1 2012-01-27 00:00:07 <Graet> where?
  2 2012-01-27 00:00:15 <luke-jr> [Tuesday, January 24, 2012] [8:57:24 PM] <conman> luke-jr continuously tries to commit patches to cgminer to make eligius not work so hard and I do exactly the opposite
  3 2012-01-27 00:00:57 <Graet> well thats about a pool specific patch for your pool. doesnt say anything about efficiency
  4 2012-01-27 00:01:22 <NxTitle> kinlo: is latency a big deal in pools? i.e. if I have 500-1000ms lag will I have problems?
  5 2012-01-27 00:01:40 <kinlo> NxTitle: there is a block on average every 10 minutes
  6 2012-01-27 00:02:07 <kinlo> so if you have let's say 10 seconds of latency, you will loose 10/600
  7 2012-01-27 00:02:08 <luke-jr> Graet: no, it isn't.
  8 2012-01-27 00:02:14 <luke-jr> Graet: none of my patches are Eligius-specific
  9 2012-01-27 00:02:29 <NxTitle> kinlo: perfect, thanks for your help
 10 2012-01-27 00:03:04 <kinlo> NxTitle: you will see that in your pool as "stales"
 11 2012-01-27 00:03:16 <NxTitle> ah, I see
 12 2012-01-27 00:03:25 <Graet> it's ok luke-jr i know you put a different interpretation on most things to most ppl
 13 2012-01-27 00:03:35 <NxTitle> I was playing around with p2pool and ~22% of my shares are stales
 14 2012-01-27 00:03:36 <luke-jr> Graet: tell me which patch is Eligius-specific
 15 2012-01-27 00:03:38 <Graet> got better to do than listen tom your fud
 16 2012-01-27 00:03:43 <NxTitle> even with the forrest mod
 17 2012-01-27 00:04:03 <Graet> l8rs
 18 2012-01-27 00:04:03 <kinlo> NxTitle: do not compare p2pool to shares on a real pool
 19 2012-01-27 00:04:10 <kinlo> p2pool is different
 20 2012-01-27 00:04:19 <luke-jr> p2pool tells you relative stales too IIRC
 21 2012-01-27 00:04:31 <luke-jr> relative stales are more or less equivalent
 22 2012-01-27 00:04:40 <kinlo> also, p2pool has a "block" every 10 seconds
 23 2012-01-27 00:04:40 <NxTitle> 2012-01-26 20:02:13.590152 Pool: 143GH/s in 17366 shares (17118/17370 verified) Recent: 9.93% >14241MH/s Shares: 330 (20 orphan, 49 dead) Peers: 10
 24 2012-01-27 00:04:43 <NxTitle> 2012-01-26 20:02:13.590271 Average time between blocks: 0.45 days
 25 2012-01-27 00:04:45 <NxTitle> 2012-01-26 20:02:13.590428 Pool stales: 11% Own: 21?4% Own efficiency: 89?5%
 26 2012-01-27 00:04:54 <kinlo> so latency is much more dangerous
 27 2012-01-27 00:05:00 <NxTitle> ah, I see
 28 2012-01-27 00:05:53 <luke-jr> NxTitle: what's with all those questionamrks
 29 2012-01-27 00:06:22 <kinlo> I'm off, bye
 30 2012-01-27 00:07:01 <NxTitle> luke-jr: I think they're supposed to be periods, but I honestly have no idea - I assume it's to do with unicode
 31 2012-01-27 00:07:07 <NxTitle> kinlo: thanks for you help
 32 2012-01-27 00:07:11 <NxTitle> s/you/&r/
 33 2012-01-27 00:07:14 <luke-jr> i c
 34 2012-01-27 00:29:02 <Auctus> is there a reason the client has no option to move the data directory?
