AllCoreDevs - Testing #103
Transcript
- Dima Gusakov | Lido
gm
- Radek | solidity
Hi All.
- Dima Gusakov | Lido
Wen Glamsrdam? This is the only thing I want to know 😁
- Radek | solidity
Reacted to "Wen Glamsrdam? This ..." with 🍿
- jochem-brouwer
gm gm
- Louis
gm gm
- danceratopz
gm everyone! https://github.com/ethereum/pm/issues/2299
- Potuz
gm
- NC
no music in ACDT??
- Mega | ethlambda
gm
- Guillaume's Circleback
Guillaume is using Circleback (https://circleback.ai) to record and generate AI meeting notes for this meeting.
- Louis
Reacted to "Guillaume is using Circleback (https://circleback.ai) to record and generate AI meeting notes for this meeting." with 😃
- spencer
Reacted to "Guillaume is using..." with 😃
- hangleang
gm gm
- Parithosh Jayanthi
Reacted to "Guillaume is using Circleback (https://circleback.ai) to record and generate AI meeting notes for this meeting." with 😃
- Barnabas
Reacted to "Guillaume is using Circleback (https://circleback.ai) to record and generate AI meeting notes for this meeting." with 😃
- joshua
Reacted to "Guillaume is using Circleback (https://circleback.ai) to record and generate AI meeting notes for this meeting." with 🤖
- Dhanush | EIPsInsight
gm gm
- danceratopz
- Stefan Starflinger
gm
- Dhruv (conduit.xyz)
gm
- Statuette
Gm 💛
- Dima Gusakov | Lido
Reacted to "Guillaume is using Circleback (https://circleback.ai) to record and generate AI meeting notes for this meeting." with 😃
- cayman
gm
- Łukasz Rozmej
Reacted to "Guillaume is using Circleback (https://circleback.ai) to record and generate AI meeting notes for this meeting." with 😃
- Filip | Biconomy
gmgm
- Rahul (EF)
Hey app teams! Very happy to see you all here and work together
- spencer
Reacted to "Hey app teams! Ver..." with ➕
- Will Corcoran
Reacted to "Hey app teams! Very happy to see you all here and work together" with ❤️
- Mario Vega
Reacted to "Hey app teams! Very happy to see you all here and work together" with ❤️
- Statuette
Reacted to "Hey app teams! Very happy to see you all here and work together" with ❤️
- Toni Wahrstaetter
Reacted to "Hey app teams! Very happy to see you all here and work together" with ❤️
- spencer
Reacted to "Hey app teams! Ver..." with ❤️
- danceratopz
Reacted to "Hey app teams! Very ..." with ❤️
- danceratopz
Reacted to "Hey app teams! Very ..." with ➕
- danceratopz
Vitalik's blog post on 8037 and other invariants: https://docs.fileverse.io/document/dUHG8ythBk3jJuQEjcE7m6#k=tVLgvLrIqyEsCh7qUmACb9ggVmybyS_aIZZsAMvHDAk
- Vitalik
https://docs.fileverse.io/document/dUHG8ythBk3jJuQEjcE7m6#k=tVLgvLrIqyEsCh7qUmACb9ggVmybyS_aIZZsAMvHDAk this is the post!
- danceratopz
Replying to "https://docs.filever..." be faster
- Statuette
Reacted to "https://docs.fileverse.io/document/dUHG8ythBk3jJuQEjcE7m6#k=tVLgvLrIqyEsCh7qUmACb9ggVmybyS_aIZZsAMvHDAk this is the post!" with 💛
- Mario Vega
Reacted to "https://docs.fileverse.io/document/dUHG8ythBk3jJuQEjcE7m6#k=tVLgvLrIqyEsCh7qUmACb9ggVmybyS_aIZZsAMvHDAk this is the post!" with 💛
- jochem-brouwer
Antwoord verzenden naar "https://docs.filev..." frontran by dan :)
- Vitalik
Reacted to "be faster" with 😂
- jochem-brouwer
Heeft gereageerd op "be faster" met 😂
- Barnabas
Reacted to "be faster" with 😂
- Mario Vega
Reacted to "be faster" with 😂
- Statuette
Reacted to "be faster" with 😂
- Chris Berry
Reacted to "be faster" with 😂
- jochem-brouwer
Heeft gereageerd op "Vitalik's blog pos..." met 👍
- spencer
Reacted to "https://docs.filev..." with 💛
- spencer
Reacted to "be faster" with 😂
- danceratopz
Reacted to "https://docs.filever..." with 💛
- Statuette
https://docs.fileverse.io/document/dUHG8ythBk3jJuQEjcE7m6#k=tVLgvLrIqyEsCh7qUmACb9ggVmybyS_aIZZsAMvHDAk Sandwich attack
- jochem-brouwer
Heeft gereageerd op "https://docs.filev..." met 💛
- danceratopz
Reacted to "https://docs.filever..." with ➕
- danceratopz
Reacted to "https://docs.filever..." with 🤣
- Mario Vega
Reacted to "https://docs.fileverse.io/document/dUHG8ythBk3jJuQEjcE7m6#k=tVLgvLrIqyEsCh7qUmACb9ggVmybyS_aIZZsAMvHDAk Sandwich attack" with 🤣
- Enter Meeting Info
Reacted to "https://docs.filev..." with 🤣
- Toni Wahrstaetter
Reacted to "https://docs.fileverse.io/document/dUHG8ythBk3jJuQEjcE7m6#k=tVLgvLrIqyEsCh7qUmACb9ggVmybyS_aIZZsAMvHDAk Sandwich attack" with 🤣
- Daniel
Reacted to "be faster" with 😂
- Chris Berry
Reacted to "https://docs.fileverse.io/document/dUHG8ythBk3jJuQEjcE7m6#k=tVLgvLrIqyEsCh7qUmACb9ggVmybyS_aIZZsAMvHDAk Sandwich attack" with 🤣
- jochem-brouwer
Heeft gereageerd op "https://docs.filev..." met 🤣
- Iván | ethrex
Reacted to "https://docs.fileverse.io/document/dUHG8ythBk3jJuQEjcE7m6#k=tVLgvLrIqyEsCh7qUmACb9ggVmybyS_aIZZsAMvHDAk this is the post!" with 💛
- Iván | ethrex
Reacted to "be faster" with 😂
- hangleang
Reacted to "frontran by dan :)" with 😂
- Ryan McPeck
Are there action items on how we are going to communicate better / faster next time?
- Iván | ethrex
Reacted to "Vitalik's blog post on 8037 and other invariants: https://docs.fileverse.io/document/dUHG8ythBk3jJuQEjcE7m6#k=tVLgvLrIqyEsCh7qUmACb9ggVmybyS_aIZZsAMvHDAk" with 👍
- Barnabas
EOF gang please rise up
- spencer
Reacted to "EOF gang please ri..." with 😿
- jtraglia
Reacted to "be faster" with 😂
- jochem-brouwer
Heeft gereageerd op "Are there action i..." met ➕
- Milos | enso.build
Reacted to "EOF gang please ri..." with 😿
- danceratopz
Reacted to "Are there action ite..." with ➕
- danceratopz
Reacted to "Are there action ite..." with 👍
- Radek | solidity
Replying to "EOF gang please rise..." at the post
- Rahul (EF)
@ryan Will briefly touch on this towards the end of the call
- Ryan McPeck
Reacted to "@ryan Will briefly touch on this towards the end of the call" with 🫡
- Parithosh Jayanthi
Reacted to "EOF gang please rise up" with 😿
- Satyajit
Reacted to "frontran by dan :)" with 😂
- Parithosh Jayanthi
Reacted to "be faster" with 😂
- Parithosh Jayanthi
Reacted to "https://docs.fileverse.io/document/dUHG8ythBk3jJuQEjcE7m6#k=tVLgvLrIqyEsCh7qUmACb9ggVmybyS_aIZZsAMvHDAk this is the post!" with 💛
- Milos | enso.build
Reacted to "Are there action i..." with ➕
- Radek | solidity
Reacted to "EOF gang please rise..." with 🫡
- Satyajit
Reacted to "https://docs.filev..." with 🤣
- Statuette
Reacted to "frontran by dan :)" with 😂
- joshua
Reacted to "https://docs.fileverse.io/document/dUHG8ythBk3jJuQEjcE7m6#k=tVLgvLrIqyEsCh7qUmACb9ggVmybyS_aIZZsAMvHDAk Sandwich attack" with 🤣
- Toni Wahrstaetter
Replying to "Are there action items on how we are going to communicate better / faster next time?" I think there's an agenda item on that topic coming later but my summary would be: * not only testing worst-case blocks but also coding pattern * define invariants that might break through repricing or pq * ensure the process only allows eips to be scheduled where the impact is well understood * align with app devs and security auditing earlier in the process
- jochem-brouwer
Heeft gereageerd op "@ryan Will briefly..." met 👍
- Marius Van Der Wijden (M)
Reacted to "EOF gang please rise up" with 😿
- Marius Van Der Wijden (M)
Reacted to "EOF gang please rise up" with 🫡
- spencer
Reacted to "I think there's an..." with ❤️
- Marius Van Der Wijden (M)
Reacted to "I think there's an agenda item on that topic coming later but my summary would be: * not only testing worst-case blocks but also coding pattern * define invariants that might break through repricing or pq * ensure the process only allows eips to be scheduled where the impact is well understood * align with app devs and security auditing earlier in the process" with ❤️
- danceratopz
Reacted to "@ryan Will briefly t..." with 💙
- Dima Gusakov | Lido
Can we get a link to the report?
