glibc/support/support_test_main.c

517 lines
15 KiB
C
Raw Normal View History

/* Main worker function for the test driver.
Copyright (C) 1998-2023 Free Software Foundation, Inc.
This file is part of the GNU C Library.
The GNU C Library is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public
License as published by the Free Software Foundation; either
version 2.1 of the License, or (at your option) any later version.
The GNU C Library is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
License along with the GNU C Library; if not, see
Prefer https to http for gnu.org and fsf.org URLs Also, change sources.redhat.com to sourceware.org. This patch was automatically generated by running the following shell script, which uses GNU sed, and which avoids modifying files imported from upstream: sed -ri ' s,(http|ftp)(://(.*\.)?(gnu|fsf|sourceware)\.org($|[^.]|\.[^a-z])),https\2,g s,(http|ftp)(://(.*\.)?)sources\.redhat\.com($|[^.]|\.[^a-z]),https\2sourceware.org\4,g ' \ $(find $(git ls-files) -prune -type f \ ! -name '*.po' \ ! -name 'ChangeLog*' \ ! -path COPYING ! -path COPYING.LIB \ ! -path manual/fdl-1.3.texi ! -path manual/lgpl-2.1.texi \ ! -path manual/texinfo.tex ! -path scripts/config.guess \ ! -path scripts/config.sub ! -path scripts/install-sh \ ! -path scripts/mkinstalldirs ! -path scripts/move-if-change \ ! -path INSTALL ! -path locale/programs/charmap-kw.h \ ! -path po/libc.pot ! -path sysdeps/gnu/errlist.c \ ! '(' -name configure \ -execdir test -f configure.ac -o -f configure.in ';' ')' \ ! '(' -name preconfigure \ -execdir test -f preconfigure.ac ';' ')' \ -print) and then by running 'make dist-prepare' to regenerate files built from the altered files, and then executing the following to cleanup: chmod a+x sysdeps/unix/sysv/linux/riscv/configure # Omit irrelevant whitespace and comment-only changes, # perhaps from a slightly-different Autoconf version. git checkout -f \ sysdeps/csky/configure \ sysdeps/hppa/configure \ sysdeps/riscv/configure \ sysdeps/unix/sysv/linux/csky/configure # Omit changes that caused a pre-commit check to fail like this: # remote: *** error: sysdeps/powerpc/powerpc64/ppc-mcount.S: trailing lines git checkout -f \ sysdeps/powerpc/powerpc64/ppc-mcount.S \ sysdeps/unix/sysv/linux/s390/s390-64/syscall.S # Omit change that caused a pre-commit check to fail like this: # remote: *** error: sysdeps/sparc/sparc64/multiarch/memcpy-ultra3.S: last line does not end in newline git checkout -f sysdeps/sparc/sparc64/multiarch/memcpy-ultra3.S
2019-09-07 05:40:42 +00:00
<https://www.gnu.org/licenses/>. */
#include <support/test-driver.h>
#include <support/check.h>
#include <support/temp_file-internal.h>
#include <support/support.h>
#include <assert.h>
#include <errno.h>
#include <getopt.h>
#include <malloc.h>
#include <signal.h>
#include <stdbool.h>
#include <stdlib.h>
#include <string.h>
#include <sys/param.h>
#include <sys/resource.h>
#include <sys/time.h>
#include <sys/types.h>
#include <sys/wait.h>
#include <time.h>
#include <unistd.h>
#include <xstdio.h>
static const struct option default_options[] =
{
TEST_DEFAULT_OPTIONS
{ NULL, 0, NULL, 0 }
};
/* Show people how to run the program. */
static void
usage (const struct option *options)
{
size_t i;
printf ("Usage: %s [options]\n"
"\n"
"Environment Variables:\n"
" TIMEOUTFACTOR An integer used to scale the timeout\n"
" TMPDIR Where to place temporary files\n"
" TEST_COREDUMPS Do not disable coredumps if set\n"
"\n",
program_invocation_short_name);
printf ("Options:\n");
for (i = 0; options[i].name; ++i)
{
int indent;
indent = printf (" --%s", options[i].name);
if (options[i].has_arg == required_argument)
indent += printf (" <arg>");
printf ("%*s", 25 - indent, "");
switch (options[i].val)
{
case 'v':
printf ("Increase the output verbosity");
break;
case OPT_DIRECT:
printf ("Run the test directly (instead of forking & monitoring)");
break;
case OPT_TESTDIR:
printf ("Override the TMPDIR env var");
break;
}
printf ("\n");
}
}
/* The PID of the test process. */
static pid_t test_pid;
/* The cleanup handler passed to test_main. */
static void (*cleanup_function) (void);
static void
