msg239890 - (view) |
Author: STINNER Victor (vstinner) *  |
Date: 2015-04-02 08:15 |
Attached patch setup an exception handler on Windows. I wrote it when investigating a failure of test_exceptions related to the issue #22977. faulthandler was not trigerred while the program crashed. I didn't test the patch yet (I tested a previous attempt, a little bit different). Maybe we can avoid setting signal handlers on Windows if an exception handler is set? |
|
|
msg240229 - (view) |
Author: STINNER Victor (vstinner) *  |
Date: 2015-04-07 20:42 |
More complete (and working) patch. Most unit tests pass, but two unit tests must be updated to match the new error message. |
|
|
msg262255 - (view) |
Author: Roundup Robot (python-dev)  |
Date: 2016-03-23 10:23 |
New changeset b2f7bb63377b by Victor Stinner in branch 'default': Issue #23848: Expose _Py_DumpHexadecimal() https://hg.python.org/cpython/rev/b2f7bb63377b New changeset b114dbbe2d31 by Victor Stinner in branch 'default': faulthandler: add Windows exception handler https://hg.python.org/cpython/rev/b114dbbe2d31 |
|
|
msg262266 - (view) |
Author: Roundup Robot (python-dev)  |
Date: 2016-03-23 13:46 |
New changeset efcc48cd5bfb by Victor Stinner in branch 'default': faulthandler: only log fatal exceptions https://hg.python.org/cpython/rev/efcc48cd5bfb |
|
|
msg262269 - (view) |
Author: Roundup Robot (python-dev)  |
Date: 2016-03-23 14:20 |
New changeset e6f00778d61f by Victor Stinner in branch 'default': Issue #23848: Try to fix test_faulthandler on ARM https://hg.python.org/cpython/rev/e6f00778d61f |
|
|
msg262270 - (view) |
Author: STINNER Victor (vstinner) *  |
Date: 2016-03-23 14:22 |
The change b114dbbe2d31 introduced a regression on the ARM buildbot. I hope that it's fixed by the change e6f00778d61f. http://buildbot.python.org/all/builders/ARMv7%20Ubuntu%203.x/builds/3800/steps/test/logs/stdio 1:06:35 [400/400/1] test_faulthandler Timeout (1:00:00)! Thread 0x40102110 (most recent call first): File "/ssd/buildbot/buildarea/3.x.gps-ubuntu-exynos5-armv7l/build/Lib/subprocess.py", line 1608 in _try_wait File "/ssd/buildbot/buildarea/3.x.gps-ubuntu-exynos5-armv7l/build/Lib/subprocess.py", line 1658 in wait File "/ssd/buildbot/buildarea/3.x.gps-ubuntu-exynos5-armv7l/build/Lib/subprocess.py", line 1002 in __exit__ File "/ssd/buildbot/buildarea/3.x.gps-ubuntu-exynos5-armv7l/build/Lib/test/test_faulthandler.py", line 64 in get_output File "/ssd/buildbot/buildarea/3.x.gps-ubuntu-exynos5-armv7l/build/Lib/test/test_faulthandler.py", line 108 in check_error File "/ssd/buildbot/buildarea/3.x.gps-ubuntu-exynos5-armv7l/build/Lib/test/test_faulthandler.py", line 115 in check_fatal_error File "/ssd/buildbot/buildarea/3.x.gps-ubuntu-exynos5-armv7l/build/Lib/test/test_faulthandler.py", line 241 in test_stack_overflow ... |
|
|
msg262271 - (view) |
Author: STINNER Victor (vstinner) *  |
Date: 2016-03-23 14:37 |
Pass succeeded again on ARM, I close the issue. |
|
|
msg262289 - (view) |
Author: Roundup Robot (python-dev)  |
Date: 2016-03-23 17:38 |
New changeset 3247f8df5514 by Victor Stinner in branch 'default': Issue #23848: Fix usage of _Py_DumpDecimal() https://hg.python.org/cpython/rev/3247f8df5514 |
|
|