Skip to content


Subversion checkout URL

You can clone with
Download ZIP
Commits on Nov 22, 2015
  1. @herbelin
Commits on Nov 16, 2015
  1. @herbelin

    v8.4: Reverting "compare stacks head first!" which is breaking the

    herbelin committed
    compilation. Bruno is obviously unresponsive about it.
    This reverts commit bd3cc60.
Commits on Nov 7, 2015
  1. @herbelin
Commits on Oct 26, 2015
  1. @letouzey
Commits on Oct 24, 2015
  1. @herbelin

    v8.4: Fixing a loop in checking hints with holes.

    herbelin committed
    For instance, "Hint Resolve (fst _ _)." was looping (bug in 841bc46).
Commits on Oct 15, 2015
  1. @barras

    compare stacks head first!

    barras committed
Commits on Oct 14, 2015
  1. @barras

    minor update of checker/include

    barras committed
  2. @barras
  3. @maximedenes @barras

    Remove dead code in lazy reduction machine.

    maximedenes committed with barras
  4. @barras
  5. @barras
Commits on Sep 9, 2015
  1. @mattam82
  2. @mattam82

    Backport fix for bug 4294

    mattam82 committed
    ... on modules and "with" dropping universe constraints.
  3. @maximedenes

    Fix a bug in 31 bit arithmetic, leading to failing conversion tests.

    maximedenes committed
    On 64 bits architectures, integers could have some of their 32 msb set to 1
    internally in the VM. When read back to a Coq term, this was not observable. But
    an equality test would fail. From the user point of view, the symptom was that
    vm_compute; reflexivity would succeed but the subsequent Qed would fail.
    Bug reported by Tahina Ramananandro.
  4. @catalin-hritcu @maximedenes

    Fixed critical bug in 31 bit arithmetic of VM

    catalin-hritcu committed with maximedenes
    ADDMULDIVINT31 was missing pops in some cases
Commits on Jul 28, 2015
  1. @herbelin
  2. @herbelin

    v8.4: Fixing bug #4289 (hash-consing only user part of constant not

    herbelin committed
    compatible with a unique bound module name counter which is not
    synchronous with the backtracking).
Commits on Jul 16, 2015
  1. @herbelin

    Fixing #4177 (find_projectable was liable to ask to instantiate an ev…

    herbelin committed
    …ar twice).
    Backport from v8.5 (which trunk follows).
Commits on Jun 23, 2015
  1. @zeldovich @letouzey
  2. @zeldovich @letouzey

    Define Any in Haskell extraction when Tunknown is used.

    zeldovich committed with letouzey
    Commit 84c2433 introduced the Any type alias as the Haskell extracted
    version of MiniML's Tunknown.  However, the code to define the Any
    type alias was generated conditional on usf.magic.  As it turns out,
    sometimes Tunknown appears even if usf.magic is false (i.e., even if
    MLmagic does not appear anywhere in the AST).  This produced Haskell
    code that would not compile; e.g.:
      % coqtop
      Coq < Extraction Language Haskell.
      Coq < Extraction Library Datatypes.
      The file Datatypes.hs has been created by extraction.
      % ghc Datatypes.hs
      [1 of 1] Compiling Datatypes        ( Datatypes.hs, Datatypes.o )
      Datatypes.hs:261:17: Not in scope: type constructor or class `Any'
      Datatypes.hs:261:24: Not in scope: type constructor or class `Any'
    The fix is straightforward: produce the code that defines the Any type
    alias if usf.tunknown is true.
  3. @silene @letouzey
  4. @silene @letouzey
  5. @silene @letouzey

    Do not revert parameter lists when extracting singleton types to Hask…

    silene committed with letouzey
    …ell. (Fix for bugs #3470 and #3694)
  6. @zeldovich @letouzey

    Haskell extraction: use explicit -XMagicHash instead of -fglasgow-exts

    zeldovich committed with letouzey
    When Coq's Haskell extraction needs to use unsafeCoerce, it passes
    the -fglasgow-exts option to GHC, but recent versions of GHC warn
    against this:
      xx.hs:1:16: Warning:
          -fglasgow-exts is deprecated: Use individual extensions instead
    This patch does as the warning suggests, replacing -fglasgow-exts
    with the specific option that the extraction needs (-XMagicHash).
  7. @zeldovich @letouzey

    Haskell extraction: put unsafeCoerce type declaration later

    zeldovich committed with letouzey
    When Haskell extraction requires magic type coersion, Coq produces
    the following code:
      unsafeCoerce :: a -> b
      #ifdef __GLASGOW_HASKELL__
      import qualified GHC.Base
      unsafeCoerce = GHC.Base.unsafeCoerce#
      -- HUGS
      import qualified IOExts
      unsafeCoerce = IOExts.unsafeCoerce
    GHC version 7.6.3 does not allow imports after a type declaration,
    and produces this error:
      xx.hs:20:1: parse error on input `import'
    (referring to the first import statement above).  This patch moves
    the unsafeCoerce type declaration to just after the import statement,
    fixing this compile error.
Commits on Jun 22, 2015
  1. @letouzey

    Extraction: no more ascii blob in type variables (fix #3227)

    letouzey committed
     Since type variables are local to the definition, we simply rename
     them in case of unicode chars. We also get rid of any ' to avoid
     Ocaml illegal 'a' type var (clash with char litteral).
Commits on Apr 21, 2015
  1. @herbelin

    v8.4: Fixing #3376 and #4191 (wrong index of maximally-inserted

    herbelin committed
    implicit argument in the presence of let-ins).
Commits on Apr 9, 2015
  1. @letouzey

    v8.4: restore compatibility with OCaml 3.11.2

    letouzey committed
     I doubt anybody is still using 3.11.2 ... except my old mingw
     cross-compiling toolchain for building Windows packages.
  2. @letouzey

    Changelog for 8.4pl6

    letouzey committed
Commits on Apr 1, 2015
  1. @bgregoir
Commits on Mar 25, 2015
  1. @herbelin

    Fully fixing bug #3491 (anomaly when building _rect scheme in the

    herbelin committed
    presence of let-ins and recursively non-uniform parameters).
    The bug was in the List.chop of Inductiveops.get_arity which was wrong
    in the presence of let-ins and recursively non-uniform parameters.
    The bug #3491 showed up because, in addition to have let-ins, it was
    wrongly detected as having recursively non-uniform parameters.
Commits on Mar 24, 2015
  1. @herbelin

    v8.4: Fixing wrong rel_context in checking positivity condition.

    herbelin committed
    Parameters were missing in the context, apparently without negative
    effects because the context was used only for whd normalization of
    types, while reduction (in was resistant to unbound rels.
    Also fixing computation of non-recursively uniform arguments in the
    presence of let-ins. This fixes #3491.
Commits on Mar 16, 2015
  1. @clarus @aspiwack
Commits on Mar 13, 2015
  1. @aspiwack

    Fix compilation with forthcoming Ocaml version 4.03.

    aspiwack committed
    Patch by Pierre Chambart, with the following comment:
     * OCaml runtime header files used to declare the int32, uint32, int64
    and uint64 type. That got removed, and uses of those types should be replaced by the standard ones: uint32_t, int32_t, int64_t, uint64_t. Those are defined in stdint.h.
    * OCaml now forbid declaring two exceptions with the same name in a compilation unit.
  2. @herbelin
Something went wrong with that request. Please try again.