Change most internal uses of __gettimeofday to __clock_gettime. Since gettimeofday will shortly be implemented in terms of clock_gettime on all platforms, internal code should use clock_gettime directly; in addition to removing a layer of indirection, this will allow us to remove the PLT-bypass gunk for gettimeofday. (We can't quite do that yet, but it'll be coming later in this patch series.) In many cases, the changed code does fewer conversions. The changed code always assumes __clock_gettime (CLOCK_REALTIME) cannot fail. Most of the call sites were assuming gettimeofday could not fail, but a few places were checking for errors. POSIX says clock_gettime can only fail if the clock constant is invalid or unsupported, and CLOCK_REALTIME is the one and only clock constant that's required to be supported. For consistency I grepped the entire source tree for any other places that checked for errors from __clock_gettime (CLOCK_REALTIME), found one, and changed it too. (For the record, POSIX also says gettimeofday can never fail.) (It would be nice if we could declare that GNU systems will always support CLOCK_MONOTONIC as well as CLOCK_REALTIME; there are several places where we are using CLOCK_REALTIME where _MONOTONIC would be more appropriate, and/or trying to use _MONOTONIC and then falling back to _REALTIME. But the Hurd doesn't support CLOCK_MONOTONIC yet, and it looks like adding it would involve substantial changes to gnumach's internals and API. Oh well.) A few Hurd-specific files were changed to use __host_get_time instead of __clock_gettime, as this seemed tidier. We also assume this cannot fail. Skimming the code in gnumach leads me to believe the only way it could fail is if __mach_host_self also failed, and our Hurd-specific code consistently assumes that can't happen, so I'm going with that. With the exception of support/support_test_main.c, test cases are not modified, mainly because I didn't want to have to figure out which test cases were testing gettimeofday specifically. The definition of GETTIME in sysdeps/generic/memusage.h had a typo and was not reading tv_sec at all. I fixed this. It appears nobody has been generating malloc traces on a machine that doesn't have a superseding definition. There are a whole bunch of places where the code could be simplified by factoring out timespec subtraction and/or comparison logic, but I want to keep this patch as mechanical as possible. Checked on x86_64-linux-gnu, i686-linux-gnu, powerpc64le-linux-gnu, powerpc64-linux-gnu, powerpc-linux-gnu, and aarch64-linux-gnu. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org> Reviewed-by: Lukasz Majewski <lukma@denx.de>
2019-08-17 00:38:22 +00:00
print_timestamp (const char *what, struct timespec tv)
{
struct tm tm;
Change most internal uses of __gettimeofday to __clock_gettime. Since gettimeofday will shortly be implemented in terms of clock_gettime on all platforms, internal code should use clock_gettime directly; in addition to removing a layer of indirection, this will allow us to remove the PLT-bypass gunk for gettimeofday. (We can't quite do that yet, but it'll be coming later in this patch series.) In many cases, the changed code does fewer conversions. The changed code always assumes __clock_gettime (CLOCK_REALTIME) cannot fail. Most of the call sites were assuming gettimeofday could not fail, but a few places were checking for errors. POSIX says clock_gettime can only fail if the clock constant is invalid or unsupported, and CLOCK_REALTIME is the one and only clock constant that's required to be supported. For consistency I grepped the entire source tree for any other places that checked for errors from __clock_gettime (CLOCK_REALTIME), found one, and changed it too. (For the record, POSIX also says gettimeofday can never fail.) (It would be nice if we could declare that GNU systems will always support CLOCK_MONOTONIC as well as CLOCK_REALTIME; there are several places where we are using CLOCK_REALTIME where _MONOTONIC would be more appropriate, and/or trying to use _MONOTONIC and then falling back to _REALTIME. But the Hurd doesn't support CLOCK_MONOTONIC yet, and it looks like adding it would involve substantial changes to gnumach's internals and API. Oh well.) A few Hurd-specific files were changed to use __host_get_time instead of __clock_gettime, as this seemed tidier. We also assume this cannot fail. Skimming the code in gnumach leads me to believe the only way it could fail is if __mach_host_self also failed, and our Hurd-specific code consistently assumes that can't happen, so I'm going with that. With the exception of support/support_test_main.c, test cases are not modified, mainly because I didn't want to have to figure out which test cases were testing gettimeofday specifically. The definition of GETTIME in sysdeps/generic/memusage.h had a typo and was not reading tv_sec at all. I fixed this. It appears nobody has been generating malloc traces on a machine that doesn't have a superseding definition. There are a whole bunch of places where the code could be simplified by factoring out timespec subtraction and/or comparison logic, but I want to keep this patch as mechanical as possible. Checked on x86_64-linux-gnu, i686-linux-gnu, powerpc64le-linux-gnu, powerpc64-linux-gnu, powerpc-linux-gnu, and aarch64-linux-gnu. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org> Reviewed-by: Lukasz Majewski <lukma@denx.de>
2019-08-17 00:38:22 +00:00
/* Casts of tv.tv_nsec below are necessary because the type of
