| 2006-01-31 |
Chris Lattner | Compile this: |
tree | commitdiff |
| 2006-01-31 |
Evan Cheng | Added custom lowering of fabs |
tree | commitdiff |
| 2006-01-31 |
Chris Lattner | add the 'lucas' optimization |
tree | commitdiff |
| 2006-01-31 |
Chris Lattner | I don't see why this optimization isn't safe, but it... |
tree | commitdiff |
| 2006-01-31 |
Chris Lattner | Another high-prio selection performance bug |
tree | commitdiff |
| 2006-01-31 |
Chris Lattner | more mumbling |
tree | commitdiff |
| 2006-01-31 |
Chris Lattner | add some notes |
tree | commitdiff |
| 2006-01-30 |
Evan Cheng | Don't generate complex sequence for SETOLE, SETOLT... |
tree | commitdiff |
| 2006-01-30 |
Chris Lattner | Fix FP constants, and the SparcV8/2006-01-22-BitConvert... |
tree | commitdiff |
| 2006-01-30 |
Evan Cheng | i64 -> f32, f32 -> i64 and some clean up. |
tree | commitdiff |
| 2006-01-30 |
Evan Cheng | Always use FP stack instructions to perform i64 to... |
tree | commitdiff |
| 2006-01-30 |
Chris Lattner | Revamp the ICC/FCC reading instructions to be parameter... |
tree | commitdiff |
| 2006-01-30 |
Chris Lattner | Compile: |
tree | commitdiff |
| 2006-01-30 |
Chris Lattner | If the target has V9 instructions, this pass is a noop... |
tree | commitdiff |
| 2006-01-30 |
Chris Lattner | When in v9 mode, emit fabsd/fnegd/fmovd |
tree | commitdiff |
| 2006-01-30 |
Chris Lattner | First step towards V9 instructions in the V8 backend... |
tree | commitdiff |
| 2006-01-30 |
Chris Lattner | Two changes: |
tree | commitdiff |
| 2006-01-30 |
Chris Lattner | When lowering SELECT_CC, see if the input is a lowered... |
tree | commitdiff |
| 2006-01-30 |
Chris Lattner | Move MaskedValueIsZero from the DAGCombiner to the... |
tree | commitdiff |
| 2006-01-30 |
Chris Lattner | Implement isMaskedValueZeroForTargetNode for the variou... |
tree | commitdiff |
| 2006-01-30 |
Chris Lattner | adjust prototype |
tree | commitdiff |
| 2006-01-29 |
Chris Lattner | Functions that are lazily streamed in from the .bc... |
tree | commitdiff |
| 2006-01-29 |
Chris Lattner | add another note |
tree | commitdiff |
| 2006-01-29 |
Chris Lattner | add some performance notes from looking at sgefa |
tree | commitdiff |
| 2006-01-29 |
Chris Lattner | add a high-priority SSE issue from sgefa |
tree | commitdiff |
| 2006-01-29 |
Chris Lattner | add a missed optimization |
tree | commitdiff |
| 2006-01-29 |
Chris Lattner | Now that OpActions is big enough, we can specify action... |
tree | commitdiff |
| 2006-01-29 |
Chris Lattner | disable this for now |
tree | commitdiff |
| 2006-01-29 |
Reid Spencer | Add a note about lowering llvm.memset, llvm.memcpy... |
tree | commitdiff |
| 2006-01-29 |
Chris Lattner | remove now-dead code, the legalizer takes care of this... |
tree | commitdiff |
| 2006-01-29 |
Chris Lattner | The FP stack doesn't support UNDEF, ask the legalizer... |
tree | commitdiff |
| 2006-01-29 |
Chris Lattner | Request expansion of ConstantVec nodes. |
tree | commitdiff |
| 2006-01-29 |
Chris Lattner | Targets all now request ConstantFP to be legalized... |
tree | commitdiff |
| 2006-01-29 |
