You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
weston/tests/weston-test-fixture-composi...

508 lines
12 KiB

tests: thread-based client harness This replaces the old test harness with a new one. The old harness relied on fork()'ing each test which makes tests independent, but makes debugging them harder. The new harness runs client code in a thread instead of a new process. A side-effect of not fork()'ing anymore is that any failure will stop running a test series short. Fortunately we do not have any tests that are expected to crash or fail. The old harness executed 'weston' from Meson, with lots of setup as both command line options and environment variables. The new harness executes wet_main() instead: the test program itself calls the compositor main function to execute the compositor in-process. Command line arguments are configured in the test program itself, not in meson.build. Environment variables aside, you are able to run a test by simply executing the test program, even if it is a plugin test. The new harness adds a new type of iteration: fixtures. For now, fixtures are used to set up the compositor for tests that need a compositor. If necessary, a fixture setup may include a data array of arbitrary type for executing the test series for each element in the array. This will be most useful for running screenshooting tests with both Pixman- and GL-renderers. The new harness outputs TAP formatted results into stdout. Meson is not switched to consume TAP yet though, because it would require a Meson version requirement bump and would not have any benefits at this time. OTOH outputting TAP is trivial and sets up a clear precedent of random test chatter belonging to stderr. This commit migrates only few tests to actually make use of the new features: roles is a basic client test, subsurface-shot is a client test that demonstrates the fixture array, and plugin-registry is a plugin test. The rest of the tests will be migrated later. Once all tests are migrated, we can remove the test-specific setup from meson.build, leaving only the actual build instructions in there. The not migrated tests and stand-alone tests suffer only a minor change: they no longer fork() for each TEST(), otherwise they keep running as before. Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.com>
5 years ago
/*
* Copyright 2019 Collabora, Ltd.
*
* Permission is hereby granted, free of charge, to any person obtaining
* a copy of this software and associated documentation files (the
* "Software"), to deal in the Software without restriction, including
* without limitation the rights to use, copy, modify, merge, publish,
* distribute, sublicense, and/or sell copies of the Software, and to
* permit persons to whom the Software is furnished to do so, subject to
* the following conditions:
*
* The above copyright notice and this permission notice (including the
* next paragraph) shall be included in all copies or substantial
* portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#include "config.h"
#include <string.h>
#include <assert.h>
#include <libudev.h>
#include <unistd.h>
#include <sys/file.h>
#include <sys/types.h>
#include <sys/stat.h>
#include <stdarg.h>
#include <stdlib.h>
#include <errno.h>
#include <fcntl.h>
tests: thread-based client harness This replaces the old test harness with a new one. The old harness relied on fork()'ing each test which makes tests independent, but makes debugging them harder. The new harness runs client code in a thread instead of a new process. A side-effect of not fork()'ing anymore is that any failure will stop running a test series short. Fortunately we do not have any tests that are expected to crash or fail. The old harness executed 'weston' from Meson, with lots of setup as both command line options and environment variables. The new harness executes wet_main() instead: the test program itself calls the compositor main function to execute the compositor in-process. Command line arguments are configured in the test program itself, not in meson.build. Environment variables aside, you are able to run a test by simply executing the test program, even if it is a plugin test. The new harness adds a new type of iteration: fixtures. For now, fixtures are used to set up the compositor for tests that need a compositor. If necessary, a fixture setup may include a data array of arbitrary type for executing the test series for each element in the array. This will be most useful for running screenshooting tests with both Pixman- and GL-renderers. The new harness outputs TAP formatted results into stdout. Meson is not switched to consume TAP yet though, because it would require a Meson version requirement bump and would not have any benefits at this time. OTOH outputting TAP is trivial and sets up a clear precedent of random test chatter belonging to stderr. This commit migrates only few tests to actually make use of the new features: roles is a basic client test, subsurface-shot is a client test that demonstrates the fixture array, and plugin-registry is a plugin test. The rest of the tests will be migrated later. Once all tests are migrated, we can remove the test-specific setup from meson.build, leaving only the actual build instructions in there. The not migrated tests and stand-alone tests suffer only a minor change: they no longer fork() for each TEST(), otherwise they keep running as before. Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.com>
5 years ago
#include "shared/helpers.h"
#include "weston-test-fixture-compositor.h"
#include "weston.h"
#include "test-config.h"
tests: thread-based client harness This replaces the old test harness with a new one. The old harness relied on fork()'ing each test which makes tests independent, but makes debugging them harder. The new harness runs client code in a thread instead of a new process. A side-effect of not fork()'ing anymore is that any failure will stop running a test series short. Fortunately we do not have any tests that are expected to crash or fail. The old harness executed 'weston' from Meson, with lots of setup as both command line options and environment variables. The new harness executes wet_main() instead: the test program itself calls the compositor main function to execute the compositor in-process. Command line arguments are configured in the test program itself, not in meson.build. Environment variables aside, you are able to run a test by simply executing the test program, even if it is a plugin test. The new harness adds a new type of iteration: fixtures. For now, fixtures are used to set up the compositor for tests that need a compositor. If necessary, a fixture setup may include a data array of arbitrary type for executing the test series for each element in the array. This will be most useful for running screenshooting tests with both Pixman- and GL-renderers. The new harness outputs TAP formatted results into stdout. Meson is not switched to consume TAP yet though, because it would require a Meson version requirement bump and would not have any benefits at this time. OTOH outputting TAP is trivial and sets up a clear precedent of random test chatter belonging to stderr. This commit migrates only few tests to actually make use of the new features: roles is a basic client test, subsurface-shot is a client test that demonstrates the fixture array, and plugin-registry is a plugin test. The rest of the tests will be migrated later. Once all tests are migrated, we can remove the test-specific setup from meson.build, leaving only the actual build instructions in there. The not migrated tests and stand-alone tests suffer only a minor change: they no longer fork() for each TEST(), otherwise they keep running as before. Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.com>
5 years ago
struct prog_args {
int argc;
char **argv;
char **saved;
int alloc;
};
static void
prog_args_init(struct prog_args *p)
{
memset(p, 0, sizeof(*p));
}
static void
prog_args_take(struct prog_args *p, char *arg)
{
assert(arg);
if (p->argc == p->alloc) {
p->alloc += 10;
p->argv = realloc(p->argv, sizeof(char *) * p->alloc);
assert(p->argv);
}
p->argv[p->argc++] = arg;
}
/*
* The program to be executed will trample on argv, hence we need a copy to
* be able to free all our args.