tv_nsec is not literally long int on all supported platforms. */
if (gmtime_r (&tv.tv_sec, &tm) == NULL)
Change most internal uses of __gettimeofday to __clock_gettime. Since gettimeofday will shortly be implemented in terms of clock_gettime on all platforms, internal code should use clock_gettime directly; in addition to removing a layer of indirection, this will allow us to remove the PLT-bypass gunk for gettimeofday. (We can't quite do that yet, but it'll be coming later in this patch series.) In many cases, the changed code does fewer conversions. The changed code always assumes __clock_gettime (CLOCK_REALTIME) cannot fail. Most of the call sites were assuming gettimeofday could not fail, but a few places were checking for errors. POSIX says clock_gettime can only fail if the clock constant is invalid or unsupported, and CLOCK_REALTIME is the one and only clock constant that's required to be supported. For consistency I grepped the entire source tree for any other places that checked for errors from __clock_gettime (CLOCK_REALTIME), found one, and changed it too. (For the record, POSIX also says gettimeofday can never fail.) (It would be nice if we could declare that GNU systems will always support CLOCK_MONOTONIC as well as CLOCK_REALTIME; there are several places where we are using CLOCK_REALTIME where _MONOTONIC would be more appropriate, and/or trying to use _MONOTONIC and then falling back to _REALTIME. But the Hurd doesn't support CLOCK_MONOTONIC yet, and it looks like adding it would involve substantial changes to gnumach's internals and API. Oh well.) A few Hurd-specific files were changed to use __host_get_time instead of __clock_gettime, as this seemed tidier. We also assume this cannot fail. Skimming the code in gnumach leads me to believe the only way it could fail is if __mach_host_self also failed, and our Hurd-specific code consistently assumes that can't happen, so I'm going with that. With the exception of support/support_test_main.c, test cases are not modified, mainly because I didn't want to have to figure out which test cases were testing gettimeofday specifically. The definition of GETTIME in sysdeps/generic/memusage.h had a typo and was not reading tv_sec at all. I fixed this. It appears nobody has been generating malloc traces on a machine that doesn't have a superseding definition. There are a whole bunch of places where the code could be simplified by factoring out timespec subtraction and/or comparison logic, but I want to keep this patch as mechanical as possible. Checked on x86_64-linux-gnu, i686-linux-gnu, powerpc64le-linux-gnu, powerpc64-linux-gnu, powerpc-linux-gnu, and aarch64-linux-gnu. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org> Reviewed-by: Lukasz Majewski <lukma@denx.de>
2019-08-17 00:38:22 +00:00
printf ("%s: %lld.%09ld\n",
what, (long long int) tv.tv_sec, (long int) tv.tv_nsec);
else
Change most internal uses of __gettimeofday to __clock_gettime. Since gettimeofday will shortly be implemented in terms of clock_gettime on all platforms, internal code should use clock_gettime directly; in addition to removing a layer of indirection, this will allow us to remove the PLT-bypass gunk for gettimeofday. (We can't quite do that yet, but it'll be coming later in this patch series.) In many cases, the changed code does fewer conversions. The changed code always assumes __clock_gettime (CLOCK_REALTIME) cannot fail. Most of the call sites were assuming gettimeofday could not fail, but a few places were checking for errors. POSIX says clock_gettime can only fail if the clock constant is invalid or unsupported, and CLOCK_REALTIME is the one and only clock constant that's required to be supported. For consistency I grepped the entire source tree for any other places that checked for errors from __clock_gettime (CLOCK_REALTIME), found one, and changed it too. (For the record, POSIX also says gettimeofday can never fail.) (It would be nice if we could declare that GNU systems will always support CLOCK_MONOTONIC as well as CLOCK_REALTIME; there are several places where we are using CLOCK_REALTIME where _MONOTONIC would be more appropriate, and/or trying to use _MONOTONIC and then falling back to _REALTIME. But the Hurd doesn't support CLOCK_MONOTONIC yet, and it looks like adding it would involve substantial changes to gnumach's internals and API. Oh well.) A few Hurd-specific files were changed to use __host_get_time instead of __clock_gettime, as this seemed tidier. We also assume this cannot fail. Skimming the code in gnumach leads me to believe the only way it could fail is if __mach_host_self also failed, and our Hurd-specific code consistently assumes that can't happen, so I'm going with that. With the exception of support/support_test_main.c, test cases are not modified, mainly because I didn't want to have to figure out which test cases were testing gettimeofday specifically. The definition of GETTIME in sysdeps/generic/memusage.h had a typo and was not reading tv_sec at all. I fixed this. It appears nobody has been generating malloc traces on a machine that doesn't have a superseding definition. There are a whole bunch of places where the code could be simplified by factoring out timespec subtraction and/or comparison logic, but I want to keep this patch as mechanical as possible. Checked on x86_64-linux-gnu, i686-linux-gnu, powerpc64le-linux-gnu, powerpc64-linux-gnu, powerpc-linux-gnu, and aarch64-linux-gnu. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org> Reviewed-by: Lukasz Majewski <lukma@denx.de>