- marc | wolovim
Reacted to "I think there's an agenda item on that topic coming later but my summary would be: * not only testing worst-case blocks but also coding pattern * define invariants that might break through repricing or pq * ensure the process only allows eips to be scheduled where the impact is well understood * align with app devs and security auditing earlier in the process" with ❤️
- Filip | Biconomy
+1
- Greg K | Lido
Reacted to "Can we get a link to the report?" with 👍
- Dmytro Khimchenko
Reacted to "Can we get a link ..." with 👍
- CPerezz
Reacted to "Can we get a link to..." with 👍
- Dragan Rakita
Reacted to "EOF gang please rise up" with 🫡
- Dragan Rakita
Reacted to "Are there action items on how we are going to communicate better / faster next time?" with ➕
- Mario Vega
Reacted to "I think there's an agenda item on that topic coming later but my summary would be: * not only testing worst-case blocks but also coding pattern * define invariants that might break through repricing or pq * ensure the process only allows eips to be scheduled where the impact is well understood * align with app devs and security auditing earlier in the process" with ❤️
- Dragan Rakita
Reacted to "be faster" with 😂
- danceratopz
Reacted to "Can we get a link to..." with 👍
- jochem-brouwer
Heeft gereageerd op "I think there's an..." met ❤️
- Hubert (ChainSecurity)
Replying to "Can we get a link ..." Happy to do so, we just need to clarify if it is fine to point out specific problems in a public report or whether the public report should only contain summaries.
- Dima Gusakov | Lido
Replying to "Can we get a link to..." Given the presentation mentioning staking protocols, I want to request it in private to act from the Lido side
- Dima Gusakov | Lido
Replying to "Can we get a link to..." dmitry.gusakov@lido.fi
- Potuz
Replying to "Can we get a link to..." this is a good question, if there's any chance the fork goes on without spec changes, some of these specific issues may perhaps be better privately informed to the affected protocols instead?
- Dima Gusakov | Lido
Reacted to "this is a good question, if there's any chance the fork goes on without spec changes, some of these specific issues may perhaps be better privately informed to the affected protocols instead?" with 👍
- Lumi | Offchain Labs
Reacted to "this is a good question, if there's any chance the fork goes on without spec changes, some of these specific issues may perhaps be better privately informed to the affected protocols instead?" with 👍
- joshua
Reacted to "this is a good question, if there's any chance the fork goes on without spec changes, some of these specific issues may perhaps be better privately informed to the affected protocols instead?" with 👍
- d
Reacted to "this is a good que..." with 👍
- Parithosh Jayanthi
Reacted to "Can we get a link to the report?" with 👍
- Satyajit
Reacted to "this is a good que..." with 👍
- Parithosh Jayanthi
Reacted to "this is a good question, if there's any chance the fork goes on without spec changes, some of these specific issues may perhaps be better privately informed to the affected protocols instead?" with 👍
- Hubert (ChainSecurity)
Reacted to "Given the presenta..." with 👍
- Marius Van Der Wijden (M)
Replying to "Can we get a link to..." We have started a big outreach campaign to privately contact affected protocols
- Parithosh Jayanthi
Reacted to "We have started a big outreach campaign to privately contact affected protocols" with ❤️
- Radek | solidity
Reacted to "We have started a bi..." with ❤️
- jochem-brouwer
Heeft gereageerd op "We have started a ..." met ❤️
- Potuz
Reacted to "We have started a bi..." with ❤️
- Statuette
Reacted to "We have started a big outreach campaign to privately contact affected protocols" with ❤️
- hangleang
Reacted to "We have started a ..." with ❤️
- Greg K | Lido
Reacted to "this is a good question, if there's any chance the fork goes on without spec changes, some of these specific issues may perhaps be better privately informed to the affected protocols instead?" with 👍
- Dragan Rakita
Reacted to "We have started a big outreach campaign to privately contact affected protocols" with ❤️
- Dima Gusakov | Lido
Replying to "Can we get a link to..." Or to security@lido.fi
- Hubert (ChainSecurity)
Reacted to "dmitry.gusakov@lid..." with 👍
- Hubert (ChainSecurity)
Reacted to "Or to security@lid..." with 👍
- Hubert (ChainSecurity)
Replying to "Can we get a link ..." @Dima Gusakov | Lido , yes we will.
- Dragan Rakita
Reacted to "I think there's an agenda item on that topic coming later but my summary would be: * not only testing worst-case blocks but also coding pattern * define invariants that might break through repricing or pq * ensure the process only allows eips to be scheduled where the impact is well understood * align with app devs and security auditing earlier in the process" with ❤️
- Tanguy (ChainSecurity)
Replying to "Can we get a link ..." Here are already the ChainSecurity slides: https://docs.google.com/presentation/d/1FyMPRzlsiOHzs4ia-3nAAaPGLmO_yzLurmvU2OIYI18/edit?usp=sharing
- Tony Chen | TokenPocket
Reacted to "Here are already the ChainSecurity slides: https://docs.google.com/presentation/d/1FyMPRzlsiOHzs4ia-3nAAaPGLmO_yzLurmvU2OIYI18/edit?usp=sharing" with 👍
- jochem-brouwer
Want to thank these researches to explicitly figure out if the flags are real instead of forwarding false positives to these protocols. Thanks so much!
- spencer
Reacted to "Want to thank thes..." with ❤️
- Louis
Reacted to "Want to thank these researches to explicitly figure out if the flags are real instead of forwarding false positives to these protocols. Thanks so much!" with ❤️
- Dhanush | EIPsInsight
Reacted to "Want to thank these researches to explicitly figure out if the flags are real instead of forwarding false positives to these protocols. Thanks so much!" with ❤️
- KowalskiDefi
Reacted to "Want to thank thes..." with ❤️
- Ilias | Dedaub
Reacted to "Want to thank thes..." with ❤️
- Toni Wahrstaetter
Reacted to "Want to thank these researches to explicitly figure out if the flags are real instead of forwarding false positives to these protocols. Thanks so much!" with ❤️
- Will Corcoran
Reacted to "Want to thank these researches to explicitly figure out if the flags are real instead of forwarding false positives to these protocols. Thanks so much!" with ❤️
- Mario Vega
Reacted to "Want to thank these researches to explicitly figure out if the flags are real instead of forwarding false positives to these protocols. Thanks so much!" with ❤️
- jtraglia
Reacted to "Want to thank thes..." with ❤️
- Potuz
Holy! I just now realized this call is much larger than ACD(E/C)... testing team is taking over coordination!
- Rahul (EF)
Antworten an "Want to thank thes..." +1
- neville (Dedaub)
Reacted to "Want to thank these researches to explicitly figure out if the flags are real instead of forwarding false positives to these protocols. Thanks so much!" with ❤️
- jtraglia
Reacted to "Holy! I just now r..." with 😂
- Milos | enso.build
Reacted to "Holy! I just now r..." with 😂
- Louis
Reacted to "Holy! I just now realized this call is much larger than ACD(E/C)... testing team is taking over coordination!" with 🥰
- spencer
Reacted to "Holy! I just now r..." with 🙌
- jochem-brouwer
Antwoord verzenden naar "Holy! I just now r..." capture risk by testing :)
- Toni Wahrstaetter
Reacted to "Holy! I just now realized this call is much larger than ACD(E/C)... testing team is taking over coordination!" with 🥰
- spencer
Replying to "Holy! I just now r..." wen can we sfi?
