1. 19 Jun, 2015 2 commits
  2. 24 May, 2015 1 commit
    • Matthias Braun's avatar
      be: Move main backend loop into backends. · 3e79d08e
      Matthias Braun authored
      Instead of having a generaic main loop in be_main with several callbacks
      into the targets, split the code in be_main up into a set of convenience
      functions and let the targets implement the main loop itself. This is
      nicer because:
      
      - The control flow into and out of the functions in the target becomes
        obvious as no external party is using callbacks anymore.
      - Targets can insert custom code at any place between the generic
        passes.
      - In the future we can parameterize the generic backend passes directly
        instead of stuffing info into generic isa_if structures.
      3e79d08e
  3. 18 Jan, 2015 4 commits
  4. 31 Dec, 2014 1 commit
  5. 29 Dec, 2014 2 commits
  6. 25 Dec, 2014 1 commit
  7. 25 Nov, 2014 2 commits
  8. 22 Nov, 2014 1 commit
  9. 13 Nov, 2014 2 commits
  10. 22 Oct, 2014 1 commit
  11. 18 Oct, 2014 2 commits
  12. 07 Oct, 2014 1 commit
    • Matthias Braun's avatar
      API change+fix for IncSP alignment · b719d82d
      Matthias Braun authored
      - alignment is set on each single IncSP now instead of a global arch_env
        setting
      - fixed ia32 not requesting any alignment anymore (even on darwin where
        we need it)
      - fixed amd64 creating IncSP nodes without flag_modifies_flags set
      b719d82d
  13. 02 Oct, 2014 1 commit
    • Matthias Braun's avatar
      cleanup/normalize be_add_missing_keeps usage · 6e9dd15b
      Matthias Braun authored
      We now add missing keeps exactly before register allocation where we
      actually need them (so that unused outputs which still need registers
      assigned are not missed by the register allocation).
      6e9dd15b
  14. 30 Sep, 2014 4 commits
  15. 19 Sep, 2014 1 commit
  16. 17 Sep, 2014 1 commit
    • Matthias Braun's avatar
      put irg link directly into node attributs · f3209aa8
      Matthias Braun authored
      The previous approach put the irg link into the block attributes, so
      while saving memory an extra indirection was necessary. Some testing
      with valgrind massif+callgrind revealed that a direct link increses peak
      memory usage by ~1% while also increasing runtime by ~1%. More
      importantly the code gets a bit simpler and some strange corner cases can be
      removed with the direct link.
      f3209aa8
  17. 15 Sep, 2014 1 commit
  18. 08 Aug, 2014 1 commit
  19. 03 Jun, 2014 1 commit
  20. 02 Jun, 2014 2 commits
  21. 30 May, 2014 3 commits
  22. 28 May, 2014 5 commits