summaryrefslogtreecommitdiffstats
path: root/llvm/lib
Commit message (Collapse)AuthorAgeFilesLines
...
* Fix CodeGen/PowerPC/2006-04-05-splat-ish.llChris Lattner2006-04-051-2/+2
| | | | llvm-svn: 27439
* Handle canonical form of e.g.Evan Cheng2006-04-053-0/+56
| | | | | | | | | | | | vector_shuffle v1, v1, <0, 4, 1, 5, 2, 6, 3, 7> This is turned into vector_shuffle v1, <undef>, <0, 0, 1, 1, 2, 2, 3, 3> by dag combiner. It would match a {p}unpckl on x86. llvm-svn: 27437
* Make a vector live across blocks have the correct Vec type. This fixesChris Lattner2006-04-051-1/+9
| | | | | | CodeGen/X86/2006-04-04-CrossBlockCrash.ll llvm-svn: 27436
* Bogus assertEvan Cheng2006-04-051-5/+8
| | | | llvm-svn: 27434
* Fallthrough to expand if a VECTOR_SHUFFLE cannot be custom lowered.Evan Cheng2006-04-052-16/+15
| | | | llvm-svn: 27433
* Exapnd a VECTOR_SHUFFLE to a BUILD_VECTOR if target asks for it to be expandedEvan Cheng2006-04-051-2/+30
| | | | | | or custom lowering fails. llvm-svn: 27432
* revert this, this is safe, if conservative. leave a note to that effectAndrew Lenharth2006-04-051-9/+7
| | | | llvm-svn: 27428
* Handle v8i16 shuffle that must be broken into a pair of pshufhw / pshuflw.Evan Cheng2006-04-051-2/+55
| | | | llvm-svn: 27427
* add vslChris Lattner2006-04-051-0/+2
| | | | llvm-svn: 27425
* add vmladduhmChris Lattner2006-04-052-2/+2
| | | | llvm-svn: 27423
* Add m[tf]vscr instructions.Chris Lattner2006-04-052-0/+34
| | | | llvm-svn: 27421
* add a noteChris Lattner2006-04-041-2/+4
| | | | llvm-svn: 27419
* Add missing byte merges.Chris Lattner2006-04-041-0/+2
| | | | llvm-svn: 27418
* Add FP -> Int ConversionsChris Lattner2006-04-041-2/+4
| | | | llvm-svn: 27417
* add average intrinsicsChris Lattner2006-04-041-0/+7
| | | | llvm-svn: 27416
* add a noteChris Lattner2006-04-041-2/+13
| | | | llvm-svn: 27414
* Fix some broken logic that would cause us to codegen ↵Chris Lattner2006-04-041-2/+2
| | | | | | {2147483647,2147483647,2147483647,2147483647} as 'vspltisb v0, -1'. llvm-svn: 27413
* Added pslldq and psrldq.Evan Cheng2006-04-041-0/+18
| | | | llvm-svn: 27412
* Minor fixes + naming changes.Evan Cheng2006-04-042-157/+158
| | | | llvm-svn: 27410
* PSHUF* encoding bugs.Evan Cheng2006-04-041-6/+6
| | | | llvm-svn: 27405
* Do not create ZEXTLOAD's unless we are before legalize or the operation isChris Lattner2006-04-041-1/+2
| | | | | | legal. llvm-svn: 27402
* Ask legalize to promote all vector shuffles to be v16i8 instead of having toChris Lattner2006-04-043-46/+69
| | | | | | | | handle all 4 PPC vector types. This simplifies the matching code and allows us to eliminate a bunch of patterns. This also adds cases we were missing, such as CodeGen/PowerPC/vec_splat.ll:splat_h. llvm-svn: 27400
* * Add supprot for SCALAR_TO_VECTOR operations where the input needs to beChris Lattner2006-04-041-21/+102
| | | | | | | | | | promoted/expanded (e.g. SCALAR_TO_VECTOR from i8/i16 on PPC). * Add support for targets to request that VECTOR_SHUFFLE nodes be promoted to a canonical type, for example, we only want v16i8 shuffles on PPC. * Move isShuffleLegal out of TLI into Legalize. * Teach isShuffleLegal to allow shuffles that need to be promoted. llvm-svn: 27399
