Annotating Deeply Embedded Languages

Presented at the Haskell Implementers Workshop 2022: https://icfp22.sigplan.org/home/hiw-2022#About
Video: TBD

Haskell's strong static typing and the ease at which the programmer can create and manipulate algebraic data types make it a good choice for implementing (embedded) programming languages. However a common problem with deeply embedded languages is that there is no strong connection between the source program the user writes; the abstract syntax tree that is generated for that embedded program; and the object code that is eventually generated, complied, and executed for that program. Once the embedded language moves beyond the prototype phase, this disconnect makes it almost impossible for a user to debug their program or determine where the execution time is being spent.

In this talk we discuss recent work on annotating the abstract syntax tree of the deeply embedded languages Accelerate. We present our approach for automatically gathering source location information for embedded expressions using a novel implicit-parameter based approach. Unlike the existing HasCallStack mechanism provided by GHC, our approach statically ensures that the required source location information is available when needed. Annotating the abstract syntax tree of the program with this information, it is now possible to map the embedded program back to the original source code. We demonstrate our current progress on using this information to improve compiler diagnostics, as well as enable profiler and debugger integration for embedded programs.

Transcript

1. Annotating Deeply Embedded Languages
   Robbert van der Helm
   Trevor L. McDonell
   Gabriele Keller
   

   View Slide

2. Embedded Languages
   A language written inside of another language
   
   
   May be domain speci
   fi
   c
   
   
   That other language is called the host language
   2
   

   View Slide

3. Deeply Embedded Languages
   3
   https://ghc.gitlab.haskell.org/ghc/doc/users_guide/exts/gadt.html
   data Exp a where
   
   
   Lit
   : :
   Int
   - >
   Exp Int
   
   
   Succ
   : :
   Exp Int
   - >
   Exp Int
   
   
   . . .
   

   View Slide

4. ans
   : :
   Exp Int
   
   
   ans = Succ (Lit 41)
   Deeply Embedded Languages
   3
   https://ghc.gitlab.haskell.org/ghc/doc/users_guide/exts/gadt.html
   data Exp a where
   
   
   Lit
   : :
   Int
   - >
   Exp Int
   
   
   Succ
   : :
   Exp Int
   - >
   Exp Int
   
   
   . . .
   

   View Slide

5. ans
   : :
   Exp Int
   
   
   ans = Succ (Lit 41)
   Deeply Embedded Languages
   3
   https://ghc.gitlab.haskell.org/ghc/doc/users_guide/exts/gadt.html
   data Exp a where
   
   
   Lit
   : :
   Int
   - >
   Exp Int
   
   
   Succ
   : :
   Exp Int
   - >
   Exp Int
   
   
   . . .
   eval
   : :
   Exp Int
   - >
   Int
   
   
   eval (Lit i) = i
   
   
   eval (Succ x) = 1 + eval x
   
   
   . . .
   

   View Slide

6. ans
   : :
   Exp Int
   
   
   ans = Succ (Lit 41)
   Deeply Embedded Languages
   3
   https://ghc.gitlab.haskell.org/ghc/doc/users_guide/exts/gadt.html
   data Exp a where
   
   
   Lit
   : :
   Int
   - >
   Exp Int
   
   
   Succ
   : :
   Exp Int
   - >
   Exp Int
   
   
   . . .
   eval
   : :
   Exp Int
   - >
   Int
   
   
   eval (Lit i) = i
   
   
   eval (Succ x) = 1 + eval x
   
   
   . . .
   

   View Slide

7. ans
   : :
   Exp Int
   
   
   ans = Succ (Lit 41)
   Deeply Embedded Languages
   3
   https://ghc.gitlab.haskell.org/ghc/doc/users_guide/exts/gadt.html
   data Exp a where
   
   
   Lit
   : :
   Int
   - >
   Exp Int
   
   
   Succ
   : :
   Exp Int
   - >
   Exp Int
   
   
   . . .
   eval
   : :
   Exp Int
   - >
   Int
   
   
   eval (Lit i) = i
   
   
   eval (Succ x) = 1 + eval x
   
   
   . . .
   

