Re: [E-devel] efl test suite timeouts in 1.26?
On Thu, 6 Jan 2022 19:32:59 -0800 Ross Vandegrift said: > On Tue, Jan 04, 2022 at 08:24:02AM +, Carsten Haitzler wrote: > > I don't see the timeouts here... this relies on check as the infra - and > > check does all the stuff before a fork happens to run the actual test... > > and check is a local system dependency... so something either really screwy > > about your os, build env or check? > > I played with this some more - it's related to check's forking on debian > unstable. The issue doesn't happen in debian stable, or if CK_FORK=no. check > is at 0.15.2 in both though. > > > [0/1] Running all tests. > > 1/39 eo-mixins OK 0.09s > > 2/39 eo-mixins-dbg OK 0.09s > > 3/39 eo-interface OK 0.09s > > 4/39 eo-interface-dbg OK 0.08s > > 5/39 eo-function-overrides OK 0.08s > > 6/39 eo-function-overrides-dbg OK 0.08s > > 7/39 eo-composite-objects OK 0.08s > > 8/39 eo-composite-objects-dbg OK 0.08s > > 9/39 eo-children OK 0.07s > > ... etc. > > Once past the timeouts, I am seeing a few things that could be issues: > > $ CK_VERBOSITY=verbose ./src/tests/eet/eet_suite "Eet File" > Running suite(s): Eet > 50%: Checks: 4, Failures: 0, Errors: 2 > ../src/tests/eet/eet_test_file.c:117:P:Eet File:eet_test_file_simple_write:0: > Passed ../src/tests/eet/eet_test_file.c:175:E:Eet File:eet_test_file_data:0: > (after this point) Received signal 11 (Segmentation > fault) ../src/tests/eet/eet_test_file.c:371:P:Eet > File:eet_test_file_data_dump:0: Passed ../src/tests/eet/eet_test_file.c: > 419:E:Eet File:eet_test_file_fp:0: (after this point) Received signal 11 > (Segmentation fault) > > $ CK_VERBOSITY=vebose ./src/tests/elementary/elementary_suite 2> /dev/null | > grep :F: ../src/tests/elementary/suite_helpers.c: > 604:F:elm_check:elm_test_check_callbacks:1: Assertion '*called == 0' failed: > *called == 1, 0 == 0 ../src/tests/elementary/elm_test_widget_focus.c: > 194:F:elm_widget_focus:elm_test_widget_focus_simple_widget:0: Assertion > 'flag_focused == ((Eina_Bool)1)' failed: flag_focused == 0, ((Eina_Bool)1) == > 1 > > $ CK_VERBOSITY=verbose ./src/tests/elementary/efl_ui_suite 2> /dev/null | > grep :F: ../src/tests/elementary/efl_ui_test_focus.c: > 454:F:efl_ui_focus:invalid_args_check:0: Assertion 'expect_error_start == > ((Eina_Bool)1)' failed: expect_error_start == 0, ((Eina_Bool)1) == 1 > > Haven't had time to dig further. fixed them in git in the meantime :) -- - Codito, ergo sum - "I code, therefore I am" -- Carsten Haitzler - ras...@rasterman.com ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] efl test suite timeouts in 1.26?