 35 2012-01-27 00:29:19 <Auctus> I try to keep my non-system things on a drive other than C
 36 2012-01-27 00:30:44 <luke-jr> Auctus: I just set HOME before I run bitcoind ;)
 37 2012-01-27 00:31:26 <luke-jr> though it would be trivial to patch in an option
 38 2012-01-27 00:32:10 <Auctus> yeah, the average end-user is not going to know how to do anything outside the client, i think it'd be a good idea, even if you can't change it from the client, at least ask at install time
 39 2012-01-27 00:35:12 <CIA-2> bitcoin: Luke Dashjr * ra9f222702d02 eloipool/merklemaker.py: Track lowest merkleroot volumes seen http://tinyurl.com/6msntzm
 40 2012-01-27 00:36:28 <sipa> Auctus: what about -datadir= ?
 41 2012-01-27 00:38:22 <luke-jr> sipa: does that work? I didn't see it in the code O.o
 42 2012-01-27 00:39:18 <sipa> sure it works
 43 2012-01-27 00:39:35 <sipa> it's how testnet-in-a-box works, eg
 44 2012-01-27 00:40:01 <gmaxwell> I use datadir all the time.. only sane way to run a bunch of copies of bitcoin.
 45 2012-01-27 00:42:49 <sipa> nice trick: symlink blkindex.dat and blk0001.dat in all bitcoin dirs to the same place
 46 2012-01-27 00:44:27 <gmaxwell> sipa: I tried using btrfs cow on blk0001.dat to split one node into several without wasting space .. but bitcoin is obviously much slower on btrfs so it wasn't worth it.
 47 2012-01-27 00:45:00 <luke-jr> sipa: that works? O.o
 48 2012-01-27 00:45:23 <luke-jr> I'd expect the nodes to conflict
 49 2012-01-27 00:45:44 <sipa> conflict?
 50 2012-01-27 00:46:04 <sipa> luke-jr: oh, not for multiple copies at the same time
 51 2012-01-27 00:46:31 <luke-jr> :/
 52 2012-01-27 00:46:42 <gmaxwell> (yea, what I did worked for multiple copies at the same time except for btrfs being slow)
 53 2012-01-27 00:47:24 <gmaxwell> I was thinking perhaps if I moved the file size cap down I'd get a blk0002.dat and blk0001.dat could just be symlinked.. but I didn't bother actually looking into that.
 54 2012-01-27 00:53:39 <forrestv> NxTitle, luke-jr, they're ???'s
 55 2012-01-27 00:54:17 <forrestv> NxTitle, want come to #p2pool ? the high amount of dead shares would indicate that your miner is misconfigured
 56 2012-01-27 00:54:49 <NxTitle> sure
 57 2012-01-27 01:50:59 <wirehead> \n2747243
 58 2012-01-27 01:59:27 <midnightmagic> gmaxwell: Have you tried it on MurderFS?
 59 2012-01-27 02:25:13 <sipa> midnightmagic: you mean reiserfs?
 60 2012-01-27 03:19:21 <Diablo-D3> gmaxwell: https://www.destroyallsoftware.com/talks/wat
 61 2012-01-27 03:19:59 <BlueMatt> Diablo-D3: wat?
 62 2012-01-27 03:20:03 <Diablo-D3> wat.
 63 2012-01-27 03:20:08 <BlueMatt> what?
 64 2012-01-27 03:21:01 <gribble> New news from bitcoinrss: sipa opened pull request 787 on bitcoin/bitcoin <https://github.com/bitcoin/bitcoin/pull/787>
 65 2012-01-27 03:21:33 <BlueMatt> Diablo-D3: wtf???
 66 2012-01-27 03:22:00 <Diablo-D3> blueMatt: wat!
 67 2012-01-27 03:22:20 <BlueMatt> javascript...oh god :O
 68 2012-01-27 03:22:43 <Diablo-D3> the last one is the best one
 69 2012-01-27 03:23:43 <BlueMatt> WATMAN
 70 2012-01-27 03:23:53 <BlueMatt> that was pretty good
 71 2012-01-27 03:26:28 <k9quaint> nananana :)
 72 2012-01-27 03:27:06 <k9quaint> the fact that most of the web is written in javascript is proof that there is a god
 73 2012-01-27 03:27:15 <k9quaint> because I died and went to hell
 74 2012-01-27 03:27:25 <k9quaint> and here I am, surrounded by javascript :(
 75 2012-01-27 03:27:56 <k9quaint> which makes you guys either my tormentors, or also dead and cast down into hell
 76 2012-01-27 03:28:14 <gmaxwell> sipa: whats the capacity like with the bias against duplication?