   View Slide

8. ans
   : :
   Exp Int
   
   
   ans = Succ (Lit 41)
   Deeply Embedded Languages
   3
   https://ghc.gitlab.haskell.org/ghc/doc/users_guide/exts/gadt.html
   data Exp a where
   
   
   Lit
   : :
   Int
   - >
   Exp Int
   
   
   Succ
   : :
   Exp Int
   - >
   Exp Int
   
   
   . . .
   eval
   : :
   Exp Int
   - >
   Int
   
   
   eval (Lit i) = i
   
   
   eval (Succ x) = 1 + eval x
   
   
   . . .
   academics
   

   View Slide

9. ans
   : :
   Exp Int
   
   
   ans = Succ (Lit 41)
   Deeply Embedded Languages
   3
   https://ghc.gitlab.haskell.org/ghc/doc/users_guide/exts/gadt.html
   data Exp a where
   
   
   Lit
   : :
   Int
   - >
   Exp Int
   
   
   Succ
   : :
   Exp Int
   - >
   Exp Int
   
   
   . . .
   eval
   : :
   Exp Int
   - >
   Int
   
   
   eval (Lit i) = i
   
   
   eval (Succ x) = 1 + eval x
   
   
   . . .
   academics
   me
   

   View Slide

10. Deeply Embedded Languages
    4
    

    View Slide

11. Deeply Embedded Languages
    Advantages:
    Integrates with the host language
    No separate parsing step
    4
    

    View Slide

12. Deeply Embedded Languages
    Advantages:
    Integrates with the host language
    No separate parsing step
    Disadvantages:
    No separate parsing step
    4
    

    View Slide

13. Deeply Embedded Languages
    Advantages:
    Integrates with the host language
    No separate parsing step
    Disadvantages:
    No separate parsing step
    Deep embeddings lack context information
    4
    

    View Slide

14. The Problem
    • Let’s write an embedded program!
    5
    https://github.com/tmcdonell/lulesh-accelerate
    

    View Slide

15. The Problem
    • Let’s write an embedded program!
    1. Write some code
    5
    https://github.com/tmcdonell/lulesh-accelerate
    

    View Slide

16. The Problem
    • Let’s write an embedded program!
    1. Write some code
    5
    https://github.com/tmcdonell/lulesh-accelerate
    

    View Slide

17. The Problem
    • Let’s write an embedded program!
    1. Write some code
    5
    https://github.com/tmcdonell/lulesh-accelerate
    

    View Slide

18. The Problem
    • Let’s write an embedded program!
    1. Write some code
    2. Run it!
    5
    https://github.com/tmcdonell/lulesh-accelerate
    

    View Slide

19. The Problem
    • Let’s write an embedded program!
    1. Write some code
    2. Run it!
    5
    https://github.com/tmcdonell/lulesh-accelerate
    

    View Slide

20. The Problem
    • Let’s write an embedded program!
    1. Write some code
    2. Run it!
    5
    https://github.com/tmcdonell/lulesh-accelerate
    

    View Slide

21. The Problem
    • Let’s write an embedded program!
    1. Write some code
    2. Run it!
    5
    https://github.com/tmcdonell/lulesh-accelerate
    

    View Slide

22. The Problem
    • Let’s write an embedded program!
    1. Write some code
    2. Run it!
    5
    https://github.com/tmcdonell/lulesh-accelerate
    👍
    

    View Slide

23. The Problem
    • Let’s write an embedded program!
    1. Write some code
    2. Run it!
    5
    https://github.com/tmcdonell/lulesh-accelerate
    👍
    

    View Slide

24. The Problem
    • Let’s write an embedded program!
    1. Write some code
    2. Run it!
    5
    https://github.com/tmcdonell/lulesh-accelerate
    👍
    

    View Slide

25. The Problem
    • Let’s write an embedded program!
    1. Write some code
    2. Run it!
    5
    https://github.com/tmcdonell/lulesh-accelerate
    👍
    

    View Slide

26. The Problem
    • Let’s write an embedded program!
    1. Write some code
    2. Run it!
    3. …
    5
    https://github.com/tmcdonell/lulesh-accelerate
    👍
    

    View Slide

27. The Problem
    • Let’s write an embedded program!
    1. Write some code
    2. Run it!
    3. …
    5
    https://github.com/tmcdonell/lulesh-accelerate
    👍
    

    View Slide

28. The Problem
    • Let’s write an embedded program!
    1. Write some code
    2. Run it!
    3. …
    5
    https://github.com/tmcdonell/lulesh-accelerate
    👍
    

    View Slide

29. The Problem
    • Let’s write an embedded program!
    1. Write some code
    2. Run it!
    3. …
    5
    https://github.com/tmcdonell/lulesh-accelerate
    👍
    

