Add ClashFloPoCo #23
Open
Add ClashFloPoCo #23
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Currently, it has the warning issue for the genBlackBoxProd function used in FloPoCoExample.hs . I will add ghc flag to disable the warning later. |
christiaanb
reviewed
Jan 14, 2025
| @@ -0,0 +1,121 @@ | |||
| # Clash library for interacting with FloPoCo | |||
| This is the library contains API to interact with FloPoCo to generate | |||
There was a problem hiding this comment.
Suggested change
| This is the library contains API to interact with FloPoCo to generate | |
| This is library contains an API to interact with [FloPoCo](https://flopoco.org/) to generate | |
| @@ -0,0 +1,121 @@ | |||
| # Clash library for interacting with FloPoCo | |||
| This is the library contains API to interact with FloPoCo to generate | |||
| hardware synthesizable floating point arithmetics function. | |||
There was a problem hiding this comment.
Suggested change
| hardware synthesizable floating point arithmetics function. | |
| synthesizable floating point arithmetic functions. | |
| # Clash library for interacting with FloPoCo | ||
| This is the library contains API to interact with FloPoCo to generate | ||
| hardware synthesizable floating point arithmetics function. | ||
| It can also be used with Float type in haskell |
There was a problem hiding this comment.
Suggested change
| It can also be used with Float type in haskell | |
| It can also be used with Haskell's regular 'Float' type. | |
| It can also be used with Float type in haskell | ||
| # How to use it | ||
| 1. Install FloPoCo library | ||
| On Linux, user can run to go to FloPoCo source code follow the instruction. |
There was a problem hiding this comment.
Suggested change
| On Linux, user can run to go to FloPoCo source code follow the instruction. | |
| On Linux, you can follow the [installation instructions](https://flopoco.org/flopoco_installation.html) to build FloPoCo from source. |
| # How to use it | ||
| 1. Install FloPoCo library | ||
| On Linux, user can run to go to FloPoCo source code follow the instruction. | ||
| For Window, user must enable Window Subsystem Linux to install FloPoCo as |
There was a problem hiding this comment.
Suggested change
| For Window, user must enable Window Subsystem Linux to install FloPoCo as | |
| For Windows, you must enable Window Subsystem Linux first and then follow the instructions for Linux as |
Comment on lines
17
to
28
| module Clash.Cores.ClashFloPoCo.GenTemDSL | ||
| ( getPipeDep, | ||
| flopocoPrim, | ||
| genFloPoCoInfoEntity, | ||
| liftInfoEntity, | ||
| genBlackBoxFunction, | ||
| genTemplateFunction, | ||
| genBlackBoxTemplateFunction, | ||
| genBlackBoxTemplateFunctionProd, | ||
| genBlackBox, | ||
| genBlackBoxProd, | ||
| ) |
There was a problem hiding this comment.
Use leading , for the export list, see for example:
clash-cores/src/Clash/Cores/Xilinx/Floating.hs
Lines 45 to 76 in 3cc8c71
| getPipeDep infoen = do | ||
| return (LitT (NumTyLit (fromIntegral (convertMaybeToInt (pipedep infoen))))) | ||
|
|
||
| -- | Function to generate the InfoEntity coresponding to the arithmetic entity generated by FloPoCo |
There was a problem hiding this comment.
Have haddock generate a link to InfoEntity
Suggested change
| -- | Function to generate the InfoEntity coresponding to the arithmetic entity generated by FloPoCo | |
| -- | Function to generate the 'InfoEntity' coresponding to the arithmetic entity generated by FloPoCo |
Comment on lines
100
to
102
| -- | ||
| -- | ||
|
|
Comment on lines
96
to
98
| liftInfoEntity :: InfoEntity -> Q Exp | ||
| liftInfoEntity (InfoEntity n f p ins outs) = | ||
| [|InfoEntity $(lift n) $(lift f) $(lift p) $(lift ins) $(lift outs)|] |
There was a problem hiding this comment.
Why not create an instance Lift InfoEntity where you define the InfoEntity type?
Comment on lines
+141
to
+143
| -- | Function to convert Maybe Int to Int | ||
| convertMaybeToInt :: Maybe Int -> Int | ||
| convertMaybeToInt = fromMaybe 0 |
There was a problem hiding this comment.
I would just inline this definition, because:
- It only has one use site
- The name
convertMaybeToIntdoesn't really inform the user that happens when the argument isNothing; the choice for0is arbitrary.
christiaanb
reviewed
Jan 14, 2025
Comment on lines
+173
to
+199
| let funcDec = | ||
| FunD | ||
| bbfName | ||
| [ Clause | ||
| [WildP, WildP, WildP, WildP] | ||
| ( NormalB | ||
| ( AppE | ||
| (VarE 'pure) | ||
| ( AppE | ||
| (ConE 'Right) | ||
| ( TupE | ||
| [ Just (RecUpdE (VarE 'emptyBlackBoxMeta) [('bbKind, ConE 'TDecl)]), | ||
| Just | ||
| ( AppE | ||
| ( AppE | ||
| (AppE (ConE 'BBFunction) (LitE (StringL (entityName <> "TF")))) -- 1st arg | ||
| (LitE (IntegerL 0)) -- 2nd arg | ||
| ) | ||
| (AppE (VarE tfName) (LitE (StringL entityName))) -- 3rd arg | ||
| ) | ||
| ] | ||
| ) | ||
| ) | ||
| ) | ||
| ) | ||
| [] | ||
| ] |
There was a problem hiding this comment.
This could be simplified to:
-- Create the function declaration
funExpr <- [|pure (Right (emptyBlackBoxMeta {bbKind = TDecl}, BBFunction (entityName <> "TF") 0 ($(varE tfName) entityName)))|]
let funcDec = FunD bbfName [Clause [WildP,WildP,WildP, WildP] (NormalB funExpr) []]
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This will ClashFloPoCo, which is the library for interacting between Clash and FloPoCo. This library provides a custom example of the implementation and binding 3 single precision floating point operations supported by FloPoCo: addition, fuse multiply addition, and exponent. It also provides a markdown doc about how to use it.