| 2005-05-09 |
Chris Lattner | Fix the syntax of the i/o instructions, these are obvio... |
tree | commitdiff |
| 2005-05-09 |
Chris Lattner | legalize readio/writeio into load/stores, fixing CodeGe... |
tree | commitdiff |
| 2005-05-09 |
Chris Lattner | legalize readio/writeio into a load/store if requested |
tree | commitdiff |
| 2005-05-09 |
Chris Lattner | legalize READPORT, WRITEPORT, READIO, WRITEIO, at least... |
tree | commitdiff |
| 2005-05-09 |
Chris Lattner | Add support for matching the READPORT, WRITEPORT, READI... |
tree | commitdiff |
| 2005-05-09 |
Chris Lattner | Add support for READPORT, WRITEPORT, READIO, WRITEIO |
tree | commitdiff |
| 2005-05-09 |
Chris Lattner | restore some non-dead code I removed last night breakin... |
tree | commitdiff |
| 2005-05-09 |
Chris Lattner | fold and (shl X, C1), C2 -> rlwinm when possible. ... |
tree | commitdiff |
| 2005-05-09 |
Chris Lattner | Fold shifts into subsequent SHL's. These shifts often... |
tree | commitdiff |
| 2005-05-09 |
Duraid Madina | fix and cleanup constmul code a bit, this fixes mediabe... |
tree | commitdiff |
| 2005-05-09 |
Chris Lattner | Wrap long lines, remove dead code that is now handled... |
tree | commitdiff |
| 2005-05-09 |
Chris Lattner | Fix FP -> bool casts |
tree | commitdiff |
| 2005-05-09 |
Chris Lattner | implement and.ll:test33 |
tree | commitdiff |
| 2005-05-09 |
Chris Lattner | Don't use the load/store instruction as the source... |
tree | commitdiff |
| 2005-05-09 |
Chris Lattner | memoize all nodes, even null Value* nodes. Do not... |
tree | commitdiff |
| 2005-05-09 |
Chris Lattner | wrap long lines |
tree | commitdiff |
| 2005-05-09 |
Chris Lattner | Print SrcValue nodes correctly |
tree | commitdiff |
| 2005-05-09 |
Chris Lattner | Fix X86/2005-05-08-FPStackifierPHI.ll: ugly gross hack. |
tree | commitdiff |
| 2005-05-09 |
Chris Lattner | Preserve CC's when linking modules |
tree | commitdiff |
| 2005-05-09 |
Chris Lattner | Preserve calling conventions when doing IPO |
tree | commitdiff |
| 2005-05-09 |
Chris Lattner | wrap long lines, preserve calling conventions when... |
tree | commitdiff |
| 2005-05-08 |
Chris Lattner | By definition, 'tail' calls cannot access the stack... |
tree | commitdiff |
| 2005-05-08 |
Chris Lattner | Verify that varargs functions all have ccc |
tree | commitdiff |
| 2005-05-08 |
Chris Lattner | Convert non-address taken functions with C calling... |
tree | commitdiff |
| 2005-05-08 |
Chris Lattner | Implement Reassociate/mul-neg-add.ll |
tree | commitdiff |
| 2005-05-08 |
Chris Lattner | Bail out earlier |
tree | commitdiff |
| 2005-05-08 |
Chris Lattner | Teach reassociate that 0-X === X*-1 |
tree | commitdiff |
| 2005-05-08 |
Chris Lattner | Fix PR557 and basictest[34].ll. |
tree | commitdiff |
| 2005-05-08 |
Chris Lattner | Add debugging information |
tree | commitdiff |
| 2005-05-08 |
Chris Lattner | eliminate gotos |
tree | commitdiff |
| 2005-05-08 |
Chris Lattner | Wrap long lines. Fix "warning: conflicting types for... |
tree | commitdiff |
| 2005-05-08 |
Chris Lattner | Improve reassociation handling of inverses, implementin... |
tree | commitdiff |
| 2005-05-08 |
Chris Lattner | clean up and modernize this pass. |
tree | commitdiff |
| 2005-05-08 |
