[tor-bugs] #17549 [Tor]: tortls.c compile error w/Tor v0.2.8.0-alpha-dev and OpenSSL 1.1.0-dev (git-19e10f95c105a698) against SSL_Dev
Tor Bug Tracker & Wiki
blackhole at torproject.org
Tue Nov 10 14:19:41 UTC 2015
#17549: tortls.c compile error w/Tor v0.2.8.0-alpha-dev and OpenSSL 1.1.0-dev (git-
19e10f95c105a698) against SSL_Dev
---------------------------------------------+-----------------------------
Reporter: yancm | Owner:
Type: defect | Status:
Priority: Medium | needs_revision
Component: Tor | Milestone: Tor:
Severity: Normal | 0.2.8.x-final
Keywords: tor-core, openssl, 027-backport | Version: Tor:
Parent ID: | unspecified
Sponsor: | Resolution:
| Actual Points:
| Points:
---------------------------------------------+-----------------------------
Changes (by nickm):
* status: needs_review => needs_revision
Comment:
Looks good, but I bet this will break with libressl. That's not _too_
horrible to do in an alpha, but we should avoid it if we can.
Also, hmm. I generally think it's better to make #ifdefs not be small
pieces of functions, but rather independent functions. So what if instead
we do something like
{{{
#if (it's an old version)
#define OpenSSL_version_num() SSLeay()
#define OpenSSL_version(v) SSLeay_version(v)
#define OPENSSL_VERSION SSLEAY_VERSION
...
#endif
}}}
and then our functions can stay nice and pretty?
> I'm getting tempted to start saying that chasing OpenSSL 1.1.0-dev is at
people's own risk, and they should be prepared to patch trivial issues
like this.
Well, I'll be a bit more moderate, but basically agree: It's cool if
somebody wants to run openssl-1.1.0-dev, but they should expect it to
break Tor pretty frequently, and they should expect that we'll fix this
kind of issue as we get around to it and no sooner. Otherwise we couldn't
try to keep up with this as all.
--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/17549#comment:5>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online
More information about the tor-bugs
mailing list