| 2009-01-30 |
Mon P Wang | When PerformBuildVectorCombine, avoid creating a X86ISD... |
tree | commitdiff |
| 2009-01-29 |
Dan Gohman | Make x86's BT instruction matching more thorough, and... |
tree | commitdiff |
| 2009-01-28 |
Mon P Wang | Fixed lowering of v816 shuffles. |
tree | commitdiff |
| 2009-01-28 |
Duncan Sands | Rename getAnalysisToUpdate to getAnalysisIfAvailable. |
tree | commitdiff |
| 2009-01-28 |
Evan Cheng | The memory alignment requirement on some of the mov... |
tree | commitdiff |
| 2009-01-28 |
Mon P Wang | Add shuffle splat pattern for x86 sse shifts. |
tree | commitdiff |
| 2009-01-27 |
Dan Gohman | Reformat the allocation-order arrays to a more conventi... |
tree | commitdiff |
| 2009-01-27 |
Dan Gohman | Simplify findNonImmUse; return the result using the... |
tree | commitdiff |
| 2009-01-27 |
Evan Cheng | Implement multiple with overflow by 2 with an add instr... |
tree | commitdiff |
| 2009-01-27 |
Dan Gohman | Eliminate unnecessary operands-list traversals. |
tree | commitdiff |
| 2009-01-27 |
Dan Gohman | Fix the Red Zone calculation for functions with frame... |
tree | commitdiff |
| 2009-01-26 |
Dan Gohman | Implement Red Zone utilization on x86-64. This is currently |
tree | commitdiff |
| 2009-01-26 |
Evan Cheng | Enhance logic in X86DAGToDAGISel::PreprocessForRMW... |
tree | commitdiff |
| 2009-01-26 |
Dan Gohman | Take the next steps in making SDUse more consistent... |
tree | commitdiff |
| 2009-01-26 |
Nate Begeman | De-identifying per sabre review |
tree | commitdiff |
| 2009-01-26 |
Nate Begeman | Map address space 256 to gs; similar mappings could... |
tree | commitdiff |
| 2009-01-26 |
Nate Begeman | Support pattern matching various x86 sse shifts. |
tree | commitdiff |
| 2009-01-25 |
Torok Edwin | should have removed the + when manually applying a... |
tree | commitdiff |
| 2009-01-25 |
Torok Edwin | revert this patch for now, because Codegen does still... |
tree | commitdiff |
| 2009-01-25 |
Torok Edwin | If user explicitly asks not to use SSE, don't force... |
tree | commitdiff |
| 2009-01-24 |
Nate Begeman | Fix an indent and a typo. |
tree | commitdiff |
| 2009-01-23 |
Chris Lattner | hopefully address PR3379 by making the P modifier work... |
tree | commitdiff |
| 2009-01-22 |
Bob Wilson | Add SelectionDAG::getNOT method to construct bitwise... |
tree | commitdiff |
| 2009-01-22 |
Evan Cheng | Eliminate a couple of fields from TargetRegisterClass... |
tree | commitdiff |
| 2009-01-21 |
Dan Gohman | Recognize inline asm for bswap on x86-64 GLIBC. This... |
tree | commitdiff |
| 2009-01-21 |
Evan Cheng | Also favors NOT64r. |
tree | commitdiff |
| 2009-01-21 |
Dan Gohman | Fix a recent regression. ClrOpcode is not set for i8... |
tree | commitdiff |
| 2009-01-21 |
Evan Cheng | Favors generating "not" over "xor -1". For example. |
tree | commitdiff |
| 2009-01-20 |
Evan Cheng | Change TargetInstrInfo::isMoveInstr to return source... |
tree | commitdiff |
| 2009-01-19 |
Evan Cheng | DIVREM isel deficiency: If sign bit is known zero,... |
tree | commitdiff |
| 2009-01-19 |
Evan Cheng | Minor tweak to LowerUINT_TO_FP_i32. Bias (after scalar_... |
tree | commitdiff |
| 2009-01-19 |
Evan Cheng | Now not UINT_TO_FP is legal (it's marked custom), dag... |
tree | commitdiff |
| 2009-01-17 |
Bill Wendling | Extend thi |
tree | commitdiff |
| 2009-01-17 |
