summaryrefslogtreecommitdiff
path: root/lib/Target/X86/X86InstrFMA.td
AgeCommit message (Expand)AuthorFilesLines
2015-12-09X86-FMA3: Defined the ExeDomain property for Scalar FMA3 opcodes.Vyacheslav Klochkov1-0/+3
2015-12-05[X86][FMA4] Explicitly set the domain of FMA4 float/double scalar instructionsSimon Pilgrim1-29/+32
2015-11-26X86-FMA3: Improved/enabled the memory folding optimization for scalar loadsVyacheslav Klochkov1-13/+5
2015-11-25fix typo; NFCSanjay Patel1-1/+1
2015-11-13X86-FMA3: Implemented commute transformations FMA*_Int instructions.Vyacheslav Klochkov1-6/+6
2015-11-06Improved the operands commute transformation for X86-FMA3 instructions.Andrew Kaylor1-53/+45
2015-11-04Created new X86 FMA3 opcodes (FMA*_Int) that are used now for lowering of sca...Andrew Kaylor1-38/+94
2015-05-25[X86] When pattern-matching scalar FMA3 intrinsics, don't re-arrange the firs...Michael Kuperstein1-2/+7
2014-11-26Replace neverHasSideEffects=1 with hasSideEffects=0 in all .td files.Craig Topper1-3/+3
2014-05-07[X86] Selectively mark the FMA variants inside a family as isCommutable.Quentin Colombet1-14/+32
2014-04-02[X86] Make the VFMA*231 variants commutable and relax the alignment restrictionsLang Hames1-0/+2
2014-02-04[X86] Only 213 FMA3 variants should be marked commutable.Lang Hames1-3/+5
2014-01-31Replace X86 FMA intrinsic pseduo-instructions with def pats.Lang Hames1-26/+21
2014-01-23Replace vfmaddxx213 instructions with their 231-type equivalents in accumulatorLang Hames1-3/+3
2014-01-05Add a new x86 specific instruction flag to force some isCodeGenOnly instructi...Craig Topper1-2/+2
2014-01-02Mark all x86 Int_ and _Int patterns as isCodeGenOnly so the disassembler tabl...Craig Topper1-13/+17
2013-09-30Various x86 disassembler fixes.Craig Topper1-7/+8
2013-09-21Remove alignment restrictions from FMA load folding.Craig Topper1-31/+31
2013-01-14Simplify nested strconcats in X86 td files since strconcat can take more than...Craig Topper1-7/+7
2012-12-26Mark all the _REV instructions as not having side effects. They aren't really...Craig Topper1-2/+2
2012-11-21Fix execution domain for packed FMA4 instructions.Craig Topper1-27/+29
2012-09-19Add explicit VEX_L tags to all 256-bit instructions. This will allow us to re...Craig Topper1-7/+8
2012-08-31Mark FMA4 instructions as commutable and add them to the folding tables.Craig Topper1-0/+4
2012-08-31Mark FMA3 instructions as commutable so that the operands to the multiply par...Craig Topper1-0/+4
2012-08-31Add support for converting llvm.fma to fma4 instructions.Craig Topper1-32/+68
2012-08-29Convert FMA4 patterns to use target specific nodes instead of intrinsics to a...Craig Topper1-32/+34
2012-08-24Remove more mayLoad workarounds.Jakob Stoklund Olesen1-7/+7
2012-08-24Custom lower FMA intrinsics to target specific nodes and remove the patterns.Craig Topper1-131/+0
2012-08-21Cleanup the scalar FMA3 definitions. Add patterns to fold loads with scalar f...Craig Topper1-44/+49
2012-08-21Merge FMA3 instructions with and without patterns into single classes using n...Craig Topper1-38/+21
2012-08-20Remove FMA3 intrinsic instructions in favor of patterns.Craig Topper1-70/+146
2012-08-20Use correct intrinsic for 256-bit VFMSUBADDPS.Craig Topper1-1/+1
2012-08-19Remove trailing white space and tab characters. No functional change.Craig Topper1-33/+33
2012-08-01Added FMA functionality to X86 target.Elena Demikhovsky1-81/+165
2012-06-04Add intrinsic forms for FMA instructions to opcode folding tables.Craig Topper1-4/+4
2012-06-03Rename fma4 intrinsics to just fma since they are now used for both FMA4 and ...Craig Topper1-54/+54
2012-06-03Use sse_load_f32/64 for scalar FMA3 intrinsic patterns instead of 128-bit loa...Craig Topper1-36/+30
2012-06-03Add neverHasSideEffects and mayLoad to FMA3 instructions.Craig Topper1-0/+5
2012-06-01Remove fadd(fmul) patterns for FMA3. This needs to be implemented by paying a...Craig Topper1-212/+0
2012-06-01Tidy up. Remove trailing spaces and fix the worst of the 80 column violations.Craig Topper1-74/+90
2012-05-31Added FMA3 Intel instructions.Elena Demikhovsky1-42/+321
2012-02-19some comment fix for X86 and ARMJia Liu1-1/+1
2012-02-18Emacs-tag and some comment fix for all ARM, CellSPU, Hexagon, MBlaze, MSP430,...Jia Liu1-1/+1
2012-01-05Mark scalar FMA4 instructions as ignoring the VEX.L bit.Craig Topper1-4/+4
2011-12-30Add FMA4 instructions to disassembler.Craig Topper1-38/+53
2011-12-30Separate the concept of having memory access in operand 4 from the concept of...Craig Topper1-6/+6
2011-12-30Combine FMA4 SS/SD patterns with the instruction definitions.Craig Topper1-97/+24
2011-12-30Combine FMA4 PS/PD patterns with the instruction definitions.Craig Topper1-219/+42
2011-12-30Change FMA4 memory forms to use memopv* instead of alignedloadv*. No need to ...Craig Topper1-58/+48
2011-12-30Fix load size for FMA4 SS/SD instructions. They need to use f32 and f64 size,...Craig Topper1-60/+43
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-18 07:01:58 +0000