 77 2012-01-27 03:29:01 <BlueMatt> k9quaint: considering I gave up trying to use javascript years ago, Id say Im not in hell
 78 2012-01-27 03:29:05 <BlueMatt> so I guess tormentors?
 79 2012-01-27 03:29:22 <BlueMatt> though boost is pretty shitty sometimes too...
 80 2012-01-27 03:31:27 <sipa> gmaxwell: i have a node running now for some time, with currently 35 connections; i have 9000 addresses
 81 2012-01-27 03:32:00 <sipa> though for a non-reachable node it make take quite some more time to reach that number, as the number of sources is limited
 82 2012-01-27 03:32:16 <sipa> oh, IT'S OVER NINE THOUSAND
 83 2012-01-27 03:32:24 <gmaxwell> hah
 84 2012-01-27 03:32:33 <gmaxwell> What was Diablo-D3's line.
 85 2012-01-27 03:32:38 <gmaxwell> er, That was.
 86 2012-01-27 03:32:43 <k9quaint> wat
 87 2012-01-27 03:33:06 <gmaxwell> sipa: what are you using for the source info on the addr.dat imports?
 88 2012-01-27 03:33:30 <phantomcircuit> sipa, did you just import your changes to what is considered a good address to store?
 89 2012-01-27 03:33:46 <sipa> gmaxwell: 127.0.0.1
 90 2012-01-27 03:34:08 <sipa> phantomcircuit: hmm?
 91 2012-01-27 03:34:25 <phantomcircuit> yeah
 92 2012-01-27 03:34:28 <phantomcircuit> 787
 93 2012-01-27 03:34:50 <phantomcircuit> Keep the address tables in-memory, and asynchronously dump the entire to able in addr.dat.
 94 2012-01-27 03:34:51 <phantomcircuit> hehe
 95 2012-01-27 03:34:52 <BlueMatt> good conversation: "hmm?" "yeah" "787"
 96 2012-01-27 03:34:52 <phantomcircuit> typo
 97 2012-01-27 03:35:47 <splatster> Diablo-D3: is that video with a real crowd
 98 2012-01-27 03:36:34 <phantomcircuit> if (nTime > nNow + 10*60) // came in a flying DeLorean
 99 2012-01-27 03:36:36 <phantomcircuit> lollll
100 2012-01-27 03:37:06 <BlueMatt> we need more similar comments like that in bitcoin
101 2012-01-27 03:37:08 <Diablo-D3> splatster: yes
102 2012-01-27 03:37:16 <splatster> this is great
103 2012-01-27 03:39:48 <sipa> please test :)
104 2012-01-27 03:41:07 <marf_away> no
105 2012-01-27 03:41:10 <marf_away> the comment is bad
106 2012-01-27 03:41:17 <marf_away> i dont understand it
107 2012-01-27 03:41:25 <marf_away> delorean fast or slow?
108 2012-01-27 03:41:37 <BlueMatt> stop trolling
109 2012-01-27 03:41:47 <sipa> marf_away: http://www.imdb.com/title/tt0096874/
110 2012-01-27 03:41:57 <marf_away> i know the movi
111 2012-01-27 03:42:04 <marf_away> but is it slow or fast?
112 2012-01-27 03:42:22 <BlueMatt> its a time-traveling car, does it matter how fast it goes?
113 2012-01-27 03:42:24 <marf_away> he needed a steamtrain = slow
114 2012-01-27 03:42:58 <marf_away> ok now i understand
115 2012-01-27 03:43:08 <marf_away> its from the future...
116 2012-01-27 03:43:19 <BlueMatt> ...
