| 2008-04-27 |
Michael Beck | - when creating a x87 memory constant, check if the...
|
commit | commitdiff | tree |
| 2008-04-27 |
Michael Beck | - added fc_can_lossless_conv_to() / tarval_ieee754_can_conv_...
|
commit | commitdiff | tree |
| 2008-04-26 |
Michael Beck | allow Cmp, Neg, Abs for floating point even if no_float...
|
commit | commitdiff | tree |
| 2008-04-26 |
Michael Beck | fixed warning
|
commit | commitdiff | tree |
| 2008-04-26 |
Michael Beck | - BugFix:
|
commit | commitdiff | tree |
| 2008-04-21 |
Michael Beck | - more SSE constants handled by instructions
|
commit | commitdiff | tree |
| 2008-04-20 |
Michael Beck | - moved the imul mem,imm32 splitting into peephole...
|
commit | commitdiff | tree |
| 2008-04-20 |
Michael Beck | typo fixed
|
commit | commitdiff | tree |
| 2008-04-20 |
Michael Beck | more SSE constants handled without float_entity
|
commit | commitdiff | tree |
| 2008-04-19 |
Michael Beck | - add optimisation for size
|
commit | commitdiff | tree |
| 2008-04-19 |
Michael Beck | - don't dump alignment 0
|
commit | commitdiff | tree |
| 2008-04-19 |
Michael Beck | - added alignment_label_max_skip
|
commit | commitdiff | tree |
| 2008-04-19 |
Michael Beck | fix r19298: offsets must be fixed for PopMem
|
commit | commitdiff | tree |
| 2008-04-19 |
Michael Beck | syntay error fixed
|
commit | commitdiff | tree |
| 2008-04-19 |
Michael Beck | - handle failure of getpwuid() more gracefully
|
commit | commitdiff | tree |
| 2008-04-19 |
Michael Beck | always align blocks if there is no fall-through
|
commit | commitdiff | tree |
| 2008-04-19 |
Michael Beck | reorganized cpu's, architectures, and features, hopefully...
|
commit | commitdiff | tree |
| 2008-04-19 |
Michael Beck | fixed indentation
|
commit | commitdiff | tree |
| 2008-04-19 |
Michael Beck | do not pad return after a fall-through Jmp
|
commit | commitdiff | tree |
| 2008-04-19 |
Michael Beck | [r19329]
|
commit | commitdiff | tree |
| 2008-04-18 |
Michael Beck | load store opt fail with partially overwritten data
|
commit | commitdiff | tree |
| 2008-04-15 |
Michael Beck | - add barcelona and generic32 targets
|
commit | commitdiff | tree |
| 2008-04-15 |
Michael Beck | - do not use imul mem, imm32 on newer AMD cpu's
|
commit | commitdiff | tree |
| 2008-04-15 |
Michael Beck | - added add esp,8 -> pop reg, pop reg
|
commit | commitdiff | tree |
| 2008-04-15 |
Michael Beck | - fixed costs for Opteron and K10
|
commit | commitdiff | tree |
| 2008-04-14 |
Michael Beck | Linux gprof need base pointer
|
commit | commitdiff | tree |
| 2008-04-14 |
Michael Beck | add Linux grof support
|
commit | commitdiff | tree |
| 2008-04-14 |
Michael Beck | removed incomplete code
|
commit | commitdiff | tree |
| 2008-04-14 |
Michael Beck | - get_irg_initial_exec()/set_irg_initial_exec() added
|
commit | commitdiff | tree |
| 2008-04-13 |
Michael Beck | - removed obstack from be_main_env_t, it was only used...