*/
static void
prog_args_save(struct prog_args *p)
{
assert(p->saved == NULL);
p->saved = calloc(p->argc, sizeof(char *));
assert(p->saved);
memcpy(p->saved, p->argv, sizeof(char *) * p->argc);
}
static void
prog_args_fini(struct prog_args *p)
{
int i;
assert(p->saved);
for (i = 0; i < p->argc; i++)
free(p->saved[i]);
free(p->saved);
free(p->argv);
prog_args_init(p);
}
static char *
get_lock_path(void)
{
const char *env_path, *suffix;
char *lock_path;
suffix = "weston-test-suite-drm-lock";
env_path = getenv("XDG_RUNTIME_DIR");
if (!env_path) {
fprintf(stderr, "Failed to compute lock file path. " \
"XDG_RUNTIME_DIR is not set.\n");
return NULL;
}
if (asprintf(&lock_path, "%s/%s", env_path, suffix) == -1)
return NULL;
return lock_path;
}
/*
* DRM-backend tests need to be run sequentially, since there can only be
* one user at a time with master status in a DRM KMS device. Since the
* test suite runs the tests in parallel, there's a mechanism to assure
* only one DRM-backend test is running at a time: tests of this type keep
* waiting until they acquire a lock (which is hold until they end).
*
* Returns -1 in failure and fd in success.
*/
static int
wait_for_lock(void)
{
char *lock_path;
int fd;
lock_path = get_lock_path();
if (!lock_path)
goto err_path;
fd = open(lock_path, O_RDWR | O_CLOEXEC | O_CREAT, 00700);
if (fd == -1) {
fprintf(stderr, "Could not open lock file %s: %s\n", lock_path, strerror(errno));
goto err_open;
}
fprintf(stderr, "Waiting for lock on %s...\n", lock_path);
/* The call is blocking, so we don't need a 'while'. Also, as we
* have a timeout for each test, this won't get stuck waiting. */
if (flock(fd, LOCK_EX) == -1) {
fprintf(stderr, "Could not lock %s: %s\n", lock_path, strerror(errno));
goto err_lock;
}
fprintf(stderr, "Lock %s acquired.\n", lock_path);
free(lock_path);
return fd;
err_lock:
close(fd);
err_open:
free(lock_path);
err_path:
return -1;
}
tests: thread-based client harness This replaces the old test harness with a new one. The old harness relied on fork()'ing each test which makes tests independent, but makes debugging them harder. The new harness runs client code in a thread instead of a new process. A side-effect of not fork()'ing anymore is that any failure will stop running a test series short. Fortunately we do not have any tests that are expected to crash or fail. The old harness executed 'weston' from Meson, with lots of setup as both command line options and environment variables. The new harness executes wet_main() instead: the test program itself calls the compositor main function to execute the compositor in-process. Command line arguments are configured in the test program itself, not in meson.build. Environment variables aside, you are able to run a test by simply executing the test program, even if it is a plugin test. The new harness adds a new type of iteration: fixtures. For now, fixtures are used to set up the compositor for tests that need a compositor. If necessary, a fixture setup may include a data array of arbitrary type for executing the test series for each element in the array. This will be most useful for running screenshooting tests with both Pixman- and GL-renderers. The new harness outputs TAP formatted results into stdout. Meson is not switched to consume TAP yet though, because it would require a Meson version requirement bump and would not have any benefits at this time. OTOH outputting TAP is trivial and sets up a clear precedent of random test chatter belonging to stderr. This commit migrates only few tests to actually make use of the new features: roles is a basic client test, subsurface-shot is a client test that demonstrates the fixture array, and plugin-registry is a plugin test. The rest of the tests will be migrated later. Once all tests are migrated, we can remove the test-specific setup from meson.build, leaving only the actual build instructions in there. The not migrated tests and stand-alone tests suffer only a minor change: they no longer fork() for each TEST(), otherwise they keep running as before. Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.com>
5 years ago
/** Initialize part of compositor setup
*
* \param setup The variable to initialize.
* \param testset_name Value for testset_name member.
