summaryrefslogtreecommitdiff
path: root/lib/Target/X86/X86InstrInfo.cpp
Commit message (Collapse)AuthorAge
* Remove isImm(), isReg(), and friends, in favor of Dan Gohman2008-09-13
| | | | | | | | | | isImmediate(), isRegister(), and friends, to avoid confusion about having two different names with the same meaning. I'm not attached to the longer names, and would be ok with changing to the shorter names if others prefer it. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@56189 91177308-0d34-0410-b5e6-96231b3b80d8
* Use static_cast instead of C style cast.Evan Cheng2008-08-29
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@55552 91177308-0d34-0410-b5e6-96231b3b80d8
* Backing out 55521. Not safe.Evan Cheng2008-08-29
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@55548 91177308-0d34-0410-b5e6-96231b3b80d8
* Swap fp comparison operands and change predicate to allow load folding.Evan Cheng2008-08-28
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@55521 91177308-0d34-0410-b5e6-96231b3b80d8
* These assertions should be return false's instead, allowing the client to ↵Owen Anderson2008-08-26
| | | | | | detect the failure. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@55377 91177308-0d34-0410-b5e6-96231b3b80d8
* Make TargetInstrInfo::copyRegToReg return a bool indicating whether the copy ↵Owen Anderson2008-08-26
| | | | | | | | | | requested was inserted or not. This allows bitcast in fast isel to properly handle the case where an appropriate reg-to-reg copy is not available. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@55375 91177308-0d34-0410-b5e6-96231b3b80d8
* Convert uses of std::vector in TargetInstrInfo to SmallVector. This change ↵Owen Anderson2008-08-14
| | | | | | had to be propoagated down into all the targets and up into all clients of this API. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@54802 91177308-0d34-0410-b5e6-96231b3b80d8
* Add an EXTRACTPSmr pattern to match the pattern thatDan Gohman2008-08-08
| | | | | | | X86ISelLowering creates. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@54544 91177308-0d34-0410-b5e6-96231b3b80d8
* Re-enable elimination of unnecessary SUBREG_TO_REG instructions inDan Gohman2008-08-07
| | | | | | | | | | | | LowerSubregs, and fix an x86-64 isel bug that this exposed. SUBREG_TO_REG for x86-64 implicit zero extension is only safe for isel to generate when the source is known to always have zeros in the high 32 bits. The EXTRACT_SUBREG instruction does not clear the high 32 bits. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@54444 91177308-0d34-0410-b5e6-96231b3b80d8
* Rename SDOperand to SDValue.Dan Gohman2008-07-27
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@54128 91177308-0d34-0410-b5e6-96231b3b80d8
* Use movaps instead of movups to spill 16-byte vector values when default ↵Evan Cheng2008-07-21
| | | | | | alignment is >= 16. This fixes some massive performance regressions. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53844 91177308-0d34-0410-b5e6-96231b3b80d8
* Use aligned stack spills, where possible. This fixes PR2549.Anton Korobeynikov2008-07-19
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53784 91177308-0d34-0410-b5e6-96231b3b80d8
* Add a new function, ReplaceAllUsesOfValuesWith, which handles bulkDan Gohman2008-07-17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | replacement of multiple values. This is slightly more efficient than doing multiple ReplaceAllUsesOfValueWith calls, and theoretically could be optimized even further. However, an important property of this new function is that it handles the case where the source value set and destination value set overlap. This makes it feasible for isel to use SelectNodeTo in many very common cases, which is advantageous because SelectNodeTo avoids a temporary node and it doesn't require CSEMap updates for users of values that don't change position. Revamp MorphNodeTo, which is what does all the work of SelectNodeTo, to handle operand lists more efficiently, and to correctly handle a number of corner cases to which its new wider use exposes it. This commit also includes a change to the encoding of post-isel opcodes in SDNodes; now instead of being sandwiched between the target-independent pre-isel opcodes and the target-dependent pre-isel opcodes, post-isel opcodes are now represented as negative values. This makes it possible to test if an opcode is pre-isel or post-isel without having to know the size of the current target's post-isel instruction set. These changes speed up llc overall by 3% and reduce memory usage by 10% on the InstructionCombining.cpp testcase with -fast and -regalloc=local. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53728 91177308-0d34-0410-b5e6-96231b3b80d8
* Add a utility function to MachineInstr for testing whether an instructionDan Gohman2008-07-12
| | | | | | | | has exactly one MachineMemOperand, and change some X86 lowering code to make use of it. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53498 91177308-0d34-0410-b5e6-96231b3b80d8