- jochem-brouwer
Heeft gereageerd op "Holy! I just now r..." met 😂
- Ben Adams
Replying to "Holy! I just now realized this call is much larger than ACD(E/C)... testing team is taking over coordination!" 🌏👨🚀🔫👨
- spencer
Reacted to "🌏👨🚀🔫..." with 😂
- Dima Gusakov | Lido
Reacted to "Holy! I just now realized this call is much larger than ACD(E/C)... testing team is taking over coordination!" with 😂
- Dima Gusakov | Lido
Reacted to " 🌏👨🚀🔫👨" with 😂
- Mega | ethlambda
Reacted to " 🌏👨🚀🔫👨" with 😂
- Potuz
Reacted to " 🌏👨🚀🔫👨" with 😂
- Radek | solidity
Replying to "Holy! I just now rea..." ACDT call when EOF was rejected was similar. ;D
- cayman
Reacted to " 🌏👨🚀🔫👨" with 😂
- Łukasz Rozmej
Reacted to "Want to thank these researches to explicitly figure out if the flags are real instead of forwarding false positives to these protocols. Thanks so much!" with ❤️
- Parithosh Jayanthi
Reacted to " 🌏👨🚀🔫👨" with 😂
- danceratopz
Reacted to "Holy! I just now rea..." with 🧪
- danceratopz
Reacted to "Holy! I just now rea..." with 🙌
- Luis Pinto
I feel we are missing what parts of 8037/8038 spec actually break contracts, in what way and how many related breakages are exactly. It could multiple things: non monotonic decreasing gas left, increased state gas, gas allocation, refund mechanics changing. How do we correlate that with these analysis? Maybe I missed something
- jochem-brouwer
Heeft gereageerd op "🌏👨🚀🔫..." met 😂
- Barnabas
Reacted to "Holy! I just now realized this call is much larger than ACD(E/C)... testing team is taking over coordination!" with 😂
- Jango
Reacted to "I feel we are missin..." with 👍
- atarpara
Reacted to "I feel we are miss..." with 👍
- Tanguy (ChainSecurity)
Replying to "I feel we are miss..." This might give more context: https://www.chainsecurity.com/blog/glamsterdam-gas/ https://github.com/ethereum/glamsterdam-gas-invariants/tree/master/invariants
- neville (Dedaub)
You'd need to look at the issues one by one. A lot of the issues are due to gas increasing and breaking constant gas calls, but these are lower severity.
- Barnabas
are there any confirmed “won’t/can’t fix” contracts?
- Marius Van Der Wijden (M)
Replying to "I feel we are missin..." Most breakages afaik are from hardcded gas values and increased costs
- Luis Pinto
Replying to "You'd need to look a..." Yeah I’m just asking this because if breakages are due to increasing state gas costs we will no longer be able to scale if core devs don’t change that
- Rahul (EF)
Some won't fix but because false positives / not as big a deal as we thought
- spencer
Reacted to "Most breakages afa..." with ➕
- neville (Dedaub)
I suspect many of them will be "won't fix"
- danceratopz
Jango shared his findings here: https://github.com/mejango/glamsterdam-pocs
- Tanguy (ChainSecurity)
Reacted to "Most breakages afa..." with 👍
- Milos | enso.build
Is this similar to LzCompose from layerZero
- Milos | enso.build
(which should be affected as well)
- Francesco D'Amato
Reacted to "Most breakages afaik are from hardcded gas values and increased costs" with 👍
- Luis Pinto
Reacted to "This might give more context: https://www.chainsecurity.com/blog/glamsterdam-gas/ https://github.com/ethereum/glamsterdam-gas-invariants/tree/master/invariants" with 🙏
- Luis Pinto
Reacted to "Most breakages afaik are from hardcded gas values and increased costs" with 👍
- Potuz
I've seen comments, mostly on twitter which is already biased towards rage baiting, about how the methodology itself to find affected contracts may be wrong. Is there any credit to those? It would *not* surprise me if it turns out that we fork and we see a bunch of projects affected simply because they do not follow the upgrade process at all and we didn't find them correctly. Example of these random comments is this one https://x.com/banteg/status/2119117966134452341?a=l7G82B2YfnU3ywpjmec&s=20
- Justin Florentine
Replying to "I've seen comments, ..." i find it a valid question considering different methods have different results - how many other methods are out there taht find more?
- spencer
Replying to "I feel we are miss..." its around ~89% of findings from hardcoded gas costs, from what i can tell now, the rest involve 8037
- Vitalik
Last part of my post at https://docs.fileverse.io/u/0/w/personal/document/dUHG8ythBk3jJuQEjcE7m6#k=tVLgvLrIqyEsCh7qUmACb9ggVmybyS_aIZZsAMvHDAk (ctrl+f for "What about the future?") touches on that question very directly
- danceratopz
Reacted to "Last part of my post..." with 👍
- Milos | enso.build
you should compare with active contracts tho
- Ansgar Dietrichs
I think this Juicebox comment is reasonable, especially the question posed at the end! I think some of the upcoming Ethereum changes (foremost pq) will make continued changes inevitable, but I think the answer can’t be undermining trust in immutability. We need to step up the work around establishing a clear shared understanding of which parts of Ethereum are guaranteed & won’t change, and which are closer to “implementation specifics” that may or may not change in the future, and reliance on them is at ones own risk. Today, this distinction is very fuzzy.
- derl
yes taking value held by the contracts into account, pure numbers is misleading stats
- Satyajit
Reacted to "Want to thank thes..." with ❤️
- spencer
Reacted to "I think this Juice..." with ➕
- danceratopz
Reacted to "I think there's an a..." with ❤️
- Guru
Reacted to "I think this Juicebo..." with ➕
- Francesco D'Amato
Replying to "I've seen comments, ..." 3 different teams worked on this fwiw
- raxhvl
Reacted to "I think this Juice..." with ➕
- Potuz
Reacted to "Last part of my post..." with 👍
- Chris Berry
Reacted to "I think this Juicebox comment is reasonable, especially the question posed at the end! I think some of the upcoming Ethereum changes (foremost pq) will make continued changes inevitable, but I think the answer can’t be undermining trust in immutability. We need to step up the work around establishing a clear shared understanding of which parts of Ethereum are guaranteed & won’t change, and which are closer to “implementation specifics” that may or may not change in the future, and reliance on them is at ones own risk. Today, this distinction is very fuzzy." with ➕
- Milos | enso.build
Reacted to "yes taking value h..." with ➕
- Jango
on fringe contracts: https://x.com/me_jango/status/2119131865357684919?a=bJhQDaKNDp-omQmDhmw&s=20
- danceratopz
Reacted to "I think this Juicebo..." with ➕
- Toni Wahrstaetter
Replying to "I've seen comments, mostly on twitter which is already biased towards rage baiting, about how the methodology itself to find affected contracts may be wrong. Is there any credit to those? It would *not* surprise me if it turns out that we fork and we see a bunch of projects affected simply because they do not follow the upgrade process at all and we didn't find them correctly. Example of these random comments is this one https://x.com/banteg/status/2119117966134452341?a=l7G82B2YfnU3ywpjmec&s=20" even more, there were 2-3 research teams + 3 auditers looking into this, more or less focused
- Rahul (EF)
Important to clarify - only 0.1% of transactions are affected
- Ansgar Dietrichs
Replying to "I think this Juicebo..." e.g. the specific behavior of an opcode is something one should be able to rely on even on a 10 year horizon. SELFDESTRUCT is an important couterexample there - will there ever be another one? but say the block header structure will likely change. things like gas prices & gas logic more broadly have been more in the gray area in between. we should strive towards making that all much more explicit.
