Skip to content

Conversation

@dunxen
Copy link
Contributor

@dunxen dunxen commented Nov 28, 2022

Adds two new payment Methods for identifying payments with custom
min_final_cltv_expiry_delta as payments with LDK or user payment
hashes.

The min_final_cltv_expiry_delta value is packed into the metadata
bytes of the payment secret, taking up 12 bits.

Fixes #1850

@dunxen dunxen force-pushed the 2022-11-config-minfinalcltvexpiry branch 3 times, most recently from a297bba to 354fab2 Compare November 29, 2022 20:07
@dunxen dunxen force-pushed the 2022-11-config-minfinalcltvexpiry branch 2 times, most recently from 64b8e36 to 91db879 Compare December 12, 2022 15:39
@dunxen dunxen changed the title DRAFT: Allow configurable min_final_cltv_expiry_delta Allow configurable min_final_cltv_expiry_delta Dec 12, 2022
@dunxen dunxen marked this pull request as ready for review December 12, 2022 15:41
@TheBlueMatt
Copy link
Collaborator

Ugh, needs rebase, sorry about that.

@dunxen
Copy link
Contributor Author

dunxen commented Dec 13, 2022

Ugh, needs rebase, sorry about that.

it's the way of the git lol

@dunxen dunxen force-pushed the 2022-11-config-minfinalcltvexpiry branch 2 times, most recently from 01197fb to 147be43 Compare December 13, 2022 10:19
@dunxen
Copy link
Contributor Author

dunxen commented Dec 13, 2022

I went the "add param onto existing methods" route because I feel it's compatible with all utility methods. Didn't feel right at the time to create more variants of each method which also seems a bit much. I know optional params are usually not ideal. Open to ideas.

@shamardy
Copy link
Contributor

I know optional params are usually not ideal. Open to ideas.

You can have min_final_cltv_expiry as optional in the functions that the user should use only and use unwrap_or(DEFAULT_MIN_FINAL_CLTV_EXPIRY) to provide non optional values to other functions.

In my opinion, having this parameter as optional in lightning-invoice crate functions (create_phantom_invoice, create_phantom_invoice_with_description_hash, create_invoice_from_channelmanager, create_invoice_from_channelmanager_with_description_hash) is enough. Other functions are more advanced, and if the user chooses to use them, he/she should provide a value for min_final_cltv_expiry.

@dunxen
Copy link
Contributor Author

dunxen commented Dec 13, 2022

I know optional params are usually not ideal. Open to ideas.

You can have min_final_cltv_expiry as optional in the functions that the user should use only and use unwrap_or(DEFAULT_MIN_FINAL_CLTV_EXPIRY) to provide non optional values to other functions.

Yeah that too, but wanted to avoid validating the value in many places and just in the lowest level internal functions, so that's why some internal functions just pass through the Option.

Copy link
Collaborator

@TheBlueMatt TheBlueMatt left a comment

Choose a reason for hiding this comment

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

Thanks! This is pretty close I think, just some notes about missing test coverage and docs.

Copy link
Collaborator

@TheBlueMatt TheBlueMatt left a comment

Choose a reason for hiding this comment

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

Oh, two more comments.

@dunxen dunxen force-pushed the 2022-11-config-minfinalcltvexpiry branch from 147be43 to 652c45d Compare December 16, 2022 11:20
@dunxen
Copy link
Contributor Author

dunxen commented Dec 16, 2022

Still working on that test coverage, but rebased an pushed some fixes so long.

@codecov-commenter
Copy link

codecov-commenter commented Dec 16, 2022

Codecov Report

Base: 90.80% // Head: 90.69% // Decreases project coverage by -0.12% ⚠️

Coverage data is based on head (0eb4550) compared to base (e0a0add).
Patch coverage: 95.45% of modified lines in pull request are covered.

