<jeffro256[m]> "I believe you can pick random..." <- It's sum(out) - sum(in 0 .. n-1) for in n.

The correct blinding factor is whatever causes in x == out x

You are provided an x in your output. You create xs in the outputs you make. You randomly select in xs for your TX, yet the final x must...

CLSAG proofs take the output you're spending's commitment and verify they have an equivalent H component for the input you're defining ("pseudo-out").

Just trying to fill in some gaps there ;)

Anyone can direct me to the proper room for a monero gui stuck transaction 17 hours now

#monero

tingels4996[m]: there is a monero gui channel on matrix, try to join it

Ok i will try that cant seem to get any help from anyone in the monero community

like I said I'll help you but join the other channel as this is off topic here

so how to join matrix as it goes to element

which one fractal ,wechat??

either search for "monero gui" in element, or use web.libera.chat and connect to #monero-gui

#monero-gui:monero.social

ok will try that

.merge+ 8384

Added

sech1: regarding binaries, i'm available to do them whenever we need

i just feel like everything is half ready, still don't know about ledger

i mean we can risk it and branch + tag now and add all remaining fixes in the .1 point release

what's half ready except multisig and ledger?

.merges

I'm getting reorgs on almost every single block on mainnet.

new top block candidate: 2646306 -> 2646307

new top block candidate: 2646307 -> 2646308

Or is that the host that's to slow ??

my node is fine, nothing like this in logs

Im all good here

You're probably connected to a bad peer

FWIW that message does not say there's a reorg. I assume you have others that do ?

sech1: the glibc issue

i spent the last days on it but wasn't able to figure it out

Sync data returned a new top block candidate: 2646307 -> 2646308 [Your node is 1 blocks (2.0 minutes) behind]

New top block candidate is saying you are currently behind and syncing, and a node sent you a newer block than what you were working on

glibc issue, is it important? Ubuntu 16.04 is EOL

sech1: not super important, but at least something we should fix for the .1 update

as time passes, we'll get more and more 3rd party dependencies that do this

I got a shitload of those messages. Up to 6 blocks behind.

someone experienced with autotools can probably solve it in a couple minutes

autoconf or whatever it's called

autoconf has always been black magic to me

the compatibility code exists already, i just have to patch the repo so that autoconf uses the compatibility code instead of the system lib on linux at least

Come over to #monero. Ask there if needed. Tldr all seems to be fine and I assume youre using a pi, hdd or bad internet resulting in slow sync

If up to 6 blocks behind, and it catxhes up, either your node loses connections a lot (and then finds a node with a higher chain), or you find confused nodes.

Or #monero-support:monero.social

Maybe on purpose, you never know. Harmless stuff on porpose happened before.

selsta: I *think* you have to modify config.h.cmake.in, add a line with your particular define you want to enable.

It's not actually using autofoo IIRC, just using cmake to generate the config it'd have.

they don't use cmake at all

we patched our old version of unbound to use cmake, but we don't use that anymore

Oh. Maybe I have an old version then.

OK, 2019. A patch by me. Must be old indeed :D

So the monero build system actually runs autofoo in external/unbound now ?

no, we completely removed the submodule

OK, then why are you trying to do :S

You know what, I clearly have no idea what I'm talking about here, so nevermind me.

everything works fine, it's just that unbound bumps our min glibc from 2.17 to 2.26

which means ubuntu 16.04 users can't run the release binaries anymore

They don't seem to be using autofoo, I just looked.

i wanted to add a patch here: github.com/monero-project/monero/bl…contrib/depends/packages/unbound.mk so that compatibility code for the function `explicit_bzero` and `reallocarray` is used.

i have an issueeeeeeeeeeee

<selsta> "i mean we can risk it and branch..." <- it would be better to risk and start reading code

build/libgdmeowletxmr.dll: src/GDLibrary.cpp src/GDMeowletXMR.cpp build/meowletxmr.a godot-cpp/bin/libgodot-cpp.windows.debug.64.a

$(CC) -shared -o $@ $(filter-out headers, $^) -static -static-libgcc -static-libstdc++ $(GODOT_INCLUDES) $(CFLAGS) $(CLIBFLAGS) -lsetupapi -lhid $(LDFLAGS)

this is my recipeeee

when i link wallet2 things to my DLL

it can't be load somehowwww

<MeowingCat> "it can't be load somehowwww" <- do you have public with these experiments ?