Evan Cheng | Fix MatchAddress bug that's preventing negative displac... |
tree | commitdiff |
| 2009-01-17 |
Bill Wendling | Temporarily revert my last change. It is causing a... |
tree | commitdiff |
| 2009-01-17 |
Bill Wendling | Implement a special algorithm for converting uint_to_fp... |
tree | commitdiff |
| 2009-01-16 |
Bill Wendling | Add support for non-zero __builtin_return_address value... |
tree | commitdiff |
| 2009-01-15 |
Mon P Wang | Expand insert/extract of a <4 x i32> with a variable... |
tree | commitdiff |
| 2009-01-15 |
Rafael Espindola | Add the private linkage. |
tree | commitdiff |
| 2009-01-15 |
Dan Gohman | Move a few containers out of ScheduleDAGInstrs::BuildSc... |
tree | commitdiff |
| 2009-01-15 |
Dan Gohman | Add load-folding table entries for BT*ri8 instructions. |
tree | commitdiff |
| 2009-01-15 |
Dan Gohman | Make getWidenVectorType const. |
tree | commitdiff |
| 2009-01-13 |
Dan Gohman | BT appears to be available on all >= i386 chips. |
tree | commitdiff |
| 2009-01-13 |
Dan Gohman | Don't use a BT instruction if the AND has multiple... |
tree | commitdiff |
| 2009-01-13 |
Dan Gohman | Disable the register+memory forms of the bt instruction... |
tree | commitdiff |
| 2009-01-13 |
Dan Gohman | Add bt instructions that take immediate operands. |
tree | commitdiff |
| 2009-01-13 |
Dan Gohman | Fix a few more JIT encoding issues in the BT instructions. |
tree | commitdiff |
| 2009-01-13 |
Devang Patel | Use DebugInfo interface to lower dbg_* intrinsics. |
tree | commitdiff |
| 2009-01-12 |
Duncan Sands | Rename getABITypeSize to getTypePaddedSize, as |
tree | commitdiff |
| 2009-01-10 |
Evan Cheng | 80 col violation. |
tree | commitdiff |
| 2009-01-09 |
Misha Brukman | Removed trailing whitespace from Makefiles. |
tree | commitdiff |
| 2009-01-09 |
Dan Gohman | Add load-folding table entries for MOVDQA. |
tree | commitdiff |
| 2009-01-09 |
Dan Gohman | Whitespace and other minor adjustments to make SSE... |
tree | commitdiff |
| 2009-01-08 |
Devang Patel | Convert DwarfWriter into a pass. |
tree | commitdiff |
| 2009-01-07 |
Dan Gohman | Add patterns to match conditional moves with loads... |
tree | commitdiff |
| 2009-01-07 |
Dan Gohman | Add load-folding table entries for cmovno too. |
tree | commitdiff |
| 2009-01-07 |
Dan Gohman | Define instructions for cmovo and cmovno. |
tree | commitdiff |
| 2009-01-07 |
Dan Gohman | X86_COND_C and X86_COND_NC are alternate mnemonics for |
tree | commitdiff |
| 2009-01-06 |
Dan Gohman | Revert r42653 and forward-port the code that lets INC64... |
tree | commitdiff |
| 2009-01-05 |
Bill Wendling | Revert r61415 and r61484. Duncan was correct that these... |
tree | commitdiff |
| 2009-01-05 |
Dan Gohman | Tidy up #includes, deleting a bunch of unnecessary... |
tree | commitdiff |
| 2009-01-05 |
Devang Patel | squash warnings. |
tree | commitdiff |
| 2009-01-05 |
Evan Cheng | Atom and Core i7 do not have same model number after... |
tree | commitdiff |
| 2009-01-03 |
Evan Cheng | Add Intel processors core i7 and atom. |
tree | commitdiff |
| 2009-01-03 |
Evan Cheng | Fix PR3210: Detect more Intel processors. Patch by... |
tree | commitdiff |
| 2009-01-02 |
Evan Cheng | Do not isel load folding bt instructions for pentium... |
tree | commitdiff |
| 2009-01-02 |
Evan Cheng | Fix x86 CPU id detection to identify Penryn (and future... |
