aboutsummaryrefslogtreecommitdiff
path: root/src
AgeCommit message (Collapse)Author
2016-09-25Change file name ReplacesSubAccesses ReplaceAccesschick
2016-09-25convert all occurencess of BigInt == Int to BigInt == BigIntchick
2016-09-25Fix Anonymous function convertible to a method valuechick
if methods has parens, then referencing without parens is a method value, you don't need following underscore
2016-09-25Minor fixes, typo in wordchick
missing declarations in scala doc
2016-09-25Use empty-parens as appropriate for f: => Unit callschick
2016-09-23use .count instead of filter and sizechick
2016-09-23Use parens on Unit methodschick
2016-09-23use .isEmpty, .nonEmpty, isDefinedchick
2016-09-23use .indiceschick
2016-09-23use .head instead of (0)chick
2016-09-22Fixed width inference for add, sub (#312)Adam Izraelevitz
Fixes #308 Fixes #193
2016-09-21Fix clock connections in InferReadWrite (#310)Donggyu
2016-09-21swap functions in MemPortUtils and MemTransformUtils properly for further ↵Donggyu Kim
refactoring
2016-09-21refactor AnnotateValidMemConfigsDonggyu Kim
2016-09-21refactor ReplaceMemMacrosDonggyu Kim
2016-09-21refactor UpdateDuplicateMemMacrosDonggyu Kim
2016-09-21clean up ReplSeqMemDonggyu Kim
2016-09-21refactor AnnotateMemMacrosDonggyu Kim
2016-09-21refactor InferReadWriteDonggyu Kim
2016-09-21generalize Analysis.getConnects for code resuseDonggyu Kim
2016-09-16fill empty module body with "begin end" (#305)Yunsup Lee
* fill empty module body with "begin end" apparently vivado treats an empty module as a black box and triggers an error * Changed empty module to use always @(*) begin end
2016-09-15Fix non-determinism bug in ExpandWhens (#303)Jack Koenig
Despite the fact that LinkedHashMaps preserve insertion order in traversal, it appears that .keys and .keySet return Sets that do not provide the same guarantee
2016-09-14fix spaces in WIR.scalaDonggyu Kim
2016-09-14style fixes for Compiler.scala, LoweringCompiler.scalaDonggyu Kim
2016-09-14Fix for more general case of getConnectOrigin with reg feedback (#301)Angie Wang
2016-09-14Added Rob.fir for regression testing (#258)Donggyu
2016-09-14fix enable signal inferecne for smems' read ports (#289)Donggyu
2016-09-14Fixed infinite loop for finding connect origin in ReplSeqMem (#300)Angie Wang
* Addressed the fact that a node can be connected to itself (updating reg)
2016-09-13Fix a lurking width-inference bug; improve adjacent style (#298)Andrew Waterman
ceil(log(x) / log(2)) does not, in general, round to ceil(log2(x)). I noticed this because of #297.
2016-09-13use BoolType for UIntType(IntWidth(1))Donggyu Kim
2016-09-13remove VIndentDonggyu Kim
2016-09-13use case object for WVoid, WInvalidDonggyu Kim
2016-09-13cache IntWidths to avoid redudant object creationsDonggyu Kim
2016-09-13use case object for KindDonggyu Kim
2016-09-13clean up PadWidthDonggyu Kim
2016-09-13clean up LowerTypesDonggyu Kim
no vars for mname, info
2016-09-13clean up Passes.scalaDonggyu Kim
2016-09-13type aliasesDonggyu Kim
2016-09-13remove Utils.{width_BANG, long_BANG}Donggyu Kim
2016-09-13remove Utils.get_typeDonggyu Kim
2016-09-13use MemPortUtils.memType for DefMemoryDonggyu Kim
2016-09-13remove Utils.{mapr, get_name} and fix spacesDonggyu Kim
2016-09-13MemPortUtils: return correct memory typesDonggyu Kim
2016-09-13clean up MemUtilsDonggyu Kim
2016-09-12Add LegalizeSpec for testing Verilog Legalization passJack
2016-09-12Add legalization of pad operation on literals.Jack
Performing a pad on SInt literals results in linting warnings in Verilator. This commit replaces pad operations on literal values with a literal of the correct width.
2016-09-12Cast bit select of SInt in PadWidths to SIntJack
Fixes #172
2016-09-12Legalize bit select. Run Legalize after PadWidths.Jack
Bit selecting a literal resulted in invalid Verilog. Legalize now deals with this by replacing any bits select of UInt or SInt literals with a new literal composed of the selected bits. Legalize also is now run after PadWidths because that pass introduces this issue. Fixes #170
2016-09-12Change Legalize Connect to respect SIntJack
Legalize will wrap the rhs of a connect statement with a bit select primop if the lhs is of smaller width than the rhs. This bit select is now wrapped in a asSInt cast if the original rhs was an SInt so that is has the correct type. Fixes #173
2016-09-12Change Legalize Shift Right to respect SIntjackkoenig
Fix bug where Legalize was generating a bit select for SInts without then casting to SInt Fixes #169