* have public repo with these

This is a particuarly useless request. There are no details, and we have no idea it's even related to monero.

Any multiline details to go to a paste site (eg, paste.debian.net).

<sech1> "autoconf has always been black..." <- gnu.org/savannah-checkouts/gnu/auto…/manual/autoconf-2.71/autoconf.html

lrde.epita.fr/~adl/dl/autotools.pdf, the most user friendly probably

i think some of libraries that wallet2 is using maybe still trying to load dynamically

MeowingCat_: provide more info or post your build script somewhere, so that it would be possible to debug it

moneromooo, would you be willing to take a glance at #8385 and see if the concept checks out ?

ooo123ooo1234567, okiiiiii i will

monero-project/monero #8384#issuecomment-1154219150, is it possible to put it into approval ?

is there a PE equivalent of readelf?

msys has binutils, any problems with them ?

readelf is not workinf for PE

objdump maybe

depends on what you're trying to read

ohh trying thank youu

goooood objdump works

something is wrong with wallet2

really interesting

MeowingCat: what did you find?

ummmmmmm

g++ -shared -o build/libgdmeowletxmr.dll src/GDLibrary.cpp src/GDMeowletXMR.cpp build/meowletxmr.a godot-cpp/bin/libgodot-cpp.windows.debug.64.a -Igodot-cpp -Igodot-cpp/godot-headers -Igodot-cpp/include -Igodot-cpp/include\gen -Igodot-cpp/include\core -Igodot-cpp/godot-headers -Igodot-cpp/include -Igodot-cpp/include\gen -Igodot-cpp/include\core -g -std=c++17 -fstack-protector -static -static-libgcc -static-libstdc++ -lsodium -lboost_serializati

on-mt -lboost_thread-mt -lboost_system-mt -lboost_date_time-mt -lboost_filesystem-mt -lboost_chrono-mt -lboost_program_options-mt -lboost_regex-mt -lboost_locale-mt -lssl -lcrypto -lcrypt32 -lz -lwsock32 -lws2_32 -lprotobuf -lusb-1.0 -lhidapi -liphlpapi -liconv -lsetupapi -lhid

thisssssssssssss

meowletxmr.a has everything my library, wallet2 and wallet2 deps

im making it with ar

is it complete list of deps to compile successfully ?

im using my static library for test executable

everything is ok

when i don't use something from that in Godot DLL it works

buttt when i use wallet2 things in DLL (it is linking my static library that contains wallet2 tooo)

DLL can't load

:(((((

not enough information to debug it

i feeeeel like i should build wallet2 things with -fPIC

or maybe not

idkkk

damn things :(((((((((((((((((

what can be wrong with using wallet2 in a dynamic library??

jeffro256: wow

xoxoxo

is there anyone using wallet2 for dynamic library??

Yes.

all windows code is PIC already

MeowingCat: just run monero binaries compilation with high verbosity and check what is going on there; why is it so difficult ?

* binaries compilation for target platform with high

my DLL can't be load if i link wallet2

Whlie I have nfi what your problem is since you persist in giving no details whatsoever, if you link a dynamic lib with a static archive, I suspect externals will be resolved in the dynamic lib, but any used by whatever runtime loads the dynamic lib won't find anything from the static archive.

