aboutsummaryrefslogtreecommitdiff
path: root/src
Commit message (Collapse)AuthorAge
* Fill and clean up *TakeIx and *DropIx functionsmvecsReplicateMikolaj Konarski5 days
|
* Add permId and permShiftNMikolaj Konarski6 days
|
* Odds and endsMikolaj Konarski6 days
|
* Eliminate all QuantifiedConstraints extensions from the codeMikolaj Konarski6 days
|
* Use shxDropSSX instead of shxSplitApp, etc.Mikolaj Konarski6 days
|
* Add pragmas to functions that -fprof-late-overloaded-calls reportsMikolaj Konarski6 days
| | | | | | | as breaking the chain of specialization (but it's currently broken, so these may be false positives, but these pragmas may be handy in the future anyway and they can't do much harm in theory and in practice they are right now neutral performance-wise).
* Remove ixsFromIxR' and simplify ixsFromIxRMikolaj Konarski6 days
|
* Make List?, except ListH, less generalMikolaj Konarski6 days
|
* De-triplicate ix?ToLinearMikolaj Konarski6 days
|
* De-triplicate sh?EnumMikolaj Konarski6 days
|
* Define ix?FromLinear without THMikolaj Konarski6 days
|
* Implement index conversions as unsafeCoerceMikolaj Konarski6 days
| | | | until maybe we make shaped and ranked newtypes over mixed?
* Optimize slightly sh?FromList and sh?ToListmild-tweaks-repackagedMikolaj Konarski6 days
|
* Implement ranked shape conversions as newtype coercesMikolaj Konarski6 days
|
* Make ShR a newtype over ShXMikolaj Konarski6 days
|
* Add the trivial Eq and Ord instances also for StaticShXMikolaj Konarski6 days
|
* Prefer newtype deriving over Generic deriving for simplicityMikolaj Konarski6 days
|
* Temporarily remove the UNPACK, until GHC is fixed not to crash with itMikolaj Konarski6 days
|
* Optimize the representation of ListHMikolaj Konarski6 days
|
* Inline SMayNat in ListHMikolaj Konarski6 days
|
* Clone ListX into ListHMikolaj Konarski6 days
|
* Simplify the signature of fooIndexMikolaj Konarski6 days
|
* Implement shxFromShS and shsFromShX as a newtype coerceMikolaj Konarski6 days
|
* Make ShS a newtype over ShXMikolaj Konarski6 days
| | | | TODO: use lemmas in place of the unsafeCoerceRefl
* Work around the following GHC bug in GHC 9.10.3:Mikolaj Konarski6 days
| | | | | | | | | | | | | | | | src/Data/Array/Nested/Mixed.hs:1045:13: error: [GHC-83865] • Couldn't match type ‘i0’ with ‘Int’ Expected: ShX sh i0 Actual: ShX sh1 Int ‘i0’ is untouchable inside the constraints: (n1 : sh1) ~ (AddMaybe (AddMaybe i n) k : sh) bound by UnkSkol (please report this as a bug) Call stack: CallStack (from HasCallStack): unkSkolAnon, called at compiler/GHC/Tc/Types/Origin.hs:331:42 in ghc-9.10.3-b4c3:GHC.Tc.Types.Origin unkSkol, called at compiler/GHC/Tc/Gen/Pat.hs:1275:49 in ghc-9.10.3-b4c3:GHC.Tc.Gen.Pat at src/Data/Array/Nested/Mixed.hs:1045:7-14
* Add a precautionary INLINEABLE to the now polymorphic shsTakeIxMikolaj Konarski7 days
|
* Add mtoListPrimLinear and friendsMikolaj Konarski7 days
|
* Add mslice to be typing-wise a counterpart of mappendMikolaj Konarski7 days
|
* Fix typo and reword docsTom Smeding7 days
|
* Add a bang not to overlap big allocationsMikolaj Konarski7 days
|
* Uniformise indentationTom Smeding7 days
|
* Save some toVector by using toVectorListTMikolaj Konarski7 days
|
* Let X.fromListOuterSN and ravelOuterN take NonEmptyMikolaj Konarski7 days
|
* Don't force a list of identical elementsMikolaj Konarski7 days
| | | | | | | | | This makes a big difference when a pair of lists is attempted to be streamed and the first one is of trivial primitive elements (e.g., when implementing a fold as a special case of mapAccum with the output list containing only ()). Forcing the first trivial list would cause the second non-trivial list to be represented as nested thunks, burdening GC greatly.
* Variable naming and doc commentsTom Smeding7 days
|
* Simplify and slightly speed up internal fromList functionsMikolaj Konarski2026-02-12
|
* Inline most lifting wrappersMikolaj Konarski2026-02-12
| | | | | | | This results in only marginal performance gain, probably because they are already small enough to be specialized and/or inlined automatically, but these pragmas ensure it remains so regardless of changes in GHC heuristics.
* Make it clear that ravelOuterN is called just onceMikolaj Konarski2026-02-12
|
* Add a potentialy helpful bang for consistencyMikolaj Konarski2026-02-12
|
* Document that mshapeTree is partialMikolaj Konarski2026-02-12
|
* Add a few more GHC warnings and fix all offendersMikolaj Konarski2026-01-31
|
* Improve the implementation of fromSNat'Mikolaj Konarski2026-01-31
| | | | leading to 6% lower allocation in Det.hs primal test.
* Fix f in SMayNat to always be SNat and UNPACK itMikolaj Konarski2026-01-31
|
* Implement Tom's ixsFromIxSMikolaj Konarski2026-01-31
| | | | | | that really helps in processing indexes using list functions and getting back to the same type of index now that KnownNat is removed from IxS and so the type information is harder to preserve.
* Make type variables consistent between shaped and mixed listsMikolaj Konarski2026-01-31
|
* Generalize shsTakeIxMikolaj Konarski2026-01-31
|
* Speed up the bad case of mvecsReplicate using mvecsWriteLinearMikolaj Konarski2026-01-31
|
* Express mvecsWrite and mvecsWritePartial using the new methodsMikolaj Konarski2026-01-31
| | | | | | | and change the type of the latter to make it possible. This slightly improves performance of horde-ad tests, before horde-ad even starts using the Linear methods, which improves performance even more.
* Add mvecsWriteLinear and mvecsWritePartialLinear (unused yet)Mikolaj Konarski2026-01-31
|
* Add mvecsReplicateMikolaj Konarski2026-01-31
|
generated by cgit v1.2.3-70-g09d2 (git 2.52.0) at 2026-03-21 22:33:06 +0100