GitLab

Previously a Bad node was returned which could hide problems and more
importantly you create new nodes (the BadBB) in an operation that apparently
only queries the graph.

Nobody uses it currently, and it is a burden for everyone writing a new
pass.

It has been empty for nearly all systems. People who used to put stuff
in config.h can still create a config.h on their own and inject
-include (gcc) or /FI (msvc) into their CPPFLAGS.

Just state the copyright and the project the files belong to. You can
then find the COPYING file for the whole project.

This is a workaround for keepalive edges which are probably broken
(adding the stuff to the End node is wrong, it should get added to the
 endless loop somehow)

The opt_manage framework didn't really do much except assuring that some
analysis info is computed before a pass starts. We can achieve this
simpler by providing an assure_irg_properties function.
This commit also fixes some places where analysis info was potentially
not invalidated before.

graph_properties are things that a transformation may destroy but which
can be easily recalculated.

graph_constraints are set after lowering phases have been running and
forbid certain constructs that no (later) transformation is allowed to construct

some were missing anyway and nobody is using the global declarations anyway

Converted: ifconv, boolopt, combo, fpvrp, ldst, code placement,
jumpthreading, deconv, parallelize mem, loop, scalar replacement,
cfopt.

GCSEing nodes into unreachable blocks is ok now. The
verifier does not check the cfg, when irg is floating.
Code placement fixes and pins the irg after gcse.
Nodes in unreachable blocks are moved into the first block
after the start block.

Reachability checks in place_late are converted into asserts,
because after place_early, there are no unreachable blocks anymore.

reverts c64fe299 and 318579c4
but also fixes opt/unreachable_block

See http://www.libfirm.org/Unreachable_Code for details

Revert "merge gcse into place_code (a better solution would be having it always on until the backend scheduler)"

This reverts commit bdddae273a9cd497dc79c2f90726482620d3c8d4.

merge gcse into place_code (a better solution would be having it always on until the backend scheduler)

[r28127]

[r28041]

[r27833]

[r27277]

[r26996]

  (with different arguments and register constraints for the same node now)
- Lots of cleanups/changes in the arm backend. We can represent all
  "shifter operands" now (but the code selection isn't optimal yet)
- More fixes all over the place - arm backend handles 164.gzip now

[r26673]

- add an irg_verify_pass()

[r26374]

- add more passes

[r26366]

- more passes implemented
- small fixes

[r26356]

- maybe we should initially mark labeled blocks as non-removable

[r24451]

[r22889]

[r22754]