On Tue, Jan 04, 2022 at 08:24:02AM +, Carsten Haitzler wrote: > I don't see the timeouts here... this relies on check as the infra - and check > does all the stuff before a fork happens to run the actual test... and check > is > a local system dependency... so something either really screwy about your os, > build env or check? I played with this some more - it's related to check's forking on debian unstable. The issue doesn't happen in debian stable, or if CK_FORK=no. check is at 0.15.2 in both though. > [0/1] Running all tests. > 1/39 eo-mixins OK 0.09s > 2/39 eo-mixins-dbg OK 0.09s > 3/39 eo-interface OK 0.09s > 4/39 eo-interface-dbg OK 0.08s > 5/39 eo-function-overrides OK 0.08s > 6/39 eo-function-overrides-dbg OK 0.08s > 7/39 eo-composite-objects OK 0.08s > 8/39 eo-composite-objects-dbg OK 0.08s > 9/39 eo-children OK 0.07s > ... etc. Once past the timeouts, I am seeing a few things that could be issues: $ CK_VERBOSITY=verbose ./src/tests/eet/eet_suite "Eet File" Running suite(s): Eet 50%: Checks: 4, Failures: 0, Errors: 2 ../src/tests/eet/eet_test_file.c:117:P:Eet File:eet_test_file_simple_write:0: Passed ../src/tests/eet/eet_test_file.c:175:E:Eet File:eet_test_file_data:0: (after this point) Received signal 11 (Segmentation fault) ../src/tests/eet/eet_test_file.c:371:P:Eet File:eet_test_file_data_dump:0: Passed ../src/tests/eet/eet_test_file.c:419:E:Eet File:eet_test_file_fp:0: (after this point) Received signal 11 (Segmentation fault) $ CK_VERBOSITY=vebose ./src/tests/elementary/elementary_suite 2> /dev/null | grep :F: ../src/tests/elementary/suite_helpers.c:604:F:elm_check:elm_test_check_callbacks:1: Assertion '*called == 0' failed: *called == 1, 0 == 0 ../src/tests/elementary/elm_test_widget_focus.c:194:F:elm_widget_focus:elm_test_widget_focus_simple_widget:0: Assertion 'flag_focused == ((Eina_Bool)1)' failed: flag_focused == 0, ((Eina_Bool)1) == 1 $ CK_VERBOSITY=verbose ./src/tests/elementary/efl_ui_suite 2> /dev/null | grep :F: ../src/tests/elementary/efl_ui_test_focus.c:454:F:efl_ui_focus:invalid_args_check:0: Assertion 'expect_error_start == ((Eina_Bool)1)' failed: expect_error_start == 0, ((Eina_Bool)1) == 1 Haven't had time to dig further. Ross ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
Re: [E-devel] efl test suite timeouts in 1.26?
On Mon, 3 Jan 2022 22:59:00 -0800 Ross Vandegrift said: I don't see the timeouts here... this relies on check as the infra - and check does all the stuff before a fork happens to run the actual test... and check is a local system dependency... so something either really screwy about your os, build env or check? [0/1] Running all tests. 1/39 eo-mixins OK 0.09s 2/39 eo-mixins-dbg OK 0.09s 3/39 eo-interface OK 0.09s 4/39 eo-interface-dbg OK 0.08s 5/39 eo-function-overrides OK 0.08s 6/39 eo-function-overrides-dbg OK 0.08s 7/39 eo-composite-objects OK 0.08s 8/39 eo-composite-objects-dbg OK 0.08s 9/39 eo-children OK 0.07s ... etc. > Hello, > > With 1.26, I'm seeing "ninja test" timeout on every single suite. Except for > the actual failures & errors I'm seeing, running the individual binaries is > fine. There seems to be something weird with forking - I hardcoded "do_fork = > 0;" at src/tests/efl_check.h:359 and most of the timeouts are gone. > > Sample comparison with default meson options: > > Summary of Failures: > > 18/44 eo-suite TIMEOUT 30.11s > 19/44 eo-suite-dbg TIMEOUT 30.10s > 20/44 eo-suite-fallback TIMEOUT 30.09s > 21/44 eina TIMEOUT 30.14s killed by signal > 15 SIGTERM 22/44 efl-suite