|
commit | commitdiff | tree |
| 2008-04-13 |
Michael Beck | C99 features removed
|
commit | commitdiff | tree |
| 2008-04-12 |
Michael Beck | revert to old cost model, seems to work better
|
commit | commitdiff | tree |
| 2008-04-12 |
Michael Beck | typo fixed
|
commit | commitdiff | tree |
| 2008-04-12 |
Michael Beck | gcc has same costs for prescott and nocona
|
commit | commitdiff | tree |
| 2008-04-12 |
Michael Beck | updated timings
|
commit | commitdiff | tree |
| 2008-04-11 |
Michael Beck | updated cost model from gcc 4.3.0
|
commit | commitdiff | tree |
| 2008-04-11 |
Michael Beck | nocona is a Pentium 4 variant
|
commit | commitdiff | tree |
| 2008-04-11 |
Michael Beck | limited support for inlining K&R functions added
|
commit | commitdiff | tree |
| 2008-04-10 |
Michael Beck | disable inline of functions which call alloca(), causes...
|
commit | commitdiff | tree |
| 2008-04-10 |
Michael Beck | - BugFix: Do not optimize dead blocks: They might even...
|
commit | commitdiff | tree |
| 2008-04-10 |
Michael Beck | - replaced brain-damaged recursive version of get_cf_op...
|
commit | commitdiff | tree |
| 2008-04-10 |
Michael Beck | - inline_method() now sets current_ir_graph itself
|
commit | commitdiff | tree |
| 2008-04-10 |
Michael Beck | BugFix: - there is only ONE proj_m to replace
|
commit | commitdiff | tree |
| 2008-04-07 |
Michael Beck | issue a debug output when inlining
|
commit | commitdiff | tree |
| 2008-04-07 |
Michael Beck | - BugFix: SSA construction requires current block set ...
|
commit | commitdiff | tree |
| 2008-04-07 |
Michael Beck | fixed unsaved checkin ...
|
commit | commitdiff | tree |
| 2008-04-07 |
Michael Beck | - optimize Tuple nodes away before calling scalar replacement
|
commit | commitdiff | tree |
| 2008-04-07 |
Michael Beck | - Do not set phase_low here. currently ssa-construction...
|
commit | commitdiff | tree |
| 2008-04-07 |
Michael Beck | - uses new SSA constructor interface, shortening and...
|
commit | commitdiff | tree |
| 2008-04-05 |
Michael Beck | updated
|
commit | commitdiff | tree |
| 2008-04-05 |
Michael Beck | completely removed old verbosity based debug output
|
commit | commitdiff | tree |
| 2008-04-05 |
Michael Beck | switch scalar replacement to new style debug output
|
commit | commitdiff | tree |
| 2008-04-05 |
Michael Beck | after inlining, the loop-info is inconsistent
|
commit | commitdiff | tree |
| 2008-04-05 |
Michael Beck | - init function for inlining added
|
commit | commitdiff | tree |
| 2008-04-04 |
Michael Beck | Improved inliner: rerun scalar_replacement() if addresses...
|
commit | commitdiff | tree |
| 2008-04-04 |
Michael Beck | Improved param weights: handle Id and Tuple nodes
|
commit | commitdiff | tree |
| 2008-04-04 |
Michael Beck | BugFix:
|
commit | commitdiff | tree |
| 2008-04-04 |
Michael Beck | BugFix: store the weigths to the right location
|
commit | commitdiff | tree |
| 2008-04-04 |
Michael Beck | fixed obstack initialization, removed unused variable
|
commit | commitdiff | tree |
| 2008-04-04 |
Michael Beck | - only local visible methods get the big benefice
|
commit | commitdiff | tree |
| 2008-04-04 |
Michael Beck | fixed indentation, deleted useless comment
|
commit | commitdiff | tree |
| 2008-04-04 |
Michael Beck | calculate benefice on real callee
|
commit | commitdiff | tree |
| 2008-04-04 |
Michael Beck | no bonus for recursive inlining
|
commit | commitdiff | tree |
| 2008-04-04 |
Michael Beck | fixed warnings
|
commit | commitdiff | tree |
| 2008-04-04 |
Michael Beck | new benefice based heuristic inliner added
|
commit | commitdiff | tree |
| 2008-04-04 |
Michael Beck | - improced param weight calculation
|
commit | commitdiff | tree |
| 2008-04-04 |
Michael Beck | cosmetic
|
commit | commitdiff | tree |
| 2008-04-04 |
Michael Beck | Moved inlining and DCE (which is implemented like inlinung...
