aboutsummaryrefslogtreecommitdiffstats
path: root/lib/Target/X86/X86InstrInfo.cpp
Commit message (Expand)AuthorAgeFilesLines
* Revert r56675 - it breaks unwinding runtime everywhere.Anton Korobeynikov2008-10-041-18/+8
* Switch the MachineOperand accessors back to the short names likeDan Gohman2008-10-031-67/+67
* Move the GlobalBaseReg field out of X86ISelDAGToDAG.cppDan Gohman2008-09-301-6/+17
* Mark lea fi# as being really rematerializable.Dan Gohman2008-09-261-2/+3
* Avoid spilling EBP / RBP twice in the prologue.Evan Cheng2008-09-261-8/+18
* Move the code for initializing the global base reg out ofDan Gohman2008-09-231-0/+29
* Remove isImm(), isReg(), and friends, in favor of Dan Gohman2008-09-131-18/+18
* Use static_cast instead of C style cast.Evan Cheng2008-08-291-1/+2
* Backing out 55521. Not safe.Evan Cheng2008-08-291-26/+1
* Swap fp comparison operands and change predicate to allow load folding.Evan Cheng2008-08-281-1/+26
* These assertions should be return false's instead, allowing the client to det...Owen Anderson2008-08-261-4/+8
* Make TargetInstrInfo::copyRegToReg return a bool indicating whether the copy ...Owen Anderson2008-08-261-15/+17
* Convert uses of std::vector in TargetInstrInfo to SmallVector. This change h...Owen Anderson2008-08-141-3/+3
* Add an EXTRACTPSmr pattern to match the pattern thatDan Gohman2008-08-081-0/+1
* Re-enable elimination of unnecessary SUBREG_TO_REG instructions inDan Gohman2008-08-071-0/+1
* Rename SDOperand to SDValue.Dan Gohman2008-07-271-7/+7
* Use movaps instead of movups to spill 16-byte vector values when default alig...Evan Cheng2008-07-211-9/+19
* Use aligned stack spills, where possible. This fixes PR2549.Anton Korobeynikov2008-07-191-19/+23
* Add a new function, ReplaceAllUsesOfValuesWith, which handles bulkDan Gohman2008-07-171-2/+2
* Add a utility function to MachineInstr for testing whether an instructionDan Gohman2008-07-121-7/+3
* Pool-allocation for MachineInstrs, MachineBasicBlocks, andDan Gohman2008-07-071-43/+47
* Make DenseMap's insert return a pair, to more closely resemble std::map.Dan Gohman2008-07-071-8/+13
* - Remove calls to copyKillDeadInfo which is an N^2 function. Instead, propaga...Evan Cheng2008-07-031-61/+89
* commuteInstruction should preserve dead markers.Evan Cheng2008-07-031-1/+3
* Make LiveVariables even more optional, by making it optional in the call to T...Owen Anderson2008-07-021-5/+7
* Split ISD::LABEL into ISD::DBG_LABEL and ISD::EH_LABEL, eliminatingDan Gohman2008-07-011-1/+2
* If it's determined safe, remat MOV32r0 (i.e. xor r, r) and others as it is in...Evan Cheng2008-06-241-11/+53
* Add option to commuteInstruction() which forces it to create a new (commuted)...Evan Cheng2008-06-161-2/+3
* Wrap MVT::ValueType in a struct to get type safetyDuncan Sands2008-06-061-3/+3
* Use PMULDQ for v2i64 multiplies when SSE4.1 is available. And addDan Gohman2008-05-231-0/+4
* Change target-specific classes to use more precise static types.Dan Gohman2008-05-141-1/+1
* Constify the machine instruction passed into theBill Wendling2008-05-121-1/+2
* Undo r50574. We are already ensuring the folded load address is 16-byte aligned.Evan Cheng2008-05-021-0/+8
* Not safe folding a load + FsXORPSrr into FsXORPSrm. It's loading a FR64 value...Evan Cheng2008-05-021-8/+0
* Don't forget to update the current operand when getting the size of an instru...Nicolas Geoffray2008-04-201-4/+14
* Don't forget about sub-register indices when rematting instructions.Evan Cheng2008-04-161-0/+13
* Infrastructure for getting the machine code size of a function and an instruc...Nicolas Geoffray2008-04-161-0/+540
* Rename MemOperand to MachineMemOperand. This was suggested byDan Gohman2008-04-071-1/+1
* ReMat of load from stub in pic mode extends the life of pic base. Currently s...Evan Cheng2008-04-011-0/+6
* Remove unnecessary and non-deterministic checking code. Re-enable remat of lo...Evan Cheng2008-04-011-11/+0
* Disabling remat of load from gv stub (temporarily) again to fix llvmgcc boots...Evan Cheng2008-04-011-0/+3
* It's not safe to fold a load from GV stub or constantpool into a two-address ...Evan Cheng2008-03-311-0/+8
* Move reMaterialize() from TargetRegisterInfo to TargetInstrInfo.Evan Cheng2008-03-311-0/+28
* Re-apply 48911.Evan Cheng2008-03-311-8/+12
* Backing out 48911 for now. It's breaking stuff.Evan Cheng2008-03-281-12/+8
* Load from stub is already re-materializable.Evan Cheng2008-03-281-8/+12
* Code clean up.Evan Cheng2008-03-271-14/+16
* Allow certain lea instructions to be rematerialized.Evan Cheng2008-03-271-35/+66
* Remove an unused command line option.Evan Cheng2008-03-271-6/+0
* Add CMP32mr and friends to the load-unfolding table. AmongDan Gohman2008-03-251-0/+4