summaryrefslogtreecommitdiff
path: root/src
Commit message (Collapse)AuthorAge
* WIP revamp accumulators again: explicit monoid typesTom Smeding9 days
| | | | | | | | No more D2 in accumulators! Paving the way for configurable sparsity of products and arrays. The idea is to make separate monoid types for a "product cotangent" and an "array cotangent" that can be lowered to either a sparse monoid or a non-sparse monoid. Downsides of this approach: lots of API duplication.
* simplify: Additional rules inspired by Example.neuralTom Smeding11 days
|
* pretty: Print build dimensionTom Smeding11 days
|
* Don't unSTyTom Smeding11 days
|
* Remove done TODOTom Smeding11 days
|
* Compile: Give IDs to C modules for easier debuggingTom Smeding12 days
|
* Compile: Reset accidentally enabled debug switchTom Smeding12 days
|
* compile: More checks, don't crash on check failTom Smeding2025-04-21
|
* test: A test for new accumMap logicTom Smeding2025-04-21
|
* idana: Unify in either fallback caseTom Smeding2025-04-21
|
* splitLets: Don't split if unnecessaryTom Smeding2025-04-21
|
* pretty: Generate type-tagged fallback namesTom Smeding2025-04-21
|
* Let peeling (let x = Just e in e' ~> let x = e in e'[Just x/x])Tom Smeding2025-04-21
|
* Add integer modulo operatorTom Smeding2025-04-21
|
* Some progress with accumMapTom Smeding2025-04-18
|
* An unused function (descrPrj)Tom Smeding2025-04-18
|
* Better simplification of onehotsTom Smeding2025-04-18
|
* pretty: Print arguments of open expressionTom Smeding2025-04-18
|
* Populate accumMapTom Smeding2025-04-06
|
* Split product lets before chadTom Smeding2025-04-05
|
* Compile: Generate slightly less redundant C codeTom Smeding2025-03-30
|
* Pass around an accumMap (but it's empty still)Tom Smeding2025-03-28
|
* Compile: Commutative fold still broken, but sum is vectorisedTom Smeding2025-03-28
|
* Compile: vectorise commutative foldsTom Smeding2025-03-27
|
* Pretty: fix printing of EMaybeTom Smeding2025-03-27
|
* Compile: Fix formatting of generated codeTom Smeding2025-03-27
|
* Fix refcount bug with indexing for nested arraysTom Smeding2025-03-27
|
* Compile: Fix debugPrintAST formattingTom Smeding2025-03-27
|
* Compile: Decrement accumval only when added, not movedTom Smeding2025-03-27
|
* Compile: remove unused functionTom Smeding2025-03-27
|
* Compile: Better alloc & refc debug facilitiesTom Smeding2025-03-27
|
* test: Start of a list of compile testsTom Smeding2025-03-26
|
* Fix various issues in Compile (still broken)Tom Smeding2025-03-26
|
* Language: support Maybe and Accum termsTom Smeding2025-03-26
|
* Compile.Exec: also remove tmpdir when compilation failsTom Smeding2025-03-26
|
* D2[Array] now has a Maybe instead of zero-size for zeroTom Smeding2025-03-26
| | | | Remaining problem: 'add' in Compile doesn't use the D2 stuff
* Compile: More checksTom Smeding2025-03-23
|
* Compile: More debugging machineryTom Smeding2025-03-22
|
* Compile: Don't toList the Bag in 'scope'Tom Smeding2025-03-21
|
* Compile.Exec: Better diagnostics on GCC errorTom Smeding2025-03-21
|
* Compile: First compilation of fold1iTom Smeding2025-03-21
|
* Compile: Decrement array elements when freeing arrayTom Smeding2025-03-21
|
* Compile: Add some forgotten refcount decrementsTom Smeding2025-03-21
|
* Compile: compileAssign helper functionTom Smeding2025-03-21
|
* Commutativity marker on fold1iTom Smeding2025-03-20
|
* Compile EAccumTom Smeding2025-03-17
|
* simplifyOneHotTermTom Smeding2025-03-16
|
* Complete accumulator revamp!Tom Smeding2025-03-15
|
* Much process with accumulator revampTom Smeding2025-03-14
|
* WIP revamp accumulator projection type reprTom Smeding2025-03-14
| | | | | | | I stopped working on this because I realised that having sparse products (and coproducts, prehaps) everywhere is a very bad idea in general, and that we need to fix that first before really being able to do anything else productive with performance.
generated by cgit v1.2.3-70-g09d2 (git 2.48.1) at 2025-05-06 22:47:31 +0200