eth_simulate Implementers' Meeting #48

eth_multicall and eth_simulate Meetings · 5 May 2025 · 19:14

A meeting of the developers working on eth_multicall and its successor eth_simulate, held on 5 May 2025. Client and tooling teams compare implementations and work on the specification for simulating transactions.

Topics

From eth_multicall and eth_simulate Meetings, hosted by Ethereum Foundation.

Transcript

Seems like the meeting Zoom link changed. We were in the wrong meeting with Michael. Hello. Yeah, I did receive an email on Friday. The meeting was cancelled and then a new one. Read it. I was just reading it now. Well, with the canceled one, there was still the one old meeting URL, but then there was a new mail that has had the new one. I wonder if it is still in the EF

calendar. I don't cheat there. Uh I I actually clicked the link from the calendar. It has update. There you do. We have any updates? I don't have any updates except we could discuss about that if we should merge this meeting with the RPC standards meeting. So um yeah I have a very small

update. Um one of the bugs that Bizu found um in G it I had a fix for it. uh it was sitting for a while and uh it was merged now and it will be released a bit later. So um it would be nice if bas I don't know if there's any here um could check the tests again and see if now we have everything matching. So yeah that's something from my side. I don't have much more unfortunately. Um, no progress on

estimate gas or tracing as I've been kind of taken with other stuff. Do you think you have time to look at those later or are you completely full with work? Well, possibly after Pectra. Yep. Which is in two days. Unless there's some bugs and then it's, you know, more work and more more stressful work as well. Yeah.

What do you feel about merging this with the other meeting since we generally don't have that many updates anymore? I think if we adopt this time, it's okay for me. Yeah, I'm not sure what was the time for that other meeting. It was later in the day. I think 5:00 p.m. Central European time or 400 p.m. Not sure. Maybe somebody else can comment.

The port is is at 3:00 in UTC. So that's like 3 hours later. They don't really know who are there. if you can get them to move to this time or not. But yeah, it's going to be more difficult the more people there to get get some time that everyone agrees on us. So something that I

um is sort of important for me personally is that we um get those tests passing all of them so that we can get the spec merged and the test merged. uh we've been kind of blocked on those two um by having at least two clients that agree fully and it feels like it's very close but I haven't heard any updates from mind for a while so I don't know where they are going with that but maybe G and BU are the closest to each other yeah possibly

But is is bas using mostly get implementation or are they separate implementations? No, they have their own implementation in Java, I guess. But maybe inspired by a guest. But if that's the guess, then we could maybe test that how close to get and be are from each other and if they matching it, then we can merge with those. But I guess we don't have anyone from BESU here at the moment. Does our test should have the BS support? I think we had it at some point, but then they were really far away and I haven't looked afterwards on that.

Now there we want probably want to merge the specs and then we want also merge the tests. Yeah, exactly. You know be happy still we are still happy making it's the classic case that 80% of the work took 20% of the time. Yeah. you wanted to say something. I would say that we have been still making changes to the spec. So that's why we haven't be kind of been done with it.

Yeah. But I guess if you if you can get two lines to agree on on the specs that then we maybe can merge them with that those as far as memory serves me. Most of the changes were uh fixes that we discovered along with the tests like the the mismatches in the tests and there was one thing related to peg try I believe other than that it was relatively stable. Yeah. Well, there was that gas limit kind of how the gas limit is calculated when it's missing and I think that was bit bigger change in a way. Well, in the it's easy to change the spec, but in a

way, but it was a logic difference, not just a kind of misunderstood. Can you see reach out to Basil to ask about that? Uh yeah, I will uh once uh it's released today, I will ping you to have a look. Yeah, I guess we don't have anything else to discuss. I do not.

Okay, see you next week. Bye. Talk to you.