Chris Lattner | Strength reduce SAR into SHR if there is no way sign... |
tree | commitdiff |
| 2005-05-08 |
Chris Lattner | Refactor some code |
tree | commitdiff |
| 2005-05-08 |
Chris Lattner | Handle some simple cases where we can see that values... |
tree | commitdiff |
| 2005-05-07 |
Chris Lattner | Fix a miscompilation of crafty by clobbering the "A... |
tree | commitdiff |
| 2005-05-07 |
Chris Lattner | Rewrite the guts of the reassociate pass to be more... |
tree | commitdiff |
| 2005-05-07 |
Reid Spencer | * Add two strlen optimizations: |
tree | commitdiff |
| 2005-05-07 |
Reid Spencer | Don't increment the counter unless the debug flag is... |
tree | commitdiff |
| 2005-05-07 |
Chris Lattner | Convert shifts to muls to assist reassociation. This... |
tree | commitdiff |
| 2005-05-07 |
Chris Lattner | Simplify the code and rearrange it. No major functiona... |
tree | commitdiff |
| 2005-05-07 |
Jeff Cohen | Silence VC++ warnings about unsafe mixing of ints and... |
tree | commitdiff |
| 2005-05-06 |
Chris Lattner | remove some dead (always dynamically false) flags |
tree | commitdiff |
| 2005-05-06 |
Chris Lattner | encode calling conventions for call/invoke instructions. |
tree | commitdiff |
| 2005-05-06 |
Chris Lattner | encode function calling convs in the bytecode file... |
tree | commitdiff |
| 2005-05-06 |
Chris Lattner | parse new calling conv specifiers |
tree | commitdiff |
| 2005-05-06 |
Chris Lattner | wrap a longline |
tree | commitdiff |
| 2005-05-06 |
Chris Lattner | add support for explicit calling conventions |
tree | commitdiff |
| 2005-05-06 |
Chris Lattner | use splice instead of remove/insert for a minor speedup |
tree | commitdiff |
| 2005-05-06 |
Chris Lattner | remove some ugly hacks that are no longer needed since... |
tree | commitdiff |
| 2005-05-06 |
Chris Lattner | BAD typeo which caused many testsuite failures last... |
tree | commitdiff |
| 2005-05-06 |
Chris Lattner | clean up the CBE output a bit |
tree | commitdiff |
| 2005-05-06 |
Chris Lattner | add tail marker as a comment |
tree | commitdiff |
| 2005-05-06 |
Chris Lattner | Make the stub functions be tail calls |
tree | commitdiff |
| 2005-05-06 |
Chris Lattner | Preserve tail marker |
tree | commitdiff |
| 2005-05-06 |
Chris Lattner | Implement Transforms/Inline/inline-tail.ll |
tree | commitdiff |
| 2005-05-06 |
Chris Lattner | preserve the tail marker |
tree | commitdiff |
| 2005-05-06 |
Chris Lattner | lex tail |
tree | commitdiff |
| 2005-05-06 |
Chris Lattner | add bytecode reader support for tail calls |
tree | commitdiff |
| 2005-05-06 |
Chris Lattner | Add a 'tail' marker for call instructions, patch contri... |
tree | commitdiff |
| 2005-05-06 |
Chris Lattner | Wrap long lines |
tree | commitdiff |
| 2005-05-06 |
Chris Lattner | DCE intrinsic instructions without side effects. |
tree | commitdiff |
| 2005-05-06 |
Chris Lattner | These intrinsics do not access memory |
tree | commitdiff |
| 2005-05-06 |
Chris Lattner | Teach instcombine propagate zeroness through shl instru... |
tree | commitdiff |
| 2005-05-06 |
Chris Lattner | Implement shift.ll:test23. If we are shifting right... |
tree | commitdiff |
| 2005-05-06 |
Chris Lattner | Implement xor.ll:test22 |
tree | commitdiff |
| 2005-05-06 |