*
* \ingroup testharness_private
*/
void
compositor_setup_defaults_(struct compositor_setup *setup,
const char *testset_name)
{
*setup = (struct compositor_setup) {
.backend = WESTON_BACKEND_HEADLESS,
.renderer = RENDERER_NOOP,
.shell = SHELL_DESKTOP,
.xwayland = false,
.width = 320,
.height = 240,
.scale = 1,
.transform = WL_OUTPUT_TRANSFORM_NORMAL,
tests: thread-based client harness This replaces the old test harness with a new one. The old harness relied on fork()'ing each test which makes tests independent, but makes debugging them harder. The new harness runs client code in a thread instead of a new process. A side-effect of not fork()'ing anymore is that any failure will stop running a test series short. Fortunately we do not have any tests that are expected to crash or fail. The old harness executed 'weston' from Meson, with lots of setup as both command line options and environment variables. The new harness executes wet_main() instead: the test program itself calls the compositor main function to execute the compositor in-process. Command line arguments are configured in the test program itself, not in meson.build. Environment variables aside, you are able to run a test by simply executing the test program, even if it is a plugin test. The new harness adds a new type of iteration: fixtures. For now, fixtures are used to set up the compositor for tests that need a compositor. If necessary, a fixture setup may include a data array of arbitrary type for executing the test series for each element in the array. This will be most useful for running screenshooting tests with both Pixman- and GL-renderers. The new harness outputs TAP formatted results into stdout. Meson is not switched to consume TAP yet though, because it would require a Meson version requirement bump and would not have any benefits at this time. OTOH outputting TAP is trivial and sets up a clear precedent of random test chatter belonging to stderr. This commit migrates only few tests to actually make use of the new features: roles is a basic client test, subsurface-shot is a client test that demonstrates the fixture array, and plugin-registry is a plugin test. The rest of the tests will be migrated later. Once all tests are migrated, we can remove the test-specific setup from meson.build, leaving only the actual build instructions in there. The not migrated tests and stand-alone tests suffer only a minor change: they no longer fork() for each TEST(), otherwise they keep running as before. Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.com>
5 years ago
.config_file = NULL,
.extra_module = NULL,
.logging_scopes = NULL,
.testset_name = testset_name,
};
}
static const char *
backend_to_str(enum weston_compositor_backend b)
{
static const char * const names[] = {
[WESTON_BACKEND_DRM] = "drm-backend.so",
[WESTON_BACKEND_FBDEV] = "fbdev-backend.so",
[WESTON_BACKEND_HEADLESS] = "headless-backend.so",
[WESTON_BACKEND_RDP] = "rdp-backend.so",
[WESTON_BACKEND_WAYLAND] = "wayland-backend.so",
[WESTON_BACKEND_X11] = "X11-backend.so",
};
assert(b >= 0 && b < ARRAY_LENGTH(names));
return names[b];
}
static const char *
renderer_to_arg(enum weston_compositor_backend b, enum renderer_type r)
{
static const char * const headless_names[] = {
[RENDERER_NOOP] = NULL,
[RENDERER_PIXMAN] = "--use-pixman",
[RENDERER_GL] = "--use-gl",
};
static const char * const drm_names[] = {
[RENDERER_PIXMAN] = "--use-pixman",
[RENDERER_GL] = NULL,
};
tests: thread-based client harness This replaces the old test harness with a new one. The old harness relied on fork()'ing each test which makes tests independent, but makes debugging them harder. The new harness runs client code in a thread instead of a new process. A side-effect of not fork()'ing anymore is that any failure will stop running a test series short. Fortunately we do not have any tests that are expected to crash or fail. The old harness executed 'weston' from Meson, with lots of setup as both command line options and environment variables. The new harness executes wet_main() instead: the test program itself calls the compositor main function to execute the compositor in-process. Command line arguments are configured in the test program itself, not in meson.build. Environment variables aside, you are able to run a test by simply executing the test program, even if it is a plugin test. The new harness adds a new type of iteration: fixtures. For now, fixtures are used to set up the compositor for tests that need a compositor. If necessary, a fixture setup may include a data array of arbitrary type for executing the test series for each element in the array. This will be most useful for running screenshooting tests with both Pixman- and GL-renderers. The new harness outputs TAP formatted results into stdout. Meson is not switched to consume TAP yet though, because it would require a Meson version requirement bump and would not have any benefits at this time. OTOH outputting TAP is trivial and sets up a clear precedent of random test chatter belonging to stderr. This commit migrates only few tests to actually make use of the new features: roles is a basic client test, subsurface-shot is a client test that demonstrates the fixture array, and plugin-registry is a plugin test. The rest of the tests will be migrated later. Once all tests are migrated, we can remove the test-specific setup from meson.build, leaving only the actual build instructions in there. The not migrated tests and stand-alone tests suffer only a minor change: they no longer fork() for each TEST(), otherwise they keep running as before. Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.com>
5 years ago
switch (b) {
case WESTON_BACKEND_HEADLESS:
assert(r >= RENDERER_NOOP && r <= RENDERER_GL);
tests: thread-based client harness This replaces the old test harness with a new one. The old harness relied on fork()'ing each test which makes tests independent, but makes debugging them harder. The new harness runs client code in a thread instead of a new process. A side-effect of not fork()'ing anymore is that any failure will stop running a test series short. Fortunately we do not have any tests that are expected to crash or fail. The old harness executed 'weston' from Meson, with lots of setup as both command line options and environment variables. The new harness executes wet_main() instead: the test program itself calls the compositor main function to execute the compositor in-process. Command line arguments are configured in the test program itself, not in meson.build. Environment variables aside, you are able to run a test by simply executing the test program, even if it is a plugin test. The new harness adds a new type of iteration: fixtures. For now, fixtures are used to set up the compositor for tests that