tree | commitdiff |
| 2009-01-02 |
Evan Cheng | Use movaps / movd to extract vector element 0 even... |
tree | commitdiff |
| 2009-01-01 |
Duncan Sands | Fix PR3274: when promoting the condition of a BRCOND... |
tree | commitdiff |
| 2008-12-29 |
Bill Wendling | Linux wants the FDE initial location and address range... |
tree | commitdiff |
| 2008-12-25 |
Chris Lattner | Add a simple pattern for matching 'bt'. |
tree | commitdiff |
| 2008-12-25 |
Chris Lattner | Fix some JIT encodings. |
tree | commitdiff |
| 2008-12-25 |
Chris Lattner | BT memory operands load from their address operand. |
tree | commitdiff |
| 2008-12-24 |
Chris Lattner | translateX86CC can never fail. Simplify it based on... |
tree | commitdiff |
| 2008-12-24 |
Bill Wendling | Darwin likes for the EH frame to be non-local. |
tree | commitdiff |
| 2008-12-24 |
Bill Wendling | GCC doesn't emit DW_EH_PE_sdata4 for the FDE encoding... |
tree | commitdiff |
| 2008-12-24 |
Dan Gohman | Fix a compiler-abort on a testcase where the stack... |
tree | commitdiff |
| 2008-12-24 |
Chris Lattner | indentation |
tree | commitdiff |
| 2008-12-23 |
Chris Lattner | simplify some control flow and reduce indentation,... |
tree | commitdiff |
| 2008-12-23 |
Dan Gohman | Add instruction patterns and encodings for the x86... |
tree | commitdiff |
| 2008-12-23 |
Devang Patel | Silence unused variable warnings. |
tree | commitdiff |
| 2008-12-23 |
Dan Gohman | Clean up the atomic opcodes in SelectionDAG. |
tree | commitdiff |
| 2008-12-23 |
Mon P Wang | Fixed code generation for v8i16 and v16i8 splats on... |
tree | commitdiff |
| 2008-12-23 |
Dan Gohman | Make the fuse-failed debug output human-readable. |
tree | commitdiff |
| 2008-12-20 |
Dan Gohman | Fix fast-isel to not emit invalid assembly when present... |
tree | commitdiff |
| 2008-12-19 |
Dan Gohman | Move the patterns which have i8 immediates before the... |
tree | commitdiff |
| 2008-12-19 |
Chris Lattner | Fix some release-assert warnings |
tree | commitdiff |
| 2008-12-19 |
Rafael Espindola | Fix bug 3202. |
tree | commitdiff |
| 2008-12-18 |
Dan Gohman | When emitting instructions that define EFLAGS and the... |
tree | commitdiff |
| 2008-12-18 |
Dan Gohman | When setting up the frame pointer, add it as a live... |
tree | commitdiff |
| 2008-12-18 |
Mon P Wang | Fixed x86 code generation of multiple for v2i64. It... |
tree | commitdiff |
| 2008-12-18 |
Dan Gohman | Mark the x86 fp stack registers as "reserved". This... |
tree | commitdiff |
| 2008-12-16 |
Dan Gohman | Add initial support for back-scheduling address computa... |
tree | commitdiff |
| 2008-12-12 |
Bill Wendling | - Use patterns instead of creating completely new instr... |
tree | commitdiff |
| 2008-12-12 |
Mon P Wang | Added support for SELECT v8i8 v4i16 for X86 (MMX) |
tree | commitdiff |
| 2008-12-12 |
Bill Wendling | Redo the arithmetic with overflow architecture. I was... |
tree | commitdiff |
| 2008-12-10 |
Evan Cheng | Some code clean up. |
tree | commitdiff |
| 2008-12-10 |
Bill Wendling | Only perform SETO/SETC to JO/JC conversion if extractva... |
tree | commitdiff |
| 2008-12-10 |
Evan Cheng | Fix MachineCodeEmitter to use uintptr_t instead of... |
tree | commitdiff |
| 2008-12-10 |
Bill Wendling | Whitespace changes. |
tree | commitdiff |
| next |
projects / oota-llvm.git / history