- spencer
i think the core devs strongly feel all contracts are treated as equal on ethereum irrespective of value, it hurts us to any break contracts
- Luis Pinto
Reacted to "i think the core devs strongly feel all contracts are treated as equal on ethereum irrespective of value, it hurts us to any break contracts" with ➕
- Ansgar Dietrichs
so like Toni is saying, we should treat the current situation as a dual challenge: handle the concrete Glamsterdam situation (maybe more importantly!) learn the right lessons and figure out how to improve the situation regarding future forks
- Milos | enso.build
Reacted to "so like Toni is sa..." with 💯
- danceratopz
Reacted to "so like Toni is sayi..." with 👍
- spencer
Reacted to "so like Toni is sa..." with 👍
- Mislav | Ethlabs
Reacted to "so like Toni is sa..." with 👍
- CPerezz
Reacted to "i think the core dev..." with 👍
- Ansgar Dietrichs
Replying to "so like Toni is sayi..." fwiw regarding Glamsterdam I personally tend to believe best course is still to go ahead with the fork as-is, but with some more time before rollout for mitigations. but ofc that needs to be a careful decision.
- Parithosh Jayanthi
Reacted to "so like Toni is saying, we should treat the current situation as a dual challenge: handle the concrete Glamsterdam situation (maybe more importantly!) learn the right lessons and figure out how to improve the situation regarding future forks" with 👍
- Łukasz Rozmej
Reacted to "so like Toni is saying, we should treat the current situation as a dual challenge: handle the concrete Glamsterdam situation (maybe more importantly!) learn the right lessons and figure out how to improve the situation regarding future forks" with 💯
- Łukasz Rozmej
Reacted to "so like Toni is saying, we should treat the current situation as a dual challenge: handle the concrete Glamsterdam situation (maybe more importantly!) learn the right lessons and figure out how to improve the situation regarding future forks" with 👍
- Potuz
Dumb question: why isn't the gas schedule exposed in the EVM? is this just that generally introspection is not desired? I've found explicit mentions against this on EIP 7069.
- Lumi | Offchain Labs
Reacted to "Dumb question: why isn't the gas schedule exposed in the EVM? is this just that generally introspection is not desired? I've found explicit mentions against this on EIP 7069." with ➕
- Daniel Gretzke
First of all, hardcoding gas oftentimes is a must to protect users from potentially malicious contracts, but also the gas changes causing the contracts to break is not the primary issue imo, it’s the fact that with the proposed implementation the user can choose to use the reservoir or not allowing the user to choose what breaks and what doesn’t alongside a couple other issues that come with it
- jochem-brouwer
Antwoord verzenden naar "Dumb question: why..." If we would not have gas introspection we would not have (most of these) problems. I guess if we would restart the chain we would not have the GAS opcode. This opcode is there since the Chainstart and has bitten us many times before
- CPerezz
Replying to "Dumb question: why i..." It is for execution gas. State gas is divided into reservoir and state gas itself. And has subtleties so it’s a much bigger change to do. Unsure if would be much useful
- Hubert (ChainSecurity)
Reacted to "If we would not ha..." with 👍
- Vitalik
Replying to "Dumb question: why..." One issue is that the main other proposed change going forward is to make gas multidimensional, which breaks the model rather than parameters
- Potuz
Replying to "Dumb question: why i..." Oh that's a great point
- Dragan Rakita
Reacted to "If we would not have gas introspection we would not have (most of these) problems. I guess if we would restart the chain we would not have the GAS opcode. This opcode is there since the Chainstart and has bitten us many times before" with 👍
- Lumi | Offchain Labs
Reacted to "One issue is that the main other proposed change going forward is to make gas multidimensional, which breaks the model rather than parameters" with 👍
- Will Corcoran
assurances
- Dragan Rakita
Replying to "Dumb question: why i..." Maybe it is worth adding mechanisms that would allow us to support pattern where now GAS is needed, and only then we can mark GAS as deprecated (or marked as broken).
- Jango
safe try/catch composability is a massive thing to break
- Milos | enso.build
Reacted to "safe try/catch com..." with 💯
- Milos | enso.build
seems there's a huge disconnect
- eric.zhong
Reacted to "safe try/catch composability is a massive thing to break" with 👍
- logical revert. Happy
https://github.com/ethereum/glamsterdam-gas-invariants/pull/4
- Statuette
There's a lot of accumulated knowledge, and not all of it gets passed down automatically. Imo that's pretty natural. Ethereum has been around for 11 years! It's like the evolution of literature, moving from oral traditions to written records. Formalizing those unwritten rules is already a big step forward. I think there's another step: turning written rules into shared understanding. And that takes ongoing communication, lots of coordination, and trust.
- Łukasz Rozmej
Reacted to "safe try/catch composability is a massive thing to break" with 💯
- Łukasz Rozmej
Reacted to "safe try/catch composability is a massive thing to break" with 👍
- Daniel Gretzke
Reacted to "safe try/catch compo..." with 💯
- Daniel Gretzke
Reacted to "safe try/catch compo..." with 👍
- spencer
Reacted to "There's a lot of a..." with ➕
- Marius Van Der Wijden (M)
Reacted to "There's a lot of accumulated knowledge, and not all of it gets passed down automatically. Imo that's pretty natural. Ethereum has been around for 11 years! It's like the evolution of literature, moving from oral traditions to written records. Formalizing those unwritten rules is already a big step forward. I think there's another step: turning written rules into shared understanding. And that takes ongoing communication, lots of coordination, and trust." with ➕
- Daniel Gretzke
Replying to "First of all, hardco..." I’ll take a look, thanks! That’s also been an issue in the past, I’m more talking about implicit branching introduced in glamsterdam
- danceratopz
Reacted to "There's a lot of acc..." with 💙
- Hubert (ChainSecurity)
Replying to "safe try/catch com..." I think it would be good if people come together to write recommendations for that. Because otherwise there might be many broken fixes for this. Happy to receive feedback: https://github.com/ethereum/glamsterdam-gas-invariants/pull/4
- danceratopz
Reacted to "I think it would be ..." with 💙
- Hubert (ChainSecurity)
Reacted to "I’ll take a look,..." with 👍
- jochem-brouwer
I think we should more actively involve the app layer with the fork process (proposed changes to the app/user layer (EVM)) and to ensure that any concerns are explicitly flagged and possibly addressed much earlier, in order to regain (lost?) trust from the app layer
- Barnabas
I don't think Ethereum can honestly promise true immutability. We already know the protocol will have to change in ways that affect deployed contracts, whether that's state repricing, post-quantum security, or whatever else the future holds.
- Milos | enso.build
Reacted to "I think we should ..." with 💯
- jochem-brouwer
Heeft gereageerd op "I don't think Ethe..." met ➕
- Marius Van Der Wijden (M)
Reacted to "I don't think Ethereum can honestly promise true immutability. We already know the protocol will have to change in ways that affect deployed contracts, whether that's state repricing, post-quantum security, or whatever else the future holds." with ❤️
- Dhruv (conduit.xyz)
Reacted to "I don't think Ethe..." with ➕
- Potuz
Reacted to "There's a lot of acc..." with ➕
- Parithosh Jayanthi
Reacted to "I don't think Ethereum can honestly promise true immutability. We already know the protocol will have to change in ways that affect deployed contracts, whether that's state repricing, post-quantum security, or whatever else the future holds." with ❤️
- Jango
appreciate it
- Parithosh Jayanthi
Reacted to "I don't think Ethereum can honestly promise true immutability. We already know the protocol will have to change in ways that affect deployed contracts, whether that's state repricing, post-quantum security, or whatever else the future holds." with ➕
- jochem-brouwer
Heeft gereageerd op "appreciate it" met ❤️
- spencer
Reacted to "appreciate it" with ❤️
- Mario Vega
Reacted to "appreciate it" with ❤️
- Marius Van Der Wijden (M)
Reacted to "appreciate it" with ❤️
- Łukasz Rozmej
Reacted to "I don't think Ethereum can honestly promise true immutability. We already know the protocol will have to change in ways that affect deployed contracts, whether that's state repricing, post-quantum security, or whatever else the future holds." with ➕
- Łukasz Rozmej
Reacted to "I don't think Ethereum can honestly promise true immutability. We already know the protocol will have to change in ways that affect deployed contracts, whether that's state repricing, post-quantum security, or whatever else the future holds." with ❤️
- jochem-brouwer
Heeft gereageerd op "I don't think Ethe..." met ❤️
- Statuette
Reacted to "I don't think Ethereum can honestly promise true immutability. We already know the protocol will have to change in ways that affect deployed contracts, whether that's state repricing, post-quantum security, or whatever else the future holds." with ➕
- Lumi | Offchain Labs
Reacted to "I don't think Ethereum can honestly promise true immutability. We already know the protocol will have to change in ways that affect deployed contracts, whether that's state repricing, post-quantum security, or whatever else the future holds." with ➕
- Lumi | Offchain Labs
Reacted to "I don't think Ethereum can honestly promise true immutability. We already know the protocol will have to change in ways that affect deployed contracts, whether that's state repricing, post-quantum security, or whatever else the future holds." with ❤️
- felipe
Reacted to "I don't think Ethereum can honestly promise true immutability. We already know the protocol will have to change in ways that affect deployed contracts, whether that's state repricing, post-quantum security, or whatever else the future holds." with ➕
- felipe
Reacted to "I don't think Ethereum can honestly promise true immutability. We already know the protocol will have to change in ways that affect deployed contracts, whether that's state repricing, post-quantum security, or whatever else the future holds." with ❤️
- Hubert (ChainSecurity)
About how to do it going forward, given that we have AI powers now: What prevents the EF from maintaining this contract/application search capabilities. Then for every proposed change, the EF runs this search so CoreDevs can directly take breakage into account.
