On Thu, May 28, 2015 at 10:00:19AM +0000, Morton, Derek J wrote: > > >-----Original Message----- > >From: Daniel Vetter [mailto:daniel.vetter@xxxxxxxx] On Behalf Of Daniel Vetter > >Sent: Thursday, May 28, 2015 10:05 AM > >To: Morton, Derek J > >Cc: intel-gfx@xxxxxxxxxxxxxxxxxxxxx; Wood, Thomas > >Subject: Re: [PATCH i-g-t v3] lib/tests/igt_segfault.c Add unit test to test segfault handling > > > >On Wed, May 27, 2015 at 04:54:24PM +0100, Derek Morton wrote: > >> Unit test to check a segfaulting subtest is handled correctly. > >> > >> v2: Added script to check subtest results > >> v3: Removed script. Updated test to use fork to monitor return status. > >> > >> Signed-off-by: Derek Morton <derek.j.morton@xxxxxxxxx> > >> --- > >> lib/tests/Makefile.sources | 1 + > >> lib/tests/igt_segfault.c | 139 +++++++++++++++++++++++++++++++++++++++++++++ > >> 2 files changed, 140 insertions(+) > >> create mode 100644 lib/tests/igt_segfault.c > >> > >> diff --git a/lib/tests/Makefile.sources b/lib/tests/Makefile.sources > >> index 10e0617..5fa0b31 100644 > >> --- a/lib/tests/Makefile.sources > >> +++ b/lib/tests/Makefile.sources > >> @@ -8,6 +8,7 @@ check_PROGRAMS = \ > >> igt_simple_test_subtests \ > >> igt_timeout \ > >> igt_invalid_subtest_name \ > >> + igt_segfault \ > >> $(NULL) > >> > >> check_SCRIPTS = \ > >> diff --git a/lib/tests/igt_segfault.c b/lib/tests/igt_segfault.c new > >> file mode 100644 index 0000000..b420b1a > >> --- /dev/null > >> +++ b/lib/tests/igt_segfault.c > >> @@ -0,0 +1,139 @@ > >> +/* > >> + * Copyright (c) 2015 Intel Corporation > >> + * > >> + * 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. > >> + * > >> + * Authors: > >> + * Derek Morton <derek.j.morton@xxxxxxxxx> > >> + * > >> + */ > >> + > >> +/* > >> + * Testcase: Test the framework catches a segfault and returns an error. > >> + * > >> + * 1. Test a crashing simple test is reported. > >> + * 2. Test a crashing subtest is reported. > >> + * 3. Test a crashing subtest following a passing subtest is reported. > >> + * 4. Test a crashing subtest preceeding a passing subtest is reported. > >> + */ > >> + > >> +#include <stdlib.h> > >> +#include <sys/wait.h> > >> +#include <sys/types.h> > >> +#include <assert.h> > >> +#include <errno.h> > >> + > >> +#include "drmtest.h" > >> +#include "igt_core.h" > >> + > >> +/* > >> + * We need to hide assert from the cocci igt test refactor spatch. > >> + * > >> + * IMPORTANT: Test infrastructure tests are the only valid places > >> +where using > >> + * assert is allowed. > >> + */ > >> +#define internal_assert assert > >> + > >> +bool simple; > >> +bool runa; > >> +bool runc; > >> +char test[] = "test"; > >> +char *argv_run[] = { test }; > >> + > >> +static int do_fork(void) > >> +{ > >> + int pid, status; > >> + int argc; > >> + void (*crashme)(void) = NULL; > >> + > >> + switch (pid = fork()) { > >> + case -1: > >> + internal_assert(0); > >> + case 0: > >> + if (simple) { > >> + argc = 1; > >> + igt_simple_init(argc, argv_run); > >> + crashme(); > >> + > >> + igt_exit(); > >> + } else { > >> + > >> + argc = 1; > >> + igt_subtest_init(argc, argv_run); > >> + > >> + if(runa) > >> + igt_subtest("A") > >> + ; > >> + > >> + igt_subtest("B") > >> + crashme(); > >> + > >> + if(runc) > >> + igt_subtest("C") > >> + ; > >> + > >> + igt_exit(); > >> + } > >> + default: > >> + while (waitpid(pid, &status, 0) == -1 && > >> + errno == EINTR) > >> + ; > >> + > >> + if(WIFSIGNALED(status)) > >> + return WTERMSIG(status) + 128; > >> + > >> + return WEXITSTATUS(status); > >> + } > >> +} > >> + > >> +int main(int argc, char **argv) > >> +{ > >> + /* Test Crash in simple test is reported */ > >> + simple = true; > >> + runa=false; > >> + runc=false; > >> + igt_info("Simple test.\n"); > >> + fflush(stdout); > >> + internal_assert(do_fork() == SIGSEGV + 128); > >> + > >> + /* Test crash in a single subtest is reported */ > >> + simple = false; > >> + igt_info("Single subtest.\n"); > >> + fflush(stdout); > >> + internal_assert(do_fork() == SIGSEGV + 128); > > > >This one here fails for me when running make check. I didn't look into details. Also something seemed off with the content-encoding of your mail, at least git apply-mbox complained. I didn't debug but applied it manually. > > > >Also please add igt_segfault to the relevant .gitignore. > >-Daniel > > > > I submitted a separate patch "[PATCH i-g-t] lib/igt_core.c: Flag the test as failing after a segfault" as segfaults were not being flagged as a failure. If that patch has not yet been merged I would expect this test to fail at this point. (I was asked to write this UT when I submitted that patch.) > Git send-email asked about encoding when I submitted the patch. I assumed it was the (c) character in the copyright header. I just selected whatever it recommended. (UTF8 I think). Previous patches have not been a problem. > I will update the .gitignore and resubmit this patch. Indeed. I've applied&pushed that one. -Daniel -- Daniel Vetter Software Engineer, Intel Corporation http://blog.ffwll.ch _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/intel-gfx
