Commit Graph

1520 Commits

Author SHA1 Message Date
Nick Mathewson
c38fa93ad1 Merge commit 'origin/maint-0.2.1' 2010-04-15 10:35:09 -04:00
Nick Mathewson
6ad09cc6af Fix renegotiation on OpenSSL versions that backport RFC5746.
Our code assumed that any version of OpenSSL before 0.9.8l could not
possibly require SSL_OP_ALLOW_UNSAFE_LEGACY_RENEGOTIATION.  This is
so... except that many vendors have backported the flag from later
versions of openssl when they backported the RFC5476 renegotiation
feature.

The new behavior is particularly annoying to detect.  Previously,
leaving SSL_OP_ALLOW_UNSAFE_LEGACY_RENEGOTIATION unset meant that
clients would fail to renegotiate.  People noticed that one fast!
Now, OpenSSL's RFC5476 support means that clients will happily talk to
any servers there are, but servers won't accept renegotiation requests
from unpatched clients unless SSL_OP_ALLOW_etc is set.  More fun:
servers send back a "no renegotiation for you!" error, which unpatched
clients respond to by stalling, and generally producing no useful
error message.

This might not be _the_ cause of bug 1346, but it is quite likely _a_
cause for bug 1346.
2010-04-13 15:05:03 -04:00
Nick Mathewson
927425150b Merge branch 'asprintf' 2010-04-02 12:30:46 -04:00
Roger Dingledine
625963d92a commit my annotations while i was hunting down the host order bug 2010-03-05 16:04:01 -05:00
Nick Mathewson
897b0ebbac better handle the case where *strp is in asprintf args 2010-02-28 21:46:46 -05:00
Nick Mathewson
da220157a9 Update copyright dates for files not in maint-0.2.1 2010-02-27 17:19:00 -05:00
Nick Mathewson
b006e3279f Merge remote branch 'origin/maint-0.2.1'
Conflicts:
	src/common/test.h
	src/or/test.c
2010-02-27 17:16:31 -05:00
Nick Mathewson
c3e63483b2 Update Tor Project copyright years 2010-02-27 17:14:21 -05:00
Sebastian Hahn
a9802d3322 Zero a cipher completely before freeing it
We used to only zero the first ptrsize bytes of the cipher. Since
cipher is large enough, we didn't zero too many bytes. Discovered
and fixed by ekir. Fixes bug 1254.
2010-02-26 05:47:25 +01:00
Nick Mathewson
f0b5f87eab Add the MIN and MAX macros for platforms that lack them 2010-02-25 16:48:39 -05:00
Nick Mathewson
eb10d441b6 Fix 64-bit printf issues in consensus-bw-weights5-merge.
For my 64-bit Linux system running with GCC 4.4.3-fc12-whatever, you
can't do 'printf("%lld", (int64_t)x);' Instead you need to tell the
compiler 'printf("%lld", (long long int)x);' or else it doesn't
believe the types match.  This is why we added U64_PRINTF_ARG; it
looks like we needed an I64_PRINTF_ARG too.
2010-02-25 16:22:40 -05:00
Nick Mathewson
6fa8dacb97 Add a tor_asprintf() function, and use it in a couple of places.
asprintf() is a GNU extension that some BSDs have picked up: it does a printf
into a newly allocated chunk of RAM.

