1. 20 Mar, 2011 1 commit
    • Ian Lynagh's avatar
      Change how we compute install paths on cygwin · b1fff745
      Ian Lynagh authored
      We used to have
          MK_INSTALL_DEST = "$(shell cygpath $1)"
      but this meant we ended up with
          "$(shell cygpath "[...]/html/`basename $$i`")"
      and the $(...) gets evaluated before the makefile rule, so the for loop
      hasn't been run, and so $i isn't defined. So we were taking the basename
      of the empty string, meaning docs weren't being installed in the right
      Now we have
          MK_INSTALL_DEST = $$(cygpath $1)
      so the evaluation happens in the shell, while the for loop is running.
  2. 17 Mar, 2011 1 commit
  3. 16 Mar, 2011 6 commits
  4. 15 Mar, 2011 1 commit
  5. 13 Mar, 2011 2 commits
  6. 11 Mar, 2011 1 commit
  7. 19 Feb, 2011 1 commit
  8. 27 Oct, 2010 1 commit
  9. 09 Mar, 2011 1 commit
  10. 07 Mar, 2011 1 commit
  11. 06 Mar, 2011 2 commits
  12. 01 Mar, 2011 2 commits
    • Sergei Trofimovich's avatar
      configure: amend sanity check · e45b7db2
      Sergei Trofimovich authored
      As we perform some mangling of original --build/--host/--target
      params we should check the result of mangling against desired triplet,
      not originally passed by user.
      Patch also adds mangled triplets to the output.
    • Sergei Trofimovich's avatar
      configure: triplet: accept i486 CPU (and more), more vendors and OSes · c00ea87f
      Sergei Trofimovich authored
      Gentoo has interesting ports:
        - freebsd with following triplet:
              ./configure --build=i686-gentoo-freebsd8 --host=i686-gentoo-freebsd8 --host=i686-gentoo-freebsd8
          should be recognized as 'i386-unknown-freebsd'
        - 'pc' vendor along with non-'i386' is very common:
              ./configure --build=i486-pc-linux-gnu --i486-pc-linux-gnu --host=i486-pc-linux-gnu
              ./configure --build=i686-pc-linux-gnu --i686-pc-linux-gnu --host=i686-pc-linux-gnu
          should be recognized as 'i386-unknown-linux'
      Patch adds:
        - 'pc' vendor                 (maps to 'unknown')
        - 'gentoo' vendor             (maps to 'unknown')
        - 'i486', 'i586', 'i686' CPUs (maps to 'i386')
        - 'freebsd8' OS               (maps to 'freebsd')
  13. 06 Mar, 2011 1 commit
  14. 05 Mar, 2011 5 commits
  15. 02 Mar, 2011 2 commits
  16. 22 Feb, 2011 1 commit
    • simonpj@microsoft.com's avatar
      Restore SPECIALISE INSTANCE pragmas · 0f409c99
      simonpj@microsoft.com authored
      For some reason I'd commented out this code, which 
      meant that we weren't getting the the goodness of
      the SPECIALISE INSTANCE pragamas in GHC.Real. And
      that in turn killed performance in nofib 'power'
  17. 01 Mar, 2011 1 commit
  18. 27 Feb, 2011 1 commit
  19. 26 Feb, 2011 1 commit
    • vivian's avatar
      :script file scripts in GHCi #1363 · eccb2d89
      vivian authored
      This patch adds the script command in GHCi
      A file is read and executed as a series of GHCi commands.
      Execution terminates on the first error.  The filename and
      line number are included in the error.
  20. 25 Feb, 2011 1 commit
  21. 08 Feb, 2011 1 commit
  22. 23 Feb, 2011 1 commit
  23. 22 Feb, 2011 1 commit
  24. 21 Feb, 2011 1 commit
    • simonpj@microsoft.com's avatar
      Fix another fundep error (fixes Trac #4969) · d1796b52
      simonpj@microsoft.com authored
      If I had a pound for every hour Dimitrios and I have spent
      making functional dependencies work right, we'd be rich!
      We had stupidly caused a 'wanted' to be rewritten by a 'derived', with
      resulting abject failure.  As well as fixing the bug, this patch
      refactors some more, adds useful assert and comments.
  25. 18 Feb, 2011 1 commit
    • Edward Z. Yang's avatar
      New codegen: GC calling convention must use registers. · 4c53d93a
      Edward Z. Yang authored
      Previously, on register-deficient architectures like x86-32,
      the new code generator would emit code for calls to stg_gc_l1,
      stg_gc_d1 and stg_gc_f1 that pushed their single argument on
      to the stack, while the functions themselves expected the
      argument to live in L1, D1 and F1 (respectively).  This was
      because cmmCall with the GC calling convention allocated real
      registers, not virtual registers.
      This patch modifies the code for assigning registers/stack slots
      to use the right calling convention for GC and adds an assertion
      to ensure it did it properly.
  26. 20 Feb, 2011 1 commit
    • chak@cse.unsw.edu.au.'s avatar
      Added a VECTORISE pragma · f2aaae97
      chak@cse.unsw.edu.au. authored
      - Added a pragma {-# VECTORISE var = exp #-} that prevents
        the vectoriser from vectorising the definition of 'var'.
        Instead it uses the binding '$v_var = exp' to vectorise
        'var'.  The vectoriser checks that the Core type of 'exp'
        matches the vectorised Core type of 'var'.  (It would be
        quite complicated to perform that check in the type checker
        as the vectorisation of a type needs the state of the VM
      - Added parts of a related VECTORISE SCALAR pragma
      - Documented -ddump-vect
      - Added -ddump-vt-trace
      - Some clean up
  27. 19 Feb, 2011 1 commit