00:50
<
phantomcircuit >
gmaxwell: ^ ping
00:54
<
gmaxwell >
phantomcircuit: concept ack. While I'm here, can you get fuzzing going on matt's compact block relay new message types?
00:55
<
phantomcircuit >
i've been fuzzing most of the deserialization but had the same issue with linking as i'm having with CheckBlock
00:56
<
phantomcircuit >
best i can tell there's an issue with -fPIC and -fPIE being mixed somewhere but it doesn't really make any sense to me yet
00:56
<
phantomcircuit >
the next step for that is to make params also a parameter
00:56
<
phantomcircuit >
brb
01:03
<
phantomcircuit >
b
01:17
<
phantomcircuit >
gmaxwell: yeah
01:18
<
phantomcircuit >
i get what the issue is, i'm just not sure how to fix it with autotools
07:17
<
jonasschnelli >
gmaxwell, sipa: re encryption: Did we agree on rekeying = cranking the initial KDF by HASH(old_sym_key)?
07:24
<
jonasschnelli >
«Why are there two key exchanges? A single shared-secret could be used
07:24
<
jonasschnelli >
to generate keys for each direction. And it would reinforce the single
07:24
<
jonasschnelli >
symmetric cipher rule.»
07:25
<
sipa >
i don't disagree; i think it's just easier to keep it seoarate
07:28
<
jonasschnelli >
Yes. Let's keep it for now.
07:29
<
jonasschnelli >
What about dropping the Symmetric Cipher Negotiation for the first version of the BIP?
07:29
<
jonasschnelli >
There is only ChaCha20-Poly1305 anyways.
07:29
<
jonasschnelli >
If we would extend the encryption to support multiple symmetric cipher suites, we could work that out in detail.
07:30
<
jonasschnelli >
(later)
11:08
<
GitHub6 >
bitcoin/master 073225c Wladimir J. van der Laan: chain: define enum used as bit field as uint32_t...
11:08
<
GitHub6 >
bitcoin/master 07e4edb Wladimir J. van der Laan: auto_ptr → unique_ptr...
11:08
<
GitHub6 >
bitcoin/master 86b800c Wladimir J. van der Laan: Merge #7964: Minor changes for c++11 consistency...
12:49
<
GitHub168 >
bitcoin/master f4ac02e Kaz Wesley: fix race that could fail to persist a ban...
12:49
<
GitHub168 >
bitcoin/master 03cf6e8 Wladimir J. van der Laan: Merge #7959: fix race that could fail to persist a ban...
14:31
<
spudowiar1 >
if I were my brother's age now, I'd be rich now
14:32
<
spudowiar1 >
anyone know where I can get proofs about Satoshi being uncovered?
14:32
<
murch >
spudowiar1: #bitcoin is chewing that piece of fat. Here is probably not the right place.
14:32
<
spudowiar1 >
yeah, just found it
14:33
<
spudowiar1 >
murch: forgot to join Bitcoin related channels on IRC
14:33
<
spudowiar1 >
murch: just remembered
18:10
<
MarcoFalke >
paveljanik, do you see the IndexError('pop from empty list',) often?
18:11
<
paveljanik >
MarcoFalke, master or 7980?
18:11
<
paveljanik >
current 7980 looks OK, no issues so far.
18:11
<
paveljanik >
but in the previous commit, almost always IIRC
18:12
<
MarcoFalke >
but good that it is now fixed, somehow
18:12
<
paveljanik >
I can't judge myself, sorry. I can only state what fails in the master and is OK in your PR.
18:14
<
MarcoFalke >
It is popping from the unspents, so if the dict was ordered wrong and you ended up with odd unspents... Maybe that caused it.
18:14
<
MarcoFalke >
I am running with the py3 patch and this dict-patch right now in a loop
18:15
<
MarcoFalke >
If there are no issues, I think it is fine
18:15
<
paveljanik >
BTW - why the difference between running from the main dir as python qa/rpc/... and ./smartfee.py?
18:16
<
paveljanik >
will be back in half an hour.
18:16
<
MarcoFalke >
There should be no difference. It is just me being in the root dir all the time ;)
18:46
<
paveljanik >
MarcoFalke, ok, but why it has different results?
18:47
<
paveljanik >
on master, python qa/rpc-tests/smartfees.py --srcdir=src almost always ends up on "pop from empty list". But ./smartfee.py never...
18:50
<
MarcoFalke >
Which means self.nodes[0].listunspent(0) is empty
18:50
<
MarcoFalke >
Corrupt cache?
18:51
<
paveljanik >
what cache? Isn't it setting everything from scratch itself?
18:52
<
paveljanik >
a, you mean cache/ directory?
18:53
<
paveljanik >
Right, removing it fixed the problem.
18:54
<
paveljanik >
I was fooled by "Initializing test directory /var/folders/65/fn0h49r55k7779vg1b_h461r0000gn/T/testOdh...." 8)
18:54
<
MarcoFalke >
Yes, it is not really intuitive, as the cache dirs end up in different locations
18:55
<
MarcoFalke >
There prop should be only one cache dir at the root
20:09
<
sdaftuar >
cfields: that's the test suite coverage? i was just wanting to look at that...
20:10
<
cfields >
sdaftuar: yes, that's all tests, excluding extended pruning rpc
20:10
<
sdaftuar >
ah ok, awesome. thanks!
20:11
<
paveljanik >
MarcoFalke, tests running now - on master, every second fails, 7980: every test was OK so far. Will report tomorrow. Thank you!
20:11
<
cfields >
I'm looking through it now to see what's interesting to test for
20:11
<
cfields >
sdaftuar: as a next step i'm trying to create a coverage diff, but i'm not sure if the tools know how to handle that
20:13
<
sdaftuar >
cool. i'm going through the PR commit by commit, so i'll see if i can just manually look at what code hasn't been tested
20:51
<
gmaxwell >
nickler: You asked me earlier about coverage, see cfiel ds above.
21:51
<
james341 >
can u help me with something?
21:51
<
james341 >
now I am in a very difficult situation and I do not have enough money even for food . who do not mind , please give me a couple of dollars . here is my Bitcoin address :"1HZAGzo7DJ4cqMuHuhjPZY1hwHhZYU52Tu" . I will be very grateful to anyone who can help me -- do not consider this topic for spam, I need help
23:39
<
achow101 >
Is gavin's commit access still revoked?
23:41
<
gmaxwell >
Yes. I think it's unlikely to be restored. It hasn't been used in a year, and the situation with CSW is still very screwy.
23:55
<
gmaxwell >
molz: he's an org member with readonly access. like luke-jr. for example. (there are others, but they're set to private)
23:55
<
molz >
gmaxwell, ah.. thanks for this info, i kept wondering about this
23:56
<
gmaxwell >
(being an org member lets you assign tickets to them-- really bitcoin core should probably make a lot more people org members)