2008-06-03 |
Matthias Braun | switch projs have long pnc not pn_Cmp |
tree | commitdiff |
2008-06-03 |
Matthias Braun | we do need conv/cast skipping |
tree | commitdiff |
2008-06-02 |
Matthias Braun | handle Bads more gracefully |
tree | commitdiff |
2008-06-02 |
Michael Beck | add missing condition |
tree | commitdiff |
2008-06-02 |
Matthias Braun | only skip float->float convs |
tree | commitdiff |
2008-06-02 |
Michael Beck | - added option -b omitleaffp, omitting the frame pointe... |
tree | commitdiff |
2008-06-02 |
Michael Beck | - better handling for strictConv's, optimized conv_orgi... |
tree | commitdiff |
2008-06-02 |
Michael Beck | test for strictConv orgie |
tree | commitdiff |
2008-06-02 |
Michael Beck | - Fixed strict-convs before Stores again: We can skip... |
tree | commitdiff |
2008-06-02 |
Matthias Braun | add missing returns |
tree | commitdiff |
2008-06-01 |
Michael Beck | - stack alignment is now power of two |
tree | commitdiff |
2008-06-01 |
Michael Beck | Do NOT ignore Bad Phi predecessors. Otherwise |
tree | commitdiff |
2008-06-01 |
Michael Beck | call optimize_cf() to delete dead-code that might arise... |
tree | commitdiff |
2008-05-31 |
Matthias Braun | use ir_type from call instead of the type from the... |
tree | commitdiff |
2008-05-31 |
Matthias Braun | outputting +,-,* in assembler makes no sense for floats |
tree | commitdiff |
2008-05-31 |
Matthias Braun | fix alloca mapper |
tree | commitdiff |
2008-05-31 |
Matthias Braun | Alloc is not a binop |
tree | commitdiff |
2008-05-31 |
Michael Beck | updated for latest libFirm changes |
tree | commitdiff |
2008-05-31 |
Matthias Braun | no, we cannot skip all convs before vfst (but just... |
tree | commitdiff |
2008-05-31 |
Michael Beck | - in x87 mode, skipp ALL conv's before stores |
tree | commitdiff |
2008-05-31 |
Michael Beck | allow x + 0.0 = x even in strict mode |
tree | commitdiff |
2008-05-30 |
Michael Beck | renamed modecode, mode_sort, mode_arithmetic to ir_* |
tree | commitdiff |
2008-05-30 |
Michael Beck | inline_method(): |
tree | commitdiff |
2008-05-30 |
Michael Beck | - removed assertion() it is allowed to ask for non... |
tree | commitdiff |
2008-05-30 |
Michael Beck | handle a call with less arguments than needed gracefull... |
tree | commitdiff |
2008-05-30 |
Michael Beck | - no more be_init(): backend is initialised in init_firm |
tree | commitdiff |
2008-05-30 |
Michael Beck | - no more be_init(): backend is initialised in init_firm |
tree | commitdiff |
2008-05-30 |
Michael Beck | add extra info for immediate float mode |
tree | commitdiff |
2008-05-30 |
Matthias Braun | cleanup space generation logic and make it more robust... |
tree | commitdiff |
2008-05-29 |
Michael Beck | fixed warning |
tree | commitdiff |
2008-05-29 |
Michael Beck | - Fixed output for IncSP nodes |
tree | commitdiff |
2008-05-29 |
Michael Beck | - Add a peephole phase that fixes all stack offsets |
tree | commitdiff |
2008-05-29 |
Michael Beck | - Emitter for IncSP emits now Add or Sub |
tree | commitdiff |
2008-05-29 |
Michael Beck | renamed SubSP to SubSPandCopy, add some constants |
tree | commitdiff |
2008-05-28 |
Michael Beck | - moved peephole_IncSP_IncSP() to bepeephole.c, as... |
tree | commitdiff |
2008-05-28 |
Michael Beck | simplify code |
tree | commitdiff |
2008-05-28 |
Michael Beck | fixed warning |
tree | commitdiff |
2008-05-28 |
Michael Beck | better version of the barrier remover (no warnings) |
tree | commitdiff |
2008-05-28 |
Michael Beck | remove barrier from schedule before peephole optimization |
tree | commitdiff |
2008-05-28 |
Michael Beck | implemened add esp, 4/8 -> push/push |
tree | commitdiff |
2008-05-28 |
Michael Beck | fixed condition for add/sub esp, 4/8 |
tree | commitdiff |
2008-05-28 |
Michael Beck | bock marker can be dumped in the block title |
tree | commitdiff |
2008-05-28 |
Michael Beck | - dump the block mark |
tree | commitdiff |
2008-05-28 |
Michael Beck | - added is_Id, is_CallBegin, is_Free |
tree | commitdiff |
2008-05-28 |
Michael Beck | - added is_Filter() |
tree | commitdiff |
2008-05-27 |
Michael Beck | Phi's are now transmitted as a Phi list, not using... |
tree | commitdiff |
2008-05-27 |
Michael Beck | BugFix: phi node might be exchanged, read the next... |
tree | commitdiff |
2008-05-27 |
Michael Beck | - BugFix: set the mark on the right block |
tree | commitdiff |
2008-05-27 |
Michael Beck | add missing non-inline versions |
tree | commitdiff |
2008-05-27 |
Michael Beck | BugFix: fixed Phi-list manipulation |
tree | commitdiff |
2008-05-27 |
