| msg60346 - (view) |
Author: Walter Dörwald (doerwalter) *  |
Date: 2003-06-17 18:17 |
| Python crashes with a floating point exception on SF's Alpha/Linux 2.2 machine in the compile farm: $ ./python Python 2.3b1+ (#2, Jun 17 2003, 10:50:58) [GCC 2.95.4 20011002 (Debian prerelease)] on linux2 Type "help", "copyright", "credits" or "license" for more information. >>> pow(1e200+1j,1e200+1j) Floating point exception $ |
|
|
| msg60347 - (view) |
Author: Martin v. Löwis (loewis) * |
Date: 2003-06-21 13:51 |
| Logged In: YES user_id=21627 Can you tell whether you were compiling with -ieee? |
|
|
| msg60348 - (view) |
Author: Walter Dörwald (doerwalter) * |
Date: 2003-06-23 11:13 |
| Logged In: YES user_id=89016 Doesn't seem that way: gcc -pthread -c -DNDEBUG -g -O3 -Wall -Wstrict-prototypes -I. -I./Include -DPy_BUILD_CORE -o Objects/complexobject.o Objects/complexobject.c Changing OPT in Makefile to include -ieee and recompiling doesn't fix the problem, I still get a "Floating point exception". |
|
|
| msg60349 - (view) |
Author: Martin v. Löwis (loewis) * |
Date: 2003-06-23 21:24 |
| Logged In: YES user_id=21627 Ah, I see: LINUX. This is probably a compiler bug, without even looking at the details. |
|
|
| msg61326 - (view) |
Author: Christian Heimes (christian.heimes) * |
Date: 2008-01-20 18:51 |
| Does the error occur with Python 2.6, too? |
|
|
| msg67345 - (view) |
Author: Ali Polatel (hawking) |
Date: 2008-05-25 12:51 |
| Probably related to this. test_pow fails on alpha with both 2.4.4 and 2.5.2. test test_pow failed -- Traceback (most recent call last): File "/var/tmp/portage/dev-lang/python-2.4.4-r9/work/Python-2.4.4/Lib/test/test_pow.py", line 109, in test_bug705231 eq(pow(a, 1.23e167), 1.0) ValueError: negative number cannot be raised to a fractional power I'll test with 2.6 and report back. |
|
|
| msg70079 - (view) |
Author: Georg Brandl (georg.brandl) * |
Date: 2008-07-20 11:23 |
| Did you get some results? |
|
|
| msg82270 - (view) |
Author: Mark Dickinson (mark.dickinson) * |
Date: 2009-02-16 20:47 |
| Closing this due to lack of feedback. Ali Polatel's report looks like an entirely separate issue; most likely an alpha libm bug. |
|
|