Chris Lattner | Update alpha to reflect recent constantfp legalize... |
tree | commitdiff |
| 2006-01-29 |
Chris Lattner | cmovle != cmovlt |
tree | commitdiff |
| 2006-01-29 |
Jeff Cohen | Fix typo. |
tree | commitdiff |
| 2006-01-28 |
Jeff Cohen | Flesh out AMD family/models. |
tree | commitdiff |
| 2006-01-28 |
Jeff Cohen | Correctly determine CPU vendor. |
tree | commitdiff |
| 2006-01-28 |
Jeff Cohen | Use union instead of reinterpret_cast. |
tree | commitdiff |
| 2006-01-28 |
Jeff Cohen | Fix recognition of Intel CPUs. |
tree | commitdiff |
| 2006-01-28 |
Chris Lattner | Is64Bit reflects the capability of the chip, not an... |
tree | commitdiff |
| 2006-01-28 |
Chris Lattner | Fix a bunch of JIT failures with the new isel |
tree | commitdiff |
| 2006-01-28 |
Jeff Cohen | Improve X86 subtarget support for Windows and AMD. |
tree | commitdiff |
| 2006-01-28 |
Chris Lattner | silence a warning |
tree | commitdiff |
| 2006-01-28 |
Chris Lattner | Fix a bug in my elimination of ISD::CALL this morning... |
tree | commitdiff |
| 2006-01-28 |
Chris Lattner | make this work on non-native hosts |
tree | commitdiff |
| 2006-01-28 |
Chris Lattner | add a note about how we should implement this FIXME... |
tree | commitdiff |
| 2006-01-28 |
Nate Begeman | Implement Promote for VAARG, and allow it to be custom... |
tree | commitdiff |
| 2006-01-28 |
Nate Begeman | Add a couple more things to the readme. |
tree | commitdiff |
| 2006-01-28 |
Chris Lattner | Remove some dead code |
tree | commitdiff |
| 2006-01-27 |
Chris Lattner | Switch to AlphaISD::CALL instead of ISD::CALL |
tree | commitdiff |
| 2006-01-27 |
Chris Lattner | Use PPCISD::CALL instead of ISD::CALL |
tree | commitdiff |
| 2006-01-27 |
Chris Lattner | Use V8ISD::CALL instead of ISD::CALL |
tree | commitdiff |
| 2006-01-27 |
Evan Cheng | A bit of wisdom from Chris on the last entry. |
tree | commitdiff |
| 2006-01-27 |
Evan Cheng | AT&T assembly convention: registers are in lower case. |
tree | commitdiff |
| 2006-01-27 |
Chris Lattner | initialize member vars |
tree | commitdiff |
| 2006-01-27 |
Chris Lattner | initialize all instance vars |
tree | commitdiff |
| 2006-01-27 |
Chris Lattner | Make llvm.frame/returnaddr not crash on ppc |
tree | commitdiff |
| 2006-01-27 |
Evan Cheng | Added notes about a x86 isel deficiency. |
tree | commitdiff |
| 2006-01-27 |
Evan Cheng | Added a temporary option -enable-x86-sse to enable... |
tree | commitdiff |
| 2006-01-27 |
Evan Cheng | Bye bye Pattern ISel, hello DAG ISel. |
tree | commitdiff |
| 2006-01-27 |
Nate Begeman | Remove TLI.LowerReturnTo, and just let targets custom... |
tree | commitdiff |
| 2006-01-27 |
Evan Cheng | A better workaround |
tree | commitdiff |
| 2006-01-27 |
Chris Lattner | force sse/3dnow off until they work. This fixes all... |
tree | commitdiff |
| 2006-01-27 |
Chris Lattner | Unbreak the JIT with SSE |
tree | commitdiff |
| 2006-01-27 |
Evan Cheng | x86 CPU detection and proper subtarget support |
tree | commitdiff |
| 2006-01-27 |
Evan Cheng | Subtarget feature can now set any variable to any value |
tree | commitdiff |
| 2006-01-27 |