Michael Beck | redo r19786, peculiarity has been fixed in the cparser FE |
tree | commitdiff |
2008-05-27 |
Matthias Braun | need includes for alloca |
tree | commitdiff |
2008-05-27 |
Matthias Braun | remove a too strict check, we can have local function... |
tree | commitdiff |
2008-05-27 |
Matthias Braun | make verifier a bit less strict about keeps |
tree | commitdiff |
2008-05-27 |
Matthias Braun | do an upconv instead of panic when we do <32bit right... |
tree | commitdiff |
2008-05-27 |
Matthias Braun | build a keep behind each pop to avoid double register... |
tree | commitdiff |
2008-05-26 |
Michael Beck | add some comments, small cosmetic |
tree | commitdiff |
2008-05-26 |
Michael Beck | small BugFix |
tree | commitdiff |
2008-05-26 |
Michael Beck | must use Lg, not Ne to check for != |
tree | commitdiff |
2008-05-26 |
Michael Beck | use block Phi lists and Block marks to get rid of block... |
tree | commitdiff |
2008-05-26 |
Michael Beck | add *_Block_mark() api to mark easily Blocks in the... |
tree | commitdiff |
2008-05-26 |
Michael Beck | used new add_Block_phi() |
tree | commitdiff |
2008-05-26 |
Michael Beck | typo fixed |
tree | commitdiff |
2008-05-26 |
Michael Beck | add_Block_phi() implemented |
tree | commitdiff |
2008-05-26 |
Michael Beck | C99 features removed |
tree | commitdiff |
2008-05-25 |
Matthias Braun | fix a few warnings |
tree | commitdiff |
2008-05-25 |
Matthias Braun | commit the beginning of advanced boolopt (had that... |
tree | commitdiff |
2008-05-25 |
Michael Beck | Special handling seems not to work, remove critical... |
tree | commitdiff |
2008-05-25 |
Michael Beck | get rid of most Optimisation warning: found add/sub... |
tree | commitdiff |
2008-05-25 |
Michael Beck | add Christoph's loop test |
tree | commitdiff |
2008-05-25 |
Michael Beck | - simplified get_effective_use_block() |
tree | commitdiff |
2008-05-25 |
Michael Beck | - do not optimize away Confirms with Constants |
tree | commitdiff |
2008-05-25 |
Michael Beck | - can now handle const-like Confirms |
tree | commitdiff |
2008-05-25 |
Michael Beck | improved test: added new Confirm/cond_eval combination |
tree | commitdiff |
2008-05-24 |
Michael Beck | arch_env_* macros updated |
tree | commitdiff |
2008-05-24 |
Michael Beck | allocate all op attributes on block instead of a lot... |
tree | commitdiff |
2008-05-24 |
Michael Beck | - no need to add two custom pointers: the tag is now... |
tree | commitdiff |
2008-05-24 |
Christoph Mallon | When testing whether generating a Psi is allowed, check... |
tree | commitdiff |
2008-05-24 |
Matthias Braun | normalize/fix ProjM numbers, avoid dead code in some... |
tree | commitdiff |
2008-05-24 |
Matthias Braun | we can have mode_b inputs to Convs as well (I hope... |
tree | commitdiff |
2008-05-23 |
Michael Beck | small improvements: cache different environment members... |
tree | commitdiff |
2008-05-23 |
Michael Beck | - not really necessary, but makes the array size again... |
tree | commitdiff |
2008-05-23 |
Michael Beck | - removed C99 features |
tree | commitdiff |
2008-05-23 |
Michael Beck | BugFix: wrong condition for Abs (this fixes Queens.c) |
tree | commitdiff |
2008-05-23 |
Matthias Braun | - Aus arch_isa_t wird arch_env_t sonst ändert sich... |
tree | commitdiff |
2008-05-23 |
Matthias Braun | we don't need no stinking selfs |
tree | commitdiff |
2008-05-23 |
Matthias Braun | - arch_get_irn_ops simplified |
tree | commitdiff |
2008-05-23 |
Michael Beck | Removed the irn_handler stack. |
tree | commitdiff |
2008-05-23 |
Michael Beck | Removed the arch_irn_handler_t. This was just an additi... |
tree | commitdiff |
2008-05-22 |
Michael Beck | check for Cmp nodes before doing doz |
tree | commitdiff |
2008-05-22 |
Michael Beck | - normalize Muxes a bit: this hopefully improves the... |
tree | commitdiff |
2008-05-22 |
Michael Beck | use is_Const() |
tree | commitdiff |
2008-05-22 |
Michael Beck | missing goto added |
tree | commitdiff |
2008-05-22 |
Michael Beck | - add more Mux optimization cases |
tree | commitdiff |
2008-05-21 |
Michael Beck | fixed warnings |
tree | commitdiff |
2008-05-21 |
Michael Beck | - placed phi_handler into the be_main environment,... |
tree | commitdiff |
2008-05-21 |
Michael Beck | - BugFix: replaced wrong cast by right code (no crash... |
tree | commitdiff |
2008-05-21 |
Michael Beck | typos fixed |
tree | commitdiff |
2008-05-21 |
Michael Beck | - fixed doxygen comments |
tree | commitdiff |
2008-05-21 |
Michael Beck | - beo_* Opcodes are now defined in irop.h, and no more... |
tree | commitdiff |
next |