117 2012-01-27 03:43:21 <marf_away> connection from the future
118 2012-01-27 03:43:39 <marf_away> ok :P comment is funny but bad
119 2012-01-27 03:43:40 <marf_away> :P
120 2012-01-27 04:21:39 <phantomcircuit> sipa, seems a bit... randomly organized :)
121 2012-01-27 05:05:13 <CIA-2> bitcoin: Luke Dashjr * r9af60344fe75 eloipool/jsonrpcserver.py: Disable early longpoll headers for PhoenixMiner < 2.8.1 http://tinyurl.com/829l3zn
122 2012-01-27 06:45:09 <CIA-2> bitcoin: jedi95 * rc1bf824af77d Phoenix-Miner/minerutil/RPCProtocol.py: Fix small mistake in x-roll-ntime RPC code. Not worth a version bump since this functionality isn't used just yet. http://tinyurl.com/7bt2b4k
123 2012-01-27 06:50:34 <Hunterbunter> what would be the most effective language to code a site like mtgox in?
124 2012-01-27 06:50:58 <Hunterbunter> I'n not really a web programmer, more c++ etc, but it's all very interesting
125 2012-01-27 07:05:25 <CIA-2> bitcoin: Luke Dashjr * r4bc3c011eaad eloipool/ (bitcoin/node.py eloipool.py merklemaker.py): Make BitcoinNode more robust, and use checksum on ver message http://tinyurl.com/7zvyyga
126 2012-01-27 09:01:48 <weex> can the timestamp of a transaction be trusted not to change(i.e. is it encoded in the transaction message)?
127 2012-01-27 09:02:29 <cjd> in general, never ever trust timestamps
128 2012-01-27 09:02:39 <josephcp> the timestamp is encoded as part of the hash so it can't be edited by other parties, however, its accuracy isn't guaranteed, from what I understand
129 2012-01-27 09:03:15 <weex> ok, i'll not use the timestamp then :)
130 2012-01-27 09:03:33 <cjd> mm if it's hashed into the tx then it's probably accurate to within about 2 hours IIRC
131 2012-01-27 16:05:30 <luke-jr> makomk: ping
132 2012-01-27 16:30:13 <CIA-2> bitcoin: Luke Dashjr * r68ca277f4259 eloipool/ (bitcoin/node.py eloipool.py): Dump block payload to log, and store it in an unbounded array just in case http://tinyurl.com/85xes2r
133 2012-01-27 16:45:50 <makomk> luke-jr: hmmmmm?
134 2012-01-27 16:46:36 <luke-jr> makomk: interested in reinventing CoiledCoin? :P
135 2012-01-27 16:46:47 <makomk> Not particularly, no.
136 2012-01-27 16:47:00 <luke-jr> oh
137 2012-01-27 16:47:13 <luke-jr> I was thinking it might make sense for BIP 16 & 17 :p
138 2012-01-27 17:03:59 <luke-jr> gavinandresen: ping
139 2012-01-27 17:04:13 <luke-jr> BIP 16 states "The rules for validating these outpoints when relaying transactions or considering them for inclusion in a new block are as follows:"
140 2012-01-27 17:04:26 <luke-jr> does that mean the blocks themselves are valid if they contain transactions violating these rules?
141 2012-01-27 17:04:40 <gavinandresen> ... after the switchover date, yes.  Not before.
142 2012-01-27 17:04:57 <luke-jr> err, you mean the inverse of that? O.o
143 2012-01-27 17:05:05 <luke-jr> they're valid before, not after?
144 2012-01-27 17:05:08 <gavinandresen> "These same rules shall be applied when validating transactions in blocks with timestamps after February 15, 2012"
145 2012-01-27 17:05:19 <luke-jr> OK
146 2012-01-27 17:05:34 <luke-jr> what if there's a backward timestamp BTW? <.<
147 2012-01-27 17:06:00 <gavinandresen> If the backwards timestamp is before the switchover time then the rules do not apply to that block
148 2012-01-27 17:06:18 <luke-jr> interesting
149 2012-01-27 17:10:05 <BlueMatt> so in other words dont use p2sh txes until a day or two after the switchover date
150 2012-01-27 17:12:30 <BlueMatt> gavinandresen: back to the sigop count stuff: why even bother with legacy/p2sh differentiation, why not just change GetSigOpCount to count p2sh sigops and not bother changing anything else?