2019-08-17 00:38:22 +00:00
printf ("%s: %04d-%02d-%02dT%02d:%02d:%02d.%09ld\n",
what, 1900 + tm.tm_year, tm.tm_mon + 1, tm.tm_mday,
Change most internal uses of __gettimeofday to __clock_gettime. Since gettimeofday will shortly be implemented in terms of clock_gettime on all platforms, internal code should use clock_gettime directly; in addition to removing a layer of indirection, this will allow us to remove the PLT-bypass gunk for gettimeofday. (We can't quite do that yet, but it'll be coming later in this patch series.) In many cases, the changed code does fewer conversions. The changed code always assumes __clock_gettime (CLOCK_REALTIME) cannot fail. Most of the call sites were assuming gettimeofday could not fail, but a few places were checking for errors. POSIX says clock_gettime can only fail if the clock constant is invalid or unsupported, and CLOCK_REALTIME is the one and only clock constant that's required to be supported. For consistency I grepped the entire source tree for any other places that checked for errors from __clock_gettime (CLOCK_REALTIME), found one, and changed it too. (For the record, POSIX also says gettimeofday can never fail.) (It would be nice if we could declare that GNU systems will always support CLOCK_MONOTONIC as well as CLOCK_REALTIME; there are several places where we are using CLOCK_REALTIME where _MONOTONIC would be more appropriate, and/or trying to use _MONOTONIC and then falling back to _REALTIME. But the Hurd doesn't support CLOCK_MONOTONIC yet, and it looks like adding it would involve substantial changes to gnumach's internals and API. Oh well.) A few Hurd-specific files were changed to use __host_get_time instead of __clock_gettime, as this seemed tidier. We also assume this cannot fail. Skimming the code in gnumach leads me to believe the only way it could fail is if __mach_host_self also failed, and our Hurd-specific code consistently assumes that can't happen, so I'm going with that. With the exception of support/support_test_main.c, test cases are not modified, mainly because I didn't want to have to figure out which test cases were testing gettimeofday specifically. The definition of GETTIME in sysdeps/generic/memusage.h had a typo and was not reading tv_sec at all. I fixed this. It appears nobody has been generating malloc traces on a machine that doesn't have a superseding definition. There are a whole bunch of places where the code could be simplified by factoring out timespec subtraction and/or comparison logic, but I want to keep this patch as mechanical as possible. Checked on x86_64-linux-gnu, i686-linux-gnu, powerpc64le-linux-gnu, powerpc64-linux-gnu, powerpc-linux-gnu, and aarch64-linux-gnu. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org> Reviewed-by: Lukasz Majewski <lukma@denx.de>
2019-08-17 00:38:22 +00:00
tm.tm_hour, tm.tm_min, tm.tm_sec, (long int) tv.tv_nsec);
}
/* Timeout handler. We kill the child and exit with an error. */
static void
__attribute__ ((noreturn))
signal_handler (int sig)
{
int killed;
int status;
/* Do this first to avoid further interference from the
subprocess. */
Change most internal uses of __gettimeofday to __clock_gettime. Since gettimeofday will shortly be implemented in terms of clock_gettime on all platforms, internal code should use clock_gettime directly; in addition to removing a layer of indirection, this will allow us to remove the PLT-bypass gunk for gettimeofday. (We can't quite do that yet, but it'll be coming later in this patch series.) In many cases, the changed code does fewer conversions. The changed code always assumes __clock_gettime (CLOCK_REALTIME) cannot fail. Most of the call sites were assuming gettimeofday could not fail, but a few places were checking for errors. POSIX says clock_gettime can only fail if the clock constant is invalid or unsupported, and CLOCK_REALTIME is the one and only clock constant that's required to be supported. For consistency I grepped the entire source tree for any other places that checked for errors from __clock_gettime (CLOCK_REALTIME), found one, and changed it too. (For the record, POSIX also says