2009-01-20 |
Michael Beck | - say debug message BEFORE the irg is killed |
tree | commitdiff |
2009-01-20 |
Michael Beck | - check for entities on the frame only (no more value_p... |
tree | commitdiff |
2009-01-20 |
Michael Beck | - set the entity state to NOT computed after inlining |
tree | commitdiff |
2009-01-20 |
Michael Beck | - undo erroneous checkin |
tree | commitdiff |
2009-01-20 |
Michael Beck | - fixed handling of outer frame access |
tree | commitdiff |
2009-01-20 |
Michael Beck | - teach the analyzer how to handle inner functions |
tree | commitdiff |
2009-01-19 |
Michael Beck | - do not kick inner functions from the frame type |
tree | commitdiff |
2009-01-19 |
Michael Beck | Get rid of pn_Start_value_param: |
tree | commitdiff |
2009-01-18 |
Michael Beck | - preliminary steps to remove the value_param Proj |
tree | commitdiff |
2009-01-18 |
Michael Beck | - fixed typos |
tree | commitdiff |
2009-01-14 |
Michael Beck | - add support for statistics and merge debug info |
tree | commitdiff |
2009-01-14 |
Michael Beck | - fixed signed/unsigned warning |
tree | commitdiff |
2009-01-03 |
Michael Beck | - BugFix: not all users of the frame are Sel nodes... |
tree | commitdiff |
2009-01-02 |
Michael Beck | - BugFix: not all users of the frame are Sel nodes |
tree | commitdiff |
2008-12-19 |
Michael Beck | - fixed CSE and Combo for Builtin nodes |
tree | commitdiff |
2008-12-15 |
Michael Beck | - fixed warning |
tree | commitdiff |
2008-12-14 |
Moritz Kroll | Added cons_flags parameter to new_Load and new_Store... |
tree | commitdiff |
2008-12-13 |
Michael Beck | - BugFix: the mode of a Load must be its load_mode... |
tree | commitdiff |
2008-12-13 |
Matthias Braun | fix new_Block users |
tree | commitdiff |
2008-12-12 |
Michael Beck | - do not remove keeps for now, only Bads, NoMem and... |
tree | commitdiff |
2008-12-09 |
Michael Beck | - fixed assertion |
tree | commitdiff |
2008-12-09 |
Michael Beck | - do not move the return out of labeled blocks, this... |
tree | commitdiff |
2008-12-09 |
Michael Beck | - do not optimize away labeled blocks |
tree | commitdiff |
2008-12-08 |
Christoph Mallon | Before deconving a node, make sure it is a downconv. |
tree | commitdiff |
2008-11-28 |
Matthias Braun | remove archop stuff which wasn't really used anyway |
tree | commitdiff |
2008-11-27 |
Moritz Kroll | Removed mode parameter from Const and Const_type constr... |
tree | commitdiff |
2008-11-25 |
Christoph Mallon | Retain debug info in normalize_one_return(), if all... |
tree | commitdiff |
2008-11-25 |
Michael Beck | - handle strictConv and remainderlessDiv for in opcode map |
tree | commitdiff |
2008-11-24 |
Christoph Mallon | Retain debug info in normalize_n_returns(). |
tree | commitdiff |
2008-11-18 |
Christoph Mallon | s/new_r_Const(current_ir_graph, /new_Const(/. |
tree | commitdiff |
2008-11-18 |
Moritz Kroll | Removed unused block parameter from Const constructors |
tree | commitdiff |
2008-11-14 |
Matthias Braun | rename mangling stuff to avoid name clashes |
tree | commitdiff |
2008-11-11 |
Michael Beck | - do not read the link of an exchanged node (although it |
tree | commitdiff |
2008-11-11 |
Michael Beck | - cosmetic changes |
tree | commitdiff |
2008-11-11 |
Michael Beck | - keep ALL mode_M nodes that have only unreachable... |
tree | commitdiff |
2008-11-07 |
Michael Beck | - fixed opcode(): Const and SymConst can happen through... |
tree | commitdiff |
2008-11-06 |
Michael Beck | - BugFix: cmp_nodes gets an ir_node ** |
tree | commitdiff |
2008-11-06 |
Michael Beck | - generalized end-block melting into generic block... |
tree | commitdiff |
2008-11-05 |
Michael Beck | - refactored code for generic block shaping |
tree | commitdiff |
2008-11-05 |
Michael Beck | - BugFix: remove critical edges to ensure that Phi... |
tree | commitdiff |
2008-10-31 |
Michael Beck | - calling normalize_n_returns() improve greatly the... |
tree | commitdiff |
2008-10-31 |
Michael Beck | - fixed allocation of resources even more |
tree | commitdiff |
2008-10-31 |
Michael Beck | - fixed allocation of resources |
tree | commitdiff |
2008-10-31 |
Sebastian Buchwald | - reserve necessary resources before collect phiprojs |
tree | commitdiff |
2008-10-30 |
Michael Beck | - reserve the IR_RESOURCE_PHI_LIST resource |
tree | commitdiff |
2008-10-30 |
Michael Beck | - invalidate tr_outs because Calls might be removed |
tree | commitdiff |
2008-10-30 |
Michael Beck | - do not treat Constants as live-ins: this assures... |
tree | commitdiff |
2008-10-30 |
Michael Beck | - BugFix: kill partitions with 0 blocks either |
tree | commitdiff |
2008-10-28 |
Michael Beck | deactivate debug output |
tree | commitdiff |
2008-10-28 |
Michael Beck | - implemented apply phase |
tree | commitdiff |
2008-10-28 |
