RandomPool initialization is slow on Windows

Bug #248522 reported by A.M. Kuchling on 2008-07-14
2
Affects Status Importance Assigned to Milestone
Python-Crypto
Undecided
John A Meinel

Bug Description

See discussion in http://article.gmane.org/gmane.comp.version-control.bazaar-ng.general/43874 .

Excerpt: In RandomPool.__init__() ...

However, time.time() on windows only has a resolution of ~15ms. So
getting 100 new ticks takes... 15ms * 100 = 1.5s (no surprise there).

It turns out that on *windows* you should be using time.clock() which
uses a high resolution timer. (On Linux, time.clock() measures CPU time,
which isn't good for this particular use.)

John A Meinel (jameinel) wrote :

I just put together a patch for this. I don't know how it fits with the overall code base, but at least it is consistent within randompool.py

Changed in pycrypto:
assignee: nobody → jameinel
status: New → Confirmed
John A Meinel (jameinel) wrote :

Just uploading the same change as a patch, since I don't know if a branch is preferred.

Dwayne Litzenberger (dlitz) wrote :

I'm not going to bother fixing this, since RandomPool is now deprecated anyway.

See Bug #249765 and http://lists.dlitz.net/pipermail/pycrypto/2008q3/000020.html.

Changed in pycrypto:
status: Confirmed → Won't Fix
To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Other bug subscribers