need a compositor. If necessary, a fixture setup may include a data array of arbitrary type for executing the test series for each element in the array. This will be most useful for running screenshooting tests with both Pixman- and GL-renderers. The new harness outputs TAP formatted results into stdout. Meson is not switched to consume TAP yet though, because it would require a Meson version requirement bump and would not have any benefits at this time. OTOH outputting TAP is trivial and sets up a clear precedent of random test chatter belonging to stderr. This commit migrates only few tests to actually make use of the new features: roles is a basic client test, subsurface-shot is a client test that demonstrates the fixture array, and plugin-registry is a plugin test. The rest of the tests will be migrated later. Once all tests are migrated, we can remove the test-specific setup from meson.build, leaving only the actual build instructions in there. The not migrated tests and stand-alone tests suffer only a minor change: they no longer fork() for each TEST(), otherwise they keep running as before. Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.com>
5 years ago
return headless_names[r];
case WESTON_BACKEND_DRM:
assert(r >= RENDERER_PIXMAN && r <= RENDERER_GL);
return drm_names[r];
tests: thread-based client harness This replaces the old test harness with a new one. The old harness relied on fork()'ing each test which makes tests independent, but makes debugging them harder. The new harness runs client code in a thread instead of a new process. A side-effect of not fork()'ing anymore is that any failure will stop running a test series short. Fortunately we do not have any tests that are expected to crash or fail. The old harness executed 'weston' from Meson, with lots of setup as both command line options and environment variables. The new harness executes wet_main() instead: the test program itself calls the compositor main function to execute the compositor in-process. Command line arguments are configured in the test program itself, not in meson.build. Environment variables aside, you are able to run a test by simply executing the test program, even if it is a plugin test. The new harness adds a new type of iteration: fixtures. For now, fixtures are used to set up the compositor for tests that need a compositor. If necessary, a fixture setup may include a data array of arbitrary type for executing the test series for each element in the array. This will be most useful for running screenshooting tests with both Pixman- and GL-renderers. The new harness outputs TAP formatted results into stdout. Meson is not switched to consume TAP yet though, because it would require a Meson version requirement bump and would not have any benefits at this time. OTOH outputting TAP is trivial and sets up a clear precedent of random test chatter belonging to stderr. This commit migrates only few tests to actually make use of the new features: roles is a basic client test, subsurface-shot is a client test that demonstrates the fixture array, and plugin-registry is a plugin test. The rest of the tests will be migrated later. Once all tests are migrated, we can remove the test-specific setup from meson.build, leaving only the actual build instructions in there. The not migrated tests and stand-alone tests suffer only a minor change: they no longer fork() for each TEST(), otherwise they keep running as before. Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.com>
5 years ago
default:
assert(0 && "renderer_to_str() does not know the backend");
}
return NULL;
}
static const char *
shell_to_str(enum shell_type t)
{
static const char * const names[] = {
[SHELL_TEST_DESKTOP] = "weston-test-desktop-shell.so",
[SHELL_DESKTOP] = "desktop-shell.so",
[SHELL_FULLSCREEN] = "fullscreen-shell.so",
[SHELL_IVI] = "ivi-shell.so",
};
assert(t >= 0 && t < ARRAY_LENGTH(names));
return names[t];
}
static const char *
transform_to_str(enum wl_output_transform t)
{
static const char * const names[] = {
[WL_OUTPUT_TRANSFORM_NORMAL] = "normal",
[WL_OUTPUT_TRANSFORM_90] = "rotate-90",
[WL_OUTPUT_TRANSFORM_180] = "rotate-180",
[WL_OUTPUT_TRANSFORM_270] = "rotate-270",
[WL_OUTPUT_TRANSFORM_FLIPPED] = "flipped",
[WL_OUTPUT_TRANSFORM_FLIPPED_90] = "flipped-rotate-90",
[WL_OUTPUT_TRANSFORM_FLIPPED_180] = "flipped-rotate-180",
[WL_OUTPUT_TRANSFORM_FLIPPED_270] = "flipped-rotate-270",
};
assert(t < ARRAY_LENGTH(names) && names[t]);
return names[t];
}
tests: thread-based client harness This replaces the old test harness with a new one. The old harness relied on fork()'ing each test which makes tests independent, but makes debugging them harder. The new harness runs client code in a thread instead of a new process. A side-effect of not fork()'ing anymore is that any failure will stop running a test series short. Fortunately we do not have any tests that are expected to crash or fail. The old harness executed 'weston' from Meson, with lots of setup as both command line options and environment variables. The new harness executes wet_main() instead: the test program itself calls the compositor main function to execute the compositor in-process. Command line arguments are configured in the test program itself, not in meson.build. Environment variables aside, you are able to run a test by simply executing the test program, even if it is a plugin test. The new harness adds a new type of iteration: fixtures. For now, fixtures are used to set up the compositor for tests that need a compositor. If necessary, a fixture setup may include a data array of arbitrary type for executing the test series for each element in the array. This will be most useful for running screenshooting tests with both Pixman- and GL-renderers. The new harness outputs TAP formatted results into stdout. Meson is not switched to consume TAP yet though, because it would require a Meson version requirement bump and would not have any benefits at this time. OTOH outputting TAP is trivial and sets up a clear precedent of random test chatter belonging to stderr. This commit migrates only few tests to actually make use of the new features: roles is a basic client test, subsurface-shot is a client test that demonstrates the fixture array, and plugin-registry is a plugin test. The rest of the tests will be migrated later. Once all tests are migrated, we can remove the test-specific setup from meson.build, leaving only the actual build instructions in there. The not migrated tests and stand-alone tests suffer only a minor change: they no longer fork() for each TEST(), otherwise they keep running as before. Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.com>
5 years ago
/** Execute compositor
*
* Manufactures the compositor command line and calls wet_main().