Michael Beck | - small cleanup |
tree | commitdiff |
2008-10-28 |
Michael Beck | - analysis part of an block merging optimization |
tree | commitdiff |
2008-10-27 |
Michael Beck | - more irp_resource_reserved() |
tree | commitdiff |
2008-10-27 |
Michael Beck | - BugFix: ensure that Convs are created in the right... |
tree | commitdiff |
2008-10-25 |
Michael Beck | - removed useless checking of control flow input for... |
tree | commitdiff |
2008-10-24 |
Michael Beck | - handling of Top mode can be switched on/off at runtim... |
tree | commitdiff |
2008-10-23 |
Michael Beck | - better solution for the "mode_M node computes Top... |
tree | commitdiff |
2008-10-23 |
Michael Beck | - more special Top handling: it is NOT allowed to repla... |
tree | commitdiff |
2008-10-23 |
Michael Beck | - Call nodes computes always bottom, even if they have... |
tree | commitdiff |
2008-10-23 |
Michael Beck | - more work on the Unknown problems: the only |
tree | commitdiff |
2008-10-23 |
Michael Beck | - fixed r23120 |
tree | commitdiff |
2008-10-22 |
Michael Beck | - put verify_type() again after the "type changed"... |
tree | commitdiff |
2008-10-22 |
Michael Beck | - kill keep-alive edges to dead blocks before doing... |
tree | commitdiff |
2008-10-22 |
Michael Beck | - BugFix: fixed monotony checker, now enabled |
tree | commitdiff |
2008-10-21 |
Michael Beck | - activated WITH_UNKNOWN |
tree | commitdiff |
2008-10-20 |
Michael Beck | - BugFix: we need an extra compute_Return(), else compu... |
tree | commitdiff |
2008-10-20 |
Michael Beck | - comaparation of x and Top produces True, as Top is... |
tree | commitdiff |
2008-10-20 |
Michael Beck | - typos fixed |
tree | commitdiff |
2008-10-18 |
Christoph Mallon | Add ALLOCAN() and ALLOCANZ(). |
tree | commitdiff |
2008-10-17 |
Michael Beck | - add comment explaining the connection betwenn the... |
tree | commitdiff |
2008-10-17 |
Michael Beck | BugFix for a rare case: |
tree | commitdiff |
2008-10-14 |
Christoph Mallon | s/\<\(LC_\)\?INLINE\>/inline/. |
tree | commitdiff |
2008-10-11 |
Matthias Braun | remove #ifdef HAVE_CONFIG_Hs |
tree | commitdiff |
2008-10-11 |
Matthias Braun | hooks are always enabled now -> one less option |
tree | commitdiff |
2008-10-11 |
Matthias Braun | add more missing config.h includes |
tree | commitdiff |
2008-10-10 |
Michael Beck | typo |
tree | commitdiff |
2008-10-09 |
Michael Beck | BugFix: works again for RAW with non twos-complement |
tree | commitdiff |
2008-10-08 |
Michael Beck | - BugFix: fix the rare case where two nodes a congruent... |
tree | commitdiff |
2008-10-08 |
Michael Beck | - fixed some comment typos |
tree | commitdiff |
2008-10-08 |
Michael Beck | - Add support for commutative congruences: Add(a,b... |
tree | commitdiff |
2008-10-07 |
Michael Beck | - fixed strange code: when did it break? |
tree | commitdiff |
2008-10-07 |
Michael Beck | - kill Bad nodes from the keep-alive list |
tree | commitdiff |
2008-10-07 |
Michael Beck | - used new irg_walk_blkwise_dom_top_down() to calculate... |
tree | commitdiff |
2008-10-07 |
Michael Beck | - kill Bad nodes from the keep-alive list: these arise... |
tree | commitdiff |
2008-10-07 |
Christoph Mallon | Add irn_visited_else_mark(), which combines irn_visited... |
tree | commitdiff |
2008-10-06 |
Matthias Braun | fix last fix (+ optimisation: don't do anything if... |
tree | commitdiff |
2008-10-06 |
Matthias Braun | fix condeval ssa recondstruction |
tree | commitdiff |
2008-10-06 |
Christoph Mallon | Typo in comment. |
tree | commitdiff |
2008-10-06 |
Christoph Mallon | s/irn_not_visited(x)/!irn_visited(x)/. |
tree | commitdiff |
2008-10-05 |
Christoph Mallon | Good day and welcome to the FIRM XMALLOC*() macros... |
tree | commitdiff |
2008-10-05 |
Moritz Kroll | Let some optimizations return non-zero, if they changed... |
tree | commitdiff |
2008-10-04 |
Christoph Mallon | Treat conversions between the same size as downconv... |
tree | commitdiff |
2008-09-30 |
Michael Beck | - BugFix: cleanup cannot work with a precalculated... |
tree | commitdiff |
2008-09-30 |
Sebastian Buchwald | Don't panic, the train is waiting for you... |
tree | commitdiff |
2008-09-30 |
Moritz Kroll | Optimize store after stores with partial overwrites... |
tree | commitdiff |
2008-09-30 |
Matthias Braun | old version was correct... |
tree | commitdiff |
2008-09-30 |
Matthias Braun | when we can do it for tuples, we can do it for Ids too |
tree | commitdiff |
2008-09-29 |
Michael Beck | - add a combo flag again, so inlining can use combo... |
tree | commitdiff |
2008-09-29 |
Michael Beck | - Look through Tuple when checking for address taken... |
tree | commitdiff |
2008-09-29 |
Michael Beck | - BugFix: removed missing debugging artefact, causing... |
tree | commitdiff |
next |