151 2012-01-27 17:13:14 <gmaxwell> BlueMatt: month or two after
152 2012-01-27 17:13:28 <gavinandresen> Yes, I'd wait at least a week or three.
153 2012-01-27 17:13:52 <gavinandresen> BlueMatt: the latest code makes the sigopcount Legacy+p2sh
154 2012-01-27 17:14:09 <BlueMatt> gavinandresen: why write code for both?
155 2012-01-27 17:14:09 <gavinandresen> BlueMatt: ... which is the most conservative way to avoid potential chain splits
156 2012-01-27 17:14:14 <BlueMatt> why not just add them together
157 2012-01-27 17:14:19 <gmaxwell> BlueMatt: p2sh is only secure if it has a forever majority, that won't be completely sure until lots of clients are updated.
158 2012-01-27 17:14:21 <gavinandresen> BlueMatt: that's what the code does now
159 2012-01-27 17:14:50 <BlueMatt> gmaxwell: true, but I was just commenting on the backwards timestamp issue
160 2012-01-27 17:14:53 <makomk> I presume the older version of the code never made it out there onto mining pools?
161 2012-01-27 17:15:03 <BlueMatt> gmaxwell: but yea, shouldnt be used for longer than that anyway
162 2012-01-27 17:15:20 <BlueMatt> gavinandresen: the code now has GetLegacySigOpCount and GetP2SHSigOpCount, why both?
163 2012-01-27 17:15:33 <BlueMatt> gavinandresen: why not just GetSigOpCount the way it used to be and add in p2sh ops there?
164 2012-01-27 17:16:41 <gavinandresen> BlueMatt: because it is somewhat expensive to fetch transaction inputs, and there's an 'early out' in CheckBlock if the block has an out-of-bounds number of legacy sigops.  Seemed wise to keep that check there for DoS prevention.
165 2012-01-27 17:17:30 <BlueMatt> I suppose you cant do it without fetching the prevouts?
166 2012-01-27 17:17:37 <gavinandresen> Nope.
167 2012-01-27 17:18:52 <BlueMatt> you cant blindly guess if the last string on the scriptSig is unserializeable and count it even if the prevout doesnt make it a p2sh tx?
168 2012-01-27 17:19:19 <BlueMatt> or is that too ugly?
169 2012-01-27 17:19:47 <gavinandresen> Uhhh.....     that seems like a good way for somebody to craft a signature that DOES look like code and get a blockchain split to happen....
170 2012-01-27 17:25:00 <Eliel> gavinandresen: what mechanism is used to do the coordinated activation of the new rules?
171 2012-01-27 17:25:34 <Eliel> does every pool have to activate them themselves or is there some kind of automation?
172 2012-01-27 17:25:52 <gavinandresen> Eliel: miners all run with the same -paytoscripthashtime= argument.  Feb 15'th is compiled into the code right now, but that'll probably get pushed
173 2012-01-27 17:26:33 <gavinandresen> Eliel: miners have a very strong incentive not to be the only ones doing the extra validation...  since if they mess up it is THEIR blocks that will be orphaned.
174 2012-01-27 17:33:38 <gmaxwell> Fedora updated their openssl, I've thrown up new F16 RPMS (for anyone who cares): http://people.xiph.org/~greg/openssl/
175 2012-01-27 17:41:19 <luke-jr> gavinandresen: btw, -paytoscripthashtime= is broken for both BIP 16 and 17 implementations right now
176 2012-01-27 17:41:46 <gavinandresen> how so?  There was a missing dash.....
177 2012-01-27 17:41:58 <gavinandresen> ... but that's fixed.
178 2012-01-27 17:44:00 <roconnor__> what appears in debug.log when BitcoinMiner finds a block?
179 2012-01-27 17:44:25 <luke-jr> gavinandresen: oh, you fixed that? nm
180 2012-01-27 17:45:09 <gavinandresen> btcmine just found a minor bug in the pre-0.4 backports, too...