    View Slide

30. The Problem
    • Let’s write an embedded program!
    1. Write some code
    2. Run it!
    3. …
    5
    https://github.com/tmcdonell/lulesh-accelerate
    👍
    

    View Slide

31. The Problem
    • Let’s write an embedded program!
    1. Write some code
    2. Run it!
    3. …
    5
    https://github.com/tmcdonell/lulesh-accelerate
    👍
    

    View Slide

32. The Problem
    • Let’s write an embedded program!
    1. Write some code
    2. Run it!
    3. …
    5
    https://github.com/tmcdonell/lulesh-accelerate
    👍
    

    View Slide

33. The Problem
    • Let’s write an embedded program!
    1. Write some code
    2. Run it!
    3. …
    4. Pro
    fi
    t?
    5
    https://github.com/tmcdonell/lulesh-accelerate
    👍
    🤔
    

    View Slide

34. The Problem
    There is a disconnect between:
    
    
    the embedded program the user writes;
    
    
    the abstract syntax tree generated for that program;
    
    
    and the optimised code that is eventually executed
    6
    

    View Slide

35. Objective
    1. Recover context in deeply embedded programs
    
    
    2. Annotate the embedded program with that information
    
    
    3. Find other uses for the annotation system
    7
    

    View Slide

36. The Idea
    The program is described by some abstract syntax tree
    
    
    This AST is built via smart constructors
    
    
    These smart constructors should generate and store the necessary annotations
    8
    

    View Slide

37. The Idea
    These smart constructors can be encountered by:
    
    
    Regular functions
    
    
    Type class methods
    
    
    Pattern synonymns
    9
    Embedded Pattern Matching, McDonell T.L., Meredith, J.D, and Keller G.
    

    View Slide

38. The Idea
    These smart constructors can be encountered by:
    
    
    Regular functions
    
    
    Type class methods
    
    
    Pattern synonymns
    9
    Embedded Pattern Matching, McDonell T.L., Meredith, J.D, and Keller G.
    constant
    : :
    Int
    - >
    Exp Int
    
    
    constant = Lit
    

    View Slide

39. The Idea
    These smart constructors can be encountered by:
    
    
    Regular functions
    
    
    Type class methods
    
    
    Pattern synonymns
    9
    Embedded Pattern Matching, McDonell T.L., Meredith, J.D, and Keller G.
    instance Num (Exp a) where
    
    
    (+) = PrimApp PrimAdd
    constant
    : :
    Int
    - >
    Exp Int
    
    
    constant = Lit
    

    View Slide

40. The Idea
    These smart constructors can be encountered by:
    
    
    Regular functions
    
    
    Type class methods
    
    
    Pattern synonymns
    9
    Embedded Pattern Matching, McDonell T.L., Meredith, J.D, and Keller G.
    instance Num (Exp a) where
    
    
    (+) = PrimApp PrimAdd
    pattern Maybe_
    : :
    Exp a
    - >
    Exp (Maybe a)
    constant
    : :
    Int
    - >
    Exp Int
    
    
    constant = Lit
    

    View Slide

41. Annotations
    Store metadata for an AST node
    
    
    Should be easily extensible
    
    
    Adding them shouldn’t change the user-facing language
    10
    

    View Slide

42. Storing Annotations
    11
    Trees that Grow, Njjd S. and Peyton-Jones S.
    data Ann = Ann {
    . . .
    }
    
    
    data Exp a where
    
    
    Lit
    : :
    Ann
    - >
    
    Succ
    : :
    Ann
    - >
    
    . . .
    
    constant
    : :
    Int
    - >
    Exp Int
    
    
    constant = Lit mkAnn
    
    
    mkAnn
    : : . . . = >
    Ann
    
    
    mkAnn = Ann {
    . . .
    }
    Int
    - >
    Exp Int
    
    
    Exp Int
    - >
    Exp Int
    

    View Slide

43. Storing Annotations
    11
    Trees that Grow, Njjd S. and Peyton-Jones S.
    data Ann = Ann {
    . . .
    }
    
    
    data Exp a where
    
    
    Lit
    : :
    Ann
    - >
    
    Succ
    : :
    Ann
    - >
    
    . . .
    
    constant
    : :
    Int
    - >
    Exp Int
    
    
    constant = Lit mkAnn
    
    
    mkAnn
    : : . . . = >
    Ann
    
    
    mkAnn = Ann {
    . . .
    }
    Int
    - >
    Exp Int
    
    
    Exp Int
    - >
    Exp Int
    

    View Slide

44. Storing Annotations
    11
    Trees that Grow, Njjd S. and Peyton-Jones S.
    data Ann = Ann {
    . . .
    }
    
    
    data Exp a where
    
    
    Lit
    : :
    Ann
    - >
    
    Succ
    : :
    Ann
    - >
    
    . . .
    
