15 posts tagged with "plutus-core"

High level summary​

The Plutus team has recently made a number of significant improvements to Plinth, including:

• Compiler improvements:
  • Added a type checker plugin that preserves source locations, resulting in significantly clearer error messages (#7640).
  • Added a second type checker plugin to detect unsupported Haskell features, further improving error reporting (#7659).
  • Added a driver plugin that automatically sets the required compiler flags and enables the Strict extension (#7687).
• Language improvements:
  • AsData now generates a destructor function for the data type (#7664). For matching on sum types, the destructor function is more efficient than the pattern synonyms.

The Plinth user guide will be updated soon. In the meantime, please refer to the linked PR descriptions for further details.

Additionally, a new UPLC optimization has been implemented: applications with three or more arguments are now transformed into case-constr form.

Low level summary​

Key Pull Requests Merged​

• Improve compile errors, and add Plinth.Plugin that preserves more source locations
• Plinth: improve error reporting for unsupported Haskell features
• Generate a destructor function from AsData
• Drop redundant unsafeCaseList calls produced by AsData
• Automatically set required flags and Strict extension in the plugin

High level summary​

The Plutus team has finished implementing and costing the Value primitives, as well as completing all other development required for the upcoming intra-era hard fork.

We've recently published a blog post recapping the recent UPLC 2025 event: https://www.iog.io/news/uplc-2025-conference-recap

In collaboration with Koz Ross from MLabs, we've implemented optimized byteStringToInteger and integerToByteString primitives. Both primitives will now take linear time instead of quadratic time. These improvements leverage the ghc-bignum library, which was made possible by dropping support for GHC 8.

In parallel, we continue to make progress across several other areas, including the initial work on Plutus V4, ongoing discussions around formally verifying the UPLC inliner in collaboration with Professor Philip Wadler, further optimization of UPLC runtime performance, and continued improvements to the usability of Plinth.

Low level summary​

Key Pull Requests Merged​

• Plinth user guide: more details on versions
• Add valueData and unValueData conformance tests
• Finalise cost models for valueData and unValueData
• Make Builtins.unsafeUncons use builtin list casing
• Improve UPLC parser error messages
• Value costing: insertCoin, unionValue, scaleValue
• Add casing on constants to the Plinth user guide
• Optimize VConstr bounds check in CEK machine
• Optimize Integer <-> ByteString conversions
• Make Enum PlutusTx.Rational plc compilable
• Fix the spec of CEK rules for constr and case
• Add deriveEnum for Plinth
• perf(builtins): optimize listToArray implementation and costing
• (Optimization) Combine Either Text (HeadSpine ...) into HeadSpine
• Support mkNil for any builtin types

Issues and Pull Requests In Progress​

• Formally verifying the UPLC inliner
• Adding a new Array constructor for UPLC, which can be accessed in Plutus V4
• End-to-end testing for intra-era Hard Fork features

High level summary​

The Plutus team is finalizing work for the upcoming intra-era hard fork.

Four of the Value primitives have now been costed, and the remaining three are currently in progress. After this, the final step before will be performing end-to-end testing for all new features to be introduced in the hard fork.

We are also preparing a blog post on UPLC 2025, which will be published soon.

In parallel, we continue progressing on other tasks, including support case analysis related to Data.

Low level summary​

Key Pull Requests Merged​

• Conformance tests for value built-ins taking into account 128-bit integers

• Fix Value.Quantity shrinker

• Additional conformance tests for insertCoin and valueContains

• Benchmark regular apply vs case apply

• Standardize repetition notation in EBNF grammar of plutus

• Cost models for LookupCoin, ValueContains, ValueData, UnValueData builtins

Issues and Pull Requests In Progress​

• End-to-end testing for dropList
• End-to-end testing for modular exponentiation
• End-to-end testing for multi-scalar multiplication
• End-to-end testing for Array primitives
• End-to-end testing for Value primitives
• End-to-end testing for unification of primitives across Plutus V1, V2 and V3

High level summary​

The Plutus team organized and attended UPLC 2025, an event dedicated to the UPLC language, held in Edinburgh last week. Session recordings will be available soon.

Based on community feedback, we have added a new Value primitive, scaleValue, to the upcoming intra-era hard fork, along with several minor refinements to other Value primitives.

We have also continued improving the performance of the UPLC evaluator, specifically by optimizing the inlining of the geq function.

Finally, we are progressing with costing for all Value primitives.

Low level summary​

Key Pull Requests Merged​

• Add ScaleValue primitive
• Restrict Value quantities to signed 128-bit integer range
• Add an inlinable version of 'geq'
• CIP-0153 | Add range limits to amounts in builtin Value
• Add ECD (Euclidean Common Divisor) synthetic benchmark scenario
• Add support for multiple input arguments in test framework

Issues and Pull Requests In Progress​

• Costing for Value primitives (in plutus-private)

High level summary​

We’ve completed the implementation of CIP-0153, introducing the Value built-in type and its associated functions. Our current focus is on finalizing the costing for these built-ins. Once this is complete, Plutus will be ready for the intra-era hard fork. All other features scheduled for the hard fork are already finished.

We’ve improved the Flat decoding of built-in lists, which has noticeably sped up Plutus script deserialization. In our benchmarks, some scripts' deserialization now runs with double-digit performance gains.

We’ve updated the Plutus Core specification and the Agda formalization to include the multi-scalar multiplication primitives (CIP-0133). Additional property and conformance tests have also been added for these primitives.

The team is also gearing up for UPLC 2025, a UPLC event scheduled to take place in approximately three weeks.

Low level summary​

Key Pull Requests Merged​

• Spec: serialise arrays as lists; miscellaneous updates
• Implement ValueData and UnValueData
• Add BLS12-381 MSM builtins to specification
• Benchmarks for flat list/array decoding
• Keep track of currency/token lengths in builtin Value
• Property tests for BLS12-381 multi-scalar multiplication
• Optimise list decoder
• Add BLS12-381 MSM builtins to metatheory
• Conformance tests for BLS12_381 multiScalarMul operations (2)
• Value decoding/parsing: reject currencies/tokens longer than 32 bytes
• Enforce currency and token lengths in insertCoin and unValueData
• Cover some missing cases in the connformance tests for wrteiBits
• Enable the Agda conformance tests for the array builtins
• valueContains: enforce no negative amounts in either Value
• Enable BLS12-381 MSM builtins at PV11
• Conformance tests for built-in functions on Value

Issues and Pull Requests In Progress​

• Built-in Value: costing (in plutus-private)