Maybe that's not what you're doing though. Check that before, er.. doing something.

im building my library with linking all wallet2 things

and using that for test executable

buttt when i build DLL and link wallet2 my DLL can't be loaded

idk what can be wrong because it should workk

ummmm yessssssss

we just tried my test executable on a different machine

it errors can't load libbooooooooost dynamic library

<kevin1> @here @everyone $1800 into $40k in the last 30 days.. If you really want to make huge profits on trading then, Join this winning chat:~~~~ discord.gg/cjaE5y5GzA

buttt im building my DLL with `-static`

when i decompile it i seeeee everything

interestinggg

i tried to use something with less dependencies from wallet2

i just used Monero::Utils::isAddressLocal()

and my DLL can be load when i use thiss

i found something

idk why buttt while building the DLL and linking wallet2 things to my DLL

i had to add these libraries: -lsetupapi -lhid

when i only use Monero::Utils::isAddressLocal() and link necessary things for that

my DLL can be loaded

when i use wallet API there is one more DLL dependency

SETUPAPI.dll

when i don't add this libraries im getting "undefined reference" errors

C:/msys64/mingw64/bin/../lib/gcc/x86_64-w64-mingw32/12.1.0/../../../../x86_64-w64-mingw32/bin/ld.exe: C:/msys64/mingw64/bin/../lib/gcc/x86_64-w64-mingw32/12.1.0/../../../../lib\libhidapi.a(hid.o):(.text+0xce8): undefined reference to `__imp_SetupDiGetClassDevsA'

while building my DLL

but i can build my test executable

-lsodium -lboost_serialization-mt -lboost_thread-mt -lboost_system-mt -lboost_date_time-mt -lboost_filesystem-mt -lboost_chrono-mt -lboost_program_options-mt -lboost_regex-mt -lboost_locale-mt -lssl -lcrypto -lcrypt32 -lz -lwsock32 -lws2_32 -lprotobuf -lusb-1.0 -lhidapi -liphlpapi -liconv

MeowingCat: google says -lsetupapi

what is SETUPAPI thing and why do i neeeeed to link that when i build a DLL but i don't neeeed when i build an executable?

selsta, yessss i already added thatt

butt while i build an executable i don't need to add that library

i need to add while building a shared library

and there is no a SETUPAPI.dll in my machine

when i build with -lsetupapi my DLL can't be loaded

and there is no even a SETUPAPI.dll in my machine :((((((((((((

mingw comes with libsetupapi.a

i think it would link that thing as statically buttt maybe it doesn't even exist?

because im using -static

but i see SETUPAPI.dll in "objdump -p ..."

what is this setupapi thing???? why do i neeed that when only building a DLL

i don't think you will be able to find the answers here, this isn't anything monero related and most people here are not windows devs

I have one, in my Windows 10, where I would expect it: SETUPAPI.dll in /windows/system32

and variants for 32 on 64, in C:\Windows\SysWOW64

If only one of the many packages you link has only one call into that, in its Windows implementation, you have a problem

And maybe they don't link statically, they may use GetProcAddress

rbrunner, $(CC) -shared -o $@ $(filter-out headers, $^) $(GODOT_INCLUDES) $(CFLAGS) -static -static-libgcc -static-libstdc++ $(CLIBFLAGS)

my GCC command

Not sure how that should be relevant

oh i found setupapi.dll in c:/windows/system32

Maybe Boost calls into that Windows API? Or Protobuf? Or whatever

hidapi?

Told you so, didn't I?

i had started a "find ..." but explorer.exe couldn't find that before lol

when i use something from wallet2 the difference in "objdump -p" is only setupapi.dll

and the DLL can't be loaded

Windows software is usually one crazy tangle, everything calls almost everything else :)

i don't like Booooooooooooooost

Me neither, but doesn't help. We can't ignore it to death.

But tell you what, I like Qt even less :)

Try linking some dev version of the Monero GUI wallet, as work-in-progress, on Windows.