gettimeofday can never fail.) (It would be nice if we could declare that GNU systems will always support CLOCK_MONOTONIC as well as CLOCK_REALTIME; there are several places where we are using CLOCK_REALTIME where _MONOTONIC would be more appropriate, and/or trying to use _MONOTONIC and then falling back to _REALTIME. But the Hurd doesn't support CLOCK_MONOTONIC yet, and it looks like adding it would involve substantial changes to gnumach's internals and API. Oh well.) A few Hurd-specific files were changed to use __host_get_time instead of __clock_gettime, as this seemed tidier. We also assume this cannot fail. Skimming the code in gnumach leads me to believe the only way it could fail is if __mach_host_self also failed, and our Hurd-specific code consistently assumes that can't happen, so I'm going with that. With the exception of support/support_test_main.c, test cases are not modified, mainly because I didn't want to have to figure out which test cases were testing gettimeofday specifically. The definition of GETTIME in sysdeps/generic/memusage.h had a typo and was not reading tv_sec at all. I fixed this. It appears nobody has been generating malloc traces on a machine that doesn't have a superseding definition. There are a whole bunch of places where the code could be simplified by factoring out timespec subtraction and/or comparison logic, but I want to keep this patch as mechanical as possible. Checked on x86_64-linux-gnu, i686-linux-gnu, powerpc64le-linux-gnu, powerpc64-linux-gnu, powerpc-linux-gnu, and aarch64-linux-gnu. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org> Reviewed-by: Lukasz Majewski <lukma@denx.de>
2019-08-17 00:38:22 +00:00
struct timespec now;
clock_gettime (CLOCK_REALTIME, &now);
struct stat64 st;
bool st_available = fstat64 (STDOUT_FILENO, &st) == 0 && st.st_mtime != 0;
assert (test_pid > 1);
/* Kill the whole process group. */
kill (-test_pid, SIGKILL);
/* In case setpgid failed in the child, kill it individually too. */
kill (test_pid, SIGKILL);
/* Wait for it to terminate. */
int i;
for (i = 0; i < 5; ++i)
{
killed = waitpid (test_pid, &status, WNOHANG|WUNTRACED);
if (killed != 0)
break;
/* Delay, give the system time to process the kill. If the
nanosleep() call return prematurely, all the better. We
won't restart it since this probably means the child process
finally died. */
struct timespec ts;
ts.tv_sec = 0;
ts.tv_nsec = 100000000;
nanosleep (&ts, NULL);
}
if (killed != 0 && killed != test_pid)
{
printf ("Failed to kill test process: %m\n");
exit (1);
}
if (cleanup_function != NULL)
cleanup_function ();
if (sig == SIGINT)
{
signal (sig, SIG_DFL);
raise (sig);
}
if (killed == 0 || (WIFSIGNALED (status) && WTERMSIG (status) == SIGKILL))
puts ("Timed out: killed the child process");
else if (WIFSTOPPED (status))
printf ("Timed out: the child process was %s\n",
strsignal (WSTOPSIG (status)));
else if (WIFSIGNALED (status))
printf ("Timed out: the child process got signal %s\n",
strsignal (WTERMSIG (status)));
else
printf ("Timed out: killed the child process but it exited %d\n",
WEXITSTATUS (status));
Change most internal uses of __gettimeofday to __clock_gettime. Since gettimeofday will shortly be implemented in terms of clock_gettime on all platforms, internal code should use clock_gettime directly; in addition to removing a layer of indirection, this will allow us to remove the PLT-bypass gunk for gettimeofday. (We can't quite do that yet, but it'll be coming later in this patch series.) In many cases, the changed code does fewer conversions. The changed code always assumes __clock_gettime (CLOCK_REALTIME) cannot fail. Most of the call sites were assuming gettimeofday could not fail, but a few places were checking for errors. POSIX says clock_gettime can only fail if the clock constant is invalid or unsupported, and CLOCK_REALTIME is the one and only clock constant that's required to be supported. For consistency I grepped the entire source tree for any other places that checked for errors from __clock_gettime (CLOCK_REALTIME), found one, and changed it too. (For the record, POSIX also says gettimeofday can never fail.) (It would be nice if we could declare that GNU systems will always support CLOCK_MONOTONIC as well as CLOCK_REALTIME; there are several places where we are using CLOCK_REALTIME where _MONOTONIC would be more appropriate, and/or trying to use _MONOTONIC and then falling back to _REALTIME. But the Hurd doesn't support CLOCK_MONOTONIC yet, and it looks like adding it would involve substantial changes to gnumach's internals and API. Oh well.) A few Hurd-specific files were changed to use __host_get_time instead of __clock_gettime, as this seemed tidier. We also assume this cannot fail. Skimming the code in gnumach leads me to believe the only way it could fail is if __mach_host_self also