Our tor_asprintf() differs from standard asprintf() in that:
  - Like our other malloc functions, it asserts on OOM.
  - It works on windows.
  - It always sets its return-field.
2010-02-25 16:09:10 -05:00
Mike Perry
f4d6315afa Remove misc unnecessary newlines found by new check. 2010-02-22 16:52:11 -08:00
Mike Perry
95aad71678 Add %lld compat defines. 2010-02-22 16:52:10 -08:00
Nick Mathewson
c084ae145e Merge remote branch 'sebastian/bug1254' 2010-02-22 12:45:01 -05:00
Nick Mathewson
d35b8dc582 Make expand_filename into a tor_strdup() alias on windows.
On Windows, we don't have a notion of ~ meaning "our homedir", so we
were deliberately using an #ifdef to avoid calling expand_filename()
in multiple places.  This is silly: The right place to turn a function
into a no-op on a single platform is in the function itself, not in
every single call-site.
2010-02-22 12:42:31 -05:00
Sebastian Hahn
f5112fa487 Zero a cipher completely before freeing it
We used to only zero the first ptrsize bytes of the cipher. Since
cipher is large enough, we didn't zero too many bytes. Discovered
and fixed by ekir. Fixes bug 1254.
2010-02-22 11:41:39 +01:00
Roger Dingledine
603432090d fix typo and garbage grammar 2010-02-21 17:18:42 -05:00
Nick Mathewson
391f75d792 Merge remote branch 'sebastian/bug1143' 2010-02-19 16:58:24 -05:00
Sebastian Hahn
408a828b1f Make the DNSPort option work with libevent 2.x
We need to use evdns_add_server_port_with_base() when configuring
our DNS listener, because libevent segfaults otherwise. Add a macro
in compat_libevent.h to pick the correct implementation depending
on the libevent version.

Fixes bug 1143, found by SwissTorExit
2010-02-19 22:36:53 +01:00
Nick Mathewson
715f104eeb Merge remote branch 'origin/maint-0.2.1'
Conflicts:
	ChangeLog
	configure.in
	contrib/tor-mingw.nsi.in
	src/win32/orconfig.h
2010-02-18 12:01:56 -05:00
Sebastian Hahn
c2c3a5a3f5 Fix compile 2010-02-18 13:08:57 +01:00
Nick Mathewson
e861b3be88 Even more conservative option-setting for SSL renegotiation.
This time, set the SSL3_FLAGS_ALLOW_UNSAFE_RENEGOTIATION flag on every
version before OpenSSL 0.9.8l.  I can confirm that the option value (0x0010)
wasn't reused until OpenSSL 1.0.0beta3.
2010-02-17 23:55:03 -05:00
Nick Mathewson
79bdfb63e9 Remove the --enable-iphone option as needless.
On or-talk, Marco Bonetti reports that recent iPhone SDKs build
Tor fine without it.
2010-02-12 23:06:05 -05:00
Sebastian Hahn
fe18275563 Add Windows version detection for Vista and 7
Vista is Windows 6.0, and 7 is Windows 6.1. Fixes bug 1097.

Also fix a coding style violation.
2010-02-10 08:40:44 +01:00
Nick Mathewson
c0d682686a Make tor_addr_copy() conform to memcpy requirements
The src and dest of a memcpy() call aren't supposed to overlap,
but we were sometimes calling tor_addr_copy() as a no-op.

Also, tor_addr_assign was a redundant copy of tor_addr_copy(); this patch
removes it.
2010-02-09 12:32:10 -05:00
Nick Mathewson
5314438799 Merge remote branch 'origin/maint-0.2.1' 2010-01-31 22:53:19 -05:00
Nick Mathewson
abd447f876 Revise OpenSSL fix to work with OpenSSL 1.0.0beta*
In brief: you mustn't use the SSL3_FLAG solution with anything but 0.9.8l,
and you mustn't use the SSL_OP solution with anything before 0.9.8m, and
you get in _real_ trouble if you try to set the flag in 1.0.0beta, since
they use it for something different.

For the ugly version, see my long comment in tortls.c
2010-01-31 22:48:29 -05:00
Nick Mathewson
1744e447a1 Decide whether to use SSL flags based on runtime OpenSSL version.
We need to do this because Apple doesn't update its dev-tools headers
when it updates its libraries in a security patch.  On the bright
side, this might get us out of shipping a statically linked OpenSSL on
OSX.

May fix bug 1225.