- marc | wolovim
Replying to "appreciate it" Naive q: are the try/catch // composability patterns impractical/impossible through higher level programming patterns? Would another evm feature/opcode be a better canonical solution to what youre trying to achieve?
- danceratopz
Reacted to "appreciate it" with ❤️
- Hubert (ChainSecurity)
I don't think it is realistic to ask AppDev Teams to keep up with EIPs and their state.
- Ryan McPeck
Replying to "Are there action ite..." I think coming back to this question, should be a retro: How can we communicate better/faster so we can get to this point sooner and we don’t have to delay the next fork
- Marius Van Der Wijden (M)
Reacted to "I think coming back to this question, should be a retro: How can we communicate better/faster so we can get to this point sooner and we don’t have to delay the next fork" with 👍
- danceratopz
Reacted to "I think coming back ..." with 👍
- Toni Wahrstaetter
Replying to "Are there action items on how we are going to communicate better / faster next time?" yeah agree. Such delays have a huge cost too (not just the lost confidence by the ecosystem) but esp the users we're letting down. We should avoid the same happenining at the future at all cost.
- danceratopz
Reacted to "yeah agree. Such del..." with 👍
- Jango
Reacted to "I think it would be ..." with 💙
- Rahul (EF)
I quite like Forkcast as a way to disseminate what invariants may break fwiw
- Mario Vega
Reacted to "I quite like Forkcast as a way to disseminate what invariants may break fwiw" with 💯
- Statuette
Reacted to "I quite like Forkcast as a way to disseminate what invariants may break fwiw" with 💯
- marc | wolovim
Reacted to "I quite like Forkcast as a way to disseminate what invariants may break fwiw" with 👀
- Marius Van Der Wijden (M)
Reacted to "I quite like Forkcast as a way to disseminate what invariants may break fwiw" with 💯
- jtraglia
Reacted to "I quite like Forkc..." with 💯
- Dragan Rakita
Changes in fork that affect user should be included in marketing campaign from Ethereum accounts to raise the awareness (not just one post). No one reads EIPs or attend ACD calls.
- jochem-brouwer
Antwoord verzenden naar "I don't think it i..." IMO if there are changes on the EVM, the EIPs should actively try to state from the app layer perspective what changes, I feel it is mostly core protocol oriented now
- Pooja Ranjan
An EIP is a public way to document Ethereum proposals, keeping information clear, organized, and open to everyone.
- Dragan Rakita
Reacted to "I don't think it is realistic to ask AppDev Teams to keep up with EIPs and their state." with ➕
- danceratopz
Replying to "Are there action ite..." - @Ryan McPeck the specs team at the EF will try to enforce the EIP process to be more tightly coupled to the app layer - Examples: - ethereum/execution-specs#3948 feat(tests): add a deployed-idiom invariant corpus - https://github.com/ethereum/execution-specs/pull/3948. - [ethereum/execution-specs#3982 feat(test-specs): check chain invariants while filling](https://github.com/ethereum/execution-specs/pull/3982).
- Barnabas
Replying to "Changes in fork that..." bit hard to do without a “marketing” department 😄
- Marius Van Der Wijden (M)
Replying to "An EIP is a public w..." The problem is that this document will always be a subjectiv view and doesn't fit the objective nature of an eip
- Ryan McPeck
Reacted to " - @Ryan McPeck the specs team at the EF will try to enforce the EIP process to be more tightly coupled to the app layer - Examples: - ethereum/execution-specs#3948 feat(tests): add a deployed-idiom invariant corpus - https://github.com/ethereum/execution-specs/pull/3948. - [ethereum/execution-specs#3982 feat(test-specs): check chain invariants while filling](https://github.com/ethereum/execution-specs/pull/3982)." with 👍
- Statuette
Reacted to "bit hard to do without a “marketing” department 😄" with 💛
- Potuz
Reacted to "I don't think it is ..." with 👍
- Dragan Rakita
Replying to "Changes in fork that..." First step to this: hire people or person
- Marius Van Der Wijden (M)
Replying to "Changes in fork that..." We have the access layer now
- Hubert (ChainSecurity)
Replying to "appreciate it"
- Devcon 5 in Osaka
The CALL opcode returns 0 on failure and 1 on success. It would be great if that would be different. If instead we had 0 on success and many different error codes (like bash). But I don't see any real way how to do that.
- Potuz
Reacted to "IMO if there are cha..." with 👍
- Toni Wahrstaetter
Reacted to "appreciate it" with ❤️
- Pooja Ranjan
Reacted to "We have the access layer now" with 👍
- Dragan Rakita
Reacted to "We have the access layer now" with 👍
- Dragan Rakita
Reacted to "bit hard to do without a “marketing” department 😄" with 💛
- danceratopz
Replying to "Are there action ite..."
- Mario
- Anders Elowsson
Specifying invariants likely to break and not to break are also important from a research development perspective. I think it can help us ”look deeper” when designing features, etc.
- Jango
Replying to "appreciate it" not impractical, but lossy. try/catch feels most natural: one bad recipient shouldn't block a batch or lock a withdrawal, so you catch and move on. the higher-level alternatives all exist and all cost something: pull payments / retry queues (a new slot per failure, 110k under glam, plus a second tx someone has to know to send), check-before-call (only works if you control the callee, which in an open system we usually dont), or a gas pre-check before the call (needs a gasReq constant, which I agree hard coding gas stuff is not beautiful code).
- Trent Van Epps
The EF could start by making much better use of the 4M follower twitter account it runs
- spencer
Reacted to "Specifying invaria..." with 👍
- Pooja Ranjan
Replying to "An EIP is a public w..." It will provide a direct reference that can be used in any community communication, whether a blog post or a future network upgrade Meta EIP.
- Guru
Reacted to "Specifying invariant..." with ➕
- Potuz
Can we code plugins for compilers like Solidity or Viper to flag some common behaviors that should be avoided like hardcoding gas limits, etc?
- danceratopz
- @Ryan McPeck the specs team at the EF will try to enforce the EIP process to be more tightly coupled to the app layer - Examples: - ethereum/execution-specs#3948 feat(tests): add a deployed-idiom invariant corpus - https://github.com/ethereum/execution-specs/pull/3948. - [ethereum/execution-specs#3982 feat(test-specs): check chain invariants while filling](https://github.com/ethereum/execution-specs/pull/3982).
- Kevaundray Wedderburn
Reacted to "Can we code plugins for compilers like Solidity or Viper to flag some common behaviors that should be avoided like hardcoding gas limits, etc?" with 👍
- Ansgar Dietrichs
Reacted to "There's a lot of accumulated knowledge, and not all of it gets passed down automatically. Imo that's pretty natural. Ethereum has been around for 11 years! It's like the evolution of literature, moving from oral traditions to written records. Formalizing those unwritten rules is already a big step forward. I think there's another step: turning written rules into shared understanding. And that takes ongoing communication, lots of coordination, and trust." with 💙
- neville (Dedaub)
Replying to "appreciate it" This is a larger UX issue not specifically an amsterdam defect.