❗ Current head 0eb4550 differs from pull request most recent head 8cd50d8. Consider uploading reports for the commit 8cd50d8 to get more accurate results

Additional details and impacted files
@@            Coverage Diff             @@
##             main    #1878      +/-   ##
==========================================
- Coverage   90.80%   90.69%   -0.12%     
==========================================
  Files          98       97       -1     
  Lines       51507    50871     -636     
  Branches    51507    50871     -636     
==========================================
- Hits        46770    46135     -635     
+ Misses       4737     4736       -1     
Impacted Files Coverage Δ
lightning/src/ln/functional_test_utils.rs 91.21% <ø> (ø)
lightning-invoice/src/de.rs 82.38% <83.33%> (ø)
lightning-invoice/src/lib.rs 87.37% <83.33%> (-0.11%) ⬇️
lightning/src/ln/inbound_payment.rs 89.75% <88.88%> (-3.74%) ⬇️
lightning-invoice/src/utils.rs 97.65% <97.29%> (-0.12%) ⬇️
lightning-invoice/src/payment.rs 89.63% <100.00%> (ø)
lightning-invoice/src/ser.rs 92.19% <100.00%> (ø)
lightning/src/ln/channelmanager.rs 87.24% <100.00%> (+0.05%) ⬆️
lightning/src/ln/functional_tests.rs 96.93% <100.00%> (-0.18%) ⬇️
lightning/src/ln/payment_tests.rs 98.73% <100.00%> (ø)
... and 17 more

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

@TheBlueMatt
Copy link
Collaborator

Let me know when you get the test coverage in place and I'll take another look. In the mean time, we should get another reviewer on this.

@dunxen
Copy link
Contributor Author

dunxen commented Dec 18, 2022

Let me know when you get the test coverage in place and I'll take another look.

Should have it up in the morning tomorrow.

@dunxen
Copy link
Contributor Author

dunxen commented Dec 19, 2022

Oops. This needs to be changed and tested, right?:

if (hop_data.outgoing_cltv_value as u64) <= current_height as u64 + HTLC_FAIL_BACK_BUFFER as u64 + 1 {

Also, I'm not sure where we ever enforced the MIN_FINAL_CLTV_EXPIRY(_DELTA) on current main branch... I just see that HTLC_FAIL_BACK_BUFFER above.

@TheBlueMatt
Copy link
Collaborator

Oops. This needs to be changed and tested, right?:

Yes, indeed, we need to compare against the inbound_payment decoded version and reject the payment.

Also, I'm not sure where we ever enforced the MIN_FINAL_CLTV_EXPIRY(_DELTA) on current main branch... I just see that HTLC_FAIL_BACK_BUFFER above.

We dont - we enforce HTLC_FAIL_BACK_BUFFER on purpose - MIN_FINAL_CLTV_EXPIRY_DELTA is defined as HTLC_FAIL_BACK_BUFFER + 3 so we just enforce the former to enforce the latter but with a bit of a buffer in case a block or two were found while the HTLC was being forwarded to us.

@dunxen
Copy link
Contributor Author

dunxen commented Dec 19, 2022

Yes, indeed, we need to compare against the inbound_payment decoded version and reject the payment.

Cool. Happy if I squash and rebase after fixing up to get it to a cleaner state for reviewers?

There's also no rush on this one since I guess 114 is coming in Q1 so I can jump on more review/dual funding while I wait.

@TheBlueMatt
Copy link
Collaborator

Yes, feel free to squash. There's no specific rush, no, but of course always better to have fewer open PRs :)

@TheBlueMatt TheBlueMatt added this to the 0.0.114 milestone Dec 21, 2022
@TheBlueMatt
Copy link
Collaborator

Looks like this needs rebase.

@dunxen
Copy link
Contributor Author

dunxen commented Jan 4, 2023

Back on this now. Rebased locally, but busy debugging a failing test.

@dunxen
Copy link
Contributor Author

dunxen commented Jan 17, 2023

The rebase conflicts were a little intense so had to squash some of the fixups otherwise things would get a little wild here.

