| 2007-11-16 |
Sebastian Hack | Some tweaks |
tree | commitdiff |
| 2007-11-16 |
Sebastian Hack | Changes to the Appel dumper |
tree | commitdiff |
| 2007-11-16 |
Sebastian Hack | Smaller changes in statistic output bechordal_main.c |
tree | commitdiff |
| 2007-11-16 |
Sebastian Hack | The tables can now also have other names. |
tree | commitdiff |
| 2007-11-15 |
Matthias Braun | fehler001 problem with endless loops |
tree | commitdiff |
| 2007-11-14 |
Michael Beck | - i_mapper_cbrt added |
tree | commitdiff |
| 2007-11-13 |
Matthias Braun | fix handling of default pns in the backend (they don... |
tree | commitdiff |
| 2007-11-11 |
Matthias Braun | include iroptimize.h |
tree | commitdiff |
| 2007-11-09 |
Christoph Mallon | Add crude tests for bool simplifications. |
tree | commitdiff |
| 2007-11-08 |
Sebastian Hack | Support ignore nodes |
tree | commitdiff |
| 2007-11-08 |
Michael Beck | more tests added |
tree | commitdiff |
| 2007-11-07 |
Sebastian Hack | Get rid of some get_irn_node_nr() which are still there. |
tree | commitdiff |
| 2007-11-07 |
Sebastian Hack | Fixed a bug with ofg iteration. |
tree | commitdiff |
| 2007-11-06 |
Michael Beck | more libc mapper functions added |
tree | commitdiff |
| 2007-11-05 |
Matthias Braun | we can't be sure that phis aren't used anymore when... |
tree | commitdiff |
| 2007-11-05 |
Matthias Braun | more people were including bespillbelady.h |
tree | commitdiff |
| 2007-11-05 |
Matthias Braun | fix after bespillbelady.h is removed |
tree | commitdiff |
| 2007-11-05 |
Matthias Braun | prefer rematerializable values when spilling, fix bug... |
tree | commitdiff |
| 2007-11-05 |
Matthias Braun | copy the belady3 changes to belady (only cleanups so... |
tree | commitdiff |
| 2007-11-05 |
Michael Beck | more libc mapper functions added |
tree | commitdiff |
| 2007-11-04 |
Matthias Braun | fixed a bunch of warnings (in OPTIMIZE mode) |
tree | commitdiff |
| 2007-11-03 |
Matthias Braun | get rid of benodesets (in favour of ir_nodeset), report... |
tree | commitdiff |
| 2007-11-03 |
Matthias Braun | report nextuse place in beuses, add a function which... |
tree | commitdiff |
| 2007-11-01 |
Matthias Braun | use putchar so the thing links |
tree | commitdiff |
| 2007-11-01 |
Matthias Braun | exit with 0 so makereport is green |
tree | commitdiff |
| 2007-11-01 |
Matthias Braun | use is_Const_0 not is_Const_null in fucom creation |
tree | commitdiff |
| 2007-11-01 |
Matthias Braun | check that Psi flags input doesn't depend on load when... |
tree | commitdiff |
| 2007-11-01 |
Matthias Braun | another fehler |
tree | commitdiff |
| 2007-10-31 |
Michael Beck | fixed one testcase |
tree | commitdiff |
| 2007-10-29 |
Matthias Braun | fix remove_empty_blocks (don't use a generic walker... |
tree | commitdiff |
| 2007-10-29 |
Matthias Braun | peephole optimisation: reuse zero flag from previous... |
tree | commitdiff |
| 2007-10-29 |
Matthias Braun | fix bugs when exchanging nodes to projs in bepeephole |
tree | commitdiff |
| 2007-10-29 |
Christoph Mallon | fehler103: Unusual 64bit shift triggers assert in const... |
tree | commitdiff |
| 2007-10-28 |
Michael Beck | testcase for rts library call optimization |
tree | commitdiff |
| 2007-10-28 |
Matthias Braun | fix bug introduces in last commit |
tree | commitdiff |
| 2007-10-28 |
Matthias Braun | compile without warnings when WITH_JVM is not set |
tree | commitdiff |
| 2007-10-28 |
Matthias Braun | compile without warnings when WITH_JVM is not set |
tree | commitdiff |
| 2007-10-28 |
Matthias Braun | more interesting lea problems |
tree | commitdiff |
| 2007-10-28 |
Matthias Braun | end block can also have 0 predecessors |
tree | commitdiff |
| 2007-10-28 |
Matthias Braun | improve asm test |
tree | commitdiff |
| 2007-10-26 |
Michael Beck | fixed handling of pn_Call_X_regular |
tree | commitdiff |
| 2007-10-26 |
Michael Beck | C99 constructs removed |
tree | commitdiff |
| 2007-10-26 |
Matthias Braun | moved ia32 architecture configuration to a separate... |
tree | commitdiff |
| 2007-10-26 |
Matthias Braun | small optimisation and cleanup |
tree | commitdiff |
| 2007-10-26 |
Matthias Braun | make IS_ macros return 0 or 1 |
tree | commitdiff |
| 2007-10-26 |
Matthias Braun | added some comments, don't make backedges fallthrough... |
tree | commitdiff |
| 2007-10-25 |
Michael Beck | updated cpu architectures |
tree | commitdiff |
| 2007-10-25 |
Michael Beck | C99 constructs removed |
tree | commitdiff |
| 2007-10-25 |
Christoph Mallon | Fix debug output. |
tree | commitdiff |
| 2007-10-25 |
Christoph Mallon | Small cleanups. |
tree | commitdiff |
| 2007-10-25 |
Christoph Mallon | Use add_ia32_flags() instead of set_ia32_flags(), which... |
tree | commitdiff |
