1. 20 Aug, 2013 5 commits
    • Jan Stolarek's avatar
      Cleanup StgCmm pass · e5374a1b
      Jan Stolarek authored
      This cleanup includes:
        * removing dead code. This includes forkStatics function,
          which was in fact one big noop, and global bindings in
        * converting functions that used FCode monad only to
          access DynFlags into functions that take DynFlags
          as a parameter and don't work in a monad,
        * addBindC function is now smarter. It extracts Id from
          CgIdInfo passed to it in the same way addBindsC does.
          Previously this was done at every call site, which was
    • Jan Stolarek's avatar
      Trailing whitespaces, code formatting, detabify · 3f279f37
      Jan Stolarek authored
      A major cleanup of trailing whitespaces and tabs in codeGen/
      directory. I also adjusted code formatting in some places.
    • Gabor Greif's avatar
      Fix typo · f661e79c
      Gabor Greif authored
    • Gabor Greif's avatar
      Comments only · 9325b18f
      Gabor Greif authored
    • thoughtpolice's avatar
      Update Win32 submodule. · 21db8038
      thoughtpolice authored
      This fixes a build failure for i386/Windows.
      Signed-off-by: thoughtpolice's avatarAustin Seipp <aseipp@pobox.com>
  2. 19 Aug, 2013 13 commits
  3. 18 Aug, 2013 1 commit
  4. 16 Aug, 2013 3 commits
    • eir@cis.upenn.edu's avatar
      Fix Trac #8138. · ac382ab9
      eir@cis.upenn.edu authored
      The code in CoreLint did not take into account the possibility of
      ~R# arguments to functions. These can arise in argToPat in SpecConstr.
      Now, CoreLint does not fail when it sees a ~R# parameter.
      This commit also updates the core-spec document accordingly.
    • Jan Stolarek's avatar
      Comments only · 82d5aa03
      Jan Stolarek authored
      I restored part of documentation that describes what is a let-no-escape
      and which was deleted 10 months ago together with the old codegen. Then
      I removed lots of Literate Haskell clutter (like empty \begin{code} -
      \end{code} blocks) and finally decided to remove all the Literate Haskell
      markup because there wasn't much of it left, but it made comments so
      difficult to read.
    • Jan Stolarek's avatar
      Comments only · ec621f3c
      Jan Stolarek authored
  5. 15 Aug, 2013 1 commit
    • thoughtpolice's avatar
      Don't delete HsTimeConfig.h.in during distclean. · 94c35ddf
      thoughtpolice authored
      sdist preps the tree via distclean before anything else, which caused
      HsTimeConfig.h.in under 'time' to be deleted - even though it should be
      included in the resulting tarball for ./configure.
      The correct target is 'maintainer-clean'.
      I'm guessing the nightlies didn't complain because they run ./boot,
      forcing regeneration. NixOS's Hydra does not, though.
      Thanks to Peter Simons and Andres Löh for pointing this out.
      Signed-off-by: thoughtpolice's avatarAustin Seipp <aseipp@pobox.com>
  6. 14 Aug, 2013 5 commits
  7. 13 Aug, 2013 1 commit
  8. 12 Aug, 2013 1 commit
  9. 11 Aug, 2013 3 commits
  10. 10 Aug, 2013 1 commit
  11. 09 Aug, 2013 5 commits
  12. 06 Aug, 2013 1 commit
    • gmainland's avatar
      Only use real XMM registers when assigning arguments. · 3e598fe0
      gmainland authored
      My original change to the calling convention mistakenly used all 6 XMM
      registers---which live in the global register table---on x86 (32 bit). This
      royally screwed up the floating point code generated for that platform because
      floating point arguments were passed in global registers instead of on the