diff options
| author | azidar | 2015-04-08 11:28:29 -0700 |
|---|---|---|
| committer | azidar | 2015-04-08 11:28:29 -0700 |
| commit | e5b9f6ec710e8573ce262330731bebc7524296e5 (patch) | |
| tree | 1494853a2939b20bb2c671d3c46daa29b76ecec3 /TODO | |
| parent | d4fdab6950b47379137fce750e4a3a6b262e750d (diff) | |
Finished expand whens. started infer widths. added pdf for people to view
Diffstat (limited to 'TODO')
| -rw-r--r-- | TODO | 25 |
1 files changed, 25 insertions, 0 deletions
@@ -1,5 +1,24 @@ TODO change parser to use <> syntax (and update all tests) + think about on-reset + make instances always male, flip the bundles on declaration + write out FIFO example pass + write out diff mem examples and how we will support it + instances can be bulk connected -> change := semantics to: + F := M -> F := M (legal) + M := F -> M := F (legal) + M := M -> (illegal) + F := F -> (illegal) + B := B -> F := M (legal) + F := U -> F := M (legal) + M := U -> M := F (legal) + U := M -> F := M (legal) + U := F -> M := F (legal) + U := U -> U := U (legal) + Address this in gender inference pass, when see M := F, replace with F := M + What if Instances were Male always? And its type is the flipped version of module?? + + talk to palmer/patrick about how writing passes is going to be supported Write lowering step for primops Figure out how widths propogate for all updated primops (Adam) @@ -16,6 +35,7 @@ TODO Update spec add new field for sequential or combinational add assertions + cannot connect directly to a mem (loc can never contain a mem) Future questions to address in spec: Introduction – motivation, and intended usage Philosophical justifications for all constructs @@ -49,3 +69,8 @@ defnode n = e ==> wire n e := n + + + +Front-end needs to guarantee unique names per module. +FIRRTL rule: No name can be a prefix of any other name. |