181 2012-01-27 17:50:19 <CIA-97> libbitcoin: genjix * rbf2c3734458b / (5 files in 4 dirs): Huge readability win in exchange for drop in speed during rare corner case. http://tinyurl.com/79nxnca
182 2012-01-27 17:52:42 <luke-jr> gavinandresen: the backports are one major advantage of BIP 17 when it comes to implementation IMO
183 2012-01-27 17:52:51 <luke-jr> BIP 16's backport is 10 times more invasive
184 2012-01-27 17:52:56 <luke-jr> (in LOC)
185 2012-01-27 17:53:39 <CIA-97> bips: genjix master * r8230665 / bip-0017.md : BIP 0017 - http://git.io/IjZEoA https://github.com/genjix/bips/commit/823066539702dcf9af226527454281919c3d8265
186 2012-01-27 17:53:40 <CIA-97> bips: genjix master * r0455253 / bip-0001.md : BIP 0001 is Active - http://git.io/jDQSxA https://github.com/genjix/bips/commit/0455253e4e74b85d21e461c528521acfc26c4247
187 2012-01-27 18:47:49 <gmaxwell> I need faster ECDSA code darnit. down to 5.5 minutes to sync 0-140,000 but going all the way to current takes almost another 15 minutes.
188 2012-01-27 18:49:23 <gmaxwell> s/15/25/
189 2012-01-27 18:49:29 <Joric> i wonder what does vanitygen use
190 2012-01-27 18:49:38 <gmaxwell> Joric: gpu code!
191 2012-01-27 18:50:10 <gmaxwell> but it takes something like three of those point multiplies to validate where generating a public key only takes one.
192 2012-01-27 18:50:29 <Joric> it has to perform ecdsa every time
193 2012-01-27 18:50:56 <Joric> maybe he optimized it somehow made it incremental i didn't see the code
194 2012-01-27 18:51:04 <gmaxwell> No it doesn't.
195 2012-01-27 18:51:26 <gmaxwell> It does a single point multiply to generate the public key, but it doesn't have to do a validation, which is harder.
196 2012-01-27 18:51:44 <gmaxwell> oh thats right and the point multiplies are cheaper because they do incremental updates. :(
197 2012-01-27 18:51:47 <gmaxwell> darn.
198 2012-01-27 19:01:40 <gavinandresen> gmaxwell: what are you optimizing?
199 2012-01-27 19:02:16 <forrestv> gmaxwell, i remember something about bulk-ecdsa validations that could give a result for whether a list of signatures were valid
200 2012-01-27 19:03:10 <gavinandresen> gmaxwell: ... or:  are you optimizing user experience?  You could just assume validation will succeed, throw the work into a buffer, and run a thread to validate at leisure (and do... something... if the thread finds out something does NOT validate)
201 2012-01-27 19:03:40 <gmaxwell> forrestv: yes, the ed25519 software does that.. its only a very modest speedup.
202 2012-01-27 19:03:52 <gmaxwell> (well, 2x while doing 64 validations at once)
203 2012-01-27 19:05:03 <gmaxwell> gavinandresen: dorking around with some of the fetching logic. But this is all running on effectively ramdisk. User expirence on normal setups is still IO bound. I'm hoping matt's block validation reorg will make batch writes more reasonsable.
204 2012-01-27 19:06:48 <forrestv> ah
205 2012-01-27 19:06:56 <forrestv> didn't know it was actually in use :P
206 2012-01-27 19:35:41 <BlueMatt> gavinandresen: re:<gavinandresen> Too ugly to assume that the last item in the scriptSig is/isn't a script if it unserializable; that seems like a good way for an attacker to make a blockchain split
207 2012-01-27 19:35:56 <BlueMatt> meh, its so easy for an attacker to make a chain split when you add p2sh anyway...
208 2012-01-27 19:36:22 <gavinandresen> BlueMatt: No, I mean a chain split after 100% of the network supports p2sh
209 2012-01-27 19:36:42 <gavinandresen> BlueMatt: ... if there was ever an implementation that did NOT do the ugly hack, it could get split off
210 2012-01-27 19:36:52 <BlueMatt> well then make the hack part of the spec ;)
211 2012-01-27 19:37:04 <BlueMatt> its already ugly
212 2012-01-27 19:37:14 <BlueMatt> you have to look up prevouts
213 2012-01-27 19:37:22 <BlueMatt> (which clients cant do and just looks ugly to me)
214 2012-01-27 19:38:13 <gmaxwell> BlueMatt: then you have weird crap like having to sniff for things which looklight checksigs in things which are not.
