log☇︎
3518 entries in 0.703s
mod6: here's the patch
mod6: ok, im gotta try to patch up this thing on openbsd and see if i can get anywhere.
mod6: i've got a script to pull archives, verify and patch on Linux/FreeBSD/OpenBSD. Will only compile it on Linux since we're not quite there yet on *BSD, give it a try, let me know how it goes:
nubbins`: seems easy enough to write a patch that changes version string
mod6: BingoBoingo: alerts were snipped via this patch: http://thebitcoin.foundation/turdmeister-alert-snip.tar.gz -- It's not posted to the btc-dev mailing list yet. This might have been because of issues with messages getting snarffed back in Oct./Nov.
jurov: BingoBoingo: depends on the patch
asciilifeform: danielpbarron: of course it will crash. esp. if you built it with max=750 (as in the patch) but with 250 likewise
danielpbarron: http://danielpbarron.com/patch_test.txt
ascii_field: redhat patch << i can't help but think that this is a 'wait, there was dioxin in my cyanide pill!111!' situation.
danielpbarron: i think i tried the portatronic patch first and it didn't work
danielpbarron: but just to clarify, i'm supposed to patch with the portatronic thing after the chicken etc.. patches?
ascii_field: (you did patch, right?)
mod6: <+punkman> running a node with this line removed: http://btc.yt/lxr/satoshi/source/src/main.cpp#0939 . works so far << check out this patch: http://therealbitcoin.org/ml/btc-dev/2014-December/000023.html ☟︎☟︎
punkman: (on debian wheezy, rev_bump patch)
assbot: [PATCH] Add support for Realtek Media Player SoCs ... ( http://bit.ly/1F6Cl4n )
asciilifeform: the bastard patch made the thing actually usable on pc for the first time (in my view)
kakobrekla: have it built in, but optionally as separate patch?
punkman: btw the patch that got lost in the tubes: http://dpaste.com/2EG2F1K.txt
asciilifeform: mircea_popescu: as it is, it's a fallback after dns (snipped in patch i proposed last night) and irc mechanizms.
mod6: well, but also, i like i was saying, it's inconsistant. because TomServo has checkpoints included and he got through the gauntlet, and I did as well lastweek before the version patch.
punkman: just sent a patch to the list, wonder if it'll go through
mod6: yeah, i wanna re-iterate here, all of the wedging issues ben and I had were /with/ checkpoints included. if we apply the rm_checkpoints patch, no longer a problem.
mod6: i've got mutiple good v0.5.3 chains saved, but ... just was regression testing to ensure full sync with the v0.5.3.1 patch before release sign & cut
mod6: but then i applied the rm_checkpoints patch and then it continued on np.
ben_vulpes: <asciilifeform> [] ben_vulpes, mircea_popescu: am i the only one who wonders why 1) checking hashes, inside a pgp-signed script, when patch sigs also are checked 2) patches & their hashes, sigs, listed explicitly, instead of iterating over directory << consider it an excess of paranoia
asciilifeform: ben_vulpes, mircea_popescu: am i the only one who wonders why 1) checking hashes, inside a pgp-signed script, when patch sigs also are checked 2) patches & their hashes, sigs, listed explicitly, instead of iterating over directory
asciilifeform: most of what you see in that patch is only there because the fuckers changed the data structure used to represent the bastard blocks
asciilifeform: http://therealbitcoin.org/ml/btc-dev/attachments/20150130/asciilifeform-orphanage-burner_80eb450469d93b270aa160e26b4cdab732add2c4.patch
asciilifeform: the skull'n'crossbones patch
asciilifeform: that means one or more of you lot are stuck with the misery of actually reading that patch.
asciilifeform: anyone who wants to try bastard, can make a temporary copy of tree merged up to 5.3.1 and apply the skull'n'crossbone patch.
assbot: Logged on 30-01-2015 05:09:29; asciilifeform: i just began live-fire test of the first really dangerous patch
asciilifeform: i just began live-fire test of the first really dangerous patch ☟︎
mod6: just make /sure/ to patch in the specified order.
asciilifeform: mod6: might have to patch by hand for now.
mod6: right now, i gotta figure out how to patch the test servers
mod6: YOu'll just have to patch it by hand.
asciilifeform: mircea_popescu: i'll patch it as soon as i get a working patcher (see last night's thread)
asciilifeform: appears to be wedged there. and has been for some time. but this is a build without the db lock patch. (ought to have wedged later, though ?)
decimation: too lazy to build patch sorry :(
asciilifeform: probably should explain patch
asciilifeform: not even the bdb locks patch
asciilifeform: oh don't forget the patch.
asciilifeform: turdatron == jurov's www patch submission gizmo
decimation: asciilifeform: did you have to patch gcc to fix the bug you posted the other day?
mod6: asciilifeform: sweet. what'd ya have to patch?
asciilifeform: note! had to patch the bitcoin
mod6: you know, before I did any of that, i thought about doing a detached sig, but then i wondered if something really funky would happen, like it would get somehow added to the patch-list automagically, and it doesn't belong there either.
mod6: - --verify chicken/bitcoin-asciilifeform.1.patch.sig
nubbins`: alf, you seem like the kind of guy who would, in another life, have a massive modular synthesizer in your basement, patch cables and eurorack modules all over the place
asciilifeform: may not even need to patch bitcoind
asciilifeform: one more patch then.