* Signed shr by a constant is not the same as sdiv by 2^kChris Lattner2006-04-041-9/+0
| | | | llvm-svn: 27395
* cmpps / cmppd encoding bugEvan Cheng2006-04-041-16/+16
| | | | llvm-svn: 27393
* Constant fold bitconvert(undef)Chris Lattner2006-04-041-0/+2
| | | | llvm-svn: 27391
* Compact some intrinsic definitions.Evan Cheng2006-04-041-119/+80
| | | | llvm-svn: 27388
* Plug in the byte and short splatsChris Lattner2006-04-041-2/+4
| | | | llvm-svn: 27387
* Revert accidentally committed hunks.Chris Lattner2006-04-031-3/+1
| | | | llvm-svn: 27386
* Make sure to mark unsupported SCALAR_TO_VECTOR operations as expand.Chris Lattner2006-04-031-1/+5
| | | | llvm-svn: 27385
* Some SSE1 intrinsics: min, max, sqrt, etc.Evan Cheng2006-04-031-78/+91
| | | | llvm-svn: 27384
* revert previous patchChris Lattner2006-04-031-1/+1
| | | | llvm-svn: 27383
* Use movlpd to: store lower f64 extracted from v2f64.Evan Cheng2006-04-032-6/+10
| | | | | | Use movhpd to: store upper f64 extracted from v2f64. llvm-svn: 27382
* Force use of a frame-pointer if there is anything on the stack that is alignedChris Lattner2006-04-031-7/+18
| | | | | | more than the OS keeps the stack aligned. llvm-svn: 27381
* The stack alignment is now computed dynamically, just verify it is correct.Chris Lattner2006-04-031-1/+2
| | | | llvm-svn: 27380
* Remove unused methodChris Lattner2006-04-031-8/+0
| | | | llvm-svn: 27379
* - More efficient extract_vector_elt with shuffle and movss, movsd, movd, etc.Evan Cheng2006-04-034-70/+151
| | | | | | - Some bug fixes and naming inconsistency fixes. llvm-svn: 27377
* Align vectors to the size in bytes, not bits.Chris Lattner2006-04-031-1/+1
| | | | llvm-svn: 27376
* Add a missing check, this fixes UnitTests/Vector/sumarray.cChris Lattner2006-04-031-2/+2
| | | | llvm-svn: 27375
* Add a missing check, which broke a bunch of vector tests.Chris Lattner2006-04-031-3/+6
| | | | llvm-svn: 27374
* Add the full set of min/max instructionsChris Lattner2006-04-031-6/+14
| | | | llvm-svn: 27372
* support x * (c1 + c2) where c1 and c2 are pow2s. special case for c2 == 4Andrew Lenharth2006-04-032-10/+32
| | | | llvm-svn: 27370
* mul by const conversion sequences. more coming soonAndrew Lenharth2006-04-033-2/+61
| | | | llvm-svn: 27368
* back this outAndrew Lenharth2006-04-031-25/+0
| | | | llvm-svn: 27367
* This should be a win of every archAndrew Lenharth2006-04-021-1/+26
| | | | llvm-svn: 27364
* This makes McCat/12-IOtest go 8x faster or soAndrew Lenharth2006-04-021-2/+14
| | | | llvm-svn: 27363
* This will be needed soonAndrew Lenharth2006-04-021-1/+146
| | | | llvm-svn: 27362
* add a noteChris Lattner2006-04-021-0/+9
| | | | llvm-svn: 27360
* Inform the dag combiner that the predicate compares only return a low bit.Chris Lattner2006-04-022-1/+39
| | | | llvm-svn: 27359
* relax assertionChris Lattner2006-04-021-1/+4
| | | | llvm-svn: 27358
OpenPOWER on IntegriCloud