    constant
    : :
    Int
    - >
    Exp Int
    
    
    constant = Lit mkAnn
    
    
    mkAnn
    : : . . . = >
    Ann
    
    
    mkAnn = Ann {
    . . .
    }
    Int
    - >
    Exp Int
    
    
    Exp Int
    - >
    Exp Int
    

    View Slide

45. Storing Annotations
    11
    Trees that Grow, Njjd S. and Peyton-Jones S.
    data Ann = Ann {
    . . .
    }
    
    
    data Exp a where
    
    
    Lit
    : :
    Ann
    - >
    
    Succ
    : :
    Ann
    - >
    
    . . .
    
    constant
    : :
    Int
    - >
    Exp Int
    
    
    constant = Lit mkAnn
    
    
    mkAnn
    : : . . . = >
    Ann
    
    
    mkAnn = Ann {
    . . .
    }
    Int
    - >
    Exp Int
    
    
    Exp Int
    - >
    Exp Int
    

    View Slide

46. Storing Annotations
    11
    Trees that Grow, Njjd S. and Peyton-Jones S.
    data Ann = Ann {
    . . .
    }
    
    
    data Exp a where
    
    
    Lit
    : :
    Ann
    - >
    
    Succ
    : :
    Ann
    - >
    
    . . .
    
    constant
    : :
    Int
    - >
    Exp Int
    
    
    constant = Lit mkAnn
    
    
    mkAnn
    : : . . . = >
    Ann
    
    
    mkAnn = Ann {
    . . .
    }
    Int
    - >
    Exp Int
    
    
    Exp Int
    - >
    Exp Int
    

    View Slide

47. Source Locations
    Associate AST fragments back to their original source location
    
    
    Use that for diagnostics, pro
    fi
    ling, debugging…
    12
    

    View Slide

48. Source Locations
    1. GHC Call Stacks: GHC.Stack
    
    
    2. RTS Execution Stacks: GHC.ExecutionStack
    13
    

    View Slide

49. Source Locations
    1. GHC Call Stacks: GHC.Stack
    
    
    2. RTS Execution Stacks: GHC.ExecutionStack
    13
    • Created at compile time
    • Functions require a HasCallStack constraint
    

    View Slide

50. Source Locations
    1. GHC Call Stacks: GHC.Stack
    
    
    2. RTS Execution Stacks: GHC.ExecutionStack
    13
    • Runtime backtraces!
    • No changes to user code required!
    • Created at compile time
    • Functions require a HasCallStack constraint
    

    View Slide

51. Source Locations
    1. GHC Call Stacks: GHC.Stack
    
    
    2. RTS Execution Stacks: GHC.ExecutionStack
    13
    • Runtime backtraces!
    • No changes to user code required!
    • Currently unusable 🙁
    • Created at compile time
    • Functions require a HasCallStack constraint
    

    View Slide

52. Source Locations
    Three scenarios:
    
    
    Regular functions: GHC Call Stacks
    
    
    (Existing) Type class methods: RTS Execution Stacks
    
    
    Pattern synonyms: GHC Call Stacks (plus some trickery)
    
    
    14
    https://gitlab.haskell.org/ghc/ghc/-/issues/19289
    

    View Slide

53. HasCallStack
    15
    printError
    : :
    HasCallStack
    = >
    String
    - >
    IO ()
    
    
    printError msg = putStrLn msg
    > >
    print callStack
    

    View Slide

54. HasCallStack
    15
    printError
    : :
    HasCallStack
    = >
    String
    - >
    IO ()
    
    
    printError msg = putStrLn msg
    > >
    print callStack
    printError
    : :
    (?callStack
    : :
    CallStack)
    = >
    String
    - >
    IO ()
    
    
    printError msg = putStrLn msg
    > >
    print ?callStack
    desugars to….
    