@dunxen dunxen force-pushed the 2022-11-config-minfinalcltvexpiry branch 2 times, most recently from d10f48c to 30772df Compare January 17, 2023 13:20
// any payments to succeed. Further, we don't want payments to fail if a block was found while
// a payment was being routed, so we add an extra block to be safe.
pub const MIN_FINAL_CLTV_EXPIRY_DELTA: u32 = HTLC_FAIL_BACK_BUFFER + 3;
pub const MIN_FINAL_CLTV_EXPIRY_DELTA: u16 = HTLC_FAIL_BACK_BUFFER as u16 + 3;
Copy link
Contributor

Choose a reason for hiding this comment

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

Is it too much hassle to make HTLC_FAIL_BACK_BUFFER a u16?

Copy link
Contributor Author

Choose a reason for hiding this comment

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

Was considering it and reckoned it might blow this PR up, but it would be more sanitary. Will wait for some more feedback on it. 👍

@dunxen dunxen force-pushed the 2022-11-config-minfinalcltvexpiry branch from 0eb4550 to f180062 Compare January 20, 2023 06:38
@wpaulino
Copy link
Contributor

Feel free to squash.

@dunxen dunxen force-pushed the 2022-11-config-minfinalcltvexpiry branch from f180062 to c2fe4d0 Compare January 21, 2023 12:45
Copy link
Collaborator

@TheBlueMatt TheBlueMatt left a comment

Choose a reason for hiding this comment

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

This looks great. Some style nits but otherwise happy here.

This matches the spec and helps avoid any confusion around
naming. We're also then consistent with `cltv_expiry` in an HTLC being
the actual block height value for the CLTV and not a delta.
All utility functions for invoice construction will now also accept an
Option<>al `min_final_cltv_expiry_delta` which is useful for things like
swaps etc. The `min_final_cltv_expiry_delta` will default back to
`MIN_FINAL_CLTV_EXPIRY_DELTA` if `None` is provided.
@dunxen dunxen force-pushed the 2022-11-config-minfinalcltvexpiry branch 3 times, most recently from 3a74c8d to 8cd50d8 Compare January 24, 2023 14:24
@TheBlueMatt
Copy link
Collaborator

Feel free to squash

@dunxen dunxen force-pushed the 2022-11-config-minfinalcltvexpiry branch from 8cd50d8 to 6fc348d Compare January 24, 2023 18:53
Adds two new payment `Method`s for identifying payments with custom
`min_final_cltv_expiry_delta` as payments with LDK or user payment
hashes.

The `min_final_cltv_expiry_delta` value is packed into the first 2
bytes of the expiry timestamp in the payment secret metadata.
@dunxen dunxen force-pushed the 2022-11-config-minfinalcltvexpiry branch from 6fc348d to 5b53670 Compare January 24, 2023 19:02
pub(super) fn verify<L: Deref>(payment_hash: PaymentHash, payment_data: &msgs::FinalOnionHopData, highest_seen_timestamp: u64, keys: &ExpandedKey, logger: &L) -> Result<Option<PaymentPreimage>, ()>
pub(super) fn verify<L: Deref>(payment_hash: PaymentHash, payment_data: &msgs::FinalOnionHopData,
highest_seen_timestamp: u64, keys: &ExpandedKey, logger: &L) -> Result<
(Option<PaymentPreimage>, Option<u16>), ()>
Copy link
Contributor

Choose a reason for hiding this comment

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

Nit: this wrapping is not consistent

Copy link
Collaborator

@TheBlueMatt TheBlueMatt left a comment

Choose a reason for hiding this comment

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

Don't bother changing it unless you need to address other feedback.

R::Target: Router,
L::Target: Logger,
{
if min_final_cltv_expiry_delta.is_some() && min_final_cltv_expiry_delta.unwrap().saturating_add(3) < MIN_FINAL_CLTV_EXPIRY_DELTA {
Copy link
Collaborator

Choose a reason for hiding this comment

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

nit: this check is redundant

Copy link
Contributor

Choose a reason for hiding this comment

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

I don't think so, the others are part of a different flow.

@TheBlueMatt TheBlueMatt merged commit 8bb6614 into lightningdevkit:main Jan 24, 2023
@dunxen dunxen deleted the 2022-11-config-minfinalcltvexpiry branch February 2, 2023 07:19
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Allow configurable per-payment min_final_cltv_expiry

6 participants