Chris Lattner | PHI and INLINEASM are now built-in instructions provide... |
tree | commitdiff |
| 2006-01-27 |
Chris Lattner | Add a default NoItinerary class for targets to use. |
tree | commitdiff |
| 2006-01-26 |
Jeff Cohen | Improve compatibility with VC2005, patch by Morten... |
tree | commitdiff |
| 2006-01-26 |
Jim Laskey | Set up MachineDebugInfo to scan for debug information... |
tree | commitdiff |
| 2006-01-26 |
Chris Lattner | Improve compatibility with VC2005, patch by Morten... |
tree | commitdiff |
| 2006-01-26 |
Evan Cheng | Added preliminary x86 subtarget support. |
tree | commitdiff |
| 2006-01-26 |
Duraid Madina | fix stack corruption! Previously, 16-byte whole-FP... |
tree | commitdiff |
| 2006-01-26 |
Duraid Madina | some hoovering |
tree | commitdiff |
| 2006-01-26 |
Chris Lattner | Rest of subtarget support, remove references to ppc |
tree | commitdiff |
| 2006-01-26 |
Chris Lattner | Add trivial subtarget support |
tree | commitdiff |
| 2006-01-26 |
Andrew Lenharth | minor renaming |
tree | commitdiff |
| 2006-01-26 |
Andrew Lenharth | allow R28 to be used for frame calculations without... |
tree | commitdiff |
| 2006-01-26 |
Evan Cheng | Work around some x86 Darwin assembler bugs |
tree | commitdiff |
| 2006-01-26 |
Evan Cheng | When trying to fold X86::SETCC into a Select, make... |
tree | commitdiff |
| 2006-01-26 |
Evan Cheng | Remove the uses of STATUS flag register. Rely on node... |
tree | commitdiff |
| 2006-01-25 |
Andrew Lenharth | oops |
tree | commitdiff |
| 2006-01-25 |
Andrew Lenharth | forgot one |
tree | commitdiff |
| 2006-01-25 |
Andrew Lenharth | make things compile again |
tree | commitdiff |
| 2006-01-25 |
Evan Cheng | Set SchedulingForLatency to be the default scheduling... |
tree | commitdiff |
| 2006-01-25 |
Nate Begeman | First part of bug 680: |
tree | commitdiff |
| 2006-01-25 |
Evan Cheng | Default scheduling preference is SchedulingForLatency. |
tree | commitdiff |
| 2006-01-25 |
Evan Cheng | X86 prefer scheduling for reduced register pressure. |
tree | commitdiff |
| 2006-01-25 |
Evan Cheng | Fix a selectcc lowering bug. Make a copy of X86ISD... |
tree | commitdiff |
| 2006-01-25 |
Chris Lattner | Loosen up these checks to allow direct uses of ESP |
tree | commitdiff |
| 2006-01-25 |
Duraid Madina | add bundling! well not really, for now it's just stop... |
tree | commitdiff |
| 2006-01-25 |
Andrew Lenharth | maintaining stackpointer alignment. Perhaps it doesn... |
tree | commitdiff |
| 2006-01-24 |
Chris Lattner | use ESP directly, not a copy of ESP into some other... |
tree | commitdiff |
| 2006-01-24 |
Chris Lattner | Emit the copies out of call return registers *after... |
tree | commitdiff |
| 2006-01-24 |
Chris Lattner | Allow jit-beta to work |
tree | commitdiff |
| 2006-01-24 |
Jim Laskey | Crude Dwarf global variable debugging. |
tree | commitdiff |
| 2006-01-23 |
Andrew Lenharth | bye bye Pattern ISEL |
tree | commitdiff |
| 2006-01-23 |
Andrew Lenharth | added stores to lsmark |
tree | commitdiff |
| 2006-01-23 |
Andrew Lenharth | fix up more lsmark stuff |
tree | commitdiff |
| next |
projects / oota-llvm.git / history