Skip to content

GitLab

Switch branch/tag
  1. 05 Jun, 2008 4 commits
    • simonpj@microsoft.com's avatar
      Fix Trac #2045: use big-tuple machiney for implication constraints · 4edf8929
      simonpj@microsoft.com authored
      4edf8929
    • simonpj@microsoft.com's avatar
      Comments only · 9aa2708b
      simonpj@microsoft.com authored
      9aa2708b
    • simonpj@microsoft.com's avatar
      Desugar multiple polymorphic bindings more intelligently · a3f24157
      simonpj@microsoft.com authored 
      Occasionally people write very large recursive groups of definitions. 
In general we desugar these to a single definition that binds tuple,
plus lots of tuple selectors.  But that code has quadratic size, which
can be bad.

This patch adds a new case to the desugaring of bindings, for the
situation where there are lots of polymorphic variables, but no
dictionaries.  (Dictionaries force us into the general case.)

See Note [Abstracting over tyvars only].  

The extra behaviour can be disabled with the (static) flag

	-fno-ds-multi-tyvar

in case we want to experiment with switching it on or off.  There is
essentially-zero effect on the nofib suite though.

I was provoked into doing this by Trac #1136.  In fact I'm not sure
it's the real cause of the problem there, but it's a good idea anyway.
      a3f24157
    • simonpj@microsoft.com's avatar
      Add non-recursive let-bindings for types · 1b1190e0
      simonpj@microsoft.com authored 
      This patch adds to Core the ability to say
	let a = Int in <body>
where 'a' is a type variable.  That is: a type-let.
See Note [Type let] in CoreSyn.

* The binding is always non-recursive
* The simplifier immediately eliminates it by substitution 

So in effect a type-let is just a delayed substitution.  This is convenient
in a couple of places in the desugarer, one existing (see the call to
CoreTyn.mkTyBind in DsUtils), and one that's in the next upcoming patch.

The first use in the desugarer was previously encoded as
	(/\a. <body>) Int
rather that eagerly substituting, but that was horrid because Core Lint
had do "know" that a=Int inside <body> else it would bleat.  Expressing
it directly as a 'let' seems much nicer.
      1b1190e0
  3. 04 Jun, 2008 9 commits
  5. 03 Jun, 2008 1 commit
  7. 30 May, 2008 2 commits
  9. 03 Jun, 2008 7 commits
  11. 02 Jun, 2008 6 commits
  13. 30 May, 2008 1 commit
    • Simon Marlow's avatar
      Fix a bug to do with recursive modules in one-shot mode · d83e1ac4
      Simon Marlow authored 
      The problem was that when loading interface files in checkOldIface, we
were not passing the If monad the mutable variable for use when
looking up entities in the *current* module, with the result that the
knots wouldn't be tied properly, and some instances of TyCons would
be incorrectly abstract.

This bug has subtle effects: for example, recompiling a module without
making any changes might lead to a slightly different result (noticed
due to the new interface-file fingerprints).  The bug doesn't lead to
any direct failures that we're aware of.
      d83e1ac4
  15. 29 May, 2008 1 commit
    • dias@eecs.harvard.edu's avatar
      Replacing copyins and copyouts with data-movement instructions · 0d80489c
      dias@eecs.harvard.edu authored 
      o Moved BlockId stuff to a new file to avoid module recursion
o Defined stack areas for parameter-passing locations and spill slots
o Part way through replacing copy in and copy out nodes
  - added movement instructions for stack pointer
  - added movement instructions for call and return parameters
    (but not with the proper calling conventions)
o Inserting spills and reloads for proc points is now procpoint-aware
  (it was relying on the presence of a CopyIn node as a proxy for
   procpoint knowledge)
o Changed ZipDataflow to expect AGraphs (instead of being polymorphic in
   the type of graph)
      0d80489c
  17. 30 May, 2008 2 commits
  19. 29 May, 2008 4 commits
  21. 28 May, 2008 1 commit
  23. 29 May, 2008 1 commit
    • dias@eecs.harvard.edu's avatar
      Cmm back end upgrades · 25628e27
      dias@eecs.harvard.edu authored 
      Several changes in this patch, partially bug fixes, partially new code:
o bug fixes in ZipDataflow
   - added some checks to verify that facts converge
   - removed some erroneous checks of convergence on entry nodes
   - added some missing applications of transfer functions
o changed dataflow clients to use ZipDataflow, making ZipDataflow0 obsolete
o eliminated DFA monad (no need for separate analysis and rewriting monads with ZipDataflow)
o started stack layout changes
   - no longer generating CopyIn and CopyOut nodes (not yet fully expunged though)
   - still not using proper calling conventions
o simple new optimizations:
   - common block elimination
      -- have not yet tried to move the Adams opt out of CmmProcPointZ
   - block concatenation
o piped optimization fuel up to the HscEnv
   - can be limited by a command-line flag
   - not tested, and probably not yet properly used by clients
o added unique supply to FuelMonad, also lifted unique supply to DFMonad
      25628e27
  25. 27 May, 2008 1 commit