- jochem-brouwer
Antwoord verzenden naar "Can we code plugin..."
- sol docs
- Louis
Reacted to "I don't think Ethereum can honestly promise true immutability. We already know the protocol will have to change in ways that affect deployed contracts, whether that's state repricing, post-quantum security, or whatever else the future holds." with ❤️
- Milos | enso.build
Replying to "Can we code plugin..." an up to date skill too, i found abou the changes in aug via AI "will glamsterdam break my contracts? "
- Mario Vega
- spencer
Reacted to "Can we code plugin..." with 👍
- spencer
Reacted to "an up to date skil..." with ❤️
- Dragan Rakita
There is big gap on public sharing, where major things are shared from various account on twitter. Why would user need to know if this person is legit Eth dev or not 😄
- marc | wolovim
Reacted to "not impractical, but lossy. try/catch feels most natural: one bad recipient shouldn't block a batch or lock a withdrawal, so you catch and move on. the higher-level alternatives all exist and all cost something: pull payments / retry queues (a new slot per failure, 110k under glam, plus a second tx someone has to know to send), check-before-call (only works if you control the callee, which in an open system we usually dont), or a gas pre-check before the call (needs a gasReq constant, which I agree hard coding gas stuff is not beautiful code)." with 👀
- danceratopz
Relevant to this discssion: pcaversaccio's take (EF board member), "ethereum's social contract: immutability requires predictable execution" - https://github.com/ethereum/pm/issues/2299#issuecomment-6913931359 - https://x.com/pcaversaccio/status/2122277228855836959
- Prayag Agrawal
With continuous patches and fixes related to glamsterdam bugs, any suggestions on how to stay updated with the fixes and the submitted bug findings they address ?
- Mario Vega
Replying to "https://github.com/e..." This is a new tool that can be used to run contracts against the latest specs, it’s work in progress but we can use this in theory to check how contracts behave against the newest specs
- Potuz
Reacted to "an up to date skill ..." with ❤️
- Ansgar Dietrichs
I would say we should not rush this invariant work now though. worst outcome would be to prematurely commit to something and then realize 3 months later that it wasn’t quite right
- Ryan McPeck
Thanks team!
- Mario Vega
Reacted to "I would say we should not rush this invariant work now though. worst outcome would be to prematurely commit to something and then realize 3 months later that it wasn’t quite right" with 👍
- jochem-brouwer
Heeft gereageerd op "so like Toni is sa..." met 👍
- Ansgar Dietrichs
Replying to "I would say we shoul..." more important to get it right, properly involve app devs, etc
- Justin Florentine
so we are still shipping 8037, yes?
- Daniel
Reacted to "I quite like Forkcast as a way to disseminate what invariants may break fwiw" with 💯
- Marius Van Der Wijden (M)
Replying to "With continuous patc..." Do you mean protocol fixes? Or contract fixes? If there are any protocol fixes, which seem unlikely at this moment, they will be announced on the ef blog
- Mario Vega
Reacted to "more important to get it right, properly involve app devs, etc" with 👍
- spencer
Replying to "so we are still sh..." its a must, so yes from me
- jochem-brouwer
Heeft gereageerd op "I would say we sho..." met 👍
- Toni Wahrstaetter
Replying to "so we are still shipping 8037, yes?" yes, imo. Vitalik put the trade-off very well in his blog post
- Marius Van Der Wijden (M)
We should decide on a path forward on ACDE
- jochem-brouwer
Heeft gereageerd op "We should decide o..." met 👍
- jtraglia
Reacted to "We should decide o..." with 👍
- Pooja Ranjan
Reacted to "We should decide on a path forward on ACDE" with 👍
- Dragan Rakita
Reacted to "yes, imo. Vitalik put the trade-off very well in his blog post" with 👍
- Iván | ethrex
Reacted to "We should decide on a path forward on ACDE" with 👍
- Dragan Rakita
Reacted to "so we are still shipping 8037, yes?" with 👍
- Potuz
Even if we go with 8037, we would still need to evaluate how much of a delay we need, we do not want to be captured also by non-upgrading apps, but also we do not want to rush a fork and force them to upgrade on a tight schedule
- Mario Vega
Reacted to "We should decide on a path forward on ACDE" with 👍
- spencer
Replying to "We should decide o..." can we consider agree-ing on a date for hoodi next ACDE? gives us a couple days to absorb everything
- Prayag Agrawal
Replying to "With continuous patc..." Protocol fixes since the commits keep on updating in the develop branch
- Marius Van Der Wijden (M)
Reacted to "can we consider agree-ing on a date for hoodi next ACDE? gives us a couple days to absorb everything" with 👍
- Ben Adams
Reacted to "We should decide on a path forward on ACDE" with 👍
- Potuz
Reacted to "We should decide on ..." with 👍
- Marius Van Der Wijden (M)
Replying to "Even if we go with 8..." Given the discussions with protocols that have signalled that their updating, they seem to be okay with a January Glam date
- Barnabas
Replying to "With continuous patc..." We have deployed the fork on sepolia already?
- Barnabas
Replying to "With continuous patc..." so which protocol changes do you mean?
- Hubert (ChainSecurity)
@danceratopz I would like to ask something, afterwards, but cannot raise my hand.
- Potuz
Reacted to "Given the discussion..." with 👍
- Milos | enso.build
so for apps affected by 8037 is better to speak to Rahul this week?
- Marius Van Der Wijden (M)
Barely audible
- Will Corcoran
What capability?
- Statuette
Reacted to "What capability?" with 👆
- Daniel
Reacted to "What capability?" with 👆
- Potuz
Replying to "What capability?" Running these profiling and auditing of contracts as it was done in this fork
- Hubert
What prevents the EF from maintaining this contract/application search capabilities. Then for every proposed change, the EF runs this search so CoreDevs can directly take breakage into account.
- danceratopz
Reacted to "Question from Hubert..." with 👍
- Statuette
Reacted to "Running these profiling and auditing of contracts as it was done in this fork" with 🙏
- Hubert (ChainSecurity)
Replying to "What capability?" For every EIP: You directly find the affected protocols. No guarantee for completeness. But gives coredevs a direct feedback.
- Hubert
What prevents the EF from maintaining this contract/application search capabilities. Then for every proposed change, the EF runs this search so CoreDevs can directly take breakage into account." with 👍
- Milos | enso.build
Reacted to "Question from Hube..." with 👍
- spencer
from my side, on EL specs, we will prioritize this in the process from the get go now, for every proposed EIP we will check "what contracts could this break"
- Jango
are we confident we’re not falling into sunk cost fallacy if we’re saying we would make different decisions wrt glam if we were to come to this earlier in the process?
- Mario Vega
Reacted to "from my side, on EL specs, we will prioritize this in the process from the get go now, for every proposed EIP we will check "what contracts could this break"" with 👍
- Francesco D'Amato
Replying to "What capability?" Fwiw we did/were doing this with quick slots too, looking at Hegota
- Francesco D'Amato
Replying to "What capability?" Could definitely be done more systematically
- Marius Van Der Wijden (M)
Replying to "are we confident we’..." I don't think I would've decided differently if we made these discoveries earlier
- Francesco D'Amato
Overall the work we (ethlabs + some EF people) did was fairly easy and could mostly be automated, I would expect this to be run for every EIP in the not too distant future
- Statuette
Reacted to "Overall the work we (ethlabs + some EF people) did was fairly easy and could mostly be automated, I would expect this to be run for every EIP in the not too distant future" with 💛
- Parithosh Jayanthi
Reacted to "Overall the work we (ethlabs + some EF people) did was fairly easy and could mostly be automated, I would expect this to be run for every EIP in the not too distant future" with 🫡
- Potuz
Replying to "are we confident we’..." is it clear that we would make a different decision? there seems to be a disconnect between this and Vitalik's point that something like 8037 is unavoidable if we want to keep state growth under control. This last comment resonates a lot with most of the community. From outside (as I am completely outside of this one) it looks to me as though we would not have decided differently, but rather have communicated these changes earlier and better
- danceratopz
Reacted to "Overall the work we ..." with 💙
- Marius Van Der Wijden (M)
Reacted to "is it clear that we would make a different decision? there seems to be a disconnect between this and Vitalik's point that something like 8037 is unavoidable if we want to keep state growth under control. This last comment resonates a lot with most of the community. From outside (as I am completely outside of this one) it looks to me as though we would not have decided differently, but rather have communicated these changes earlier and better" with 👍
- Will Corcoran
Reacted to "is it clear that we would make a different decision? there seems to be a disconnect between this and Vitalik's point that something like 8037 is unavoidable if we want to keep state growth under control. This last comment resonates a lot with most of the community. From outside (as I am completely outside of this one) it looks to me as though we would not have decided differently, but rather have communicated these changes earlier and better" with 👍
- Hubert (ChainSecurity)
For Daniel's comment: What are the downsides of having a new transaction type that allows specifying the reservoir explicitly?
