Change by Terry J. Reedy : ---------- versions: -Python 3.5, Python 3.6 _______________________________________ Python tracker https://bugs.python.org/issue37205 _______________________________________

Mariatta Wijaya added the comment:

can you explain why you removed 3.5 and 3.6 from the versions list?

3.5 and 3.6 are closed for regular bug fix maintenance. We're only fixing issues in 3.7 and 3.8 now. Only security fixes will be applied to 3.5 or 3.6, and this issue is not considered a security issue. More details in https://devguide.python.org/#status-of-python-branches ---------- nosy: +Mariatta _______________________________________ Python tracker https://bugs.python.org/issue37205 _______________________________________

STINNER Victor added the comment: Since I added time.perf_counter_ns() to Python 3.7, it would be acceptable to reduce the "t0" variable and suggest to use time.perf_counter_ns() instead of time.perf_counter() for best precision. ---------- _______________________________________ Python tracker https://bugs.python.org/issue37205 _______________________________________

Eryk Sun added the comment:

suggest to use time.perf_counter_ns() instead of time.perf_counter() for best precision.

QPC typically has a frequency of 1e7, which requires 24 bits for the fraction of a second. So a system can be up for years before the 53-bit precision of a float is an issue. What am I missing? ---------- nosy: +eryksun _______________________________________ Python tracker https://bugs.python.org/issue37205 _______________________________________

Daniel Hrisca added the comment: Under the hood perf_counter_ns already uses the two winapi calls (see https://github.com/python/cpython/blob/41761933c1c30bb6003b65eef1ba23a83db4e...) just that it also sets up a static variable as a time origin which makes it process-wide instead of system-wide ---------- _______________________________________ Python tracker https://bugs.python.org/issue37205 _______________________________________

STINNER Victor added the comment: I wrote PR 23284 to make time.perf_counter() on Windows and time.monotonic() on macOS system-wide (remove the offset computed at startup). I also suggested the usage of the _ns() variant functions in the documentation. ---------- _______________________________________ Python tracker https://bugs.python.org/issue37205 _______________________________________

Change by STINNER Victor : ---------- pull_requests: +22205 pull_request: https://github.com/python/cpython/pull/23314 _______________________________________ Python tracker https://bugs.python.org/issue37205 _______________________________________

STINNER Victor added the comment: I close the issue, it's now fixed in master. Thanks for the bug report Ken Healy. While Python 3.8 and 3.9 are also affected, I prefer to leave them unchanged to avoid bad surprises during a minor update (3.x.y bugfix release). If you disagree, please speak up! The regression was introduced in Python 3.7 by: commit bdaeb7d237462a629e6c85001317faa85f94a0c6 Author: Victor Stinner Date: Mon Oct 16 08:44:31 2017 -0700 bpo-31773: _PyTime_GetPerfCounter() uses _PyTime_t (GH-3983) * Rewrite win_perf_counter() to only use integers internally. * Add _PyTime_MulDiv() which compute "ticks * mul / div" in two parts (int part and remaining) to prevent integer overflow. * Clock frequency is checked at initialization for integer overflow. * Enhance also pymonotonic() to reduce the precision loss on macOS (mach_absolute_time() clock). This change was related to the implementation of the PEP 564. If you are curious about this work, I wrote two articles: * https://vstinner.github.io/python37-pep-564-nanoseconds.html * https://vstinner.github.io/python37-perf-counter-nanoseconds.html * https://vstinner.github.io/pytime.html ---------- resolution: -> fixed stage: patch review -> resolved status: open -> closed _______________________________________ Python tracker https://bugs.python.org/issue37205 _______________________________________

Change by Eryk Sun : ---------- Removed message: https://bugs.python.org/msg380944 _______________________________________ Python tracker https://bugs.python.org/issue37205 _______________________________________