Chris Lattner | implement and.ll:test30 and set.ll:test21 |
tree | commitdiff |
| 2005-05-06 |
Chris Lattner | implement or.ll:test20 |
tree | commitdiff |
| 2005-05-05 |
Misha Brukman | * Order #includes alphabetically |
tree | commitdiff |
| 2005-05-05 |
Misha Brukman | Remove extra blank line |
tree | commitdiff |
| 2005-05-05 |
Misha Brukman | Remove vim settings from source code; people should... |
tree | commitdiff |
| 2005-05-05 |
Chris Lattner | add support for undef values of opaque type, addressing... |
tree | commitdiff |
| 2005-05-05 |
Chris Lattner | Add some extra checks. Opaque types don't have a null... |
tree | commitdiff |
| 2005-05-05 |
Chris Lattner | When hitting an unsupported intrinsic, actually print it |
tree | commitdiff |
| 2005-05-05 |
Andrew Lenharth | ctpop lowering in legalize |
tree | commitdiff |
| 2005-05-05 |
Chris Lattner | Fix a bug compimling Ruby, fixing this testcase: |
tree | commitdiff |
| 2005-05-04 |
Andrew Lenharth | fix typo |
tree | commitdiff |
| 2005-05-04 |
Andrew Lenharth | Well, add support for ct* for 21264 only. |
tree | commitdiff |
| 2005-05-04 |
Andrew Lenharth | Make promoteOp work for CT* |
tree | commitdiff |
| 2005-05-04 |
Chris Lattner | Instcombine: cast (X != 0) to int, cast (X == 1) to... |
tree | commitdiff |
| 2005-05-04 |
Reid Spencer | Implement the IsDigitOptimization for simplifying calls... |
tree | commitdiff |
| 2005-05-04 |
Reid Spencer | * Correct the function prototypes for some of the funct... |
tree | commitdiff |
| 2005-05-03 |
Andrew Lenharth | Implement count leading zeros (ctlz), count trailing... |
tree | commitdiff |
| 2005-05-03 |
Chris Lattner | fix a bug in the 1 index GEP handling code |
tree | commitdiff |
| 2005-05-03 |
Reid Spencer | Implement optimizations for the strchr and llvm.memset... |
tree | commitdiff |
| 2005-05-03 |
Chris Lattner | add direct support for making GEP instrs with one index |
tree | commitdiff |
| 2005-05-03 |
Jeff Cohen | Use ANSI-approved way of getting the value infinity... |
tree | commitdiff |
| 2005-05-03 |
Reid Spencer | Avoid garbage output in the statistics display by ensur... |
tree | commitdiff |
| 2005-05-03 |
Reid Spencer | Add the StrNCmpOptimization which is similar to strcmp. |
tree | commitdiff |
| 2005-05-02 |
Reid Spencer | Implement the fprintf optimization which converts calls... |
tree | commitdiff |
| 2005-05-02 |
Andrew Lenharth | fold fp div by 0 to inf, the way gcc does. This is... |
tree | commitdiff |
| 2005-05-02 |
Andrew Lenharth | Remove support for 1.0 style varargs |
tree | commitdiff |
| 2005-05-02 |
John Criswell | Fixed a comment. |
tree | commitdiff |
| 2005-05-02 |
Duraid Madina | support multiplication by constant negative integers |
tree | commitdiff |
| 2005-05-02 |
Duraid Madina | add support for bools to SELECT, this fixes Prolangs... |
tree | commitdiff |
| 2005-05-01 |
Chris Lattner | Implement getelementptr.ll:test11 |
tree | commitdiff |
| 2005-05-01 |
Chris Lattner | Check for volatile loads only once. |
tree | commitdiff |
| 2005-05-01 |
Tanya Lattner | SMS for superblocks. |
tree | commitdiff |
| 2005-05-01 |
Tanya Lattner | Added extra constructor for superblocks. |
tree | commitdiff |
| next |
projects / oota-llvm.git / history