[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]
Re: [tor-bugs] #23061 [Core Tor/Tor]: crypto_rand_double() should produce all possible outputs on platforms with 32-bit int
#23061: crypto_rand_double() should produce all possible outputs on platforms with
32-bit int
-------------------------------------------------+-------------------------
Reporter: teor | Owner: nickm
Type: defect | Status:
| needs_review
Priority: Medium | Milestone: Tor:
| 0.3.2.x-final
Component: Core Tor/Tor | Version: Tor:
| 0.2.2.14-alpha
Severity: Normal | Resolution:
Keywords: tor-relay, security-low, privcount, | Actual Points: 0.5
031-backport, 030-backport, 029-backport, 028 |
-backport-maybe, 027-backport-maybe, 026 |
-backport-maybe, review-group-22 |
Parent ID: | Points: 0.1
Reviewer: | Sponsor:
| SponsorQ
-------------------------------------------------+-------------------------
Comment (by yawning):
Replying to [comment:21 nickm]:
> but can we do better?
On a more serious attempt at giving an answer, as a non-portable "better"
(that I still think is largely pointless), you can probably use a variant
of my solution with `long double` and cast to `double`.
Note that the behavior of such a routine is entirely compiler and
architecture dependent, and it'll probably be a lot slower when it works.
If it were up to me, I would fully document the behavior and be done with
it, under the assumption that anyone doing something that requires a less
leaky floating point abstraction can either scale their values correctly
(and use integer math) or write their own sampler.
--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/23061#comment:29>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online
_______________________________________________
tor-bugs mailing list
tor-bugs@xxxxxxxxxxxxxxxxxxxx
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs