[issue26934] android: test_faulthandler fails

Changes by Xavier de Gaye : -- status: open -> closed ___ Python tracker ___ ___

[issue26934] android: test_faulthandler fails

Roundup Robot added the comment: New changeset 2639afcedaad by Xavier de Gaye in branch '3.6': Issue #26934: Handle _ANDROID_API_LEVEL is None on Windows https://hg.python.org/cpython/rev/2639afcedaad New changeset 4012e28539c4 by Xavier de Gaye in branch 'default': Issue #26934: Merge 3.6

[issue26934] android: test_faulthandler fails

SilentGhost added the comment: Buildbots fail since _ANDROID_API_LEVEL could end up being None -- nosy: +SilentGhost status: closed -> open ___ Python tracker

[issue26934] android: test_faulthandler fails

Changes by Xavier de Gaye : -- resolution: -> fixed stage: commit review -> resolved status: open -> closed ___ Python tracker ___

[issue26934] android: test_faulthandler fails

Xavier de Gaye added the comment: New changeset f37ac1a003f3 in branch 3.6. New changeset 892f13827219 in branch default. -- ___ Python tracker ___

[issue26934] android: test_faulthandler fails

Roundup Robot added the comment: New changeset 66aac9676a28 by Xavier de Gaye in branch 'default': Issue #26934: Fix test_faulthandler on Android where raise() exits with 0, https://hg.python.org/cpython/rev/66aac9676a28 -- nosy: +python-dev ___

[issue26934] android: test_faulthandler fails

STINNER Victor added the comment: Oh, you didn't push your patch yet? Go ahead. buggy_raise_5.patch LGTM and is better than perfect :-) -- ___ Python tracker

[issue26934] android: test_faulthandler fails

Serhiy Storchaka added the comment: LGTM. -- nosy: +serhiy.storchaka ___ Python tracker ___ ___

[issue26934] android: test_faulthandler fails

Xavier de Gaye added the comment: New patch using a plain 'requires_raise' decorator without argument. -- Added file: http://bugs.python.org/file45459/buggy_raise_5.patch ___ Python tracker

[issue26934] android: test_faulthandler fails

Xavier de Gaye added the comment: > Thanks for your patience :-) No problem, your suggestions and code reviews are always very much welcome :) I will push this patch with the other Android patches after 3.6 is released. The patch at issue 26936 will also use @requires_android_level.

[issue26934] android: test_faulthandler fails

STINNER Victor added the comment: buggy_raise_4.patch LGTM. Thanks for your patience :-) Maybe later you can also modify existing tests to use the new @requires_android_level? -- ___ Python tracker

[issue26934] android: test_faulthandler fails

Xavier de Gaye added the comment: New patch. -- Added file: http://bugs.python.org/file45404/buggy_raise_4.patch ___ Python tracker ___

[issue26934] android: test_faulthandler fails

STINNER Victor added the comment: buggy_raise_3.patch LGTM but I don't think that it's worth it to add @requires_raise to support directly. I suggest to only add it to test_faulthandler.py. -- ___ Python tracker

[issue26934] android: test_faulthandler fails

Xavier de Gaye added the comment: New patch following Victor code review. -- Added file: http://bugs.python.org/file45380/buggy_raise_3.patch ___ Python tracker

[issue26934] android: test_faulthandler fails

Xavier de Gaye added the comment: Thanks for the suggestion Victor. The decorator may be used in other tests as well, and referring to the concern raised by Chi Hsuan in msg280147 and my answer, it makes it more clear that the build time _android_api_level should not be used in the standard

[issue26934] android: test_faulthandler fails

STINNER Victor added the comment: Can you try to write a decorator to not suplicate the same long skipIf()? -- ___ Python tracker ___

[issue26934] android: test_faulthandler fails

Xavier de Gaye added the comment: The problem: raise() does not cause the program to exit in a Python interactive session and a spawned Python process exits with 0. Android fixed this problem at API level 24 (Android 6.0 aka Marshmallow). The attached patch fixes this (tested at API levels 21

[issue26934] android: test_faulthandler fails

Xavier de Gaye added the comment: root@generic_x86:/data/local/tmp # python Python 3.6.0a0 (default:eee959fee5f5+, May 14 2016, 13:43:41) [GCC 4.9 20150123 (prerelease)] on linux Type "help", "copyright", "credits" or "license" for more information. >>> import faulthandler >>>

[issue26934] android: test_faulthandler fails

Chi Hsuan Yen added the comment: > Please try without "sigaction(SIGSEGV, NULL, NULL);". Does the program still > quit? Nope. int main() { raise(SIGSEGV); } Still exits normally with 0. > Does faulthandler._read_null() crash Python? Yes. int main() { int *x = 0; int y = *x;

[issue26934] android: test_faulthandler fails

STINNER Victor added the comment: Does faulthandler._read_null() crash Python? If yes, I should patch test_faulthandler to replace all _sigsegv() with faulthandler._read_null(). I was too lazy to do that because currently the unit tests checks the exact error message, whereas _read_null()

[issue26934] android: test_faulthandler fails

STINNER Victor added the comment: > This is due to something mysterious in Android's bionic & kernel. For the > following C program: > (...) Please try without "sigaction(SIGSEGV, NULL, NULL);". Does the program still quit? -- ___ Python tracker

[issue26934] android: test_faulthandler fails

Chi Hsuan Yen added the comment: This is due to something mysterious in Android's bionic & kernel. For the following C program: #include #include #include int main() { sigaction(SIGSEGV, NULL, NULL); raise(SIGSEGV); printf("Good evening!\n"); return 0; } Compiled with:

[issue26934] android: test_faulthandler fails

Xavier de Gaye added the comment: The gdb backtrace of running interactively 'faulthandler._sigsegv()'. After the SIGSEGV and the last gdb 'continue', python is still alive with a '>>>' prompt and ready for the next user input: (gdb) b faulthandler_sigsegv Breakpoint 1 at 0xb752159f: file

[issue26934] android: test_faulthandler fails

Xavier de Gaye added the comment: Reproducing the code that is executed on the first failure (test_disable) gives: On linux: $ python Python 3.6.0a0 (default:47fa003aa9f1, Feb 24 2016, 13:09:02) [GCC 5.3.0] on linux Type "help", "copyright", "credits" or "license" for more information. >>>

[issue26934] android: test_faulthandler fails

Terry J. Reedy added the comment: The 8 failures are all because exitcode is 0 when it should not be. self.assertNotEqual(exitcode, 0) AssertionError: 0 == 0 -- nosy: +terry.reedy ___ Python tracker

[issue26934] android: test_faulthandler fails

New submission from Xavier de Gaye: test_faulthandler fails on an android emulator running an x86 system image at API level 21. See the attached test_output.txt file. -- components: Cross-Build, Library (Lib) files: test_output.txt messages: 264735 nosy: Alex.Willmer, haypo, xdegaye