215 2012-01-27 19:38:17 <gavinandresen> okey doke.  I think it is ugly you can create a block right now that has zero sigops but takes a ton of sigops to validate....
216 2012-01-27 19:38:28 <gavinandresen> (that COUNTS as zero sigops....)
217 2012-01-27 19:38:59 <BlueMatt> though I see the point, they were already counted so I kinda disagree with that...
218 2012-01-27 19:39:23 <gavinandresen> Counted in previous blocks, yes, where they shouldn't have been because those previous blocks didn't have to do the validation.
219 2012-01-27 19:40:01 <BlueMatt> true...
220 2012-01-27 19:40:11 <gavinandresen> In other words: counting sigops in the outputs is broken.  A patient attacker can plant lots of very expensive transactions in lots of previous blocks and then combine them all into one killer block
221 2012-01-27 19:40:16 <CIA-97> bitcoin: various namecoin_mmm * r8f3f29..ffb5ff bitcoind-personal/ (9 files in 3 dirs): (16 commits) http://tinyurl.com/84zt6e5
222 2012-01-27 19:40:19 <BlueMatt> but then clients have yet another check that full nodes do that clients cant do
223 2012-01-27 19:40:50 <gavinandresen> BlueMatt: if clients aren't fetching inputs and validating transactions then they have no reason to care about sigop count
224 2012-01-27 19:41:08 <gavinandresen> ... if they are, then they can count....
225 2012-01-27 19:42:21 <BlueMatt> meh, sigop count is also a network block-verification rule, not just a dos prevention (though the reason for the rule is dos prevention)
226 2012-01-27 19:43:58 <BlueMatt> meh, I guess its fine...I was just hoping there was an easier way to check sigops without needing inputs...
227 2012-01-27 19:44:07 <BlueMatt> no way to do it that is much better
228 2012-01-27 19:50:30 <diki> damn, the maps,templates,vectors,static/dynamic_cast
229 2012-01-27 19:50:55 <diki> namespaces, all that nonsense
230 2012-01-27 20:00:11 <CIA-97> bitcoin: Luke Dashjr * r230cdc71bf7e eloipool/eloipool.py: Bugfix: Skip coinbaser if CoinbaserCmd is None http://tinyurl.com/7wuj6ly
231 2012-01-27 20:20:09 <CIA-97> bitcoin: Luke Dashjr * r574ed88c8aaa eloipool/interactivemode.py: interactivemode: Add exit() function to SIGTERM the entire process http://tinyurl.com/7uoy7c3
232 2012-01-27 21:17:33 <chaligo> sup bros
233 2012-01-27 21:20:03 <BlueMatt> bros?
234 2012-01-27 21:21:02 <amiller> hey bra
235 2012-01-27 21:23:25 <DrHaribo> How do I get the bitcoind 0.5.2 sources? The git repo doesn't seem to have a 0.5.2 tag?
236 2012-01-27 21:23:46 <gmaxwell> DrHaribo: 0.5.2 is in the stable repo maintained by luke, not the main repo.
237 2012-01-27 21:24:12 <BlueMatt> we really need to put that in the topic or something...
238 2012-01-27 21:24:26 <DrHaribo> gmaxwell: Thank you
239 2012-01-27 21:24:41 <gmaxwell> BlueMatt: on the github page perhaps.
240 2012-01-27 21:24:57 <diki> how many blocks are there including the ones from split chains?
241 2012-01-27 21:25:05 <BlueMatt> sipa: want to do that?