failed, and our Hurd-specific code consistently assumes that can't happen, so I'm going with that. With the exception of support/support_test_main.c, test cases are not modified, mainly because I didn't want to have to figure out which test cases were testing gettimeofday specifically. The definition of GETTIME in sysdeps/generic/memusage.h had a typo and was not reading tv_sec at all. I fixed this. It appears nobody has been generating malloc traces on a machine that doesn't have a superseding definition. There are a whole bunch of places where the code could be simplified by factoring out timespec subtraction and/or comparison logic, but I want to keep this patch as mechanical as possible. Checked on x86_64-linux-gnu, i686-linux-gnu, powerpc64le-linux-gnu, powerpc64-linux-gnu, powerpc-linux-gnu, and aarch64-linux-gnu. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org> Reviewed-by: Lukasz Majewski <lukma@denx.de>
2019-08-17 00:38:22 +00:00
print_timestamp ("Termination time", now);
if (st_available)
Change most internal uses of __gettimeofday to __clock_gettime. Since gettimeofday will shortly be implemented in terms of clock_gettime on all platforms, internal code should use clock_gettime directly; in addition to removing a layer of indirection, this will allow us to remove the PLT-bypass gunk for gettimeofday. (We can't quite do that yet, but it'll be coming later in this patch series.) In many cases, the changed code does fewer conversions. The changed code always assumes __clock_gettime (CLOCK_REALTIME) cannot fail. Most of the call sites were assuming gettimeofday could not fail, but a few places were checking for errors. POSIX says clock_gettime can only fail if the clock constant is invalid or unsupported, and CLOCK_REALTIME is the one and only clock constant that's required to be supported. For consistency I grepped the entire source tree for any other places that checked for errors from __clock_gettime (CLOCK_REALTIME), found one, and changed it too. (For the record, POSIX also says gettimeofday can never fail.) (It would be nice if we could declare that GNU systems will always support CLOCK_MONOTONIC as well as CLOCK_REALTIME; there are several places where we are using CLOCK_REALTIME where _MONOTONIC would be more appropriate, and/or trying to use _MONOTONIC and then falling back to _REALTIME. But the Hurd doesn't support CLOCK_MONOTONIC yet, and it looks like adding it would involve substantial changes to gnumach's internals and API. Oh well.) A few Hurd-specific files were changed to use __host_get_time instead of __clock_gettime, as this seemed tidier. We also assume this cannot fail. Skimming the code in gnumach leads me to believe the only way it could fail is if __mach_host_self also failed, and our Hurd-specific code consistently assumes that can't happen, so I'm going with that. With the exception of support/support_test_main.c, test cases are not modified, mainly because I didn't want to have to figure out which test cases were testing gettimeofday specifically. The definition of GETTIME in sysdeps/generic/memusage.h had a typo and was not reading tv_sec at all. I fixed this. It appears nobody has been generating malloc traces on a machine that doesn't have a superseding definition. There are a whole bunch of places where the code could be simplified by factoring out timespec subtraction and/or comparison logic, but I want to keep this patch as mechanical as possible. Checked on x86_64-linux-gnu, i686-linux-gnu, powerpc64le-linux-gnu, powerpc64-linux-gnu, powerpc-linux-gnu, and aarch64-linux-gnu. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org> Reviewed-by: Lukasz Majewski <lukma@denx.de>
2019-08-17 00:38:22 +00:00
print_timestamp ("Last write to standard output", st.st_mtim);
/* Exit with an error. */
exit (1);
}
/* This must be volatile as it will be modified by the debugger. */
static volatile int wait_for_debugger = 0;
/* Run test_function or test_function_argv. */
static int
run_test_function (int argc, char **argv, const struct test_config *config)
{
const char *wfd = getenv("WAIT_FOR_DEBUGGER");
if (wfd != NULL)
wait_for_debugger = atoi (wfd);
if (wait_for_debugger)
{
pid_t mypid;
FILE *gdb_script;
char *gdb_script_name;
int inside_container = 0;
const char *outside_pid = getenv("PID_OUTSIDE_CONTAINER");
if (outside_pid)
{
mypid = atoi (outside_pid);
inside_container = 1;
}
else
mypid = getpid();
gdb_script_name = (char *) xmalloc (strlen (argv[0]) + strlen (".gdb") + 1);
sprintf (gdb_script_name, "%s.gdb", argv[0]);
gdb_script = xfopen (gdb_script_name, "w");
fprintf (stderr, "Waiting for debugger, test process is pid %d\n", mypid);
fprintf (stderr, "gdb -x %s\n", gdb_script_name);
if (inside_container)
fprintf (gdb_script, "set sysroot %s/testroot.root\n", support_objdir_root);
fprintf (gdb_script, "file\n");