- marc | wolovim
Reacted to "Overall the work we (ethlabs + some EF people) did was fairly easy and could mostly be automated, I would expect this to be run for every EIP in the not too distant future" with 💙
- Dragan Rakita
Replying to "are we confident we’..." There is no silver bullet here, any kind of 2d gas version would break something.
- spencer
Reacted to "There is no silver..." with 💯
- Marius Van Der Wijden (M)
Replying to "For Daniel's comment..." Nothing, a new tx type is already planned once we go to full multidimensional gas
- Mislav | Ethlabs
Replying to "What capability?" e.g. Frames break contracts holding ~$2.8B in deposits. Got detected by agent swarms I ran after a tipoff from Morpho devs.
- Barnabas
we did increase the max contract size limit in glamsterdam
- Mislav | Ethlabs
Replying to "What capability?" But its good we know it now so far ahead of H
- Marius Van Der Wijden (M)
Replying to "For Daniel's comment..." Its just not something we can ship in a reasonable timeframe
- Ben Adams
Moving refund to end of tx breaks 4337 bundling; so causes issue for current AA
- Hubert (ChainSecurity)
Reacted to "Its just not somet..." with 👍
- Tanguy (ChainSecurity)
Reacted to "Overall the work w..." with 💙
- spencer
Reacted to "Moving refund to e..." with 👍
- Mario Vega
Reacted to "Moving refund to end of tx breaks 4337 bundling; so causes issue for current AA" with 👍
- spencer
Reacted to "Its just not somet..." with 👍
- Kevaundray Wedderburn
Reacted to "Moving refund to end of tx breaks 4337 bundling; so causes issue for current AA" with 👍
- Marius Van Der Wijden (M)
Reacted to "There is no silver bullet here, any kind of 2d gas version would break something." with 💯
- Potuz
Reacted to "There is no silver b..." with 💯
- Marius Van Der Wijden (M)
Reacted to "Moving refund to end of tx breaks 4337 bundling; so causes issue for current AA" with 👍
- Guru
Reacted to "Moving refund to end..." with 👍
- Jango
Replying to "are we confident we’..." what’s the assumption of what would happen if we did nothing?
- Hubert (ChainSecurity)
Replying to "Moving refund to e..." But isn't 4337 anyway broken with Glamsterdam? Example: you cannot measure gasleft() any more.
- Justin Florentine
introspection got us in this mess, hard disagree
- Justin Florentine
we need to better costrain the state function, introspection works against that
- jochem-brouwer
REVERT puts 0 on stack and puts "reason" (bytes) on the returndata buffer?
- jochem-brouwer
Antwoord verzenden naar "REVERT puts 0 on s..." ah ok its about subcalls
- Ben Adams
EOF did give a revert reason 🥲
- Ansgar Dietrichs
+1 to the comment on better mechanisms, a lot of the “hacky” introspection happening rn that leads to reliance on system behavior that are not meant to be invariants could be made unnecessary by explicit native functionality
- Milos | enso.build
is that bad to ship it 3/4 months later at this point?
- Jango
Reacted to "is that bad to ship ..." with 💯
- Jango
why the rush?
- jochem-brouwer
Antwoord verzenden naar "REVERT puts 0 on s..." REVERT EIP-140: https://eips.ethereum.org/EIPS/eip-140
- Barnabas
Replying to "EOF did give a rever..." EOF gang please rise up
- Tanguy (ChainSecurity)
Replying to "+1 to the comment ..." +1 as well, as of today, or after glamsterdam, there is no way of making a try catch detect OOG without any knowledge on the callee or hardcoding gas values
- Ansgar Dietrichs
Replying to "is that bad to ship ..." it’s not about the Glamsterdam date, it’s about the hard fork after. It would mean 4 months later for qost quantum accounts with Hegota
- marc | wolovim
For those interested, the next Frame tx breakout call is tmrw. Add to the agenda if appropriate: https://github.com/ethereum/pm/issues/2295
- danceratopz
Reacted to "For those interested..." with 💙
- Milos | enso.build
Replying to "is that bad to shi..." gotcha
- Barnabas
We can’t afford to sit on glamsterdam for 4 months when PQ/AI security concerns are present.
- spencer
Replying to "why the rush?" pq is a threat, there is some pressure ship it sooner over later -> we want to make sure ethereum is still around in a few years 😅
- Ansgar Dietrichs
Replying to "is that bad to ship ..." still ofc reasonable to ask and even to argue that that price would be worth paying to make changes
- Jango
Reacted to "pq is a threat, ther..." with 👍
- Jango
Replying to "why the rush?" and glam is a precondition?
- Milos | enso.build
k yeah agree on ai and post quantum, but we can de couple
- Jango
Reacted to "k yeah agree on ai a..." with 👍
- Ansgar Dietrichs
Replying to "is that bad to ship ..." one big problem with that is that it’s not clear that changes could fully fix the problem here
- Dragan Rakita
+1, and no need to deprecate old mechanics but marking them as broken is enough (use with warning)
- Ansgar Dietrichs
Replying to "why the rush?" yes the problem is pq depends on some other parts of Glamsterdam (e.g. ePBS), and splitting out a partial Glamsterdam itself would be very costly
- Dragan Rakita
Reacted to "+1 to the comment on better mechanisms, a lot of the “hacky” introspection happening rn that leads to reliance on system behavior that are not meant to be invariants could be made unnecessary by explicit native functionality" with ➕
- Dragan Rakita
Reacted to "EOF gang please rise up" with 😂
- Ansgar Dietrichs
Replying to "why the rush?" all of these are not insane ideas though, just very costly. always possible to argue it is worth the cost
- Statuette
Unrelated to the technical aspects, I want to emphasize and give a shoutout to EF on something they are doing right. As someone who's been dealing with capture in a smaller ecosystem (even having to organize a peaceful action at Devcon 😅) I love that we actually DO have a space where people can show up, challenge decisions, and make their voices heard without fear of being ousted. This call made me appreciate ethereum's openness even more. Disagreement is essential to progress. We shouldn't take that for granted.
- jochem-brouwer
Heeft gereageerd op "Unrelated to the t..." met ❤️
- Mario Vega
Reacted to "Unrelated to the technical aspects, I want to emphasize and give a shoutout to EF on something they are doing right. As someone who's been dealing with capture in a smaller ecosystem (even having to organize a peaceful action at Devcon 😅) I love that we actually DO have a space where people can show up, challenge decisions, and make their voices heard without fear of being ousted. This call made me appreciate ethereum's openness even more. Disagreement is essential to progress. We shouldn't take that for granted." with ❤️
- Ben Edgington
Reacted to "Unrelated to the technical aspects, I want to emphasize and give a shoutout to EF on something they are doing right. As someone who's been dealing with capture in a smaller ecosystem (even having to organize a peaceful action at Devcon 😅) I love that we actually DO have a space where people can show up, challenge decisions, and make their voices heard without fear of being ousted. This call made me appreciate ethereum's openness even more. Disagreement is essential to progress. We shouldn't take that for granted." with ❤️
- Barnabas
Reacted to "Unrelated to the technical aspects, I want to emphasize and give a shoutout to EF on something they are doing right. As someone who's been dealing with capture in a smaller ecosystem (even having to organize a peaceful action at Devcon 😅) I love that we actually DO have a space where people can show up, challenge decisions, and make their voices heard without fear of being ousted. This call made me appreciate ethereum's openness even more. Disagreement is essential to progress. We shouldn't take that for granted." with ❤️
- Marius Van Der Wijden (M)
Reacted to "Unrelated to the technical aspects, I want to emphasize and give a shoutout to EF on something they are doing right. As someone who's been dealing with capture in a smaller ecosystem (even having to organize a peaceful action at Devcon 😅) I love that we actually DO have a space where people can show up, challenge decisions, and make their voices heard without fear of being ousted. This call made me appreciate ethereum's openness even more. Disagreement is essential to progress. We shouldn't take that for granted." with ❤️
- spencer
is there a place in the eth r&d discord for others to leave comments after the call?
