[2017-04-14]How Greg Maxwell Exploited Bitcoin Unlimited in Every Way Possible

[joshy23]

How Greg Maxwell Exploited Bitcoin Unlimited in Every Way Possible

Segregated Witness (SegWit) support has been on the rise ever since the disclosure of the covert usage of AsicBoost by Bitcoin Core developer Greg Maxwell.

On April 5, Maxwell sent out a letter entitled “Inhibiting a covert attack on the Bitcoin POW function” to the Bitcoin development mailing list. In it, Maxwell explained the possibility of utilizing a method called AsicBoost to exploit a flaw in the Proof of Work function of Bitcoin.

Maxwell wrote:
“There are two major ways of exploiting the underlying vulnerability: One obvious way which is highly detectable and is not in use on the network today and a covert way which has significant interaction and potential interference with the Bitcoin protocol. The covert mechanism is not easily detected except through its interference with the protocol. In particular, the protocol interactions of the covert method can block the implementation of virtuous improvements such as segregated witness.”

Bitmain reaction

Upon the release of Maxwell’s letter, the vast majority of the community assumed that the company or the ASIC manufacturer in question was Bitmain. In fact, Bitcoin journalist Kyle Torpey confirmed with a source that Bitmain was the company mentioned in Maxwell’s letter.

As expected, Bitmain refuted the claims of Maxwell with a response statement. In it, Bitmain denied the usage of AsicBoost on the mainnet of Bitcoin but admitted to testing the technology on testnet, which further validated the statement of Maxwell.

“I never alleged that I could prove they were using it on mainnet, only that their hardware secretly supported it. They've admitted to putting it in their hardware and to using it on testnet. They claim that haven't used it on mainnet "for the good of the network"-- but the constantly produce empty blocks (a potential asicboost sign) and argued that it was fine because the protocol permits it, when people complained,” Maxwell said in a separate statement.

Controversy deepens

Controversy on the Bitmain-AsicBoost discussion deepened when various users discovered several loopholes in the public statement of Bitmain. Although the company denied the actual usage of AsicBooost on Bitcoin’s mainnet, it did not specify whether their statement applied to its subsidiaries and other companies they own.


"More importantly, Bitmain admitted their ownership of a patent on AsicBoost."

However, the company was criticized for failing to credit the original inventors of AsicBoost Timo Hanke and Sergio Lerner, who reportedly hold a patent on AsicBoost elsewhere.

Whalepanda, a cryptocurrency trader, further emphasized in his blog post that Bitmain CEO Jihan Wu’s strange behavior and his support for AsicBoost-compatible solutions or proposals such as a BCoin’s extension block proves his intent to protect AsicBoost from being disabled permanently with solutions like SegWit.

Whalepanda wrote:


“Yesterday BCoin’s proposal came with Extension Blocks which is compatible with ASICBoost. BCoin didn’t follow the normal rules of Bitcoin proposals but got the media immediately involved. EB was nothing new, they just needed another excuse to stall.”

Apart from his AsicBoost discovery, Maxwell and the rest of the Bitcoin Core development team privately disclosed bugs and technical errors in the Bitcoin Unlimited software to its development team. In return, the Bitcoin Unlimited development team has been attempting for weeks to publicize supposed bugs in Core’s software although they turned out to be non-existent.

“Next time you have a suspected vulnerability in Bitcoin Core, it would be helpful if told us immediately instead of discussing it in public for 13 days first. There are vulnerabilities in unlimited which have been privately reported to you in Unlimited by Bitcoin Core folks which you have not acted on, sadly. More severe than this one, in fact,” Maxwell said.

So far, Maxwell has made a point in disclosing bugs and errors in the Bitcoin Unlimited software and unraveling the intent behind some mining pools and companies which opposed SegWit.

https://cointelegraph.com/news/how-greg-maxwell-exploited-bitcoin-unlimited-in-every-way-possible

[European Central Bank]

crappy headline.

anything that wants to become the prime software for bitcoin should be exploited in every way possible in search of flaws, and if they are found then it should be shouted loud and clear.

nothing deserves an easy ride. i hope segwit's been given the same scrutiny.

[numismatist]

“I never alleged that I could prove they were using it on mainnet, only that their hardware secretly supported it. They've admitted to putting it in their hardware and to using it on testnet. They claim that haven't used it on mainnet "for the good of the network"-- but the constantly produce empty blocks (a potential asicboost sign) and argued that it was fine because the protocol permits it, when people complained,” Maxwell said in a separate statement.

Are there statistical data published anywhere on which pools do produce these empty blocks? IP's should be geolocateable. That could shine some light onto things.

Or maybe that has been a typo and Maxwell wrote "they constantly produce" and can prove that by data?

[TraderTimm]

“I never alleged that I could prove they were using it on mainnet, only that their hardware secretly supported it. They've admitted to putting it in their hardware and to using it on testnet. They claim that haven't used it on mainnet "for the good of the network"-- but the constantly produce empty blocks (a potential asicboost sign) and argued that it was fine because the protocol permits it, when people complained,” Maxwell said in a separate statement.

Are there statistical data published anywhere on which pools do produce these empty blocks? IP's should be geolocateable. That could shine some light onto things.

Or maybe that has been a typo and Maxwell wrote "they constantly produce" and can prove that by data?

The data is out there - and yes, while most pools can produce empty blocks, the Chinese mining concerns are the ones pumping out empty blocks above the statistically expected norms. They're complicit and its been discussed in depth elsewhere.

[veleten]

empty blocks only suit BU crew and their "low block size is the root of evil" agenda
there is no point in creating empty blocks,you lose on miner fees
but if you want to make a statement,meanwhile spamming the blockchain with heavy transactions
it all makes sense,remember "create a problem-offer solution"

[Quantus]

Greg Maxwell should post a detailed and technical description of how ASIC-boost works.
To my understanding they reorder transaction IDs and create lots of fake transactions, reordering them to create lots of different merkle root values and thus lots of different yet acceptable blocks so they can hash all of them at the same time not just one.
Someone refereed to it as increasing the key space. So instead of building just one block and hashing it till you find an output that meets the difficulty level you build 100 or 1000 acceptable blocks and then hash all of them simultaneously.
But this is detrimental to the network as a whole.  It creates massive and unnecessary bloat in the blockchain. It will lead to higher costs for full node operators and lead to consolidation of the mining community. 

[TraderTimm]

Greg Maxwell should post a detailed and technical description of how ASIC-boost works.
To my understanding they reorder transaction IDs and create lots of fake transactions, reordering them to create lots of different merkle root values and thus lots of different yet acceptable blocks so they can hash all of them at the same time not just one.
Someone refereed to it as increasing the key space. So instead of building just one block and hashing it till you find an output that meets the difficulty level you build 100 or 1000 acceptable blocks and then hash all of them simultaneously.
But this is detrimental to the network as a whole.  It creates massive and unnecessary bloat in the blockchain. It will lead to higher costs for full node operators and lead to consolidation of the mining community. 

This is pretty comprehensive -- https://lists.linuxfoundation.org/pipermail/bitcoin-dev/2017-April/013996.html

[Quantus]

Read it twice still don't understand it, I think its just beyond me.

Question, what kind of support does this BIP have?  Will it really help?

[TraderTimm]

Read it twice still don't understand it, I think its just beyond me.

Question, what kind of support does this BIP have?  Will it really help?

I really don't know how to help you. First you wanted a "detailed and technical description of how ASIC-boost works", and when one is provided you can't "get it".

Be sure to ask for what you really want? You confuse me.