fprintf (gdb_script, "file %s\n", argv[0]);
fprintf (gdb_script, "symbol-file %s\n", argv[0]);
fprintf (gdb_script, "exec-file %s\n", argv[0]);
fprintf (gdb_script, "attach %ld\n", (long int) mypid);
fprintf (gdb_script, "set wait_for_debugger = 0\n");
fclose (gdb_script);
free (gdb_script_name);
}
/* Wait for the debugger to set wait_for_debugger to zero. */
while (wait_for_debugger)
usleep (1000);
elf: Testing infrastructure for ld.so DSO sorting (BZ #17645) This is the first of a 2-part patch set that fixes slow DSO sorting behavior in the dynamic loader, as reported in BZ #17645. In order to facilitate such a large modification to the dynamic loader, this first patch implements a testing framework for validating shared object sorting behavior, to enable comparison between old/new sorting algorithms, and any later enhancements. This testing infrastructure consists of a Python script scripts/dso-ordering-test.py' which takes in a description language, consisting of strings that describe a set of link dependency relations between DSOs, and generates testcase programs and Makefile fragments to automatically test the described situation, for example: a->b->c->d # four objects linked one after another a->[bc]->d;b->c # a depends on b and c, which both depend on d, # b depends on c (b,c linked to object a in fixed order) a->b->c;{+a;%a;-a} # a, b, c serially dependent, main program uses # dlopen/dlsym/dlclose on object a a->b->c;{}!->[abc] # a, b, c serially dependent; multiple tests generated # to test all permutations of a, b, c ordering linked # to main program (Above is just a short description of what the script can do, more documentation is in the script comments.) Two files containing several new tests, elf/dso-sort-tests-[12].def are added, including test scenarios for BZ #15311 and Redhat issue #1162810 [1]. Due to the nature of dynamic loader tests, where the sorting behavior and test output occurs before/after main(), generating testcases to use support/test-driver.c does not suffice to control meaningful timeout for ld.so. Therefore a new utility program 'support/test-run-command', based on test-driver.c/support_test_main.c has been added. This does the same testcase control, but for a program specified through a command-line rather than at the source code level. This utility is used to run the dynamic loader testcases generated by dso-ordering-test.py. [1] https://bugzilla.redhat.com/show_bug.cgi?id=1162810 Signed-off-by: Chung-Lin Tang <cltang@codesourcery.com> Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
2021-10-21 13:41:21 +00:00
if (config->run_command_mode)
{
/* In run-command-mode, the child process executes the command line
arguments as a new program. */
char **argv_ = xmalloc (sizeof (char *) * argc);
memcpy (argv_, &argv[1], sizeof (char *) * (argc - 1));
argv_[argc - 1] = NULL;
execv (argv_[0], argv_);
printf ("error: should not return here\n");
exit (1);
}
if (config->test_function != NULL)
return config->test_function ();
else if (config->test_function_argv != NULL)
return config->test_function_argv (argc, argv);
else
{
printf ("error: no test function defined\n");
exit (1);
}
}
static bool test_main_called;
const char *test_dir = NULL;
unsigned int test_verbose = 0;
/* If test failure reporting has been linked in, it may contribute
additional test failures. */
static int
adjust_exit_status (int status)
{
if (support_report_failure != NULL)
return support_report_failure (status);
return status;
}
int
support_test_main (int argc, char **argv, const struct test_config *config)
{
if (test_main_called)
{
printf ("error: test_main called for a second time\n");
exit (1);
}
test_main_called = true;
const struct option *options;
if (config->options != NULL)
options = config->options;
else
options = default_options;
cleanup_function = config->cleanup_function;
int direct = 0; /* Directly call the test function? */
int status;
int opt;
unsigned int timeoutfactor = TIMEOUTFACTOR;
pid_t termpid;
/* If we're debugging the test, we need to disable timeouts and use
the initial pid (esp if we're running inside a container). */
if (getenv("WAIT_FOR_DEBUGGER") != NULL)
direct = 1;
if (!config->no_mallopt)
{
/* Make uses of freed and uninitialized memory known. Do not
pull in a definition for mallopt if it has not been defined
already. */
extern __typeof__ (mallopt) mallopt __attribute__ ((weak));
if (mallopt != NULL)
mallopt (M_PERTURB, 42);
}
while ((opt = getopt_long (argc, argv, config->optstring, options, NULL))
!= -1)
switch (opt)
{
case '?':
usage (options);
exit (1);
case 'v':
++test_verbose;
break;
case OPT_DIRECT:
direct = 1;
break;
case OPT_TESTDIR:
test_dir = optarg;
break;
default:
if (config->cmdline_function != NULL)
config->cmdline_function (opt);
}
/* If set, read the test TIMEOUTFACTOR value from the environment.