ben_vulpes: mkay, well i published a patch that bumps to 0.5.3.1, don't know if you saw that
ben_vulpes: it's amusing to watch everyone manually patch the closed source turdball to cope with each successive os x upgrade
asciilifeform: ben_vulpes, et al : these boxes won't be ready for battlefield until seeding patch.
mod6: here's a script that will build everything without the rm_checkpoints patch, with exception to the version number update:
asciilifeform: ben_vulpes, mod6: one patch that probably needs to happen in time for the 'fork war' is manual seed config.
mod6: and so should you. I'll be tabling that, updating the "Patch guide" and pulling the tarball with that patch down from the website as soon as my other vm hits current block.
mod6: in my most recent regression tests ( this past week, i've been leaving out the rm_checkpoints patch )
mod6: cool. in this case, im running v0.5.3[base]+patches{1,rm_rf_upnp,2,3,4,6(db_config),unreleased-version-update-patch}
assbot: Logged on 21-01-2015 02:59:39; Luke-Jr: I think the db_config patch may explain why it works, but could be potentially inadequate for the current consensus rules
assbot: Oracle Critical Patch Update - January 2015 ... ( http://bit.ly/1yI8rmj )
ben_vulpes: gavinandresen: i'm pointing out that the only acceptable patch for wallets is total removal.
gavinandresen: ben_vulpes: “patches welcome” Alex Morcos has a patch pending with better fee estimation code.
thestringpuller: i think asciilifeform submitted patch to gut all of it out.
ben_vulpes: <Luke-Jr> [] not sure why this patch is reducing set_lg_max.. << lol i love the stream of consciousness as Luke-Jr reads through source
nubbins`: it's all in easily-digestible chunks, no sweat to eyeball each patch & fully understand what's going on
Luke-Jr: not sure why this patch is reducing set_lg_max..
Luke-Jr: I think the db_config patch may explain why it works, but could be potentially inadequate for the current consensus rules ☟︎
Luke-Jr: which patch contains the post-0.5.3 hardforks?
mod6: Well, that's what there is: the v0.5.3 base + patch files.
ben_vulpes: you might consider using the patch names as a handle
ben_vulpes: or to say, run an lxr instance for each successive patch?
ben_vulpes: do you plan to run lxr on each patch?
mod6: so there was a check in 0.5.3 that requires checkpoints for this, which i removed in the rm_checkpoints patch
mod6: pulls all the patch packages, checks the hashes, checks the signatures.
mod6: asciilifeform: ah ok. hmm. well, too late to go back and edit the 1st patch. but i guess you can submit another that resolves that issue.
mod6: if persons submitting patches do not have L2 trust with assbot they may be submitted directly to me, and I will review. and will probably have to sponsor said patch myself.
mod6: joecool: here's a guide to submitting a patch: http://therealbitcoin.org/ml/btc-dev/2014-December/000022.html
asciilifeform: mod6: here's something i was gonna put in 'chicken' (init patch) but somehow forgot: version string.
joecool: mod6: are there plans to patch things like change predictability?
BingoBoingo: <joecool> mod6: solid, i'll open a bug to get it included in gentoo (and troll Luke-Jr) << He prolly has a historical version of his patch set s should be fine with it.
mod6: I have written a guide on how to patch the reference implementation. is that one too complex?
mircea_popescu: <assbot> Logged on 06-12-2014 02:50:54; decimation: apparently they kicked the guy off the project because of his refusal to commit a pronoun patch << well he wasn't being paid, so technically they uh.
assbot: Logged on 06-12-2014 02:50:54; decimation: apparently they kicked the guy off the project because of his refusal to commit a pronoun patch
kakobrekla: ethical_cleansing.patch
asciilifeform: BingoBoingo: we're discussing perhaps a ~1kb patch to kako's bot, rather than a serious project
mod6: here's a snippit of what my .bitcoin/database dir looks like with the BDB patch compiled in:
mod6: I was able to get upto the current block 2 different times with patches 1->5 (which includes ascii's patches, ben's patch, and my removal of checkpoints) with two seperate bdb configs. The first one just changed this: dbenv.set_lk_max_locks(40000); But ended up with a huge amount of BDB tx log files.
mod6: This month I've been doing testing with the patched version of the reference implementation. Main object here is to get from genesis block to current block, while being able to get past the wedge block 252450. I was able to do that, but not only did we test with ascii's patches and ben's UPNP patch, I added a patch (http://dpaste.com/1G3XY64) that removes the checkpoints, and a patch (http://dpaste.com/1K5M2TN) to configure BDB so the R.I. won't
undata: you decided your patch was a sensible default, though the "core devs" did not
adlai: Luke-Jr: why not make a patch that lets people dynamically enter filters into a running, compiled, program?
Luke-Jr: adlai: patch names?
adlai: probably the kind that doesn't like when patch names aren't consistent across package versions :)
Luke-Jr: so they can just modify the patch and drop it in /etc
Luke-Jr: I guess you could build the array at startup, but it's not really much easier to modify bitcoin.conf than to modify the patch
fluffypony: rithm: a black labrador wipes it's bum on a dirty patch of grass
decimation: apparently they kicked the guy off the project because of his refusal to commit a pronoun patch ☟︎
jurov: oh mod6 i forgot to suggest to send it as "patch" so that people can just add signatures.. nm, i'll add it there somehow
assbot: Outside in - Involvements with reality » Blog Archive » Chaos Patch (#38) ... ( http://bit.ly/1FLantI )