*
* Returns RESULT_SKIP if the given setup contains features that were disabled
* in the build, e.g. GL-renderer or DRM-backend.
*
* \ingroup testharness_private
*/
int
execute_compositor(const struct compositor_setup *setup,
struct wet_testsuite_data *data)
{
struct prog_args args;
char *tmp;
const char *ctmp, *drm_device;
int ret, lock_fd = -1;
tests: thread-based client harness This replaces the old test harness with a new one. The old harness relied on fork()'ing each test which makes tests independent, but makes debugging them harder. The new harness runs client code in a thread instead of a new process. A side-effect of not fork()'ing anymore is that any failure will stop running a test series short. Fortunately we do not have any tests that are expected to crash or fail. The old harness executed 'weston' from Meson, with lots of setup as both command line options and environment variables. The new harness executes wet_main() instead: the test program itself calls the compositor main function to execute the compositor in-process. Command line arguments are configured in the test program itself, not in meson.build. Environment variables aside, you are able to run a test by simply executing the test program, even if it is a plugin test. The new harness adds a new type of iteration: fixtures. For now, fixtures are used to set up the compositor for tests that need a compositor. If necessary, a fixture setup may include a data array of arbitrary type for executing the test series for each element in the array. This will be most useful for running screenshooting tests with both Pixman- and GL-renderers. The new harness outputs TAP formatted results into stdout. Meson is not switched to consume TAP yet though, because it would require a Meson version requirement bump and would not have any benefits at this time. OTOH outputting TAP is trivial and sets up a clear precedent of random test chatter belonging to stderr. This commit migrates only few tests to actually make use of the new features: roles is a basic client test, subsurface-shot is a client test that demonstrates the fixture array, and plugin-registry is a plugin test. The rest of the tests will be migrated later. Once all tests are migrated, we can remove the test-specific setup from meson.build, leaving only the actual build instructions in there. The not migrated tests and stand-alone tests suffer only a minor change: they no longer fork() for each TEST(), otherwise they keep running as before. Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.com>
5 years ago
if (setenv("WESTON_MODULE_MAP", WESTON_MODULE_MAP, 0) < 0 ||
setenv("WESTON_DATA_DIR", WESTON_DATA_DIR, 0) < 0) {
fprintf(stderr, "Error: environment setup failed.\n");
return RESULT_HARD_ERROR;
}
tests: thread-based client harness This replaces the old test harness with a new one. The old harness relied on fork()'ing each test which makes tests independent, but makes debugging them harder. The new harness runs client code in a thread instead of a new process. A side-effect of not fork()'ing anymore is that any failure will stop running a test series short. Fortunately we do not have any tests that are expected to crash or fail. The old harness executed 'weston' from Meson, with lots of setup as both command line options and environment variables. The new harness executes wet_main() instead: the test program itself calls the compositor main function to execute the compositor in-process. Command line arguments are configured in the test program itself, not in meson.build. Environment variables aside, you are able to run a test by simply executing the test program, even if it is a plugin test. The new harness adds a new type of iteration: fixtures. For now, fixtures are used to set up the compositor for tests that need a compositor. If necessary, a fixture setup may include a data array of arbitrary type for executing the test series for each element in the array. This will be most useful for running screenshooting tests with both Pixman- and GL-renderers. The new harness outputs TAP formatted results into stdout. Meson is not switched to consume TAP yet though, because it would require a Meson version requirement bump and would not have any benefits at this time. OTOH outputting TAP is trivial and sets up a clear precedent of random test chatter belonging to stderr. This commit migrates only few tests to actually make use of the new features: roles is a basic client test, subsurface-shot is a client test that demonstrates the fixture array, and plugin-registry is a plugin test. The rest of the tests will be migrated later. Once all tests are migrated, we can remove the test-specific setup from meson.build, leaving only the actual build instructions in there. The not migrated tests and stand-alone tests suffer only a minor change: they no longer fork() for each TEST(), otherwise they keep running as before. Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.com>
5 years ago
#ifndef BUILD_DRM_COMPOSITOR
if (setup->backend == WESTON_BACKEND_DRM) {
fprintf(stderr, "DRM-backend required but not built, skipping.\n");
return RESULT_SKIP;
}
#endif
#ifndef BUILD_FBDEV_COMPOSITOR
if (setup->backend == WESTON_BACKEND_FBDEV) {
fprintf(stderr, "fbdev-backend required but not built, skipping.\n");
return RESULT_SKIP;
}
#endif
#ifndef BUILD_RDP_COMPOSITOR
if (setup->backend == WESTON_BACKEND_RDP) {
fprintf(stderr, "RDP-backend required but not built, skipping.\n");
return RESULT_SKIP;
}
#endif
#ifndef BUILD_WAYLAND_COMPOSITOR
if (setup->backend == WESTON_BACKEND_WAYLAND) {
fprintf(stderr, "wayland-backend required but not built, skipping.\n");
return RESULT_SKIP;
}
#endif
#ifndef BUILD_X11_COMPOSITOR
if (setup->backend == WESTON_BACKEND_X11) {
fprintf(stderr, "X11-backend required but not built, skipping.\n");
return RESULT_SKIP;
}
#endif
#ifndef ENABLE_EGL
if (setup->renderer == RENDERER_GL) {
fprintf(stderr, "GL-renderer required but not built, skipping.\n");
return RESULT_SKIP;
}
#endif
#if !TEST_GL_RENDERER
if (setup->renderer == RENDERER_GL) {
fprintf(stderr, "GL-renderer disabled for tests, skipping.\n");
return RESULT_SKIP;
}
#endif