| 2007-10-25 |
Matthias Braun | fix parameter loads not being rematerialized |
tree | commitdiff |
| 2007-10-25 |
Christoph Mallon | Handle integer Abs for modes which are not 32bits wide. |
tree | commitdiff |
| 2007-10-25 |
Christoph Mallon | fehler102: ia32 backend cannot handle integer Abs with... |
tree | commitdiff |
| 2007-10-25 |
Christoph Mallon | In match_arguments() assert that match_am is set when... |
tree | commitdiff |
| 2007-10-25 |
Christoph Mallon | Remove unused variable. |
tree | commitdiff |
| 2007-10-24 |
Michael Beck | more Phi optimizations for Div, Mod, DivMod |
tree | commitdiff |
| 2007-10-24 |
Michael Beck | fixed test |
tree | commitdiff |
| 2007-10-24 |
Michael Beck | add test for new phi optimization |
tree | commitdiff |
| 2007-10-24 |
Sebastian Hack | A more general color priorization. |
tree | commitdiff |
| 2007-10-24 |
Christoph Mallon | Emit the "t" suffix for f96 numbers in the ia32 backend. |
tree | commitdiff |
| 2007-10-24 |
Christoph Mallon | Pad the names of fehler*.c with leading zeros for sorting. |
tree | commitdiff |
| 2007-10-24 |
Christoph Mallon | Give direct visual feedback by printing "FAILED" when... |
tree | commitdiff |
| 2007-10-24 |
Michael Beck | long double x87 modes don't support AM |
tree | commitdiff |
| 2007-10-24 |
Michael Beck | improved fehler101.c. Now with printing! |
tree | commitdiff |
| 2007-10-24 |
Michael Beck | cannot output suffix for long double store |
tree | commitdiff |
| 2007-10-24 |
Christoph Mallon | Restore the AM fold behavior of prior to r16324. It... |
tree | commitdiff |
| 2007-10-24 |
Christoph Mallon | Actually implement what r16324 promised. |
tree | commitdiff |
| 2007-10-24 |
Sebastian Hack | Prioritize chunks with constraint nodes more |
tree | commitdiff |
| 2007-10-23 |
Christoph Mallon | Improve the AM folding heuristics: Do not fold AM if... |
tree | commitdiff |
| 2007-10-23 |
Christoph Mallon | Remove dead code, remove pointless marking of memory... |
tree | commitdiff |
| 2007-10-23 |
Christoph Mallon | Remove the unused function mode_needs_gp_reg(). |
tree | commitdiff |
| 2007-10-23 |
Christoph Mallon | Clarify a comment. |
tree | commitdiff |
| 2007-10-23 |
Christoph Mallon | - Allow an arbitrary (for arbitrary < 32) number of... |
tree | commitdiff |
| 2007-10-23 |
Christoph Mallon | Do no parse {,!}out_rBAR constraints anymore. The const... |
tree | commitdiff |
| 2007-10-23 |
Michael Beck | Add a main to let it compile. |
tree | commitdiff |
| 2007-10-23 |
Michael Beck | added initialization to suppress (wrong) warning |
tree | commitdiff |
| 2007-10-23 |
Michael Beck | temporary disable "manual" spill due to probles with... |
tree | commitdiff |
| 2007-10-23 |
Michael Beck | now even buggier |
tree | commitdiff |
| 2007-10-23 |
Michael Beck | wrong spill ? |
tree | commitdiff |
| 2007-10-22 |
Michael Beck | heur5 not available yet ;-) |
tree | commitdiff |
| 2007-10-22 |
Sebastian Hack | Some heur4 tweaks |
tree | commitdiff |
| 2007-10-22 |
Matthias Braun | some more code I played around with lately |
tree | commitdiff |
| 2007-10-22 |
Michael Beck | transform the macro block header |
tree | commitdiff |
| 2007-10-21 |
Matthias Braun | don't aggressively fold some nodes when we know it... |
tree | commitdiff |
| 2007-10-21 |
Matthias Braun | fix address mode calculation: we can not eat immediates... |
tree | commitdiff |
| 2007-10-21 |
Matthias Braun | - fix a conceptual bug in peephole, we need a callback... |
tree | commitdiff |
| 2007-10-20 |
Christoph Mallon | Fix r16293. |
tree | commitdiff |
| 2007-10-19 |
Christoph Mallon | Make the emit sequences of ShrD and ShlD probably more... |
tree | commitdiff |
| 2007-10-19 |
Christoph Mallon | Now even correct! (see r16284) |
tree | commitdiff |
| 2007-10-19 |
Christoph Mallon | A do-while loop is nicer than a for loop here. |
tree | commitdiff |
| 2007-10-19 |
Christoph Mallon | Fix fehler99: always skip at least one node and check... |
tree | commitdiff |
| 2007-10-19 |
Christoph Mallon | fehler99: defective copykeep when ShrD has to identical... |
tree | commitdiff |
| 2007-10-19 |
Michael Beck | to shut up gcc in release mode |
tree | commitdiff |
| 2007-10-19 |
Matthias Braun | new testcases |
tree | commitdiff |
| 2007-10-19 |
Matthias Braun | - improve set transformation |
tree | commitdiff |
| 2007-10-19 |
Christoph Mallon | Fix fehler 98. clarify comments. |
tree | commitdiff |
| 2007-10-19 |
Christoph Mallon | make fehler98.c behave properly. |
tree | commitdiff |
| 2007-10-19 |
Christoph Mallon | fehler98: gen_Cmp tries to build test with two immediates. |
tree | commitdiff |
| 2007-10-18 |
Michael Beck | fixed release build |
tree | commitdiff |
| next |
projects / libfirm / history