summaryrefslogtreecommitdiffstats
path: root/llvm/lib/CodeGen/SelectionDAG
Commit message (Expand)AuthorAgeFilesLines
* A simplification for checking whether the signs of the operands and sum diffe...Bill Wendling2008-11-251-15/+14
* Now with the correct type for the 0.Bill Wendling2008-11-251-1/+1
* Get rid of unused variable.Bill Wendling2008-11-251-1/+0
* Hacker's Delight says, "Signed integer overflow of addition occurs if and onlyBill Wendling2008-11-251-4/+48
* Initial support for anti-dependence breaking. Currently this code does notDan Gohman2008-11-251-238/+26
* - Make lowering of "add with overflow" customizable by back-ends.Bill Wendling2008-11-242-18/+29
* Check in the rest of this change. The isAntiDep flag needs to be passedDan Gohman2008-11-242-2/+2
* If the type legalizer actually legalized anythingDuncan Sands2008-11-244-195/+234
* Eliminate some unused variable compile time warnings.Evan Cheng2008-11-243-0/+9
* Add support for llvm.uadd.with.overflow.Bill Wendling2008-11-241-5/+5
* Rename SetCCResultContents to BooleanContents. InDuncan Sands2008-11-235-23/+23
* Added check to avoid generating extract subvector beyond the end of the vecto...Mon P Wang2008-11-231-1/+2
* Cleanup of the [SU]ADDO type legalization code. Patch by Duncan!Bill Wendling2008-11-223-22/+24
* - Move conversion of [SU]ADDO from DAG combiner into legalizer.Bill Wendling2008-11-224-38/+53
* Update comments.Dan Gohman2008-11-211-3/+3
* reapply Sanjiv's patch to genericize memcpy/memset/memmove to take anChris Lattner2008-11-211-6/+3
* Revert r59802. It was breaking the build of llvm-gcc:Bill Wendling2008-11-211-3/+6
* Make mem[cpy,move,set] intrinsics overloaded.Sanjiv Gupta2008-11-211-6/+3
* Default to converting UADDO to the generic form that SADDO is converted to.Bill Wendling2008-11-211-1/+1
* Clean up normalization of shufflesMon P Wang2008-11-211-30/+31
* Combine the two add with overflow intrinsics lowerings. They differ only in D...Bill Wendling2008-11-211-18/+5
* Generate code for llvm.uadd.with.overflow intrinsic. No conversion support yet.Bill Wendling2008-11-211-1/+12
* Add a flag to SDep for tracking which edges are anti-dependence edges.Dan Gohman2008-11-212-8/+8
* Remove chains. Unnecessary.Bill Wendling2008-11-212-14/+8
* Rename SDep's isSpecial to isArtificial, to make this field a littleDan Gohman2008-11-212-40/+40
* Rename "ADDO" to "SADDO" and "UADDO". The "UADDO" isn't equivalent to "ADDC"Bill Wendling2008-11-213-7/+14
* Implement the sadd_with_overflow intrinsic. This is converted intoBill Wendling2008-11-213-2/+56
* Use ComputeLatency in the MachineInstr scheduler.Dan Gohman2008-11-211-0/+1
* Remove the CycleBound computation code from the ScheduleDAGRRListDan Gohman2008-11-211-58/+5
* Implement ComputeLatency for MachineInstr ScheduleDAGs. FactorDan Gohman2008-11-211-7/+9
* Add UADDO and SADDO nodes. These will be used for determining an overflowBill Wendling2008-11-211-0/+2
* Change these schedulers to not emit no-ops. It turns out thatDan Gohman2008-11-212-9/+3
* Add some documentation.Duncan Sands2008-11-201-1/+28
* 80-column violation.Bill Wendling2008-11-201-2/+4
* Remove a remnant of list-burr's fast mode.Dan Gohman2008-11-201-20/+0
* Factor out the SethiUllman numbering logic from the list-burr andDan Gohman2008-11-201-159/+47
* Remove the "fast" form of the list-burr scheduler, and use theDan Gohman2008-11-202-74/+15
* Facter AddPseudoTwoAddrDeps and associated infrasructure out ofDan Gohman2008-11-201-43/+60
* Factor out the code for verifying the work of the scheduler,Dan Gohman2008-11-202-75/+3
* Simplify this code a little. In the fast scheduler, CreateNewSUnitDan Gohman2008-11-191-17/+5
* Experimental post-pass scheduling support. Post-pass schedulingDan Gohman2008-11-1911-980/+301
* Move the code for printing a graph node label for an SUnit intoDan Gohman2008-11-192-22/+19
* Convert SUnit's dump method into a print method and implementDan Gohman2008-11-191-6/+10
* Rearrange code to reduce the nesting level. No functionality change.Dan Gohman2008-11-191-36/+36
* Fix debug printing of flagged SDNodes in SUnits so that theyDan Gohman2008-11-191-12/+12
* Tidy up ScheduleNodeBottomUp methods, and make them moreDan Gohman2008-11-182-10/+8
* Update a comment to reflect the current code.Dan Gohman2008-11-181-3/+2
* Remove integer promotion support for FP_EXTENDDuncan Sands2008-11-182-16/+1
* Simplify code using helper routines. There is notDuncan Sands2008-11-182-67/+53
* Whitespace cleanups.Dan Gohman2008-11-181-2/+2
OpenPOWER on IntegriCloud