    View Slide

55. HasCallStack
    16
    main
    : :
    HasCallStack
    = >
    IO ()
    
    
    main = foo
    
    
    foo
    : :
    IO ()
    - -
    silent error: no HasCallStack constraint!
    
    
    foo = bar
    
    
    bar
    : :
    HasCallStack
    = >
    IO ()
    - -
    only prints ‘bar’
    
    
    bar = print callStack
    

    View Slide

56. HasCallStack
    16
    main
    : :
    HasCallStack
    = >
    IO ()
    
    
    main = foo
    
    
    foo
    : :
    IO ()
    - -
    silent error: no HasCallStack constraint!
    
    
    foo = bar
    
    
    bar
    : :
    HasCallStack
    = >
    IO ()
    - -
    only prints ‘bar’
    
    
    bar = print callStack
    

    View Slide

57. HasCallStack
    16
    main
    : :
    HasCallStack
    = >
    IO ()
    
    
    main = foo
    
    
    foo
    : :
    IO ()
    - -
    silent error: no HasCallStack constraint!
    
    
    foo = bar
    
    
    bar
    : :
    HasCallStack
    = >
    IO ()
    - -
    only prints ‘bar’
    
    
    bar = print callStack
    

    View Slide

58. SourceMapped
    17
    data OpaqueType = NotExported
    
    
    type SourceMapped =
    
    
    ( ?requiresSourceMapping
    : :
    OpaqueType, HasCallStack )
    
    
    - -
    Throws an error if the caller did not have the HasCallStack constraint
    
    
    sourceMap
    : :
    HasCallStack
    = >
    (SourceMapped
    = >
    a)
    - >
    a
    
    
    sourceMap k =
    . . .
    

    View Slide

59. SourceMapped
    17
    data OpaqueType = NotExported
    
    
    type SourceMapped =
    
    
    ( ?requiresSourceMapping
    : :
    OpaqueType, HasCallStack )
    
    
    - -
    Throws an error if the caller did not have the HasCallStack constraint
    
    
    sourceMap
    : :
    HasCallStack
    = >
    (SourceMapped
    = >
    a)
    - >
    a
    
    
    sourceMap k =
    . . .
    The only way to satisfy
    the SourceMapped constraint
    

    View Slide

60. SourceMapped
    18
    main
    : :
    HasCallStack
    = >
    IO ()
    
    
    main = foo
    
    
    foo
    : :
    IO ()
    - -
    silent error: no HasCallStack constraint
    
    
    foo = bar
    
    
    bar
    : :
    HasCallStack
    = >
    IO ()
    
    
    bar = print callStack
    
    
    qux
    : :
    HasCallStack
    = >
    
    qux = sourceMap bar
    - -
    Runtime error: no HasCallStack
    IO ()
    

    View Slide

61. SourceMapped
    18
    main
    : :
    HasCallStack
    = >
    IO ()
    
    
    main = foo
    
    
    foo
    : :
    IO ()
    - -
    silent error: no HasCallStack constraint
    
    
    foo = bar
    
    
    bar
    : :
    HasCallStack
    = >
    IO ()
    
    
    bar = print callStack
    
    
    qux
    : :
    HasCallStack
    = >
    
    qux = sourceMap bar
    SourceMapped
    - -
    Runtime error: no HasCallStack
    IO ()
    

    View Slide

62. SourceMapped
    18
    main
    : :
    HasCallStack
    = >
    IO ()
    
    
    main = foo
    
    
    foo
    : :
    IO ()
    - -
    silent error: no HasCallStack constraint
    
    
    foo = bar
    
    
    bar
    : :
    HasCallStack
    = >
    IO ()
    
    
    bar = print callStack
    
    
    qux
    : :
    HasCallStack
    = >
    
    qux = sourceMap bar
    SourceMapped
    - -
    Compilation error: unbound implicit parameter
    - -
    Runtime error: no HasCallStack
    IO ()
    

    View Slide

63. SourceMapped
    18
    main
    : :
    HasCallStack
    = >
    IO ()
    
    
    main = foo
    
    
    foo
    : :
    IO ()
    - -
    silent error: no HasCallStack constraint
    
    
    foo = bar
    
    
    bar
    : :
    HasCallStack
    = >
    IO ()
    
    
    bar = print callStack
    
    
    qux
    : :
    HasCallStack
    = >
    
    qux = sourceMap bar
    SourceMapped
    - -
    Compilation error: unbound implicit parameter
    - -
    Runtime error: no HasCallStack
    IO ()
    

    View Slide

64. SourceMapped
    18
    main
    : :
    HasCallStack
    = >
    IO ()
    
    
    main = foo
    
    
    foo
    : :
    IO ()
    - -
    silent error: no HasCallStack constraint
    
    
    foo = bar
    
    
    bar
    : :
    HasCallStack
    = >
    IO ()
    
    
    bar = print callStack
    
    
    qux
    : :
    HasCallStack
    = >
    
    qux = sourceMap bar
    SourceMapped
    - -
    Compilation error: unbound implicit parameter
    IO ()
    - -
    Works!
    

