verifySysconf(_SC_SYNCHRONIZED_IO, ret);
verifySysconf(_SC_TIMERS, ret);
verifySysconf(_SC_GETGR_R_SIZE_MAX, ret);
verifySysconf(_SC_GETPW_R_SIZE_MAX, ret);
verifySysconf(_SC_LOGIN_NAME_MAX, ret);
verifySysconf(_SC_THREAD_DESTRUCTOR_ITERATIONS, ret);
verifySysconf(_SC_THREAD_KEYS_MAX, ret);
verifySysconf(_SC_THREAD_STACK_MIN, ret);
verifySysconf(_SC_THREAD_THREADS_MAX, ret);
verifySysconf(_SC_TTY_NAME_MAX, ret);
ASSERT_GT(ret, 0);
verifySysconf(_SC_THREADS, ret);
verifySysconf(_SC_THREAD_PRIO_INHERIT, ret);
verifySysconf(_SC_THREAD_PRIO_PROTECT, ret);
verifySysconf(_SC_NPROCESSORS_CONF, ret);
verifySysconf(_SC_NPROCESSORS_ONLN, ret);
verifySysconf(_SC_PHYS_PAGES, ret);
verifySysconf(_SC_AVPHYS_PAGES, ret);
// TODO: keep following names check here
// so that we can enable them when the error fixed
/* when call sysconf with following name,
the system call will report error "Function not implemented"
verifySysconf(_SC_BC_BASE_MAX, ret);
verifySysconf(_SC_BC_DIM_MAX, ret);
verifySysconf(_SC_BC_SCALE_MAX, ret);
verifySysconf(_SC_BC_STRING_MAX, ret);
verifySysconf(_SC_COLL_WEIGHTS_MAX, ret);
verifySysconf(_SC_EXPR_NEST_MAX, ret);
verifySysconf(_SC_XOPEN_SHM, ret);
verifySysconf(_SC_XBS5_ILP32_OFF32, ret);
verifySysconf(_SC_XBS5_ILP32_OFFBIG, ret);
verifySysconf(_SC_XBS5_LP64_OFF64, ret);
verifySysconf(_SC_XBS5_LPBIG_OFFBIG, ret);
verifySysconf(_SC_AIO_LISTIO_MAX, ret);
verifySysconf(_SC_AIO_MAX, ret);
verifySysconf(_SC_AIO_PRIO_DELTA_MAX, ret);
verifySysconf(_SC_ASYNCHRONOUS_IO, ret);
verifySysconf(_SC_MEMLOCK, ret);
verifySysconf(_SC_MEMLOCK_RANGE, ret);
verifySysconf(_SC_MEMORY_PROTECTION, ret);
verifySysconf(_SC_MESSAGE_PASSING, ret);
verifySysconf(_SC_PRIORITIZED_IO, ret);
verifySysconf(_SC_SHARED_MEMORY_OBJECTS, ret);
verifySysconf(_SC_THREAD_SAFE_FUNCTIONS, ret);
*/
/* when following names are checked,
the return value of sysconf is -1,
which means following name are invalid.
verifySysconf(_SC_2_C_DEV, ret);
verifySysconf(_SC_2_FORT_DEV, ret);
verifySysconf(_SC_2_FORT_RUN, ret);
verifySysconf(_SC_2_LOCALEDEF, ret);
verifySysconf(_SC_2_SW_DEV, ret);
verifySysconf(_SC_2_UPE, ret);
verifySysconf(_SC_2_VERSION, ret);
verifySysconf(_SC_XOPEN_CRYPT, ret);
verifySysconf(_SC_XOPEN_ENH_I18N, ret);
verifySysconf(_SC_XOPEN_XCU_VERSION, ret);
verifySysconf(_SC_XOPEN_REALTIME, ret);
verifySysconf(_SC_XOPEN_REALTIME_THREADS, ret);
verifySysconf(_SC_XOPEN_LEGACY, ret);
verifySysconf(_SC_THREAD_ATTR_STACKADDR, ret);
verifySysconf(_SC_THREAD_ATTR_STACKSIZE, ret);
verifySysconf(_SC_REALTIME_SIGNALS, ret);
*/
}