tests: thread-based client harness This replaces the old test harness with a new one. The old harness relied on fork()'ing each test which makes tests independent, but makes debugging them harder. The new harness runs client code in a thread instead of a new process. A side-effect of not fork()'ing anymore is that any failure will stop running a test series short. Fortunately we do not have any tests that are expected to crash or fail. The old harness executed 'weston' from Meson, with lots of setup as both command line options and environment variables. The new harness executes wet_main() instead: the test program itself calls the compositor main function to execute the compositor in-process. Command line arguments are configured in the test program itself, not in meson.build. Environment variables aside, you are able to run a test by simply executing the test program, even if it is a plugin test. The new harness adds a new type of iteration: fixtures. For now, fixtures are used to set up the compositor for tests that need a compositor. If necessary, a fixture setup may include a data array of arbitrary type for executing the test series for each element in the array. This will be most useful for running screenshooting tests with both Pixman- and GL-renderers. The new harness outputs TAP formatted results into stdout. Meson is not switched to consume TAP yet though, because it would require a Meson version requirement bump and would not have any benefits at this time. OTOH outputting TAP is trivial and sets up a clear precedent of random test chatter belonging to stderr. This commit migrates only few tests to actually make use of the new features: roles is a basic client test, subsurface-shot is a client test that demonstrates the fixture array, and plugin-registry is a plugin test. The rest of the tests will be migrated later. Once all tests are migrated, we can remove the test-specific setup from meson.build, leaving only the actual build instructions in there. The not migrated tests and stand-alone tests suffer only a minor change: they no longer fork() for each TEST(), otherwise they keep running as before. Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.com>
5 years ago
prog_args_init(&args);
/* argv[0] */
asprintf(&tmp, "weston-%s", setup->testset_name);
prog_args_take(&args, tmp);
asprintf(&tmp, "--backend=%s", backend_to_str(setup->backend));
prog_args_take(&args, tmp);
if (setup->backend == WESTON_BACKEND_DRM) {
drm_device = getenv("WESTON_TEST_SUITE_DRM_DEVICE");
if (!drm_device) {
fprintf(stderr, "Skipping DRM-backend tests because " \
"WESTON_TEST_SUITE_DRM_DEVICE is not set. " \
"See test suite documentation to learn how " \
"to run them.\n");
return RESULT_SKIP;
}
asprintf(&tmp, "--drm-device=%s", drm_device);
prog_args_take(&args, tmp);
prog_args_take(&args, strdup("--seat=weston-test-seat"));
prog_args_take(&args, strdup("--continue-without-input"));
lock_fd = wait_for_lock();
if (lock_fd == -1)
return RESULT_FAIL;
}
tests: thread-based client harness This replaces the old test harness with a new one. The old harness relied on fork()'ing each test which makes tests independent, but makes debugging them harder. The new harness runs client code in a thread instead of a new process. A side-effect of not fork()'ing anymore is that any failure will stop running a test series short. Fortunately we do not have any tests that are expected to crash or fail. The old harness executed 'weston' from Meson, with lots of setup as both command line options and environment variables. The new harness executes wet_main() instead: the test program itself calls the compositor main function to execute the compositor in-process. Command line arguments are configured in the test program itself, not in meson.build. Environment variables aside, you are able to run a test by simply executing the test program, even if it is a plugin test. The new harness adds a new type of iteration: fixtures. For now, fixtures are used to set up the compositor for tests that need a compositor. If necessary, a fixture setup may include a data array of arbitrary type for executing the test series for each element in the array. This will be most useful for running screenshooting tests with both Pixman- and GL-renderers. The new harness outputs TAP formatted results into stdout. Meson is not switched to consume TAP yet though, because it would require a Meson version requirement bump and would not have any benefits at this time. OTOH outputting TAP is trivial and sets up a clear precedent of random test chatter belonging to stderr. This commit migrates only few tests to actually make use of the new features: roles is a basic client test, subsurface-shot is a client test that demonstrates the fixture array, and plugin-registry is a plugin test. The rest of the tests will be migrated later. Once all tests are migrated, we can remove the test-specific setup from meson.build, leaving only the actual build instructions in there. The not migrated tests and stand-alone tests suffer only a minor change: they no longer fork() for each TEST(), otherwise they keep running as before. Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.com>
5 years ago
asprintf(&tmp, "--socket=%s", setup->testset_name);
prog_args_take(&args, tmp);
asprintf(&tmp, "--modules=%s%s%s", TESTSUITE_PLUGIN_PATH,
setup->extra_module ? "," : "",
setup->extra_module ? setup->extra_module : "");
prog_args_take(&args, tmp);
if (setup->backend != WESTON_BACKEND_DRM &&
setup->backend != WESTON_BACKEND_FBDEV) {
asprintf(&tmp, "--width=%d", setup->width);
prog_args_take(&args, tmp);