    View Slide

65. Putting it together
    19
    data Ann = Ann { locations
    : :
    HashSet CallStack,
    . . .
    }
    
    
    mkAnn
    : :
    SourceMapped
    = >
    Ann
    
    
    mkAnn = Ann { locations = capture ?callStack }
    
    
    where
    
    
    capture =
    . . .
    
    constant
    : :
    HasCallStack
    = >
    Int
    - >
    Exp Int
    
    
    constant = sourceMap $ Lit mkAnn
    

    View Slide

66. Putting it together
    19
    data Ann = Ann { locations
    : :
    HashSet CallStack,
    . . .
    }
    
    
    mkAnn
    : :
    SourceMapped
    = >
    Ann
    
    
    mkAnn = Ann { locations = capture ?callStack }
    
    
    where
    
    
    capture =
    . . .
    
    constant
    : :
    HasCallStack
    = >
    Int
    - >
    Exp Int
    
    
    constant = sourceMap $ Lit mkAnn
    • inlining
    • loop unrolling
    • [no] fast math
    • …
    

    View Slide

67. Case study: Accelerate
    • Deeply embedded language for data-parallel array computations
    
    - Multiple backends (CPU, GPU, …)
    - Multiple expression types (collective array and scalar expression)
    - Multiple AST types (surface language in HOAS, internal language is
    fi
    rst-order)
    20
    

    View Slide

68. Accelerate: Sharing recovery
    • Finds shared parts of the program based on stable names; moves those parts to explicit binders
    
    • NEW:
    
    - Keep track of annotation state during sharing recovery
    - Enable terms to be inlining by ignoring sharing
    21
    

    View Slide

69. Accelerate: Array fusion
    • Producer/consumer fusion rewrites the program to combine operations
    22
    

    View Slide

70. Accelerate: Array fusion
    • Producer/consumer fusion rewrites the program to combine operations
    22
    map f . map g
    

    View Slide

71. Accelerate: Array fusion
    • Producer/consumer fusion rewrites the program to combine operations
    22
    map f . map g
    map (f . g)
    rewries into…
    

    View Slide

72. Accelerate: Array fusion
    • Producer/consumer fusion rewrites the program to combine operations
    • NEW:
    
    - Multiple AST nodes may be merged; new annotation contains the union of the call stack sets
    - Source locations may be disjoint
    - Optimisation
    fl
    ags might differ…
    22
    map f . map g
    map (f . g)
    rewries into…
    

    View Slide

73. DEMO
    23
    

    View Slide

74. Pro
    fi
    ling LULESH
    • One kernel accounts for 66% of the runtime
    
    - A good candidate to experiment with loop unrolling…
    

    

    

    

    

    24
    Ryzen 9 5900x: 8.66 s
    - >
    9.6 s
    RTX 2080 Super: 3.13 s
    - >
    3.07 ± 0.04 s
    

    View Slide

75. Pro
    fi
    ling LULESH
    • One kernel accounts for 66% of the runtime
    
    - A good candidate to experiment with loop unrolling…
    

    

    

    

    

    24
    Ryzen 9 5900x: 8.66 s
    - >
    9.6 s
    RTX 2080 Super: 3.13 s
    - >
    3.07 ± 0.04 s
    

    View Slide

76. Pro
    fi
    ling LULESH
    • One kernel accounts for 66% of the runtime
    
    - A good candidate to experiment with loop unrolling…
    

    

    

    

    

    24
    Ryzen 9 5900x: 8.66 s
    - >
    9.6 s
    RTX 2080 Super: 3.13 s
    - >
    3.07 ± 0.04 s
    • 14x higher L1 inst. cache miss rate
    • 7x higher LL inst. cache miss rate
    

    View Slide

77. Summary
    New annotation system opens the door to a better developer experience
    
    
    Future work:
    
    
    1. Expression-level debugging and pro
    fi
    ling
    
    
    2. Granular loop optimisations
    
    
    3. …?
    
    
    4. Merge it
    25
    

    View Slide

78. AccelerateHS.org
    https://github.com/AccelerateHS/
    Robbert van der Helm Trevor L. McDonell Gabriele Keller
    

    View Slide