|
commit | commitdiff | tree |
| 2008-04-04 |
Michael Beck | - code cleanup
|
commit | commitdiff | tree |
| 2008-04-04 |
Michael Beck | C99 features removed
|
commit | commitdiff | tree |
| 2008-04-04 |
Michael Beck | handle the macro block properly
|
commit | commitdiff | tree |
| 2008-04-04 |
Michael Beck | some cosmetic
|
commit | commitdiff | tree |
| 2008-04-03 |
Michael Beck | improved Cmp(x, 0) optimizations
|
commit | commitdiff | tree |
| 2008-04-02 |
Michael Beck | switch off usage of global_pointer again: globals are...
|
commit | commitdiff | tree |
| 2008-04-02 |
Michael Beck | add list_head typedef
|
commit | commitdiff | tree |
| 2008-04-02 |
Michael Beck | used new is_Global() and get_Global_entity() instead...
|
commit | commitdiff | tree |
| 2008-04-02 |
Michael Beck | add is_Global() and get_Global_ent()
|
commit | commitdiff | tree |
| 2008-04-02 |
Michael Beck | C99 features removed
|
commit | commitdiff | tree |
| 2008-04-02 |
Michael Beck | some comments added and fixed
|
commit | commitdiff | tree |
| 2008-04-02 |
Michael Beck | C99 features removed
|
commit | commitdiff | tree |
| 2008-04-02 |
Michael Beck | fixed Win32 build
|
commit | commitdiff | tree |
| 2008-04-02 |
Michael Beck | skip Bitfield Sels when doing memory disambiguation
|
commit | commitdiff | tree |
| 2008-04-01 |
Michael Beck | renamed functions to clarify its usage
|
commit | commitdiff | tree |
| 2008-03-20 |
Michael Beck | testcase for new Confirm code
|
commit | commitdiff | tree |
| 2008-03-20 |
Michael Beck | small improvements:
|
commit | commitdiff | tree |
| 2008-03-20 |
Michael Beck | fixed comment
|
commit | commitdiff | tree |
| 2008-03-20 |
Michael Beck | when propagating constants in the x == Const case,...
|
commit | commitdiff | tree |
| 2008-03-20 |
Michael Beck | is_Abs() added
|
commit | commitdiff | tree |
| 2008-03-19 |
Michael Beck | BugFix: scalar replacement should not remove volatile...
|
commit | commitdiff | tree |
| 2008-03-12 |
Michael Beck | use .word instead of .value, older gas don't support it
|
commit | commitdiff | tree |
| 2008-03-12 |
Michael Beck | need suffixes for mvf, mnf
|
commit | commitdiff | tree |
| 2008-03-11 |
Michael Beck | set the isa right
|
commit | commitdiff | tree |
| 2008-03-11 |
Michael Beck | instructions with immediates are not commutative
|
commit | commitdiff | tree |
| 2008-03-11 |
Michael Beck | refactored immediates:
|
commit | commitdiff | tree |
| 2008-03-11 |
Michael Beck | refactored immediates:
|
commit | commitdiff | tree |
| 2008-03-11 |
Michael Beck | fixed CondJmp attributes
|
commit | commitdiff | tree |
| 2008-03-11 |
Michael Beck | set the global isa object
|
commit | commitdiff | tree |
| 2008-03-11 |
Michael Beck | use assure_irg_outs()
|
commit | commitdiff | tree |
| 2008-03-11 |
Michael Beck | updated be_AddSP semantics
|
commit | commitdiff | tree |
| 2008-03-11 |
Michael Beck | reset the layout to undefined before new members are...
|
commit | commitdiff | tree |
| next |
projects / libfirm / search