242 2012-01-27 21:35:13 <luke-jr> (and tags ofc)
243 2012-01-27 21:35:17 <CIA-97> bitcoin: Luke Dashjr checkhashverify * r78ef2cd15653 bitcoind-personal/src/ (6 files in 2 dirs): Merge branch 'checkhashverify_backport' into checkhashverify http://tinyurl.com/7wbjehg
244 2012-01-27 21:35:18 <CIA-97> bitcoin: Luke Dashjr checkhashverify * r09ff1903b999 bitcoind-personal/src/ (script.cpp test/rpc_tests.cpp test/script_P2SH_tests.cpp): Support for receiving and redeeming BIP 17 transactions http://tinyurl.com/8a28ndk
245 2012-01-27 21:35:48 <BlueMatt> probably
246 2012-01-27 21:36:21 <luke-jr> I'm sure Gavin will say no to me doing it, and no to him doing it& maybe he can OK sipa doing it :P
247 2012-01-27 21:36:40 <BlueMatt> I dont see why anyone would complain adding another repo to github/bitcoin
248 2012-01-27 21:36:49 <BlueMatt> github/bitcoin/stable-releases or smth
249 2012-01-27 21:37:01 <luke-jr> BlueMatt: why another repo? git supports multiple branches :p
250 2012-01-27 21:37:14 <luke-jr> unless another repo would let me have access w/o affective the main one
251 2012-01-27 21:37:24 <luke-jr> but IIRC github won't let you have multiple clones of the same repo under the same user :/
252 2012-01-27 21:37:26 <BlueMatt> thats the point
253 2012-01-27 21:37:40 <BlueMatt> you cant mark it as a clone, but that doesnt mean you cant have the same content
254 2012-01-27 21:38:29 <luke-jr> dunno, I suspect people would still ask
255 2012-01-27 21:38:54 <luke-jr> point of mirroring to github in the first place is so they can see/get the tags from the repo they already pull from IMO
256 2012-01-27 21:39:51 <DrHaribo> A stable branch sounds nice
257 2012-01-27 21:40:08 <BlueMatt> there is one
258 2012-01-27 21:40:10 <BlueMatt> well multiple
259 2012-01-27 21:41:31 <DrHaribo> bitcoin.org front page links to https://github.com/bitcoin/bitcoin which has no stable branch with a v0.5.2
260 2012-01-27 21:41:47 <BlueMatt> its in another place
261 2012-01-27 21:41:49 <BlueMatt> gitorious
262 2012-01-27 21:42:17 <DrHaribo> Yeah I know that now... I would have found it more intuitive if it was just a branch at the mentioned URL, but no big deal
263 2012-01-27 22:05:13 <CIA-97> bitcoin: Con Kolivas debug * r761ff5e6d1cc cgminer/ (adl.c main.c miner.h util.c): Merge branch 'master' into debug http://tinyurl.com/6vnvx4l
264 2012-01-27 22:05:14 <CIA-97> bitcoin: Con Kolivas debug * r37e5ca8cd8e6 cgminer/adl.c: For unique entries. http://tinyurl.com/6st73te
265 2012-01-27 22:05:15 <CIA-97> bitcoin: Con Kolivas debug * rbed75ac25715 cgminer/adl.c: Debug device output. http://tinyurl.com/7jo4mrd
266 2012-01-27 22:05:19 <CIA-97> bitcoin: Con Kolivas reorder * r5a0b4f62d096 cgminer/ (adl.c adl.h miner.h): Map GPU devices to virtual devices in their true physical order based on BusNumber. http://tinyurl.com/7aoyr87
267 2012-01-27 22:05:22 <CIA-97> bitcoin: various luke * r7db294..e5e1ec cgminer/ (util.c main.c): (5 commits) http://tinyurl.com/7ueqvnh
268 2012-01-27 22:05:25 <CIA-97> bitcoin: Con Kolivas reorder * r371e5f688a2b cgminer/ (adl.c adl.h main.c): Reorder displayed devices to map to physical locations and initialise according to logical location instead. http://tinyurl.com/7my6xgx
269 2012-01-27 23:30:15 <CIA-97> bitcoin: Con Kolivas reorder * r3bc0583454b0 cgminer/adl.c: Iterate and change virtual device order instead of shuffling ram. http://tinyurl.com/7zxcuop
270 2012-01-27 23:40:10 <CIA-97> bitcoin: Con Kolivas reorder * r059701272c75 cgminer/adl.c: Carry virtual gpu number across. http://tinyurl.com/6tobx77