:((((((((((((((

can't i link setupapi thing statically????

or is the problem about that???????

idkkk

last reply from ledger was on 2022-04-29, where they said they will get back to us

Oh, yeah, we also have that little Ledger problem

i predicted it months ago :D

Right

ruh roh

perhaps ledger specific code doesnt belong in monero mainline?

i'm missing context but it seemed relevant

both ledger and trezor have integration into monero-wallet-cli and monero-gui

it's not ideal yes, but people wanted hw wallets and ledger and trezor didn't want to make their own wallet

i guess that's a good point

trezor has a dedicated dev for monero (i think he is a freelancer and they pay him when monero work has to be done), ledger doesn't

<selsta> "i predicted it months ago :D" <- if no one will work on unbound then it's predictable result too

* result too, but it doesn't mean that it's impossible to write update and merge into their repo

* result too, but it doesn't mean that it's impossible to write ledger monero-app update and merge into their repo

yes, i said it already we will have to do it ourself

Question: I know monero stores its blockchain on a node locally in lmdb. How does it implement decentralization? Is there (third-party) library used for that or there some additional (native) code interacting with lmbd that does that?

* Question: I know monero stores its blockchain on a node locally in lmdb. How does it implement decentralization? Is there a (third-party) library used for that or there are some additional (native) code interacting with lmdb that does that?

no, monero does not use a third party library for "decentralization"

i mean it's based on cryptonote, if you consider that a library

he is probably asking about p2p technology

blockchain sync in general, bulletproofs etc

jozsef Can you expand upon what you mean by "implement decentralization"?

jeffro256[m]: We want to know if its possible to implement a decentralized database or would we need a blockchain to do that?

layter[m]: there is no standalone lib that adds decentralization on top of lmdb without any other overhead of cryptocurrency

layter[m]: layter: You might want to research what OpenBazaar did or look into Particl

chesterfield[m]: Alright, thanks. I'll look into those projects

I have no idea what you're building, but i'm assuming it's decentralized marketplace related?

chesterfield[m]: Correct 😅

pony_montana[m]: Hi guys, that's a paper about wealth inequality discriminations in monero, bitcoin and gold. I will be happy to know what do you think about! :)

pony_montana[m]: not a #monero-dev topic

jozsef[m]: that's what nakamoto consensus, mining/pow, and p2p networks are for

lmdb is used by p2p nodes locally only

decentralized state is basically emergent

should ask why you even want a decentralized database in the first place. as a general rule, they suck. slow performance, terrible scaling.

pony_montana[m]: try #monero-community. And maybe #monero-research-lab.

I'd probably use -lounge instead.

thanks! :)

anyway, there is no way to quantify wealth inequality in monero, so it doesn't sound like a very useful topic

The correct place is #monero-community-dev:monero.social :)

<chesterfield[m]> "I have no idea what you're..." <- github.com/neroshop/neroshop, indeed

Right, good question @hyc. Maybe instead it is better to state the problem to be solved: decentralized marketplace with a database (pics, text, etc.) that replicates in a p2p fashion. How would you suggest approaching an implementation?

I'm thinking locally anything, lmbd, sql, whatever, then data must be replicated p2p somehow so everyone sees the same eventually. Sounds like a decentralized (p2p-synced) database.

<jozsef[m]> "Right, good question @hyc. Maybe..." <- that's too complex question, try to study suggested alternatives and their pros/cons firstly

this thing made me crazy

<cryptogrampy[m]> "The correct place is #monero-..." <- 😄👌

which IRC clients are you using????

jozsef[m]: also doesn't sound like a #monero-dev question. but I guess given your problem statement, still need to ask whether the p2p participants are trustworthy or not

hyc: Correct, not trictly a monero-dev question, other than being curious how this is done in monero. Nevertheless, I think the responses were useful in that I think one could simply try implementing a p2p commhnication algorithm on top of a local database without anything fancy. Perhaps sending hashes around to query what each peer has.

s/commhnication/communication/