Proposal: Performance improvements for Data.IntMap

[Antoine Latter]

On Tue, Aug 31, 2010 at 5:09 AM, Don Stewart <dons at galois.com> wrote:
>
> #4279: Proposal: Performance improvements for Data.IntMap
> http://hackage.haskell.org/trac/ghc/ticket/4279
>
>  Following on from ticket #4277 here is a similar patch for Data.IntMap.
>
>  This proposal provides a patch that improves performance for many parts of
>  the API. Two standard techniques are applied to the code:
>
>  * worker/wrapper transformations of functions with multiple static arguments
>  * inlining of non-recursive wrappers
>
> Three complementary, but orthogonal patches are provided.
>
>  1. Add a testsuite, with
>         [http://code.haskell.org/~dons/tests/containers/intmap/hpc_index.html coverage data]
>         (currently 82% of top level functions, and all core functions).
>  2. Add a micro-benchmark suite based on Criterion, for empirical evidence
>         of improvements to each function.
>  3. The optimization patch itself.
>
> All 3 patches should be applied, under this proposal.
>

+1

All three look like a good thing.

Is this sort of optimization specific to the way GHC compiles things,
or will other compilers likely either benefit or not care?

Antoine