tests: thread-based client harness This replaces the old test harness with a new one. The old harness relied on fork()'ing each test which makes tests independent, but makes debugging them harder. The new harness runs client code in a thread instead of a new process. A side-effect of not fork()'ing anymore is that any failure will stop running a test series short. Fortunately we do not have any tests that are expected to crash or fail. The old harness executed 'weston' from Meson, with lots of setup as both command line options and environment variables. The new harness executes wet_main() instead: the test program itself calls the compositor main function to execute the compositor in-process. Command line arguments are configured in the test program itself, not in meson.build. Environment variables aside, you are able to run a test by simply executing the test program, even if it is a plugin test. The new harness adds a new type of iteration: fixtures. For now, fixtures are used to set up the compositor for tests that need a compositor. If necessary, a fixture setup may include a data array of arbitrary type for executing the test series for each element in the array. This will be most useful for running screenshooting tests with both Pixman- and GL-renderers. The new harness outputs TAP formatted results into stdout. Meson is not switched to consume TAP yet though, because it would require a Meson version requirement bump and would not have any benefits at this time. OTOH outputting TAP is trivial and sets up a clear precedent of random test chatter belonging to stderr. This commit migrates only few tests to actually make use of the new features: roles is a basic client test, subsurface-shot is a client test that demonstrates the fixture array, and plugin-registry is a plugin test. The rest of the tests will be migrated later. Once all tests are migrated, we can remove the test-specific setup from meson.build, leaving only the actual build instructions in there. The not migrated tests and stand-alone tests suffer only a minor change: they no longer fork() for each TEST(), otherwise they keep running as before. Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.com>
5 years ago
asprintf(&tmp, "--height=%d", setup->height);
prog_args_take(&args, tmp);
}
tests: thread-based client harness This replaces the old test harness with a new one. The old harness relied on fork()'ing each test which makes tests independent, but makes debugging them harder. The new harness runs client code in a thread instead of a new process. A side-effect of not fork()'ing anymore is that any failure will stop running a test series short. Fortunately we do not have any tests that are expected to crash or fail. The old harness executed 'weston' from Meson, with lots of setup as both command line options and environment variables. The new harness executes wet_main() instead: the test program itself calls the compositor main function to execute the compositor in-process. Command line arguments are configured in the test program itself, not in meson.build. Environment variables aside, you are able to run a test by simply executing the test program, even if it is a plugin test. The new harness adds a new type of iteration: fixtures. For now, fixtures are used to set up the compositor for tests that need a compositor. If necessary, a fixture setup may include a data array of arbitrary type for executing the test series for each element in the array. This will be most useful for running screenshooting tests with both Pixman- and GL-renderers. The new harness outputs TAP formatted results into stdout. Meson is not switched to consume TAP yet though, because it would require a Meson version requirement bump and would not have any benefits at this time. OTOH outputting TAP is trivial and sets up a clear precedent of random test chatter belonging to stderr. This commit migrates only few tests to actually make use of the new features: roles is a basic client test, subsurface-shot is a client test that demonstrates the fixture array, and plugin-registry is a plugin test. The rest of the tests will be migrated later. Once all tests are migrated, we can remove the test-specific setup from meson.build, leaving only the actual build instructions in there. The not migrated tests and stand-alone tests suffer only a minor change: they no longer fork() for each TEST(), otherwise they keep running as before. Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.com>
5 years ago
if (setup->scale != 1) {
asprintf(&tmp, "--scale=%d", setup->scale);
prog_args_take(&args, tmp);
}
if (setup->transform != WL_OUTPUT_TRANSFORM_NORMAL) {
asprintf(&tmp, "--transform=%s",
transform_to_str(setup->transform));
prog_args_take(&args, tmp);
}
tests: thread-based client harness This replaces the old test harness with a new one. The old harness relied on fork()'ing each test which makes tests independent, but makes debugging them harder. The new harness runs client code in a thread instead of a new process. A side-effect of not fork()'ing anymore is that any failure will stop running a test series short. Fortunately we do not have any tests that are expected to crash or fail. The old harness executed 'weston' from Meson, with lots of setup as both command line options and environment variables. The new harness executes wet_main() instead: the test program itself calls the compositor main function to execute the compositor in-process. Command line arguments are configured in the test program itself, not in meson.build. Environment variables aside, you are able to run a test by simply executing the test program, even if it is a plugin test. The new harness adds a new type of iteration: fixtures. For now, fixtures are used to set up the compositor for tests that need a compositor. If necessary, a fixture setup may include a data array of arbitrary type for executing the test series for each element in the array. This will be most useful for running screenshooting tests with both Pixman- and GL-renderers. The new harness outputs TAP formatted results into stdout. Meson is not switched to consume TAP yet though, because it would require a Meson version requirement bump and would not have any benefits at this time. OTOH outputting TAP is trivial and sets up a clear precedent of random test chatter belonging to stderr. This commit migrates only few tests to actually make use of the new features: roles is a basic client test, subsurface-shot is a client test that demonstrates the fixture array, and plugin-registry is a plugin test. The rest of the tests will be migrated later. Once all tests are migrated, we can remove the test-specific setup from meson.build, leaving only the actual build instructions in there. The not migrated tests and stand-alone tests suffer only a minor change: they no longer fork() for each TEST(), otherwise they keep running as before. Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.com>
5 years ago
if (setup->config_file) {
asprintf(&tmp, "--config=%s", setup->config_file);
prog_args_take(&args, tmp);
free(setup->config_file);