* Pool-allocation for MachineInstrs, MachineBasicBlocks, andDan Gohman2008-07-07
| | | | | | | | | | | | MachineMemOperands. The pools are owned by MachineFunctions. This drastically reduces the number of calls to malloc/free made during the "Emit" phase of scheduling, as well as later phases in CodeGen. Combined with other changes, this speeds up the "instruction selection" phase of CodeGen by 10% in some cases. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53212 91177308-0d34-0410-b5e6-96231b3b80d8
* Make DenseMap's insert return a pair, to more closely resemble std::map.Dan Gohman2008-07-07
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53177 91177308-0d34-0410-b5e6-96231b3b80d8
* - Remove calls to copyKillDeadInfo which is an N^2 function. Instead, ↵Evan Cheng2008-07-03
| | | | | | | | | propagate kill / dead markers as new instructions are constructed in foldMemoryOperand, convertToThressAddress, etc. - Also remove LiveVariables::instructionChanged, etc. Replace all calls with cheaper calls which update VarInfo kill list. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53097 91177308-0d34-0410-b5e6-96231b3b80d8
* commuteInstruction should preserve dead markers.Evan Cheng2008-07-03
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53060 91177308-0d34-0410-b5e6-96231b3b80d8
* Make LiveVariables even more optional, by making it optional in the call to ↵Owen Anderson2008-07-02
| | | | | | | | | TargetInstrInfo::convertToThreeAddressInstruction Also, if LV isn't around, then TwoAddr doesn't need to be updating flags, since they won't have been set in the first place. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@53058 91177308-0d34-0410-b5e6-96231b3b80d8
* Split ISD::LABEL into ISD::DBG_LABEL and ISD::EH_LABEL, eliminatingDan Gohman2008-07-01
| | | | | | | | | | | | | | | | | the need for a flavor operand, and add a new SDNode subclass, LabelSDNode, for use with them to eliminate the need for a label id operand. Change instruction selection to let these label nodes through unmodified instead of creating copies of them. Teach the MachineInstr emitter how to emit a MachineInstr directly from an ISD label node. This avoids the need for allocating SDNodes for the label id and flavor value, as well as SDNodes for each of the post-isel label, label id, and label flavor. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52943 91177308-0d34-0410-b5e6-96231b3b80d8
* If it's determined safe, remat MOV32r0 (i.e. xor r, r) and others as it is ↵Evan Cheng2008-06-24
| | | | | | instead of using the longer MOV32ri instruction. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52670 91177308-0d34-0410-b5e6-96231b3b80d8
* Add option to commuteInstruction() which forces it to create a new ↵Evan Cheng2008-06-16
| | | | | | (commuted) instruction. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52308 91177308-0d34-0410-b5e6-96231b3b80d8
* Wrap MVT::ValueType in a struct to get type safetyDuncan Sands2008-06-06
| | | | | | | | | | | | | | | | | and better control the abstraction. Rename the type to MVT. To update out-of-tree patches, the main thing to do is to rename MVT::ValueType to MVT, and rewrite expressions like MVT::getSizeInBits(VT) in the form VT.getSizeInBits(). Use VT.getSimpleVT() to extract a MVT::SimpleValueType for use in switch statements (you will get an assert failure if VT is an extended value type - these shouldn't exist after type legalization). This results in a small speedup of codegen and no new testsuite failures (x86-64 linux). git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@52044 91177308-0d34-0410-b5e6-96231b3b80d8
* Use PMULDQ for v2i64 multiplies when SSE4.1 is available. And addDan Gohman2008-05-23
| | | | | | | load-folding table entries for PMULDQ and PMULLD. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@51489 91177308-0d34-0410-b5e6-96231b3b80d8
* Change target-specific classes to use more precise static types.Dan Gohman2008-05-14
| | | | | | | | This eliminates the need for several awkward casts, including the last dynamic_cast under lib/Target. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@51091 91177308-0d34-0410-b5e6-96231b3b80d8
* Constify the machine instruction passed into theBill Wendling2008-05-12
| | | | | | | "is{Trivially,Really}ReMaterializable" methods. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@51001 91177308-0d34-0410-b5e6-96231b3b80d8
* Undo r50574. We are already ensuring the folded load address is 16-byte aligned.Evan Cheng2008-05-02
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@50578 91177308-0d34-0410-b5e6-96231b3b80d8
* Not safe folding a load + FsXORPSrr into FsXORPSrm. It's loading a FR64 ↵Evan Cheng2008-05-02
| | | | | | value but the load folding variant expects a 16-byte aligned address. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@50574 91177308-0d34-0410-b5e6-96231b3b80d8
