Skip to content

ACP-226: add initial delay excess #4300

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 4 commits into from
Sep 19, 2025
Merged

ACP-226: add initial delay excess #4300

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
d62a912
add acp-226 default delay excess
ceyonur Sep 19, 2025
5f0de47
Update vms/evm/acp226/acp226.go
ceyonur Sep 19, 2025
397204a
rename default to initial
ceyonur Sep 19, 2025
4a08450
rename test
ceyonur Sep 19, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions vms/evm/acp226/acp226.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,6 +21,10 @@ const (
// MaxDelayExcessDiff (Q) is the maximum change in excess per update
MaxDelayExcessDiff = 200

// InitialDelayExcess represents the initial (≈2000ms) delay excess.
// Formula: ConversionRate (2^20) * ln(2000) + 1
InitialDelayExcess = 7_970_124
Copy link
Contributor

@joshua-kim joshua-kim Sep 19, 2025 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

These feel like magic numbers even though we use a godoc to document them... could we define these through calculation in code? i.e

const conversionRate = 1 << 20 // 2^20 fixed-point scale

var InitialDelayExcess = int(float64(conversionRate)*math.Log(2000)) + 1

Copy link
Contributor

@StephenButtolph StephenButtolph Sep 19, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

These values are consensus critical, so I don't think it is reasonable to perform floating point operations to generate them. (The actual number is more important than the mechanism used to derive them)

Copy link
Contributor Author

@ceyonur ceyonur Sep 19, 2025 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

(same reason as what Stephen posted above while I writing this)
I think that would actually end up what we try to avoid with binary search. I think math.Log precision can differ from machine to machine and end-up having different initial values.

Copy link
Contributor

@michaelkaplan13 michaelkaplan13 Sep 19, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

If we need to make this configurable by minimum delay setting in the future, an approach @StephenButtolph recommended to me for ACP-224 is to use some rational approximation for ln(x), but that would basically be a magic number itself here anyways.


maxDelayExcess = 46_516_320 // ConversionRate * ln(MaxUint64 / MinDelayMilliseconds) + 1
)

Expand Down
4 changes: 2 additions & 2 deletions vms/evm/acp226/acp226_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -49,8 +49,8 @@ var (
delay: 1000,
},
{
name: "2000ms_delay",
excess: 7_970_124, // ConversionRate (2^20) * ln(2000) + 1
name: "2000ms_delay_initial",
excess: InitialDelayExcess, // ConversionRate (2^20) * ln(2000) + 1
delay: 2000,
},
{
Expand Down