[issue22624] Bogus usage of floatclock in timemodule

[Roundup Robot]

Roundup Robot added the comment:

New changeset 1452520dfe7b by Victor Stinner in branch 'default':
Issue #22624: Python 3 requires clock() to build
https://hg.python.org/cpython/rev/1452520dfe7b

--
nosy: +python-dev

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue22624] Bogus usage of floatclock in timemodule

[STINNER Victor]

STINNER Victor added the comment:

Sorry but the Wii console is not supported. You have to maintain your 
fork/patch.

Python 3 don't support platforms without clock().

--
resolution:  -> wont fix
status: open -> closed

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue22624] Bogus usage of floatclock in timemodule

[Link Mauve]

Changes by Link Mauve :


Added file: http://bugs.python.org/file36911/pyconfig.h

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue22624] Bogus usage of floatclock in timemodule

[STINNER Victor]

STINNER Victor added the comment:

Please attach your pyconfig.h file.

--

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue22624] Bogus usage of floatclock in timemodule

[Link Mauve]

Link Mauve added the comment:

I’m building against the Newlib libc, for the Wii, and it seems the issue 
leading to all of those functions not working is:
(.text.clock+0x18): undefined reference to `_times_r'

Threading is disabled, and reentrant functions as well, which could explain 
that.

But still, when pyconfig.h doesn’t define any HAVE_ macro to get the system 
time, I would expect py_process_time() to return None or something instead of 
failing to build.

--

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue22624] Bogus usage of floatclock in timemodule

[STINNER Victor]

STINNER Victor added the comment:

What is your platform?

time.process_time() supports many functions:

 - GetProcessTimes() (Windows only)
 - clock_gettime(CLOCK_PROF)
 - clock_gettime(CLOCK_PROCESS_CPUTIME_ID)
 - getrusage(RUSAGE_SELF)
 - times()
 - clock()

On POSIX, clock() is always tried as a fallback. Does your platform support at 
least one of these functions?

--

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue22624] Bogus usage of floatclock in timemodule

[STINNER Victor]

STINNER Victor added the comment:

See also the PEP 418:
http://legacy.python.org/dev/peps/pep-0418/#time-process-time

--

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue22624] Bogus usage of floatclock in timemodule

[Antoine Pitrou]

Changes by Antoine Pitrou :


--
nosy: +haypo

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue22624] Bogus usage of floatclock in timemodule

[Link Mauve]

New submission from Link Mauve:

In Modules/timemodule.c, py_process_time() still uses floatclock() even when 
HAVE_CLOCK isn’t defined.

--
components: Build
messages: 229260
nosy: Link Mauve
priority: normal
severity: normal
status: open
title: Bogus usage of floatclock in timemodule
type: compile error
versions: Python 3.5

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com