* Don't forget to update the current operand when getting the size of an ↵Nicolas Geoffray2008-04-20
| | | | | | instruction. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@50007 91177308-0d34-0410-b5e6-96231b3b80d8
* Don't forget about sub-register indices when rematting instructions.Evan Cheng2008-04-16
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@49830 91177308-0d34-0410-b5e6-96231b3b80d8
* Infrastructure for getting the machine code size of a function and an ↵Nicolas Geoffray2008-04-16
| | | | | | instruction. X86, PowerPC and ARM are implemented git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@49809 91177308-0d34-0410-b5e6-96231b3b80d8
* Rename MemOperand to MachineMemOperand. This was suggested byDan Gohman2008-04-07
| | | | | | | | review feedback from Chris quite a while ago. No functionality change. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@49348 91177308-0d34-0410-b5e6-96231b3b80d8
* ReMat of load from stub in pic mode extends the life of pic base. Currently ↵Evan Cheng2008-04-01
| | | | | | spiller doesn't do a good job of estimating the impact. Disable for now. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@49059 91177308-0d34-0410-b5e6-96231b3b80d8
* Remove unnecessary and non-deterministic checking code. Re-enable remat of ↵Evan Cheng2008-04-01
| | | | | | load from gv stub. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@49054 91177308-0d34-0410-b5e6-96231b3b80d8
* Disabling remat of load from gv stub (temporarily) again to fix llvmgcc ↵Evan Cheng2008-04-01
| | | | | | bootstrap miscompare. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@49037 91177308-0d34-0410-b5e6-96231b3b80d8
* It's not safe to fold a load from GV stub or constantpool into a two-address ↵Evan Cheng2008-03-31
| | | | | | use. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@49002 91177308-0d34-0410-b5e6-96231b3b80d8
* Move reMaterialize() from TargetRegisterInfo to TargetInstrInfo.Evan Cheng2008-03-31
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@48995 91177308-0d34-0410-b5e6-96231b3b80d8
* Re-apply 48911.Evan Cheng2008-03-31
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@48977 91177308-0d34-0410-b5e6-96231b3b80d8
* Backing out 48911 for now. It's breaking stuff.Evan Cheng2008-03-28
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@48922 91177308-0d34-0410-b5e6-96231b3b80d8
* Load from stub is already re-materializable.Evan Cheng2008-03-28
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@48911 91177308-0d34-0410-b5e6-96231b3b80d8
* Code clean up.Evan Cheng2008-03-27
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@48856 91177308-0d34-0410-b5e6-96231b3b80d8
* Allow certain lea instructions to be rematerialized.Evan Cheng2008-03-27
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@48855 91177308-0d34-0410-b5e6-96231b3b80d8
* Remove an unused command line option.Evan Cheng2008-03-27
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@48854 91177308-0d34-0410-b5e6-96231b3b80d8
* Add CMP32mr and friends to the load-unfolding table. AmongDan Gohman2008-03-25
| | | | | | | | | other things, this allows the scheduler to unfold a load operand in the 2008-01-08-SchedulerCrash.ll testcase, so it now successfully clones the comparison to avoid a pushf+popf. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@48777 91177308-0d34-0410-b5e6-96231b3b80d8
* Add support for calls that return two FP values inChris Lattner2008-03-21
| | | | | | | ST(0)/ST(1). git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@48634 91177308-0d34-0410-b5e6-96231b3b80d8
* Make insert_subreg a two-address instruction, vastly simplifying ↵Christopher Lamb2008-03-16
| | | | | | LowerSubregs pass. Add a new TII, subreg_to_reg, which is like insert_subreg except that it takes an immediate implicit value to insert into rather than a register. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@48412 91177308-0d34-0410-b5e6-96231b3b80d8
* Get rid of a pseudo instruction and replace it with subreg based operation ↵Christopher Lamb2008-03-13
| | | | | | | | | | on real instructions, ridding the asm printers of the hack used to do this previously. In the process, update LowerSubregs to be careful about eliminating copies that have side affects. Note: the coalescer will have to be careful about this too, when it starts coalescing insert_subreg nodes. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@48329 91177308-0d34-0410-b5e6-96231b3b80d8
* coalesce away 80-bit floating point copies.Chris Lattner2008-03-11
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@48241 91177308-0d34-0410-b5e6-96231b3b80d8
* convert a massive if statement to a switch.Chris Lattner2008-03-11
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@48240 91177308-0d34-0410-b5e6-96231b3b80d8
* Missed part of recommit.Christopher Lamb2008-03-11
| | | | git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@48224 91177308-0d34-0410-b5e6-96231b3b80d8
generated by cgit v1.2.3 (git 2.25.1) at 2024-05-28 09:21:32 +0000