TIMEOUT 30.01s > 23/44 emile-suite TIMEOUT 30.01s > 24/44 eet-suite TIMEOUT 30.02s exit status 1 > 27/44 eldbus-suite TIMEOUT 30.01s killed by signal > 15 SIGTERM 28/44 ecore-suite TIMEOUT 30.04s > 29/44 efl-app TIMEOUT 30.04s > 30/44 ecore_con-suite TIMEOUT 30.03s > 31/44 eeze-suiteTIMEOUT 30.02s > 32/44 efreet-suite TIMEOUT 30.01s > 33/44 edje-suiteTIMEOUT 30.01s > 37/44 elementary-suite TIMEOUT 30.03s (exit status 255 > or signal 127 SIGinvalid) 38/44 efl-ui-suite TIMEOUT > 30.03s (exit status 255 or signal 127 SIGinvalid) 39/44 > eina_cxx-suiteTIMEOUT 30.03s 40/44 > evas-suiteTIMEOUT 60.02s 41/44 > eolian_cxx-suite TIMEOUT 30.05s 43/44 > eldbus_cxx-suite TIMEOUT 30.04s (exit status 255 or signal > 127 SIGinvalid) 44/44 efl_ui_spec-suite TIMEOUT 60.02s > > Ok: 23 > Expected Fail: 0 > Fail: 0 > Unexpected Pass:0 > Skipped:0 > Timeout:21 > ... > $ time EFL_RUN_IN_TREE=1 ./src/tests/eina/eina_suite > ... > 100%: Checks: 3, Failures: 0, Errors: 0 > 100%: Checks: 11, Failures: 0, Errors: 0 > 100%: Checks: 7, Failures: 0, Errors: 0 > 100%: Checks: 9, Failures: 0, Errors: 0 > 100%: Checks: 3, Failures: 0, Errors: 0 > 100%: Checks: 7, Failures: 0, Errors: 0 > 100%: Checks: 10, Failures: 0, Errors: 0 > 100%: Checks: 3, Failures: 0, Errors: 0 > > real0m0.180s > user0m0.387s > sys 0m0.469s > $ echo $? > 0 > > > Ross > > > ___ > enlightenment-devel mailing list > enlightenment-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/enlightenment-devel > -- - Codito, ergo sum - "I code, therefore I am" -- Carsten Haitzler - ras...@rasterman.com ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
[E-devel] efl test suite timeouts in 1.26?
Hello, With 1.26, I'm seeing "ninja test" timeout on every single suite. Except for the actual failures & errors I'm seeing, running the individual binaries is fine. There seems to be something weird with forking - I hardcoded "do_fork = 0;" at src/tests/efl_check.h:359 and most of the timeouts are gone. Sample comparison with default meson options: Summary of Failures: 18/44 eo-suite TIMEOUT 30.11s 19/44 eo-suite-dbg TIMEOUT 30.10s 20/44 eo-suite-fallback TIMEOUT 30.09s 21/44 eina TIMEOUT 30.14s killed by signal 15 SIGTERM 22/44 efl-suite TIMEOUT 30.01s 23/44 emile-suite TIMEOUT 30.01s 24/44 eet-suite TIMEOUT 30.02s exit status 1 27/44 eldbus-suite TIMEOUT 30.01s killed by signal 15 SIGTERM 28/44 ecore-suite TIMEOUT 30.04s 29/44 efl-app TIMEOUT 30.04s 30/44 ecore_con-suite TIMEOUT 30.03s 31/44 eeze-suiteTIMEOUT 30.02s 32/44 efreet-suite TIMEOUT 30.01s 33/44 edje-suiteTIMEOUT 30.01s 37/44 elementary-suite TIMEOUT 30.03s (exit status 255 or signal 127 SIGinvalid) 38/44 efl-ui-suite TIMEOUT 30.03s (exit status 255 or signal 127 SIGinvalid) 39/44 eina_cxx-suiteTIMEOUT 30.03s 40/44 evas-suiteTIMEOUT 60.02s 41/44 eolian_cxx-suite TIMEOUT 30.05s 43/44 eldbus_cxx-suite TIMEOUT 30.04s (exit status 255 or signal 127 SIGinvalid) 44/44 efl_ui_spec-suite TIMEOUT 60.02s Ok: 23 Expected Fail: 0 Fail: 0 Unexpected Pass:0 Skipped:0 Timeout:21 ... $ time EFL_RUN_IN_TREE=1 ./src/tests/eina/eina_suite ... 100%: Checks: 3, Failures: 0, Errors: 0 100%: Checks: 11, Failures: 0, Errors: 0 100%: Checks: 7, Failures: 0, Errors: 0 100%: Checks: 9, Failures: 0, Errors: 0 100%: Checks: 3, Failures: 0, Errors: 0 100%: Checks: 7, Failures: 0, Errors: 0 100%: Checks: 10, Failures: 0, Errors: 0 100%: Checks: 3, Failures: 0, Errors: 0 real 0m0.180s user 0m0.387s sys 0m0.469s $ echo $? 0 Ross ___ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