[backported]
2010-01-29 17:17:47 -05:00
Nick Mathewson
4905eaa38c Detect the correct versions of openssl for tls negotiation fix
Since it doesn't seem to hurt, we should use _both_ fixes whenever
we see OpenSSL 0.9.7L .. 0.9.8, or OpenSSL 0.9.8L..
2010-01-29 17:11:20 -05:00
Nick Mathewson
8d68e5c748 Decide whether to use SSL flags based on runtime OpenSSL version.
We need to do this because Apple doesn't update its dev-tools headers
when it updates its libraries in a security patch.  On the bright
side, this might get us out of shipping a statically linked OpenSSL on
OSX.

May fix bug 1225.
2010-01-29 17:02:17 -05:00
Nick Mathewson
ab87b61a9d Don't unlock a new log until done logging the tor version.
This might please coverity scan.
2010-01-25 14:09:18 -05:00
Nick Mathewson
3b4b6009a0 Merge remote branch 'origin/maint-0.2.1' 2010-01-23 20:46:57 -05:00
Nick Mathewson
4ad5094c90 Avoid a possible crash in tls_log_errors.
We were checking for msg==NULL, but not lib or proc.  This case can
only occur if we have an error whose string we somehow haven't loaded,
but it's worth coding defensively here.

Spotted by rieo on IRC.
2010-01-22 16:32:15 -05:00
Sebastian Hahn
4728bd904f Fix build on Solaris by disabling support for DisableAllSwap
Fixes bug 1198. Solaris doesn't have RLIMIT_MEMLOCK for get/setrlimit,
so disable support because we don't know if all memory can be locked.
2010-01-19 05:04:50 +01:00
Roger Dingledine
356c927476 don't list windows capabilities in windows uname
we never used them, and maybe it's a bad idea to publish them
2010-01-15 15:56:53 -05:00
Nick Mathewson
05a2473b7f Merge branch 'ewma' 2009-12-18 22:33:02 -05:00
Karsten Loesing
f80672d747 Remove duplicate words and a duplicate newline. 2009-12-18 12:55:05 +01:00
Nick Mathewson
235f1e1a96 Refactor out the 'find string at start of any line' logic.
We do this in too many places throughout the code; it's time to start
clamping down.

Also, refactor Karsten's patch to use strchr-then-strndup, rather than
malloc-then-strlcpy-then-strchr-then-clear.
2009-12-17 18:29:37 -05:00
Nick Mathewson
616cbb31c7 Merge commit 'origin/maint-0.2.1' 2009-12-15 17:11:40 -05:00
Nick Mathewson
1c87a27574 Fix bug 1173: remove an assert(unsigned >= 0). 2009-12-15 15:51:59 -05:00
Nick Mathewson
e56747f9cf Refactor a bit so that it is safe to include math.h, and mostly not needed. 2009-12-15 14:40:49 -05:00
Nick Mathewson
2c672f73bf Fix comment typos in container.c 2009-12-15 13:20:02 -05:00
Nick Mathewson
c210db0d41 Enhance pqueue so we can remove items from the middle.
This changes the pqueue API by requiring an additional int in every
structure that we store in a pqueue to hold the index of that structure
within the heap.
2009-12-12 19:06:38 -05:00
Nick Mathewson
d086c9a7f7 Merge commit 'sebastian/fixes' 2009-12-12 02:10:57 -05:00
Nick Mathewson
9e6225ae16 Merge commit 'sebastian/coverity' 2009-12-12 02:10:19 -05:00
Nick Mathewson
0c1b3070cf Now that FOO_free(NULL) always works, remove checks before calling it. 2009-12-12 02:07:59 -05:00
Sebastian Hahn
3807db001d *_free functions now accept NULL
Some *_free functions threw asserts when passed NULL. Now all of them
accept NULL as input and perform no action when called that way.

This gains us consistence for our free functions, and allows some
code simplifications where an explicit null check is no longer necessary.
2009-12-12 03:29:44 +01:00
Sebastian Hahn
28b29e0fd7 Fix typo in a comment 2009-12-12 02:53:27 +01:00
Nick Mathewson
b51a33e527 Merge commit 'origin/maint-0.2.1' 2009-12-04 14:31:17 -05:00