- gary
已使用❤️回应"Unrelated to the t..."
- Parithosh Jayanthi
Reacted to "Unrelated to the technical aspects, I want to emphasize and give a shoutout to EF on something they are doing right. As someone who's been dealing with capture in a smaller ecosystem (even having to organize a peaceful action at Devcon 😅) I love that we actually DO have a space where people can show up, challenge decisions, and make their voices heard without fear of being ousted. This call made me appreciate ethereum's openness even more. Disagreement is essential to progress. We shouldn't take that for granted." with ❤️
- spencer
Reacted to "Unrelated to the t..." with ❤️
- Dragan Rakita
Replying to "are we confident we’..." You couldn’t increase block gas, and just repricing state opcodes would not allow you to deploy big contracts.
- Jango
yes, just dont want us to fall into sunk cost fallacy. the bits of glam that are pivotal mechanically for quantum stuff, ok. but the bits that may not be and have big app dev confidence and immutability tradeoffs, might be worth it. id argue def worth it, though i still may be missing info
- danceratopz
Reacted to "Unrelated to the tec..." with ❤️
- Statuette
Reacted to "is there a place in the eth r&d discord for others to leave comments after the call?" with 👆
- jochem-brouwer
- Statuette
Reacted to "pq is a threat, there is some pressure ship it sooner over later -> we want to make sure ethereum is still around in a few years 😅" with 💯
- Łukasz Rozmej
Reacted to "EOF did give a revert reason 🥲" with 👍
- spencer
Reacted to "https://github.com..." with ❤️
- Marius Van Der Wijden (M)
Replying to "is there a place in ..." Execution-dev?
- Pooja Ranjan
Reacted to "https://github.com/ethereum/pm/issues/2288" with ❤️
- Dragan Rakita
Reacted to "Unrelated to the technical aspects, I want to emphasize and give a shoutout to EF on something they are doing right. As someone who's been dealing with capture in a smaller ecosystem (even having to organize a peaceful action at Devcon 😅) I love that we actually DO have a space where people can show up, challenge decisions, and make their voices heard without fear of being ousted. This call made me appreciate ethereum's openness even more. Disagreement is essential to progress. We shouldn't take that for granted." with ❤️
- Ansgar Dietrichs
Replying to "why the rush?" yes I think that’s a valid take. it’s a hard call to make
- spencer
Replying to "is there a place i..." sgtm!
- jochem-brouwer
Antwoord verzenden naar "https://github.com..." Wed 18 nov 16 UTC
- spencer
"the only constant in life is change"... the only thing that is immutable is mutability? 🤔
- felipe
Reacted to "is it clear that we would make a different decision? there seems to be a disconnect between this and Vitalik's point that something like 8037 is unavoidable if we want to keep state growth under control. This last comment resonates a lot with most of the community. From outside (as I am completely outside of this one) it looks to me as though we would not have decided differently, but rather have communicated these changes earlier and better" with 👍
- felipe
Reacted to "There is no silver bullet here, any kind of 2d gas version would break something." with 💯
- jochem-brouwer
Heeft gereageerd op ""the only constant..." met 🧐
- Barnabas
hoodi fork wen?
- Jango
Replying to ""the only constant i..." the ocean is nice
- spencer
Reacted to "the ocean is nice" with 😂
- Dragan Rakita
Reacted to "the ocean is nice" with 😂
- Statuette
Reacted to "the ocean is nice" with 😂
Call summary
Targets
- •Glamsterdam mainnet targeting January 2027 (per protocol signals from affected teams) - 01:33:29
- •EIPIP call: Wednesday November 18, 16:00 UTC — AI integration into EIP review process - 01:51:19
Decisions
- •No decision on EIP-8037 today; formal path-forward decision deferred to ACDE on Thursday - 01:30:30
- •Glamsterdam delay was justified to prioritize app-layer mitigation time before mainnet - 01:17:50
Highlights
- Eip Process:
- ·jochem-brouwer: large influx of AI-generated EIPs; EIPIP call Wednesday Nov 18 16:00 UTC to discuss AI integration into review process - 01:50:21
- Audit Findings:
- ·ChainSecurity: 152 findings across mainnet; reservoir/refill mechanism is minority of breakage - 00:28:45
- ·Dedaub: 12 high + 70 medium confirmed findings from ~25K contracts with non-trivial gas patterns - 00:37:30
- ·EF internal: 3 highs (1 fixed, 2 fixing within a month); at least 25 confirmed mediums; 16 actively triaging - 00:46:52
- ·OP Stack chains independently affected; OP has fixed invariants, live on OP Sepolia - 00:48:25
- Process Improvements:
- ·Toni: core devs must clearly define which invariants are guaranteed vs. likely to change - 00:59:00
- ·Marius: proposal to publish EIP documenting stable vs. expected-to-change protocol invariants - 01:20:21
- ·Specs team (Mario/danceratopz): two open PRs adding deployed-idiom invariant corpus and chain invariant checks - 01:22:57
- ·EF/Ethlabs agent-swarm analysis was largely automatable; will be run systematically for future EIPs - 01:39:25
- ·EF Access Layer (Rahul): builder enablement team surveyed 50+ teams; direct contacts for 3,000+ L1 devs - 01:16:18
- App Developer Concerns:
- ·Jango (Juicebox): EIP-8037 allows users to tune gas and trigger catch blocks, breaking composability - 00:50:07
- ·Daniel (Uniswap): reservoir mechanism introduces implicit user-controlled branching; fundamentally breaks try-catch - 01:01:38
- ·Neville (Dedaub): no reliable way for callers to distinguish OOG revert from logical revert - 00:59:00
- ·Barnabas: Ethereum cannot honestly promise true immutability given known upcoming PQ/state changes - 01:14:14
- Glamsterdam Eip8037 Impact:
- ·EIP-8037 and EIP-8038 break smart contract invariants; central topic of call - 00:17:49
- ·EIPs enable 3.3x gas scaling (60M → 200M gas/block) but break hard-coded gas patterns - 00:21:37
- ·Delays in audits and outreach caused by EF reorgs and budget cuts; Marius accepts responsibility - 00:21:43
- ·Combined search: 200 total findings (152 ChainSecurity + EF internal); coverage not complete - 00:30:31
- ·Majority of findings from Invariants 4 and 7: hard-coded gas budgets insufficient post-Glamsterdam - 00:31:04
- Protocol Rationale And Tradeoffs:
- ·~10–15K contracts affected out of ~90M on Ethereum; immutable broken contracts hold ~$300–500K - 00:54:33
- ·No silver bullet: any 2D gas version breaks something; removing 8037 also blocks large contract deployment - 01:39:54
- ·Affected protocols signaling willingness to fix indicate a January Glamsterdam date is acceptable - 01:33:29
- ·Glamsterdam is a precondition for Hegota (PQ); removing it delays post-quantum accounts by ~4 months - 01:32:09
Action Items
- •EF specs team - Review and merge specs PRs #3948 and #3982 (deployed-idiom invariant corpus and chain invariant checks) - 01:22:57
- •All core devs (ACDE Thursday) - Decide on Glamsterdam path forward (ship 8037 as-is vs. changes) and agree on Hoodi date - 01:30:00
- •Rahul (EF Access Layer) - Coordinate outreach to remaining affected L2/app teams not yet triaging; use direct security contacts - 01:17:50
Key decisions
Glamsterdam delay was justified to prioritize app-layer mitigation time before mainnet
The delay in the Glamsterdam upgrade was deemed necessary to allow application developers sufficient time to address breaking changes caused by new EIPs.Glamsterdam mainnet targeting January 2027
SFIThe upgrade is targeted for January 2027 based on signals from affected protocols indicating readiness to implement fixes.
EIPs discussed
AI Disclaimer: Some content or metadata on EIPsInsight may be AI-inferred or automatically compiled. If you find any discrepancy, please contact us at dev@avarch.org.