This value is used to scale the default test timeout values. */
char *envstr_timeoutfactor = getenv ("TIMEOUTFACTOR");
if (envstr_timeoutfactor != NULL)
{
char *envstr_conv = envstr_timeoutfactor;
unsigned long int env_fact;
env_fact = strtoul (envstr_timeoutfactor, &envstr_conv, 0);
if (*envstr_conv == '\0' && envstr_conv != envstr_timeoutfactor)
timeoutfactor = MAX (env_fact, 1);
}
/* Set TMPDIR to specified test directory. */
if (test_dir != NULL)
{
setenv ("TMPDIR", test_dir, 1);
if (chdir (test_dir) < 0)
{
printf ("chdir: %m\n");
exit (1);
}
}
else
{
test_dir = getenv ("TMPDIR");
if (test_dir == NULL || test_dir[0] == '\0')
test_dir = "/tmp";
}
if (support_set_test_dir != NULL)
support_set_test_dir (test_dir);
int timeout = config->timeout;
if (timeout == 0)
timeout = DEFAULT_TIMEOUT;
/* Make sure we see all message, even those on stdout. */
if (!config->no_setvbuf)
setvbuf (stdout, NULL, _IONBF, 0);
/* Make sure temporary files are deleted. */
if (support_delete_temp_files != NULL)
atexit (support_delete_temp_files);
/* Correct for the possible parameters. */
argv[optind - 1] = argv[0];
argv += optind - 1;
argc -= optind - 1;
/* Call the initializing function, if one is available. */
if (config->prepare_function != NULL)
config->prepare_function (argc, argv);
const char *envstr_direct = getenv ("TEST_DIRECT");
if (envstr_direct != NULL)
{
FILE *f = fopen (envstr_direct, "w");
if (f == NULL)
{
printf ("cannot open TEST_DIRECT output file '%s': %m\n",
envstr_direct);
exit (1);
}
fprintf (f, "timeout=%u\ntimeoutfactor=%u\n",
config->timeout, timeoutfactor);
if (config->expected_status != 0)
fprintf (f, "exit=%u\n", config->expected_status);
if (config->expected_signal != 0)
fprintf (f, "signal=%s\n", strsignal (config->expected_signal));
if (support_print_temp_files != NULL)
support_print_temp_files (f);
fclose (f);
direct = 1;
}
bool disable_coredumps;
{
const char *coredumps = getenv ("TEST_COREDUMPS");
disable_coredumps = coredumps == NULL || coredumps[0] == '\0';
}
/* If we are not expected to fork run the function immediately. */
if (direct)
return adjust_exit_status (run_test_function (argc, argv, config));
/* Set up the test environment:
- prevent core dumps
- set up the timer
- fork and execute the function. */
2016-12-15 15:41:04 +00:00
test_pid = fork ();
if (test_pid == 0)
{
/* This is the child. */
if (disable_coredumps)
{
/* Try to avoid dumping core. This is necessary because we
run the test from the source tree, and the coredumps
would end up there (and not in the build tree). */
struct rlimit core_limit;
core_limit.rlim_cur = 0;
core_limit.rlim_max = 0;
setrlimit (RLIMIT_CORE, &core_limit);
}
/* We put the test process in its own pgrp so that if it bogusly
generates any job control signals, they won't hit the whole build. */
if (setpgid (0, 0) != 0)
printf ("Failed to set the process group ID: %m\n");
/* Execute the test function and exit with the return value. */
exit (run_test_function (argc, argv, config));
}
else if (test_pid < 0)
{
printf ("Cannot fork test program: %m\n");
exit (1);
}
/* Set timeout. */
signal (SIGALRM, signal_handler);
2016-12-10 15:10:53 +00:00
alarm (timeout * timeoutfactor);
/* Make sure we clean up if the wrapper gets interrupted. */
signal (SIGINT, signal_handler);
/* Wait for the regular termination. */
termpid = TEMP_FAILURE_RETRY (waitpid (test_pid, &status, 0));
if (termpid == -1)
{
printf ("Waiting for test program failed: %m\n");
exit (1);
}
if (termpid != test_pid)
{
printf ("Oops, wrong test program terminated: expected %ld, got %ld\n",
(long int) test_pid, (long int) termpid);
exit (1);
}
/* Process terminated normally without timeout etc. */
if (WIFEXITED (status))
{
if (config->expected_status == 0)
{
if (config->expected_signal == 0)
/* Exit with the return value of the test. */
return adjust_exit_status (WEXITSTATUS (status));
else
{
printf ("Expected signal '%s' from child, got none\n",
strsignal (config->expected_signal));
exit (1);
}
}
else
{
/* Non-zero exit status is expected */
if (WEXITSTATUS (status) != config->expected_status)
{
printf ("Expected status %d, got %d\n",
config->expected_status, WEXITSTATUS (status));
exit (1);
}
}
return adjust_exit_status (0);
}
/* Process was killed by timer or other signal. */
else
{
if (config->expected_signal == 0)
{
printf ("Didn't expect signal from child: got `%s'\n",
strsignal (WTERMSIG (status)));
exit (1);
}
else if (WTERMSIG (status) != config->expected_signal)
{
printf ("Incorrect signal from child: got `%s', need `%s'\n",
strsignal (WTERMSIG (status)),
strsignal (config->expected_signal));
exit (1);
}
return adjust_exit_status (0);
}
}