tests: thread-based client harness This replaces the old test harness with a new one. The old harness relied on fork()'ing each test which makes tests independent, but makes debugging them harder. The new harness runs client code in a thread instead of a new process. A side-effect of not fork()'ing anymore is that any failure will stop running a test series short. Fortunately we do not have any tests that are expected to crash or fail. The old harness executed 'weston' from Meson, with lots of setup as both command line options and environment variables. The new harness executes wet_main() instead: the test program itself calls the compositor main function to execute the compositor in-process. Command line arguments are configured in the test program itself, not in meson.build. Environment variables aside, you are able to run a test by simply executing the test program, even if it is a plugin test. The new harness adds a new type of iteration: fixtures. For now, fixtures are used to set up the compositor for tests that need a compositor. If necessary, a fixture setup may include a data array of arbitrary type for executing the test series for each element in the array. This will be most useful for running screenshooting tests with both Pixman- and GL-renderers. The new harness outputs TAP formatted results into stdout. Meson is not switched to consume TAP yet though, because it would require a Meson version requirement bump and would not have any benefits at this time. OTOH outputting TAP is trivial and sets up a clear precedent of random test chatter belonging to stderr. This commit migrates only few tests to actually make use of the new features: roles is a basic client test, subsurface-shot is a client test that demonstrates the fixture array, and plugin-registry is a plugin test. The rest of the tests will be migrated later. Once all tests are migrated, we can remove the test-specific setup from meson.build, leaving only the actual build instructions in there. The not migrated tests and stand-alone tests suffer only a minor change: they no longer fork() for each TEST(), otherwise they keep running as before. Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.com>
5 years ago
} else {
prog_args_take(&args, strdup("--no-config"));
}
ctmp = renderer_to_arg(setup->backend, setup->renderer);
if (ctmp)
prog_args_take(&args, strdup(ctmp));
asprintf(&tmp, "--shell=%s", shell_to_str(setup->shell));
prog_args_take(&args, tmp);
if (setup->logging_scopes) {
asprintf(&tmp, "--logger-scopes=%s", setup->logging_scopes);
prog_args_take(&args, tmp);
}
if (setup->xwayland)
prog_args_take(&args, strdup("--xwayland"));
wet_testsuite_data_set(data);
prog_args_save(&args);
ret = wet_main(args.argc, args.argv);
prog_args_fini(&args);
/* We acquired a lock (if this is a DRM-backend test) and now we can
* close its fd and release it, as it has already been run. */
if (lock_fd != -1)
close(lock_fd);
tests: thread-based client harness This replaces the old test harness with a new one. The old harness relied on fork()'ing each test which makes tests independent, but makes debugging them harder. The new harness runs client code in a thread instead of a new process. A side-effect of not fork()'ing anymore is that any failure will stop running a test series short. Fortunately we do not have any tests that are expected to crash or fail. The old harness executed 'weston' from Meson, with lots of setup as both command line options and environment variables. The new harness executes wet_main() instead: the test program itself calls the compositor main function to execute the compositor in-process. Command line arguments are configured in the test program itself, not in meson.build. Environment variables aside, you are able to run a test by simply executing the test program, even if it is a plugin test. The new harness adds a new type of iteration: fixtures. For now, fixtures are used to set up the compositor for tests that need a compositor. If necessary, a fixture setup may include a data array of arbitrary type for executing the test series for each element in the array. This will be most useful for running screenshooting tests with both Pixman- and GL-renderers. The new harness outputs TAP formatted results into stdout. Meson is not switched to consume TAP yet though, because it would require a Meson version requirement bump and would not have any benefits at this time. OTOH outputting TAP is trivial and sets up a clear precedent of random test chatter belonging to stderr. This commit migrates only few tests to actually make use of the new features: roles is a basic client test, subsurface-shot is a client test that demonstrates the fixture array, and plugin-registry is a plugin test. The rest of the tests will be migrated later. Once all tests are migrated, we can remove the test-specific setup from meson.build, leaving only the actual build instructions in there. The not migrated tests and stand-alone tests suffer only a minor change: they no longer fork() for each TEST(), otherwise they keep running as before. Signed-off-by: Pekka Paalanen <pekka.paalanen@collabora.com>
5 years ago
return ret;
}
static void
write_cfg(va_list entry_list, FILE *weston_ini)
{
char *entry = va_arg(entry_list, char *);
int ret;
assert(entry);
while (entry) {
ret = fprintf(weston_ini, "%s\n", entry);
assert(ret >= 0);
free(entry);
entry = va_arg(entry_list, char *);
}
}
static FILE *
open_ini_file(struct compositor_setup *setup)
{
char *wd, *tmp_path = NULL;
FILE *weston_ini = NULL;
assert(!setup->config_file);
wd = realpath(".", NULL);
assert(wd);
if (asprintf(&tmp_path, "%s/%s.ini", wd, setup->testset_name) == -1) {
fprintf(stderr, "Fail formating Weston.ini file name.\n");
goto out;
}
weston_ini = fopen(tmp_path, "w");
assert(weston_ini);
setup->config_file = tmp_path;
out:
free(wd);
return weston_ini;
}
void
weston_ini_setup_(struct compositor_setup *setup, ...)
{
FILE *weston_ini = NULL;
int ret;
va_list entry_list;
weston_ini = open_ini_file(setup);
assert(weston_ini);
va_start(entry_list, setup);
write_cfg(entry_list, weston_ini);
va_end(entry_list);
ret = fclose(weston_ini);
assert(ret != EOF);
}
char *
cfgln(const char *fmt, ...)
{
char *str;
int ret;
va_list ap;
va_start(ap, fmt);
ret = vasprintf(&str, fmt, ap);
assert(ret >= 0);
va_end(ap);
return str;
}